# Evidence-based Positron Emission Tomography

Summary of Recent Meta-analyses on PET

Giorgio Treglia Luca Giovanella *Editors*

Evidence-based Positron Emission Tomography

Giorgio Treglia • Luca Giovanella Editors

## Evidence-based Positron Emission Tomography

Summary of Recent Meta-analyses on PET

*Editors* Giorgio Treglia Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland Ente Ospedaliero Cantonale Bellinzona Switzerland

Luca Giovanella Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland Ente Ospedaliero Cantonale Bellinzona Switzerland

ISBN 978-3-030-47700-4 ISBN 978-3-030-47701-1 (eBook) https://doi.org/10.1007/978-3-030-47701-1 This book is an open access publication.

© The Editor(s) (if applicable) and The Author(s) 2020

**Open Access** This book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

## **Preface**

Positron emission tomography (PET), by using different radiopharmaceuticals evaluating different metabolic pathways or receptor expression, is a functional imaging method widely available worldwide.

In particular, hybrid tomographs as positron emission tomography/computed tomography (PET/CT) and positron emission tomography/magnetic resonance imaging (PET/MRI) combining morphological and functional information are currently used in the clinical practice.

Even if a large amount of literature is available about PET, the number of evidence-based articles on this imaging method, such as systematic reviews and meta-analyses, is relatively limited.

A meta-analysis is a statistical analysis that combines the results of multiple scientific studies. Meta-analysis can be performed when there are multiple scientific studies addressing the same question, with each individual study reporting measurements that are expected to have some degree of error. The aim then is to use approaches from statistics to derive a pooled estimate closest to the unknown common truth. Existing methods for meta-analysis yield a weighted average from the results of the individual studies. In addition to provide an estimate of the unknown common truth, meta-analysis has the capacity to identify sources of disagreement among the different study results, or other interesting relationships that may come to light in the context of multiple studies. A key benefit of this approach is the aggregation of information leading to a higher statistical power and more robust point estimate than is possible from the measure derived from any individual study.

This unique evidence-based book summarizes the findings or recent metaanalyses on the use of PET for different clinical indications. These metaanalyses on PET have been selected by the editors after a systematic literature search performed by using PubMed databases (last search: January 2019). Meta-analytic articles published from 2012 to the date of the last literature search were selected.

About the structure of this book, after a section introducing PET and metaanalyses, respectively, several sections describe the results of meta-analyses on PET for different indications including the following medical fields: oncology, cardiology, neurology, infectious and inflammatory diseases.

The different chapters are written by researchers who are both expert in PET and familiar with meta-analytic methodology.

This book provides evidence-based information on PET, which can be very useful for clinicians of different specialties and for international scientific societies. In particular, the evidence-based information provided by this book could help international scientific societies and national regulatory bodies on healthcare in approving the use of PET for several emerging clinical indications.

Furthermore, the updated information provided by this book could help worldwide clinicians of different specialties in prescribing PET with several radiotracers for different clinical indications.

Bellinzona and Lugano, Switzerland Giorgio Treglia Zurich, Switzerland Luca Giovanella

### **Contents**

#### **Part I Introduction**


#### **Part III Evidence-Based PET in Cardiology**


Alberto Miceli, Selene Capitanio, Maria Isabella Donegani, Stefano Raffa, Anna Borra, Matteo Bauckneht, and Silvia Morbelli

#### **Part VI Miscellaneous**

**14 Meta-Analyses on Technical Aspects of PET** . . . . . . . . . . . . . . . . 139 Luca Ceriani

**Part I**

**Introduction**

**1**

## **Introduction to Different PET Radiopharmaceuticals and Hybrid Modalities (PET/CT and PET/MRI)**

Luca Giovanella, Lisa Milan, and Arnoldo Piccardo

#### **1.1 Physical Principles of Positron Emission Tomography and Hybrid Modalities**

Positron Emission Tomography (PET) is an imaging technique performed by using positron emitting radiotracers. Positron decay occurs with neutron-poor radionuclides and consists in the conversion of a proton into a neutron with the simultaneous emission of a positron (*β*+) and a

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

Laboratory of Radiomics and Predictive Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

Clinic of Nuclear Medicine, University Hospital and University of Zurich, Zurich, Switzerland e-mail: luca.giovanella@eoc.ch

#### L. Milan

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

Laboratory of Radiomics and Predictive Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland neutrino (*ν*). The positron has a very short lifetime, and after the annihilation with an electron simultaneously produces two high-energy photons (*E* = 511 keV) in approximately opposite directions that are detected by an imaging camera. The PET scanning is based on the so-called annihilation coincidence detection (ACD) of the 511 keV γ-rays after the annihilation. Tomographic images are formed collecting data from many angles around the patient by scintillating crystals optically coupled to a photon detectors used to localize the position of the interaction and the amount of absorbed energy in the crystals (Table 1.1) [1].



L. Giovanella (\*)

A. Piccardo

Division of Nuclear Medicine, Ente Ospedaliero "Ospedali Galliera", Genoa, Italy


**Table 1.2** The PET scanner performance and the intrinsic PET limitations

The key properties that characterize the PET scanner performances are the spatial resolution, the sensitivity, the Noise-Equivalent Count Rate (NECR) and the contrast [2]. The projection data acquired in the form of sinograms are affected by a number of factors that contribute to the degradation of the final images and hence to the PET scanner performances, as reported in Table 1.2.

Two classes of reconstruction techniques exist: the analytical and the iterative methods [3]. The most used analytical method is the backprojection. To compensate the blurring, a filter is applied to the projections before they are back-projected onto the image [i.e. filtered backprojection (FBP)]. In modern scanners, the image reconstruction algorithms are based on iterative methods, which approach the true image by means of successive estimations, in order to converge to an image that best represents the original object. These algorithms are known as expectation maximization (EM) and Ordered Subset Expectation Maximization (OSEM) algorithm [4].

#### **1.2 Hybrid Scanners: PET/CT and PET/MRI**

Combined PET/CT systems were commercially available from 2001 and in a very short time the dedicated PET scanner was completely replaced by hybrid PET/CT. The ability of hybrid PET/CT systems to accurately identify the anatomic location of diseases and to provide attenuationcorrected images are the main causes of their rapid success and diffusion [5]. Modern clinical PET/CT consists in a high-performance PET scanner in-line with a high-performance CT scanner arranged in sequential gantries. The scanner table moves along the gantry axis in order to subsequently acquire CT and then PET data. A software integrated in the system has to check if the patient bed undergoes some deflections during the translation [6]. Images of tissue attenuation from the CT scan are used to derive the PET attenuation correction factors. The latter depends on the energy of the photons: since CT X-rays and PET γ-rays have an energy of 70 keV and 511 keV,


**Table 1.3** The characteristics of the three commercially available PET/MRI scanners

respectively, the attenuation correction factor obtained from CT must be scaled to the 511 keV photons applying a scaling factor defined by the ratio of the *μ* of the 511 keV photons to that of the 70 keV X-rays in a given tissue [1].

PET/MRI is a multi-modality technology combining the functional information of PET with the soft-tissue contrast of MRI. Actually, two approaches are implemented in the commercial PET/MRI scanners: sequential PET/MRI [7–9]. The characteristics of the three commercial PET/MRI scanners are summarized in Table 1.3.

#### **1.3 Positron Emission Tomography Radiopharmaceuticals**

Radiopharmaceuticals are radiolabelled molecules consisting in a molecular structure and a radioactive nuclide. The first one defines the pharmacokinetics and dynamics within the organism, while the latter is responsible for a detectable signal and for the consequent image visualization [10]. To maintain the stability of these two components, a linker may be necessary. The most important PET nuclides and their physical characteristics are summarized below:


nuclide Oxygen-18 (18O). β+ average energy is 250 keV, corresponding to a mean range in water of 0.6 mm. 18F can be produced by different nuclear reactions; however, the main production mode is targeting Oxygen-18 with cyclotron accelerated protons: 18O(p,n)18F.


to a mean range in water of 0.7 mm. The main 64Cu production modes are the following: 63Cu(n,γ) 64Cu, 65Cu(n,2n)64Cu, 64Zn(n,p)64Cu, 64Zn (d,2p)64Cu.

The wide and feasible availability of positron emitters radionuclides is a prerequisite for successful application on a routine basis. Fluorine-18 and Gallium-68 are the most used in a clinical setting, so far. Due to its versatility, 18F-Fluorodeoxyglucose (FDG), namely a radiolabelled analogue of glucose, is the by far most widely used PET radiopharmaceutical worldwide. FDG is very useful to detect malignant tumours characterized by increased glucose metabolism. However, FDG remains a nonspecific tracer and its uptake is also been observed in many benign conditions, such as infective and inflammatory processes. Therefore, over the last decade, there is a growing interest in researching and using new radiopharmaceuticals, such as radiolabelled amino acids, nucleoside derivatives, choline derivatives, nitroimidazole derivatives and peptides, able to carefully target specific biomarkers. These new generation radiopharmaceuticals allow the analysis of several molecular pathways in tumour biology including metabolism, proliferation, oxygen delivery and protein synthesis as well as receptor and gene expression (Tables 1.4, 1.5 and 1.6). Some examples of PET images with different radiopharmaceuticals are showed in Figs. 1.1 and 1.2.


**Table 1.4** (continued)






12

**Table 1.6**

Brain PET tracers


**Fig. 1.1** Biodistribution of PET tracers: 18F-FDG (**a**), 18F-FCH (**b**), 18F-DOPA (**c**), 18F-Fluociclovine (**d**)

**Fig. 1.2** Biodistribution of PET tracers: 18F-NaF (**a**), 64CuCl2 (**b**) 68Ga-DOTATOC (**c**), 68Ga-PSMA (**d**), 18F-FES (**e**)

#### **References**


ToF-PET component of the GE SIGNA PET/MR system. Med Phys. 2016;43(5):2334.

10. Wadsak W, Mitterhauser M. Basics and principles of radiopharmaceuticals for PET/CT. Eur J Radiol. 2010;73:461–9.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **A Practical Guideline on Diagnostic and Prognostic Meta-Analyses**

**2**

Ramin Sadeghi and Giorgio Treglia

#### **2.1 Introduction**

Evidence based medicine is defined as using the best available evidence for everyday clinical practice [1–3]. Synthetic literature including systematic reviews and meta-analyses plays an important role in evidence based medicine. Essentially systematic reviews and meta-analyses are the cornerstone of evidence based practice. The main difference between a systematic review and a narrative review is the clear method of the former including a clear search and predefined inclusion criteria. The methodology of systematic reviews makes them reproducible which is not the case in narrative reviews [1–3]. The number of systematic reviews and meta-analyses on nuclear medicine diagnostic and prognostic studies is increasing [4,

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

5]. In the current chapter, a practical guideline has been prepared for the researchers who intend to perform a systematic review or meta-analysis of diagnostic and prognostic studies.

#### **2.2 A Clear Topic for Systematic Review: Formulating the Question**

The single most important step in preparing a systematic review is to have a clear topic. The topic is usually divided into several aspects including: patients (the population of the study), intervention (the diagnostic test under study or a prognostic factor which is being evaluated), comparison (the procedures comparative to the index test), outcome (the outcome which is going to be evaluated which are usually sensitivity and specificity for diagnostic studies and overall survival (OR) and progression free survival (PFS) in prognostic ones).

The abovementioned method is called patients-intervention-comparison-outcome (PICO) [6, 7]. The search strategy for systematic reviews is based on the PICO question.

Here are two examples:

1. How does positron emission tomography (PET) [Intervention] work for detection of recurrence [Outcome] in endometrial carcinoma [Patients]?

R. Sadeghi (\*)

Nuclear Medicine Research Center, Mashhad University of Medical Sciences, Mashhad, Iran e-mail: sadeghir@mums.ac.ir

G. Treglia

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

Health Technology Assessment Unit, Academic Education, Research and Innovation Area, Ente Ospedaliero Cantonale, Bellinzona, Switzerland e-mail: giorgio.treglia@eoc.ch

2. What is the prognostic significance of PET [Intervention] for predicting survival [Outcome] in renal cell carcinoma [Patients]?

#### **2.3 Which Articles Should Be Included? Search Strategy, Inclusion and Exclusion Criteria**

Search strategy is based on our PICO question. The keywords and databases which are used for searching should minimize the chance of missing any relevant article. Using Boolean operators (i.e., AND, OR, NOT) is highly recommended. This makes your search as sensitive as possible.

For example for the abovementioned PICO questions the following keywords seem to be optimal:


At least two databases should be included in the search strategy. PubMed/Medline and SCOPUS (or EMBASE) are two main sources for any systematic review.

The inclusion and exclusion criteria should be as clear as possible too. The following factors should be considered to set useful inclusion criteria:


For example for the abovementioned PICO questions, the following inclusion criteria can be set:


Full texts of all relevant studies should be retrieved. The reference of primary studies and all relevant reviews should be checked to search for additional primary studies that could have been missed (backward searching of the citations). In addition, articles citing the relevant included articles can be used to find any other missing articles (forward searching of the citations). The citing articles can be found easily using Google Scholar (https://scholar.google. com/), SCOPUS, or Dimensions (a free newly launched application with many useful options: https://app.dimensions.ai/discover/publication).

Remember to keep the records of all the searches, as well as included and excluded studies.

#### **2.4 Quality Assessment of the Included Studies**

Not all included studies are of same quality. Quality of each study should be checked and reported. Several checklists are available for diagnostic studies [8, 9].

Two of the most commonly used checklists are:


QUADAS and consists of four dimensions (patient selection, index test, reference standard, and finally, flow and timing), the first three of which require an answer among the three available responses (yes/high, no/low, and unclear).

Several checklists are also available for prognostic studies [11].

Two of the most commonly used checklists are:


Checklists are usually equivalent to each other; however, each domain or dimension for all included studies should be explained in detail to give the reader of systematic reviews a clue regarding the quality of the included studies. Reporting only based on numbers (quality scores) should be discouraged.

#### **2.5 Data Extraction**

All relevant data should be extracted from the included studies. Detailed information regarding the study population, method of the diagnostic or prognostic test, gold standard test, follow-up times, methods of ascertaining outcomes, outcome variables such as false and true negative (FN, TN), false and true positive (FP, TP) cases for diagnostic studies, and hazard ratios (HR) for OS and PFS for prognostic studies should be extracted. Extraction of data should be as complete as possible to allow reconstruction of 2 × 2 diagnostic tables or HR of prognostic factors as well as sub-group analyses [11, 13, 14].

Extraction of data in prognostic studies can be very tricky: not all studies reported HR, and only Kaplan Meier (KM) curves and associated log rank tests are usually reported. HR can be extracted from KM curves. Usually the survival data can be extracted manually from KM curves using special software such as getdata graph digitizer (available at http://getdata-graph-digitizer. com/download.php). Finally the extracted survival data can be converted to HR by Parmar method using a special Excel file provided by Parmar et al. [15].

Another important aspect of extraction data in prognostic systematic reviews is type of prognostic factor (quantitative vs. qualitative factors) and evaluation of other prognostic factors (multivariate vs. univariate analysis). HR of quantitative variables (such as SUVmax) can be provided in two ways: first, the prognostic factor can be used as a quantitative variable and a HR using Cox regression is provided. The second type of HR can be calculated by categorizing a quantitative variable into two ranks (for example, SUVmax >7 and ≤7). These two types of HR cannot be pooled with each other even for a same prognostic factor. In addition, only univariate or multivariate HR should be used for pooling data across studies. Pooling univariate HR with a multivariate HR is discouraged as the latter (but not the former) takes into account other potential prognostic factors.

#### **2.6 Pooling Diagnostic Indices Across Studies and Reporting Heterogeneity**

In this final step, the numerical results of the included studies would be pooled together. First of all, diagnostic or prognostic indices of each included study should be presented.

The following diagnostic indices should be reported:


The following prognostic indices should be reported:


Meta-analysis is a special statistical method for pooling data across different studies and giving pooled diagnostic indices. For this purpose, a weight is attributed to each study and the weighted diagnostic indices are pooled together. Special software are available for this purpose, including SAS, R, and STATA.

For diagnostic studies, two free software are available:


For prognostic studies, usually hazard ratios should be pooled across included studies. Several software are available in this regard, such as R, SAS, and Comprehensive Meta-Analysis (CMA).

The least required data to be provided in a meta-analysis are:

1. Pooled indices: They can be perfectly reported by forest plots which give all included studies as well as the pooled data in one view.


5. Publication bias: Although there is substantial literature relating to publication bias in systematic reviews and meta-analyses of randomized controlled trials, little research has been done in the context of systematic reviews and meta-analyses of diagnostic studies. However, publication bias can be visually presented by funnel plots and can be quantified by several methods such as Egger's regression intercept or trim and fill method [24, 25].

#### **2.7 Discussion and Conclusion of Systematic Reviews**

The discussion and final conclusion of a systematic review and meta-analysis should be as objective as possible. The authors should discuss the main results of the systematic review and metaanalysis. Final conclusion should be based on the main results of the systematic review. Any heterogeneity of the included studies should be explained and the possible reasons should be discussed.

Standard method of reporting systematic reviews and meta-analyses Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) provides a minimum requirement for reporting systematic reviews and meta-analyses [21, 26]. Although it is originally prepared for systematic reviews of randomized clinical trials, systematic reviews of diagnostic accuracy studies can be reported using PRISMA too. PRISMA statement and checklist can be found in the following link: http://www.prisma-statement.org/.

#### **2.8 Final Comment**

To publish a high quality systematic review or meta-analysis of diagnostic test accuracy or prognostic studies, certain methodology should be followed. Only methodologically sound systematic reviews and meta-analyses are worth publication and can change or support clinical use of a diagnostic test or a prognostic factor. Hopefully, the abovementioned methodology could help the researchers through the process of systematic review and meta-analysis preparation.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part II**

**Evidence-Based PET in Oncology**

## **Evidence-Based PET for Brain Tumours**

Giorgio Treglia and Barbara Muoio

#### **3.1 Background**

Positron emission tomography (PET) using different radiotracers evaluating different metabolic patterns is able to early detect pathophysiological changes in oncological patients, including those with brain tumours. These functional changes usually occur before the development of morphological changes detected by conventional radiological imaging techniques such as computed tomography (CT) and magnetic resonance imaging (MRI) [1]. MRI is the standard neuroimaging method used for diagnosis of brain tumours, for performing stereotactic biopsy and surgical planning in neuro-oncology [2]. Currently, hybrid

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

Health Technology Assessment Unit, Academic Education, Research and Innovation Area, Ente Ospedaliero Cantonale, Bellinzona, Switzerland e-mail: giorgio.treglia@eoc.ch

#### B. Muoio

Clinic of Oncology and Internal Medicine, San Giovanni Hospital, Oncology Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

imaging techniques as PET/CT and PET/MRI, providing a combination of both functional and morphological information, may be useful methods for early diagnosis of brain tumours [1, 2].

Different PET radiotracers have been used to evaluate brain tumours including fluorine-18 fluorodeoxyglucose (18F-FDG), carbon-11 methionine (11C-methionine), fluorine-18 fluoroethyltyrosine (18F-FET), fluorine-18 fluorodihydroxyphenylalanine (18F-FDOPA), fluorine-18 fluorothymidine (18F-FLT) and radiolabelled choline (11C-choline or 18F-choline).

Enough literature data already exist about the diagnostic performance and prognostic value of PET with different tracers in brain tumours. In particular, 24 meta-analyses on the use of PET or PET/CT with different tracers in brain tumours, published from 2012, were selected through a comprehensive computer literature search [3–26]. The findings of the selected meta-analyses on the diagnostic performance are presented in Table 3.1. Here below we have summarized the main findings of meta-analytic studies based on the different clinical indications of PET or PET/CT.

#### **3.2 Evaluation of Suspicious Primary Brain Tumour**

Four meta-analyses have assessed the diagnostic performance of PET or PET/CT with different tracers in patients for whom primary brain tumours are suspected [3, 20, 23, 26].

G. Treglia (\*)

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland


**Table 3.1** Characteristics and main findings of included meta-analyses on the diagnostic performance of PET or PET/ CT with different tracers in patients with brain tumours

#### **Table 3.1** (continued)


27

(continued)


**Table 3.1** (continued)

*LR+* positive likelihood ratio; *LR−* negative likelihood ratio; *DOR* diagnostic odds ratio; *95%CI* 95% confidence interval; *AA*∗ radiolabelled amino acid PET including radiolabelled methionine, fluoroethyltyrosine and fluorodihydroxyphenylalanine; *NR* not reported; *HGG* high-grade gliomas only; *PCNSL* primary central nervous system lymphoma; *mean TBR* mean tumour-to-background uptake ratio; *max TBR* maximum tumour-to-background uptake ratio

#### **3.2.1 18F-FDG**

A meta-analysis including patients with suspicious primary brain tumours showed that 18F-FDG PET or PET/CT has a moderate sensitivity and specificity for differentiating brain tumours from non-tumour lesions. False-positive findings were often due to inflammatory lesions or other non-tumour tissues; on the other hand, reduced 18F-FDG uptake in brain tumours is usually influenced by the high physiological glucose metabolism in surrounding normal brain tissue, leading to a decreased sensitivity [23]. Another metaanalysis also demonstrated that 18F-FDG PET or PET/CT has a moderate diagnostic performance in distinguishing between tumour and nontumour lesions in the brain, lower than amino acid PET [20].

#### **3.2.2 11C-Methionine**

A meta-analysis by Zhao et al. demonstrated a good diagnostic performance of 11C-methionine PET or PET/CT in detecting brain tumours (pooled sensitivity and specificity were of 95% and 83%, respectively) with higher diagnostic accuracy values compared to 18F-FDG PET or PET/CT, likely due to the higher 11C-methionine uptake in brain tumours and lower accumulation in normal brain tissue [23].

#### **3.2.3 18F-FET**

For initial assessment of patients with a newly diagnosed brain lesion, 18F-FET PET or PET/CT demonstrated a good performance in the diagnosis of a brain tumour with a pooled sensitivity and specificity of 82% and 76%, respectively. A mean tumour-to-background uptake ratio (TBR) threshold of at least 1.6 and a maximum TBR of at least 2.1 had the best diagnostic value for differentiating brain tumours from non-tumour brain lesions. For the diagnosis of glioma versus non-glioma brain lesions, 18F-FET PET or PET/ CT demonstrated a good sensitivity (84%) but a not adequate specificity (62%) [26]. In a head-tohead comparative meta-analysis, the diagnostic performance of 18F-FET PET or PET/CT in distinguishing between tumour and non-tumour lesions in the brain was found significantly higher compared to that of 18F-FDG PET or PET/CT performed in the same patients [20].

#### **3.2.4 18F-FDOPA**

18F-FDOPA PET or PET/CT revealed a moderate sensitivity (71%) and a good specificity (86%) in detecting newly diagnosed gliomas [3].

#### **3.3 Glioma Grading**

Gliomas are the most frequent primary brain tumours. High-grade gliomas like glioblastomas are the most common gliomas in adults, with a poor prognosis with any current therapy. Conversely, low-grade gliomas, the second most common type of gliomas, are potentially curable with appropriate treatment. Several metaanalyses have evaluated the role of PET or PET/ CT with different tracers in differentiating between high-grade and low-grade gliomas [3, 6, 14, 20].

#### **3.3.1 18F-FDG**

18F-FDG uptake is significantly higher in highgrade gliomas compared with low-grade gliomas. According to the meta-analysis of Dunet et al., a mean TBR of at least 1.4 and a maximum TBR of at least 1.8 at 18F-FDG PET had the best value to distinguish between low- and high-grade gliomas, with a sensitivity, specificity and accuracy of 60%, 91% and 74%, respectively, for mean TBR and 72%, 73% and 72%, respectively, for maximum TBR [20]. A recent meta-analysis demonstrated a lower sensitivity of 18F-FDG PET or PET/CT in differentiating between high-grade and low-grade gliomas compared to radiolabelled amino acid PET (11C-methionine and 18F-FET) but with higher specificity [6].

#### **3.3.2 11C-Methionine**

11C-methionine PET or PET/CT had a moderate diagnostic accuracy in differentiating between high-grade and low-grade gliomas, according to data provided by a recent meta-analysis (pooled sensitivity and specificity of 80% and 72%, respectively) [14]. Another meta-analysis demonstrated that 11C-methionine PET or PET/CT has a higher sensitivity compared to 18F-FDG PET or PET/CT in differentiating between highgrade and low-grade gliomas but with lower specificity; diagnostic performance values were similar to those of 18F-FET PET or PET/CT in this setting [6].

#### **3.3.3 18F-FET**

18F-FET uptake is significantly higher in highgrade gliomas compared with low-grade gliomas. Dunet et al. reported that a mean TBR of at least 2.0 and a maximum TBR of at least 3.0 at 18F-FET PET reached a sensitivity, specificity and accuracy of 88%, 73% and 81%, respectively, for mean TBR, and 80%, 82% and 81%, respectively, for maximum TBR [20]. A recent metaanalysis demonstrated that 18F-FET PET or PET/ CT has a higher sensitivity compared to 18F-FDG PET or PET/CT in differentiating between highgrade and low-grade gliomas but with lower specificity; diagnostic performance values were similar to those of 11C-methionine PET or PET/ CT in this setting [6].

#### **3.3.4 18F-FDOPA**

For differentiating high-grade from low-grade gliomas, 18F-FDOPA PET or PET/CT showed a pooled sensitivity of 88% and a pooled specificity of 73% [3].

#### **3.4 Delineation of Gliomas**

For surgical and radiation therapy planning in patients with glioma, a correct delineation of the target volume is needed. A recent evidence-based article suggested that radiolabelled amino acid PET may ameliorate the delineation of highgrade gliomas compared to standard MRI [16].

#### **3.5 Diagnosis of Recurrent Brain Tumours**

Distinguishing recurrent brain tumours from non-tumour lesions after radiation therapy and/or chemotherapy is a crucial clinical issue, because the different diagnosis will lead to divergent treatments. Several meta-analyses have assessed the diagnostic performance of PET with different tracers in this setting [3, 4, 7, 9, 11, 13, 21–25].

#### **3.5.1 18F-FDG**

A meta-analysis of Zhao et al. demonstrated a moderate diagnostic accuracy of 18F-FDG PET or PET/CT in detecting brain tumour recurrence [23]. This finding was confirmed by another meta-analysis which showed a pooled sensitivity and specificity of 78% and 77%, respectively [21]. Furuse et al. showed that the diagnostic performance of 18F-FDG PET or PET/CT in detecting recurrent brain tumours was lower compared to that of radiolabelled amino acid PET or PET/ CT [7]. Nihashi et al. showed that, when considering both low- and high-grade gliomas, pooled sensitivity and specificity of 18F-FDG PET or PET/CT in detecting glioma recurrence were 77% and 78%, respectively. In subgroup analyses limited to high-grade gliomas, pooled sensitivity and specificity were 79% and 70%, respectively [25]. Wang et al. reported a moderate sensitivity (70%) but a good specificity (88%) of 18F-FDG PET or PET/CT in detecting recurrent glioma; however, the diagnostic accuracy was lower compared to that of 11C-methionine PET or PET/CT and magnetic resonance spectroscopy in this setting [22]. Another meta-analysis demonstrated that the diagnostic performance of 18F-FDG PET or PET/CT in detecting recurrent glioma is not optimal, in particular if compared with other available neuroimaging methods [7].

#### **3.5.2 11C-Methionine**

11C-methionine PET or PET/CT demonstrated good diagnostic performance in detecting brain tumour recurrence (pooled sensitivity and specificity of 92% and 87%, respectively), with higher values compared to 18F-FDG PET or PET/CT [23]. For high-grade gliomas, pooled sensitivity and specificity of 11C-methionine PET or PET/ CT in detecting glioma recurrence were 70% and 93%, respectively [25]. Compared to dynamic susceptibility contrast-enhanced MRI, 11C-methionine PET or PET/CT demonstrated comparable pooled sensitivity and specificity in detecting glioma recurrence, with pooled values of 87% and 81.3%, respectively [24]. Similar values of sensitivity and specificity (85% and 83%, respectively) were described by Wang et al., which demonstrated that the diagnostic performance of 11C-methionine PET or PET/CT in detecting glioma recurrence was similar to that of magnetic resonance spectroscopy [22]. A large meta-analysis including 29 studies confirmed the good diagnostic performance of 11C-methionine PET or PET/CT in this setting with a pooled sensitivity and specificity of 88% and 85%, respectively [13].

#### **3.5.3 18F-FET**

A recent meta-analysis demonstrated that 18F-FET PET or PET/CT has a good diagnostic accuracy in differentiating between brain tumour recurrence and radiation necrosis after treatment, with pooled sensitivity and specificity values of 82% and 80%, respectively. In the subgroup of patients with suspicious glioma recurrence, sensitivity and specificity of 18F-FET PET or PET/ CT were 83% and 81%, respectively [11]. The good diagnostic performance of 18F-FET PET or PET/CT in this setting was also confirmed by Furuse et al. who reported increased diagnostic performance of 18F-FET PET or PET/CT compared to 18F-FDG and 11C-methionine PET or PET/CT [7]. Kim et al. found that amino acid PET or PET/CT, including 18F-FET PET, has a good diagnostic performance in differentiating residual or recurrent brain tumour from treatmentrelated changes (pseudoprogression) in patients with high-grade gliomas [4].

#### **3.5.4 18F-FDOPA**

A recent meta-analysis indicated that 18F-FDOPA PET or PET/CT has a good diagnostic accuracy in differentiating between brain tumour recurrence and radiation necrosis after treatment, with pooled sensitivity and specificity values of 85% and 77%, respectively. In the subgroup of patients with suspicious glioma recurrence, sensitivity and specificity of 18F-FDOPA PET or PET/CT were 94% and 89%, respectively [11]. Xiao et al. reported a good sensitivity of 18F-FDOPA PET and PET/CT in detecting recurrent glioma (92%) and a moderate specificity (76%) [3].

#### **3.5.5 18F-FLT**

18F-FLT PET or PET/CT demonstrated a similar diagnostic performance in detecting brain tumour recurrence compared to 18F-FDG PET or PET/ CT with pooled sensitivity and specificity of 82% and 76%, respectively [21].

#### **3.5.6 18C-Choline**

A recent meta-analysis indicated that 11C-choline PET or PET/CT has a good diagnostic accuracy for differentiating glioma recurrence from radiation induced necrosis after treatment, with a pooled sensitivity and specificity of 87% and 82%, respectively [9].

#### **3.6 Diagnosis of Brain Metastases**

The reliability of PET or PET/CT with different tracers in detecting brain metastases has been evaluated to a less extent compared to primary brain tumours. A meta-analysis demonstrated that the pooled sensitivity and specificity of 18F-FDG PET or PET/CT in detecting brain metastases in patients with lung cancer were 21% and 100%, respectively. In particular, the sensitivity of this method is lower compared to that of contrast-enhanced MRI [19].

#### **3.7 Diagnosis of Recurrent Brain Metastases**

The meta-analysis of Li et al., focused on the use of PET or PET/CT with different tracers in differentiating recurrent brain metastasis from radionecrosis after radiation therapy, demonstrated a good diagnostic accuracy of PET or PET/CT with both 18F-FDG and radiolabelled amino acid tracers (11C-methionine, 18F-FET, 18F-FDOPA) in this setting [12]. MRI and PET with different tracers showed similar diagnostic performance for the detection of recurrent brain metastasis after stereotactic radiosurgery; nevertheless, advanced MRI methods showed a significantly higher diagnostic performance in this setting compared to PET [8].

#### **3.8 Diagnosis of Primary Central Nervous System Lymphoma (PCNSL)**

18F-FDG PET and PET/CT showed considerable accuracy in identifying PCNSL among various brain lesions in immunocompetent patients (pooled sensitivity and specificity of 88% and 86%, respectively), therefore, 18F-FDG PET/CT could be a valuable diagnostic imaging method in this setting [17]. High diagnostic accuracy of 18F-FDG PET and PET/CT has also been demonstrated in identifying PCNSL among various brain lesions in patients with human immunodeficiency virus (HIV) infection [18].

#### **3.9 Prognostic Value in Patients with Glioma**

Beyond the diagnostic accuracy, PET/CT parameters, and particularly the TBR, may be significant prognostic factors in patients with glioma. A recent meta-analysis demonstrated that increased TBR at 18F-FDG PET, 11C-methionine PET and 18F-FET PET could indicate poor overall survival (pooled hazard ratios were 3.05 for 18F-FDG PET, 1.59 for 11C-methionine PET and 1.15 for 18F-FET PET) [5]. Another meta-analysis showed that the TBR and metabolic tumour volume at 11C-methionine PET are significant prognostic parameters for patients with gliomas. Patients with a high TBR have a higher risk of death, and patients with a high metabolic tumour volume have a higher risk of adverse events or death [10].

#### **3.10 Conclusions**

Evidence-based data demonstrated good diagnostic performance of PET with different tracers in detecting brain tumours, in particular radiolabelled amino acid tracers showed the highest diagnostic performance values. All the PET tracers evaluated had significant prognostic value in patients with glioma [27].

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

**4**

## **Evidence-Based PET for Head and Neck Tumours**

Gaetano Paone

#### **4.1 Introduction**

Head and neck cancer (HNC) accounts for approximately 5% of all malignant tumours with a continuous growing incidence. Head and neck squamous cell carcinoma (HNSCC) represents the majority of HNC [1, 2]. Nodal involvement is frequent in HNC patients, whereas distant metastases are rather uncommon at the time of initial diagnosis and are found approximately in 10% of patients. There is a clear association with lifestyle and factors as alcoholism, smoking, alimentary factors and viruses for the etiological role, while increasing T and N stages remain the most important adverse prognostic factor [3, 4]. Diagnosis of HNC is usually achieved clinically with endoscopy to obtain direct tissue biopsies. Conventional Imaging (CI), including ultrasound (US), computed tomography (CT) and magnetic resonance imaging (MRI) is important for the evaluation of local extension and to provide information about infiltration, involvement of surrounding structures and regional nodal involvement. There is growing evidence, however, that these modalities have limitations for the diagnostic accuracy of nodal involvement and distant metastases. 18F-FDG PET/CT, allowing the analysis of both metabolic and anatomic features, is a very useful imaging tool in HNC, in particular for disease staging, detection of carcinoma of unknown primary (CUP), treatment monitoring, evaluation of residual or recurrent disease and for prognostic information [5–7].

#### **4.2 Staging**

Accurate staging of disease extension at the time of diagnosis is the most important factor for treatment planning and patients prognosis. Furthermore, providing information in early stage of disease is extremely useful for selecting high-risk patients with impact on specifictreatment selection.

#### **4.2.1 T Staging**

18F-FDG PET/CT has high accuracy in detecting the primary tumour but a moderate diagnostic performance than CI to identify the real tumour extension and infiltration of surrounding tissue and structures. These data are necessary for adequate therapeutic strategy and patient prognosis. False-negative results occur on 18F-FDG PET/CT when the primary tumour is superficial or small, but also in areas of high physiologic activity such as in pharyngeal lymphoid tissue. False-positive results of 18F-FDG PET/CT may be due to

G. Paone (\*)

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland e-mail: gaetano.paone@eoc.ch

<sup>©</sup> The Author(s) 2020 35

G. Treglia, L. Giovanella (eds.), *Evidence-based Positron Emission Tomography*, https://doi.org/10.1007/978-3-030-47701-1\_4

inflammatory processes. In literature, we found conflicting data; preliminary studies have shown divergent results of 18F-FDG PET/CT in diagnosis and staging of HNSCC. Rohde et al. compared the diagnostic accuracy of 18F-FDG PET/ CT for diagnosing HNSCC in comparison with standard CI showing a pooled sensitivity of 89.3% and specificity of 89.5% for 18F-FDG PET/CT and a pooled sensitivity and specificity of 71.6% and 78.0%, respectively, for CI. The authors concluded that 18F-FDG PET/CT is highly accurate in diagnosing patients suffering from HNSCC [8]. Chen et al. compared MRI, CT and 18F-FDG PET/CT in the diagnosis of local and metastatic nasopharyngeal carcinomas. Their analysis suggested that MRI has good accuracy in diagnosis of T stage, whereas CT has a good performance in diagnosis of N stage and 18F-FDG PET/CT shows a good accuracy in diagnosis of M stage [9]. Similarly for evaluation of extracapsular spread (ECS), CT and MRI may be similarly effective, whereas evidence was lacking for 18F-FDG PET/CT and US [10]. 18F-FDG PET/CT can provide, instead, more useful clinical information and higher sensitivity and specificity (pooled sensitivities and specificity 90% and 89%, respectively) to delineate the presence and extent of mandibular involvement in patients with oral cavity cancer, especially in cases of contextual dental artefacts [11, 12]. For evaluation of precancerous and tumour lesions of larynx, Mannelli et al. expressed the need to integrate different imaging methods, proposing a flow chart that allows to stratify patients and select the most appropriate procedure [13].

Overall, the current practice is not in favour of 18F-FDG PET/CT as gold standard for T staging in HNC in exception of cases with suspect mandibular involvement in oral cavity cancer. The preliminary data about 18F-FDG PET/MRI demonstrated high sensitivity and moderate specificity of this technique in the diagnosis of HNC lesions, showing also a better tumour delineation. Further investigations are needed to define the real impact of 18F-FDG PET/MRI in HNC and whether the technique can improve the detection rate of occult primary HNC [14].

#### **4.2.2 Nodal and Distant Metastases Detection**

Lymph nodal involvement is the most important prognostic factor in patients with HNSCC with a significant impact on outcome in terms of disease free survival and overall survival. Lymph nodal (N) metastases occur in approximately 50% of HNC patients at the time of diagnosis with a consequent survival decrease. An accurate N staging is therefore a fundamental step. Similarly, the detection of distant metastases at initial staging influences the prognosis avoiding unnecessary radical treatments. Metastases (M) are frequently found in the lungs, followed by the liver and bone.

Several data in the literature confirm an excellent diagnostic accuracy of 18F-FDG PET/CT in N and M staging. A meta-analysis of Vellayappan et al. assessed the diagnostic accuracy of 18F-FDG PET/CT for staging nasopharyngeal carcinoma (NPC), showing good accuracy of 18F-FDG PET/CT for N staging (pooled sensitivity and specificity were 84% and 90%, respectively) and for M staging (pooled sensitivity and specificity were 87% and 98%, respectively), but not for T classification [15]. Similarly, Shen et al. confirmed in their meta-analysis an excellent diagnostic performance of 18F-FDG PET/CT for detecting lymph node and distant metastases in patients with NPC with a pooled sensitivity and specificity of 89% and 96%, respectively [16]. Considering only the detection accuracy for regional nodal metastases in HNC before treatment, 18F-FDG PET/CT showed good diagnostic performance [17, 18]. Moreover, compared with CI, 18F-FDG PET/CT may have higher per-necklevel sensitivity [19]. These values are even more significant excluding clinically N0 patients with greater accuracy values for 18F-FDG PET/ CT. Several data showed moderate sensitivity of 18F-FDG PET/CT for detection of cervical lymph nodal metastases in clinical N0 HNSCC patients with absence of significant better diagnostic accuracy compared to CI; conversely, 18F-FDG PET/CT has a higher specificity and negative predictive value for the detection of cervical metastatic lymph nodes compared to the other imaging modalities in clinical N0 HNSCC [20–22]. Avoiding elective neck dissection is a fundamental step in the diagnostic-therapeutic flow chart of these patients in order to minimize morbidity and health costs. At present elective neck dissection in patients with clinical N0 should not be based upon cross-sectional imaging. A combination of CI and sentinel node biopsy seems to be the preferred staging strategy to reduce the risk of occult metastases in clinical N0 HNSCC [23].

On the other hand, the excellent diagnostic performance of 18F-FDG PET/CT for detecting distant metastases is clearly underlined in the literature [24–29]. Xu et al. showed a pooled sensitivity and specificity of 85.7% and 98.1%, respectively, for 18F-FDG PET/CT, resulting in a significantly better M staging than CI [26]. This was mainly due to the superior diagnostic performance of 18F-FDG PET/CT compared to CI in detecting bone metastases [27]. In this setting, 18F-FDG PET/CT has higher sensitivity compared to bone scintigraphy [28]. On the other hand, for detection of liver metastases 18F-FDG PET/CT requires further optimization and integration with CI, especially contrast-enhanced CT and MRI [25]. About lung metastases, a metaanalysis demonstrated that 18F-FDG PET/CT is a valuable diagnostic tool for diagnosing lung malignancies in patients with HNSCC [29].

#### **4.3 Prognostic Value**

The prognostic value of 18F-FDG PET/CT has been widely discussed with controversial results. Relevant limiting factors are the variability and reproducibility of each individual parameter. Overall, maximum standardized uptake value (SUVmax), metabolic tumour volume (MTV) and total lesion glycolysis (TLG) were significant prognostic predictors in patients with HNC [30–36].

No significant correlation was found between metabolic parameters of 18F-FDG PET/CT in HNC and human papillomavirus (HPV) status [37]. Furthermore, the semi-quantitative PET/CT parameters were not related to histopathological parameters in HNSCC, as Ki67 and p53 [38].

#### **4.4 Post-treatment Evaluation**

Relevant applications of 18F-FDG PET/CT in HNC are delineation of the tumour volume for radiation treatment planning, discrimination of post-treatment changes, evaluation of response to multimodality therapy and detection of recurrence.

About radiation therapy planning, Jeong et al. found that 18F-FDG-avid HNC apparently require 10–30% more radiation dose than FDG-non-avid tumours, supporting radiotherapy boosts for 18F-FDG-avid tumours; prospective studies are still required in this field [39].

The role of intra-therapy and post-therapy 18F-FDG PET/CT in predicting long-term survival outcomes in patients treated for HNC has been widely studied. Sheikhbahaei et al. reported that positive results of intra-therapy or post-therapy 18F-FDG PET/CT could significantly predict the 2- and 5-year risk of death or disease progression [40]. The same group confirmed the high diagnostic performance of 18F-FDG PET/CT in detecting local, regional and distant recurrences in curatively treated patients with HNC. The pooled sensitivity and specificity of follow-up 18F-FDG PET/CT for detection of recurrence were 92% and 87%, respectively [41]. These data support its use in clinical practice as confirmed also by other studies that highlight the high accuracy of 18F-FDG PET/CT performed after the completion of therapy both in NPC and HNSCC before salvage treatment [42–44]. 18F-FDG PET/ CT is also superior to MRI in distinguishing recurrent NPC from fibrosis or scar tissue after radiotherapy in irradiated fields with distortion of normal architecture [45]. Treatment-to-time scan remains a debated aspect. Several works have indicated that early 18F-FDG PET/CT was less accurate than more delayed imaging after therapy, particularly Cheung and coauthors supported the use of 18F-FDG PET/CT more than 12 weeks after radiotherapy with or without chemotherapy for the assessment of residual or recurrent HNC [46]. Recently, Helsen et al. confirmed that 18F-FDG PET/CT performed within 6 months after chemo-radiotherapy in HNSCC patients is the method of choice for ruling out residual/recurrent nodal disease reducing the need for therapeutic intervention [47]. Finally, sensitivity and specificity of 18F-FDG PET/CT in identifying local failure following curative radiotherapy or surgery for HNSCC were significantly improved when imaging was performed 3 months after end of treatment [48].

#### **4.5 Carcinoma of Unknown Origin and Incidental Findings**

Several studies have investigated the accuracy of 18F-FDG PET/CT to identify carcinoma of unknown origin (CUP) in patients with cervical lymph nodal metastases. Generally, the most common sites of detection include the palatine tonsils and the base of the tongue, with increase of falsenegative results when the primary tumour is small or adjacent to physiological uptake sites. Zhu et al. showed a high sensitivity (97%) and a moderate specificity (68%) for the detection of CUP in patients with cervical nodal metastases [49].

Finally, Treglia et al. calculated the pooled prevalence and risk of malignancy of incidental focal 18F-FDG uptake in the parotid glands. The pooled prevalence of this finding is about 1% of all 18F-FDG PET/CT. Although these incidental findings are benign in most of the cases, complementary evaluation is needed to exclude malignant lesions or with possible malignant degeneration [50].

#### **References**


pharyngeal cancer: a meta-analysis. Oncotarget. 2017;8(35):59740–7.


PET or PET/CT: systematic review and meta-analysis. Otolaryngol Head Neck Surg. 2016;154(3):421–32.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Evidence-Based PET for Thoracic Tumours**

Filippo Lococo, Alfredo Cesario, Stefano Margaritora, and Giorgio Treglia

#### **5.1 Introduction**

Fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is a robust imaging tool that is currently used in daily clinical practice for the evaluation of thoracic malignancies. This chapter provides an overview of the current evidencebased data on the usefulness of PET/CT for the evaluation of patients with thoracic tumours including lung cancer, pleural and thymic tumours, and esophageal cancer.

#### **5.2 Evidence-Based Data on PET in Primary Lung Tumours**

Herein we reviewed recent evidence-based data on the usefulness of 18F-FDG PET/CT for: (1) characterization of solitary pulmonary nodules (SPNs), (2) non-small cell lung cancer (NSCLC) staging, (3) restaging after induction therapy and systemic therapy response assessment in NSCLC, (4) radiation therapy planning, (5) diagnosis of lung cancer recurrence in NSCLC, (6) prognostic evaluation, (7) management of small cell lung cancer (SCLC).

#### **5.2.1 Characterization of Solitary Pulmonary Nodules (SPNs)**

Characterizing a SPN detected incidentally or, as is the case more recently, on CT screening for lung cancer is a major public health issue. 18F-FDG PET/CT is not indicated for characterization of SPNs of less than 8 mm in diameter according to current guidelines [1]. This threshold was set to take into account the spatial resolution of PET systems, due to the significant risk of false-negative findings for small lesions. However, over the last decade, the spatial resolution of 18F-FDG PET/CT has significantly increased and future analysis could verify if this threshold will be modified accordingly.

F. Lococo (\*) · A. Cesario · S. Margaritora Department of General Thoracic Surgery, Catholic University of the Sacred Heart, Fondazione Policinico A. Gemelli, Rome, Italy

G. Treglia

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

Health Technology Assessment Unit, Academic Education, Research and Innovation Area, Ente Ospedaliero Cantonale, Bellinzona, Switzerland e-mail: giorgio.treglia@eoc.ch

#### **5.2.1.1 Single-Time-Point 18F-FDG PET or PET/CT**

In the last decade, a robust evidence has been produced on the potential use of 18F-FDG PET/ CT in early diagnosis of lung cancer (Table 5.1). Chien and colleagues [2] in 2013 conducted a systematic review on this topic reporting evidence of lung cancer screening programmes with 18F-FDG PET, in which the estimated pooled sensitivity and specificity were 83% and 91%, respectively. At that moment, despite PET appeared to have high sensitivity and specificity as a selective screening modality, the role of primary PET screening for lung cancer remained unknown and still undefined.

Subsequently, a further systemic analysis [3] reported a very high (98.7%) pooled sensitivity of 18F-FDG PET/CT in this setting while specificity was suboptimal (58.2%).

In 2016, the research team headed by Madsen [4] suggested that 18F-FDG PET/CT can rule out malignancy in most SPNs due to high sensitivity (recommendation level A) but at the same time the sensitivity of 18F-FDG PET/CT in general is insufficient to rule out mediastinal lymph node metastasis (recommendation level A). Therefore, with few exceptions (lesions <1 cm and non-solid lesions), they concluded that SPNs could be presumptively considered benign if 18F-FDG PET is negative. In addition, lymph node metastasis in the mediastinum cannot be ruled out on the basis of a negative 18F-FDG PET/CT, and confirmative (mini)invasive staging should be performed in most patients.

More recently, a further meta-analysis [5] showed that the pooled sensitivity and specificity of 18F-FDG PET/CT in characterizing SPNs were 82% and 81%, respectively, demonstrating moderate accuracy for 18F-FDG PET/CT in differentiating malignant from benign SPNs.

A further meta-analysis exploring the value of 18F-FDG PET/CT in the diagnosis of SPNs was reported in 2018 [6]. Pooled results indicated a sensitivity of 89% and a specificity of 70%.



*95%CI* 95% confidence interval, *STP* single-time-point PET, *DTP* dual-time-point PET, *N/A* not available

Considering the unsatisfactory results, especially in terms of specificity, the authors stated that 18F-FDG PET/CT cannot replace the "gold standard" pathology by resection or biopsy.

Not dissimilar results have been reported in a further recent meta-analysis performed by Divisi and co-workers [7]. The authors concluded that despite 18F-FDG PET/CT presents a fairly good diagnostic accuracy in SPNs evaluation, it should not be considered as a discriminatory test rather than a method to be included in a clinical and diagnostic pathway.

Interestingly, Deppen and co-workers [8] evaluated the accuracy of 18F-FDG PET in diagnosing lung cancer comparing populations with or without a risk for endemic infectious lung disease. They observed a 16% lower average adjusted specificity in regions with endemic infectious lung disease (61%) compared with non-endemic regions (77%). On the other hand, the sensitivity did not change appreciably by endemic infection status, even after adjusting for relevant factors. On the light of these results, the authors did not suggest the use of 18F-FDG PET to diagnose lung cancer in regions with endemic pulmonary infections unless an institution achieves test performance accuracy similar to that found in non-endemic regions.

Lastly, a meta-analysis investigates the diagnostic performance of 18F-FDG PET/CT compared with diffusion-weighted magnetic resonance imaging (DW-MRI) for distinguishing malignant versus benign SPNs [9]. DW-MRI had a pooled sensitivity and specificity of 83% and 91%, respectively, compared with 78% and 81%, respectively, for PET/CT. The authors concluded that the diagnostic performance of DW-MRI is comparable or superior to that of 18F-FDG PET/ CT in the differentiation of malignant and benign pulmonary lesions.

#### **5.2.1.2 Dual-Time-Point (DTP) PET**

Several authors have also explored the potential use of a DTP 18F-FDG PET in differentiating malignant from benign SPNs (Table 5.1). In 2012, a meta-analysis was performed by Lin and co-workers [10] exploring the diagnostic performance of both single-time-point (STP) and DTP 18F-FDG PET techniques. Sensitivity was higher with DTP imaging at moderate levels of specificity. This potential advantage of DTP over initial STP scanning was diminished at higher levels of specificity. Although there was no clear evidence to support the routine use of DTP imaging with 18F-FDG PET in the differential diagnosis of pulmonary nodules, the authors suggested as such technique may provide additional information in selected cases with equivocal results from initial scanning. Other meta-analyses [11–13] reported similar diagnostic accuracy among DTP and STP 18F-FDG PET or PET/CT in the diagnosis of SPNs. According to these results, the additional value of DTP compared to STP 18F-FDG-PET/ CT resulted to be questionable.

#### **5.2.1.3 18F-FLT PET for Evaluation of Pulmonary Lesions**

The potential use of fluorine-18 fluorothymidine ( 18F-FLT) PET in patients with pulmonary lesions was evaluated by two meta-analyses [14, 15] (Table 5.1), which showed that 18F-FLT PET had a higher specificity but lower sensitivity compared to 18F-FDG PET in the evaluation of SPNs. Therefore, the authors assumed that 18F-FLT and 18F-FDG together could add diagnostic confidence for pulmonary lesions.

#### **5.2.2 NSCLC Staging**

Nodal (N) and distant metastases (M) staging is one of the major prognostic factors of survival in NSCLC patients. Accurate staging of distant metastases is crucial, as the treatment strategy is directly dependent on tumour stage. Although many studies have been reported in the last decades evaluating the performance of 18F-FDG PET/CT in lung cancer staging, the results among studies are still almost controversial.

#### **5.2.2.1 N Staging**

Zhao and associates [16] performed a metaanalysis about 18F-FDG PET/CT for detecting mediastinal nodal metastases in patients with NSCLC. The pooled sensitivity and specificity with 95% confidence interval values (95%CI) on a per-patient analysis were 71.9% (95%CI: 68.3– 75.3%) and 89.8% (95%CI: 88.2–91.2%), respectively.

A second meta-analysis on the same issue [17] showed a pooled sensitivity of 62% for 18F-FDG PET/CT (95%CI: 54–70%) and a pooled specificity of 92% (95%CI: 88–95%) on a node-based analysis. The pooled sensitivity and specificity were 67% (95%CI: 54–79%) and 87% (95%CI: 82–91%), respectively, on a patient-based analysis. Interestingly, those studies from tuberculosis endemic countries showed lower sensitivity and also lower specificity compared to nontuberculosis endemic countries [17, 18].

Two meta-analyses were specifically limited to early-stage NSCLC cases. In detail, Wang and co-workers [19] found that the negative predictive value (NPV) of 18F-FDG PET/CT for lymph nodal mediastinal metastases was 94% for T1 disease and 89% for T2 disease. Including both T1 disease and T2 disease, the NPV was 93% for mediastinal metastases and 87% for overall nodal metastases. Interestingly, adenocarcinoma histology type and high 18F-FDG uptake in the primary lesion were associated with greater risk of occult nodal metastases.

Similarly, a second meta-analysis [20] focused on patients with resectable NSCLC revealed that 18F-FDG PET/CT had a pooled sensitivity and specificity for N staging of 81.3% (95%CI: 70.2– 88.9%) and 79.4% (95%CI: 70–86.5%), respectively. The authors assumed that accuracy of 18F-FDG PET/CT in N staging was insufficient to allow management and strategy of care based on 18F-FDG PET/CT findings alone.

Shen et al. [21] also investigated the diagnostic value of DTP 18F-FDG PET/CT versus STP imaging for detection of mediastinal nodal metastases in NSCLC patients. Pooled sensitivity and specificity for DTP PET/CT were 85% (95%CI: 78–91%) and 75% (95%CI: 68–82%), respectively, and for STP imaging the same values were 79% (95%CI: 70–85%) and 73% (95%CI: 65–79%), respectively. The authors were very cautious in supporting the implementation of DTP imaging in routine PET protocols for mediastinal lymph node staging of NSCLC.

Lastly, two meta-analyses compared 18F-FDG PET/CT and DW-MRI for detection of mediastinal nodal metastases in NSCLC [22, 23] reporting similar results in terms of diagnostic accuracy among these two imaging methods.

#### **5.2.2.2 M Staging**

A meta-analysis by Li and co-workers [24] showed the excellent diagnostic performance of 18F-FDG PET/CT for diagnosis of distant metastases in patients with NSCLC with a pooled sensitivity and specificity of 93% (95%CI: 88–96%) and 96% (95%CI: 95–96), respectively. Similar results were reported by Yu et al. [25] who found a pooled sensitivity of 81% (95%CI: 63–92%) and 96% (95%CI: 94–98%), respectively. A further meta-analysis on the same topic [26] demonstrated that concerning extra-thoracic metastases of NSCLC, the pooled sensitivities and specificities of 18F-FDG PET/CT were 77% (95%CI: 47–93%) and 95% (95%CI: 92–97%) for all extra-thoracic metastases, whereas the same values were 91% (95%CI: 80–97%) and 98% (95%CI: 94–99%), respectively, for bone metastases. Conversely, 18F-FDG PET/CT showed low sensitivity in detecting brain metastases.

Concerning the latter issue, a comparative meta-analysis MRI and 18F-FDG PET/CT for the diagnosis of brain metastases in NSCLC [27] revealed that MRI had higher sensitivity (77%) than 18F-FDG PET/CT (21%) for the diagnosis of brain metastases.

Chang et al. [28] found a higher sensitivity and specificity of 18F-FDG PET/CT compared to bone scintigraphy (BS) in detecting bone metastases from NSCLC. A further more robust metaanalysis [29] showed that 18F-FDG PET/CT is a better imaging method in terms of sensitivity and specificity compared to MRI and BS for detecting bone metastases from NSCLC, with a pooled sensitivity and specificity of 92% (95%CI: 88–95%) and 98% (95%CI: 97–98), respectively.

Finally, the diagnostic performance of 18F-FDG PET/CT in detecting adrenal metastases from NSCLC was recently evaluated by Wu and co-workers [30]. The pooled sensitivity and specificity of 18F-FDG PET/CT in this setting were 88.7% (95%CI: 85.2–91.7%) and 90.8% (95%CI: 87.5–93.4%), respectively, suggesting excellent performance.

#### **5.2.3 Restaging After Induction Therapy and Prediction of Treatment Response**

The ability to identify potential responders to induction treatment may improve patient selection or surgery and may help in the development of response criteria suitable for routine monitoring of response. By providing information on the metabolic activity of tumour cells, 18F-FDG PET/ CT has become a powerful tool in assessing treatment response. Zhang and colleagues [31] performed a meta-analysis to evaluate the value of 18F-FDG PET in predicting the pathological tumour response of lung cancer to induction therapy. The authors found that 18F-FDG PET could play an important role in predicting nonresponders to induction therapy in cases of lung cancer: indeed, the pooled sensitivity, specificity, positive predictive value, and negative predictive value for PET-predicted response were 83% (95%CI: 76–89%), 84% (95%CI: 79–88%), 74% (95%CI: 67–81%), and 91% (95%CI: 87–94%), respectively.

A recent evidence-based article assessed the use of 18F-FDG PET/CT for both assessing the efficacy of treatment response and performing post-treatment follow-up of lung cancer [32]. PET metabolic response (PERCIST criteria) has been shown to be a better predictor of histopathologic response than anatomic response metrics (WHO and RECIST criteria). 18F-FDG PET/CT was indicated for treatment response assessment when it is performed within 6 months from treatment completion, though evidence for its comparative effectiveness with chest CT is still evolving.

#### **5.2.4 Radiation Therapy Pretreatment Planning in NSCLC**

18F-FDG PET/CT may also increase the likelihood of correctly delineating tumour tissue before radiotherapy dose planning. In 2017, Hallqvist and colleagues [33] reported the results of a meta-analysis on the use of 18F-FDG PET/ CT for radiotherapy dose planning. According to this meta-analysis, a change in target definition was 36% in patients with a former staging PET, and 43% and 26% in patients without a staging PET for NSCLC and SCLC, respectively. The corresponding summary estimates of a change in treatment intent from curative to palliative treatment were 20% and 22% and 9%, respectively. Another recent meta-analysis demonstrated that functional lung imaging, including PET, may have potential utility in radiation therapy planning and delivery [34].

#### **5.2.5 Diagnosis of Lung Cancer Recurrence**

Although there are no conclusive data to support the survival benefits of early detection or early treatment for recurrence of lung cancer, an early and accurate diagnosis of recurrence is critical to optimize therapy. A meta-analysis [35] was performed to assess the diagnostic value of 18F-FDG PET and PET/CT for cases of recurrent lung cancer. In the patient-based analysis performed, 18F-FDG PET and PET/CT were found to provide better detection of lung cancer recurrence compared to CT. Indeed, the pooled sensitivity for 18F-FDG PET, PET/CT, and CT were 94% (95%CI: 91–97%), 90% (95%CI: 84–95), and 78% (95%CI: 71–84%), respectively while the pooled specificity for 18F-FDG PET, PET/CT, and CT were 84% (95%CI: 77–89%), 90% (95%CI: 87–93%), and 80% (95%CI: 75–84%), respectively.

#### **5.2.6 Prognostic Evaluation in NSCLC**

In their meta-analysis, Paesmans et al. [36] assessed the prognostic value of primary tumour maximum standardized uptake value (SUVmax) at 18F-FDG PET for overall survival (OS) of NSCLC patients. At multivariate analysis, SUVmax was found to be independently associated with survival. The hazard ratio (HR) for SUVmax was 1.58 (95%CI: 1.27–1.96).

Despite the SUVmax represents the most widely applied semi-quantitative PET parameter in clinical practice, volumetric PET parameters, including metabolic tumour volume (MTV) and total lesion glycolysis (TLG), have been also used to reflect disease burden and tumour aggressiveness in NSCLC. A first meta-analysis performed by Liu et al. [37] explored the prognostic value of SUVmax, MTV, and TLG on diseasefree survival (DFS) and OS in surgical NSCLC patients. The pooled HRs for OS were 1.52 for SUVmax, 1.91 for MTV, and 1.94 for TLG. On the basis of these results, the authors stated that high values of SUVmax, MTV, and TLG are able to predict a higher risk of recurrence or death in patients with surgical NSCLC, suggesting the use of 18F-FDG PET/CT to select patients who are at high risk of disease recurrence or death as the best candidates from aggressive treatments. Other authors [38] conducted a meta-analysis on the prognostic value of MTV and TLG in NSCLC patients. A worse prognosis was observed in patients with high MTV (HR: 2.31) and with high TLG (HR: 2.43).

Han and colleagues [39] performed a metaanalysis exploring prognostic value of texture parameters derived by 18F-FDG PET in patients with lung cancer. They concluded that there is insufficient evidence to support the prognostic value of texture analysis in 18F-FDG PET in lung cancer.

Another interesting application of 18F-FDG PET is the ability to predict long-term results after radiation therapy. Dong and co-workers [40] explored the prognostic relevance of SUVmax at 18F-FDG PET for early-stage NSCLC patients receiving stereotactic body radiation therapy (SBRT). The authors found that those NSCLC patients presenting with high levels of pre-SBRT SUVmax had poorer OS and local control and higher risk of distant metastases. These findings were confirmed by another metaanalysis [41] showing that both pre-radiotherapy and post-radiotherapy primary tumour SUVmax can predict the outcome of patients with NSCLC treated with radiotherapy.

Other authors [42] have summarized the prognostic value of early response at 18F-FDG PET in NSCLC patients treated with tyrosine-kinase inhibitors (TKI). Early response of patients with NSCLC treated with TKIs identified on 18F-FDG PET was found to be associated with improved OS and progression-free survival (PFS).

#### **5.2.7 Management of SCLC**

The role of 18F-FDG PET in the management of SCLC has been largely investigated in the last decades. A systematic review and meta-analysis performed by Lu et al. [43] to evaluate the diagnostic accuracy of 18F-FDG PET/CT in the pretherapeutic staging of patients with SCLC demonstrated a pooled sensitivity and specificity of 97.5% (95%CI: 94.2–99.2%) and 98.2% (95%CI: 94.9–99.6%), respectively, for the detection of extensive disease in SCLC patients. Therefore, evidence-based data suggest the role of 18F-FDG PET/CT for discriminating between limited and extensive disease in SCLC.

The prognostic value of the SUVmax of primary SCLC at 18F-FDG PET was recently investigated through a meta-analytic study [44]: the pooled HR for OS was 1.13 (95%CI:1.05–1.22), thus indicating that SCLC patients with high SUVmax may have poorer prognosis.

#### **5.3 Evidence-Based Data on PET in Pleural Tumours**

Three meta-analyses assessed the role of 18F-FDG PET or PET/CT in the characterization of pleural lesions [45–47], whereas meta-analyses on the role of 18F-FDG PET/CT in staging, restaging, prognostic or treatment response evaluation of pleural tumours are currently lacking.

18F-FDG-PET and PET/CT demonstrated to be accurate diagnostic imaging methods in the differential diagnosis between malignant and benign pleural lesions in patients with or without known cancer; nevertheless, possible sources of falsenegative and false-positive results should be kept in mind [45, 46]. In patients without known cancer, sensitivity and specificity of 18F-FDG-PET and PET/CT were 95% (95%CI: 92–97%) and 82% (95%CI: 76–88%), respectively [45]. In patients with known cancer, pooled sensitivity was 86% (95%CI: 80–91%) and pooled specificity was 80% (95%CI: 73–85%) [46]. Porcel et al. in their meta-analysis [47] demonstrated that semi-quantitative PET assessment had a significantly lower sensitivity for diagnosing malignant pleural effusions than visual assessments. The pooled sensitivity and specificity of 18F-FDG PET/CT using semi-quantitative interpretation for identifying malignant pleural effusions were 81% and 74%, respectively. The moderate accuracy of semi-quantitative PET assessment precludes its routine recommendation for discriminating malignant from benign pleural effusions.

#### **5.4 Evidence-Based Data on PET in Thymic Epithelial Tumours**

One meta-analysis [48] showed that 18F-FDG PET may predict the WHO grade of malignancy in thymic epithelial tumours (TETs), reporting a statistically significant difference of SUVmax between the different TETs (low-grade thymomas, high-grade thymomas, and thymic carcinomas). In detail, the pooled mean difference of SUVmax between high-risk and low-risk thymomas was 1.2 (95%CI: 0.4–2.0), that between thymic carcinomas and low-risk thymomas was 4.8 (95%CI: 3.4–6.1), and that thymic carcinomas and high-risk thymomas was 3.5 (95%CI: 2.7–4.3).

Notably, meta-analyses on the role of 18F-FDG PET/CT in staging, restaging, prognostic or treatment response evaluation of TETs are currently lacking.

#### **5.5 Evidence-Based Data on PET in Esophageal Tumours**

#### **5.5.1 Staging**

The real and unquestionable additional diagnostic value of 18F-FDG PET/CT in comparison to conventional imaging methods is in evaluating distant metastases (M staging) of esophageal cancer [49], whereas recent evidence-based articles have addressed the performance of 18F-FDG PET/CT for detecting lymph nodal metastases (N staging).

Jiang et al. [50] found that the pooled sensitivity and specificity estimates of 18F-FDG PET/CT for detecting regional lymph nodal metastases at staging were 66% (95%CI: 51–78%) and 96% (95%CI: 92–98%), respectively. The corresponding values on a per-patient analysis were 65% (95% CI: 49–78%) and 81% (95%CI: 69–89%), respectively. Overall, 18F-FDG PET/CT has a moderate to low sensitivity and a high to moderate specificity for detection of regional nodal metastases in esophageal cancer. Therefore, extending the extent of lymph node dissection or radiotherapy target volume is necessary after the diagnosis of regional nodal metastases by 18F-FDG PET/CT.

In another meta-analysis [51], Hu et al. evaluated the diagnostic performance of 18F-FDG PET/ CT for the assessment of preoperative lymph node metastases in patients with esophageal cancer. In patients without neoadjuvant treatment, 18F-FDG PET/CT had a pooled sensitivity and specificity of 57% (95%CI: 45–69%) and 91% (95%CI: 85–95), respectively. In patients who received neoadjuvant treatment, 18F-FDG PET/ CT had a pooled sensitivity and specificity of 53% (95%CI: 35–70%) and 96% (95%CI: 86–99%), respectively. Therefore, 18F-FDG PET/CT has a high specificity but a low sensitivity; thus, it cannot accurately detect the lymph nodal involvement in patients with esophageal cancer.

Shi et al. [52] also demonstrated that 18F-FDG PET/CT had lower sensitivity and accuracy for detection of regional nodal metastases in patients with esophageal cancer before surgery. The pooled sensitivity and specificity were 62% (95%CI: 40–79%) and 96% (95%CI: 93–98%), respectively, on a per-station analysis; the corresponding values on a per-patient analysis were 55% (95%CI: 34–74%) and 76% (95%CI: 66–83%), respectively.

In this setting, cervical ultrasonography has very limited additional diagnostic value as supplement to a negative 18F-FDG PET/CT in the detection of cervical lymph node metastases during the initial staging of patients with esophageal cancer, as demonstrated by Goense et al. [53].

#### **5.5.2 Restaging**

Restaging after neoadjuvant therapy aims to reduce the number of patients undergoing oesophagectomy in case of distant (interval) metastases. Kroese et al. [54] assessed the diagnostic performance of 18F-FDG PET or PET/CT for the detection of distant interval metastases after neoadjuvant therapy in patients with esophageal cancer. The pooled proportion of patients in whom true distant interval metastases were detected by 18F-FDG PET or PET/CT at restaging was 8% (95%CI: 5–13%). The pooled proportion of patients in whom false-positive distant findings were detected by 18F-FDG PET or PET/CT at restaging was 5% (95%CI: 3–9%). In conclusion,18F-FDG PET or PET/CT at restaging after neoadjuvant therapy for esophageal cancer can considerably impact on treatment decisionmaking. However, pathological confirmation of suspected lesions is needed.

Cong et al. [55] assessed the value of 18F-FDG PET or PET/CT for response prediction of primary tumour in patients with esophageal cancer during (group A) or after (group B) neoadjuvant chemoradiotherapy. The pooled sensitivity and specificity were 85% (95%CI: 76–91%) and 59% (95%CI: 48–69%), respectively, in group A. The equivalent values were 67% (95%CI: 60–73%) and 69% (95%CI: 63–74%), respectively, in group B. Interestingly, the pooled sensitivity was 90% in the studies that enrolled patients with esophageal squamous cell carcinoma merely in group B. According to the present data, 18F-FDG PET/ CT should not be used routinely to guide treatment strategy in esophageal cancer patients, but an additional value is expected in patients with esophageal squamous cell carcinoma treated with neoadjuvant chemoradiotherapy.

Goense et al. [56] assessed the diagnostic performance of 18F-FDG PET or PET/CT for diagnosing recurrent esophageal cancer after initial treatment with curative intent. Pooled estimates of sensitivity and specificity for 18F-FDG PET and PET/CT in this setting were 96% (95%CI: 93–97%) and 78% (95%CI: 66–86%), respectively. Therefore, 18F-FDG PET and PET/CT are reliable imaging modalities with a high sensitivity and moderate specificity for detecting recurrent esophageal cancer after treatment with curative intent. However, histopathologic confirmation of PET/CT-suspected lesions is required, because a considerable false-positive rate is noticed.

#### **5.5.3 Predictive and Prognostic Value**

Han et al. [57] performed a meta-analysis on the prognostic value of volumetric parameters (MTV and TLG) derived from pretreatment 18F-FDG PET/CT in patients with esophageal cancer. The pooled HRs of MTV and TLG for OS were 2.26 (95%CI: 1.73–2.96) and 2.23 (95%CI: 1.73– 2.87), respectively. Regarding event-free survival, the pooled HRs of MTV and TLG were 2.03 (95%CI: 1.66–2.49) and 2.57 (95%CI: 1.82–3.62), respectively. Therefore, in patients with esophageal cancer, MTV and TLG derived from pretreatment 18F-FDG PET are significant prognostic factors.

Schollaert et al. [58] performed a metaanalysis on the predictive value of 18F-FDG PET for assessing DFS and OS in esophageal and oesophagogastric junction cancer after neoadjuvant chemoradiation therapy. The pooled HRs for complete metabolic response versus no response were 0.51 for OS (95%CI: 0.4–0.64) and 0.47 for DFS (95%CI: 0.38–0.57), respectively. Therefore, metabolic response on 18F-FDG PET is a significant predictor of long-term survival.

Lastly, Zhu et al. [59] performed a metaanalysis on the prognostic significance of SUVmax on 18F-FDG PET/CT in patients with localized oesophagogastric junction cancer receiving neoadjuvant chemotherapy/chemoradiation therapy. Significant prognostic values of SUVmax before and during therapy in localized oesophagogastric junction cancer were not found. Conversely, relative changes in 18F-FDG-uptake after therapy are significant prognostic markers for OS and DFS.

#### **References**


the detection of distant metastases in patients with non-small cell lung cancer at initial staging: a metaanalysis. Cancer Manag Res. 2018;10:1859–64.


of oesophageal cancer using FDG-PET—a critical appraisal. Eur J Radiol. 2012;81(1):21–30.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

**6**

## **Evidence-Based PET for Breast Cancer**

Giorgio Treglia

#### **6.1 Introduction**

Fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is currently used in daily clinical practice for the evaluation of breast cancer (BC) patients. This chapter provides an overview of the current evidence-based data on the usefulness of PET/CT (using 18F-FDG and other radiotracers) for different indications in patients with BC.

#### **6.2 Staging**

A recent network meta-analysis comparing 19 different imaging methods demonstrated the relatively higher specificity of 18F-FDG PET/CT compared to other imaging methods for the detection of BC lesions [1].

Liang et al. [2] evaluated through a metaanalytic approach the accuracy of magnetic resonance imaging (MRI) and 18F-FDG PET/CT for lymph nodal (N) staging of early BC. The pooled specificities of MRI and PET/CT for diagnosing regional lymph nodal status in BC patients were similar (93%); however, the pooled sensitivity of MRI was significantly greater than PET/CT (82% versus 64%), respectively.

Hong et al. [3] performed a meta-analysis to evaluate the value of 18F-FDG PET/CT for diagnosis of distant metastases of BC. Pooled sensitivity and specificity of 18F-FDG PET/CT were 96% (95%CI: 90–98%) and 95% (95%CI: 92–97%), respectively. Compared with conventional imaging, 18F-FDG PET/CT has higher sensitivity for diagnosis of distant metastases in BC patients.

Similar findings were reported in another meta-analysis by Sun et al. [4]: pooled sensitivity and specificity of 18F-FDG PET or PET/CT were 99% (95%CI: 88–100%) and 95% (95%CI: 89–98%), respectively, confirming the excellent diagnostic performance of 18F-FDG PET/CT for distant metastasis staging in BC patients compared to conventional imaging.

Rong et al. [5] found that the pooled sensitivity and specificity of 18F-FDG PET/CT for detecting bone metastases of BC were 93% (95%CI: 82–98%) and 99% (95%CI: 95–100%), respectively. Compared with bone scintigraphy,

G. Treglia (\*)

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

Health Technology Assessment Unit, Academic Education, Research and Innovation Area, Ente Ospedaliero Cantonale, Bellinzona, Switzerland e-mail: giorgio.treglia@eoc.ch

18F-FDG PET/CT has higher sensitivity and accuracy for detection of bone metastases in BC patients.

#### **6.3 Restaging and Assessment of Response to Neoadjuvant Therapy**

Evangelista et al. [6] performed a meta-analysis on the use of tumour markers in BC patients as a guide for 18F-FDG PET imaging. The metaanalysis provided the following results: pooled sensitivity 87.8% (95%CI: 83.8–90.9%) and pooled specificity 69.3% (95%CI: 55.3–80.5%), confirming the role of 18F-FDG PET/CT in detecting metastases in the presence of a progressive increase of serum tumour markers in BC patients.

Xiao et al. [7] found that the pooled sensitivity and specificity of 18F-FDG PET or PET/CT in detecting BC recurrence were 90% (95%CI: 88–90%) and 81% (95%CI: 78–84), respectively. Therefore, 18F-FDG PET/CT is a valuable imaging method to detect relapse in suspected recurrent BC patients.

Several meta-analyses evaluated the usefulness of 18F-FDG PET/CT in predicting the response to neoadjuvant therapy in BC patients. According to Wang et al. [8], the pooled sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 18F-FDG PET/CT in this setting were 84% (95%CI: 78–88%), 66% (95%CI: 62–70%), 50% (95%CI: 44–55%) and 91% (95%CI: 87–94%), respectively. For regional lymph nodes, sensitivity and NPV of 18F-FDG PET/CT were 92% (95%CI: 83–97%) and 88% (95%CI: 76–95%), respectively. Overall, 18F-FDG PET/CT is useful to predict neoadjuvant therapy response in BC patients, but the relatively low specificity and PPV still call for caution. Cheng et al. [9] found similar results of 18F-FDG PET/CT in this setting reporting a pooled sensitivity and specificity of 84.7% (95%CI: 79.3–89.2%) and 66.1% (95%CI: 59.8– 72.0%), respectively, indicating that 18F-FDG PET/CT has reasonable sensitivity in evaluating response to neoadjuvant chemotherapy in BC, but the specificity is relatively low. Mghanga et al. [10] found that 18F-FDG PET has moderately high sensitivity (80.5%; 95%CI: 75.9– 84.5%) and specificity (78.8%; 95%CI: 74.1–83.0%) in early detection of responders from nonresponders, and it can be used for the evaluation of response to neoadjuvant chemotherapy in BC patients. Another meta-analysis [11] reported that the pooled sensitivity and specificity of 18F-FDG PET/CT in this setting were 81.9% (95%CI: 76.0–86.6%) and 79.3% (95%CI: 72.1–85.1%), respectively, confirming the moderate accuracy of 18F-FDG PET/CT in predicting neoadjuvant therapy response in BC patients.

Several meta-analyses compared 18F-FDG PET/CT and MRI for evaluation of treatment response to neoadjuvant chemotherapy (NAC) in BC patients. Liu et al. [12] reported that 18F-FDG PET/CT has a higher sensitivity and MRI has a higher specificity in assessing pathological complete response (pCR) after NAC in BC patients. The pooled sensitivity and specificity of 18F-FDG PET/CT were 86% (95%CI: 76–93%) and 72% (95%CI: 49–87%), respectively. Therefore, the combined use of these two imaging modalities may have great potential to improve the diagnostic performance in assessing pCR after NAC. Another meta-analysis [13] indicates that the timing of imaging for NAC-response assessment exerts a major influence on the estimates of diagnostic accuracy: 18F-FDG PET/CT outperformed MRI in intra-NAC assessment, whereas the overall performance of MRI was higher after completion of NAC, before surgery. The pooled estimates of sensitivity and specificity were 71% and 77% for 18F-FDG PET/CT and 88% and 55% for MRI, respectively. Chen et al. [14] found that the diagnostic performance of MRI is similar to that of 18F-FDG PET/CT for the assessment of BC response to NAC. For 18F-FDG PET/CT, the pooled sensitivity was 87% (95%CI: 71–95%) and pooled specificity was 85% (95%CI: 70–93%). For MRI, the pooled sensitivity was 79% (95%CI: 68–87%) and the pooled specificity was 82% (95%CI: 72–89%). However, 18F-FDG PET/CT is more sensitive than conventional contrast-enhanced MRI and more specific if the second imaging scan is performed before three cycles of NAC. Lastly, Li et al. [15] found that MRI had a higher sensitivity and 18F-FDG PET/CT had a higher specificity in predicting the pathologic response after NAC in patients with BC, with similar accuracy among the two methods. The pooled sensitivity and specificity of MRI were 88% (95%CI: 78–94%), and 69% (95%CI: 51–83%), respectively. The corresponding values for 18F-FDG PET/CT were 77% (95%CI: 58–90%) and 78% (95%CI: 63–88%), respectively.

#### **6.4 Prognostic Value**

Diao et al. [16] evaluated the prognostic value of maximum standardized uptake values (SUVmax) measured in the primary lesion and axillary lymph nodes (ALN) by pretreatment 18F-FDG PET or PET/CT in patients with BC. For eventfree survival (EFS), patients with higher SUVmax in primary tumour and ALN showed a poorer prognosis with pooled hazard ratio (HR) of 1.96 (95%CI: 1.40–2.73) and 1.89 (95%CI: 0.70– 5.07), respectively. In analysing invasive ductal carcinoma (IDC) patients, the pooled HR was 1.91 (95%CI: 1.40–2.64). For overall survival (OS), the pooled HR of SUVmax in primary lesion and ALN were 0.64 (95%CI: 0.23–1.84) and 1.09 (95%CI 0.07–16.53), respectively. Therefore, patients with BC and higher SUVmax in primary lesion or ALN may experience a higher risk for recurrence or a poor progression.

#### **6.5 Incidental 18F-FDG Uptake**

A meta-analysis calculated the prevalence and clinical significance of breast incidental 18F-FDG uptake (BIU) detected by PET or PET/CT in patients performing PET for other reasons than BC evaluation [17]. The pooled prevalence of BIU on all PET scans was 0.4% (95%CI: 0.23– 0.61%), the pooled prevalence on PET scans on female patients only was 0.82% (95%CI: 0.51– 1.2%), the pooled risk of malignancy of BIU when further evaluated was 48% (95%CI: 38–58%) and the pooled risk of malignancy of BIU with histological examination was 60% (95%CI: 53–66%). Despite being uncommon, the identification of BIU frequently signals the presence of an unsuspected subclinical lesion and the risk of malignancy is very high.

#### **6.6 18F-FDG Positron Emission Mammography**

The diagnostic performance of dedicated 18F-FDG positron emission mammography (PEM) in evaluating suspicious BC has been investigated by a meta-analytic study [18]: pooled sensitivity and specificity of 18F-FDG PEM in women with suspected breast malignancy were 85% (95%CI: 83–88%) and 79% (95%CI: 74–83%), respectively, on a per-lesion-based analysis. The detection of additional breast lesions and extensive intraductal involvement is improved by PEM, with comparable accuracy over that of MRI in the depiction of invasive BC.

#### **6.7 PET/MRI**

Lin et al. [19] performed a meta-analysis to assess the staging/restaging performance of hybrid 18F-FDG PET/MRI in BC patients. The pooled sensitivity and specificity of 18F-FDG PET/MRI for staging/restaging BC were 98% (95%CI: 95–99%) and 87% (95%CI: 76–95%), respectively, on a per-patient analysis and 91% (95%CI: 88–94%) and 95% (95%CI: 92–97%), respectively, on a per-lesion analysis. Overall, 18F-FDG PET/MRI has excellent diagnostic performance in staging/restaging BC patients.

#### **6.8 Other PET Tracers Beyond 18F-FDG**

Evangelista et al. [20] assessed the role of 18F-fluoroestradiol (18F-FES) PET in patients with BC. A pooled sensitivity of 82% (95%CI: 74–88%) and a pooled specificity of 95% (95%CI: 86–99%) for the evaluation of oestrogen receptor status in BC by 18F-FES PET were found, demonstrating a good accuracy of this method in this setting. Conversely, the pooled sensitivity and specificity of 18F-FES PET in predicting the response to hormonal therapy in patients with locally advanced or metastatic BC were unsatisfactory.

Deng et al. [21] evaluated the diagnostic performance of 18F-fluorothymidine (18F-FLT) PET and PET/CT for evaluating the response to chemotherapy in patients with BC. The pooled sensitivity and specificity of 18F-FLT PET in this setting were 77.3% (95%CI: 59.4– 90%) and 68.5% (95%CI: 47.9–84.9%), respectively, with a moderate diagnostic accuracy.

#### **References**


tomography/magnetic resonance imaging in breast cancer: a review and meta-analysis. Eur J Radiol. 2018;107:158–65.

20. Evangelista L, Guarneri V, Conte PF. 18F-Fluoroestradiol positron emission tomography in breast cancer patients: systematic review of the literature & meta-analysis. Curr Radiopharm. 2016;9(3):244–57.

21. Deng SM, Zhang W, Zhang B, Wu YW. Assessment of tumor response to chemotherapy in patients with breast cancer using (18)F-FLT: a meta-analysis. Chin J Cancer Res. 2014;26(5):517–24.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

**7**

## **Evidence-Based PET for Abdominal and Pelvic Tumours**

Salvatore Annunziata, Daniele Antonio Pizzuto, and Federica Galiandro

#### **7.1 Introduction**

Evidence-based data about the usefulness of positron emission tomography (PET) and hybrid imaging methods (PET/CT and PET/MRI) in abdominal and pelvic tumours have been collected and discussed in this chapter. These data were divided in three sections: (1) gastrointestinal tumours, (2) uro-genital tumours, (3) gynaecological tumours. Several pooled data (diagnostic and prognostic data), clinical settings (e.g. staging, restaging, treatment evaluation) and radiotracers as fluorine-18 fluorodeoxyglucose ( 18F-FDG), radiolabelled choline and prostatespecific membrane antigen (PSMA) were considered.

#### **7.2 PET in Gastrointestinal Tumours**

Fifty-two meta-analyses on the role of PET imaging in gastrointestinal tumours have been selected [1–52]. Pooled data about PET/CT in colorectal cancer, gastric cancer, anal cancer, stromal tumours, hepato-biliary tumours, liver metastases and pancreatic cancer have been reported in Table 7.1.

#### **7.2.1 Colorectal Cancer**

Fourteen meta-analyses about 18F-FDG PET/CT in colorectal cancer have been found [1–14]. Two meta-analyses evaluated the role of this imaging method in a staging setting, showing good specificity but low sensitivity [4, 13]. Similarly, two studies showed high accuracy in a restaging setting [7, 12]. Some meta-analyses assessed sub-optimal accuracy in treatment evaluation [3, 6, 8, 10, 14]. Recent meta-analyses found that focal colorectal incidental uptake at 18F-FDG PET/CT is observed in a not negligible number of patients who undergo this imaging method with a high risk of malignant or premalignant lesions [2, 9]. Finally, poor predictive or prognostic role of 18F-FDG PET/CT in colorectal cancer emerged [1, 5, 11].

S. Annunziata (\*)

Nuclear Medicine Unit, IRCCS Regina Elena National Cancer Institute, Rome, Italy

D. A. Pizzuto

Department of Nuclear Medicine, University Hospital Zurich and University of Zurich, Zurich, Switzerland

F. Galiandro

Department of Digestive Surgery, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy

<sup>©</sup> The Author(s) 2020 59 G. Treglia, L. Giovanella (eds.), *Evidence-based Positron Emission Tomography*, https://doi.org/10.1007/978-3-030-47701-1\_7


**Table 7.1** Main findings of meta-analyses about the role of PET imaging in gastrointestinal tumours



*HR* hazard ratio, *PFS* progression free survival, *OS* overall survival

#### **7.2.2 Gastric Cancer**

Seven meta-analyses analysed the role of 18F-FDG PET/CT in gastric cancer [15–21]. Three meta-analyses found a good accuracy in a staging setting, but with low sensitivity in detecting lymph nodal (N) involvement [15, 19, 21]. Conversely, other meta-analyses showed a good accuracy in a restaging setting [17, 18, 20]. Only one evidence-based article demonstrated a suboptimal prognostic value of 18F-FDG PET/CT in gastric cancer [16].

#### **7.2.3 Anal Cancer**

Five meta-analyses about 18F-FDG PET/CT in anal cancer have been included [22–26]. Some meta-analyses evaluated the role in a staging setting, with discordant accuracy values [24–26]. One meta-analysis found a strong prognostic power of 18F-FDG PET parameters for progression free survival (PFS) and overall survival (OS) [22]. Finally, another meta-analysis assessed the role of 18F-FDG PET/CT in radiotherapy planning [23].

#### **7.2.4 Stromal Tumours (GIST)**

Two meta-analyses about the role of 18F-FDG PET/CT in treatment evaluation and prediction of malignant potential in patients with GIST have been found and included [27, 28], suggesting a role of this imaging method in these settings.

#### **7.2.5 Hepato-biliary Tumours**

Nine meta-analyses about 18F-FDG PET/CT in hepatic and biliary tumours have been included [29–37]. Some meta-analyses found a role of 18F-FDG PET/CT in a staging setting, in particular about detection of distant metastases (M) [30, 32, 33, 35–37]. One meta-analysis found low sensitivity in a restaging setting [29]. Conversely, another meta-analysis showed high prognostic power for PFS by 18F-FDG PET/CT in hepatobiliary tumours [31]. Beyond 18F-FDG, radiolabelled choline PET/CT showed a good detection rate of tumour lesions in patients with hepatocellular carcinoma [34].

#### **7.2.6 Liver Metastases**

Seven meta-analyses about the role of 18F-FDG PET/CT in detecting liver metastases from different primary tumours have been found [38–44]. Some studies showed high specificity in a staging setting [38, 40, 41]. One study found a suboptimal sensitivity also in a restaging setting [43]. The role in treatment evaluation improved in a recent meta-analysis [42, 44].

#### **7.2.7 Pancreatic Cancer**

Eight meta-analyses about 18F-FDG PET/CT in pancreatic cancer have been published and included [45–52]. Interestingly, some papers showed good sensitivity in a staging setting [46, 48, 51, 52]. Two studies demonstrated a good accuracy of this imaging method in characterizing pancreatic lesions [49, 50]. Finally, two meta-analyses found a prognostic power for 18F-FDG PET/CT in pancreatic cancer [47, 51].

#### **7.3 PET in Gynaecological Tumours**

Thirty-three meta-analyses on the role of 18F-FDG PET imaging in gynaecological tumours have been selected [53–82]. Pooled data about 18F-FDG PET/CT in cervical cancer, endometrial cancer, ovarian cancer and peritoneal carcinomatosis have been reported in Table 7.2.

#### **7.3.1 Cervical Cancer**

Twelve meta-analyses about the role of 18F-FDG PET/CT in cervical cancer have been included [53–64]. Some studies evaluated the role of 18F-FDG PET/CT in staging cervical cancer, showing low sensitivity and high specificity in N staging [53, 56, 64]. Several studies evaluated the role of 18F-FDG PET/CT in a restaging setting, with high values of sensitivity and specificity [55, 58–

61, 63]. Some meta-analyses found a prognostic role of 18F-FDG PET/CT in cervical cancer [54, 57, 62].

#### **7.3.2 Endometrial Cancer**

Seven meta-analyses about the role of 18F-FDG PET/CT in endometrial cancer have been included [65–71]. Some meta-analyses evaluated the role of 18F-FDG PET/CT in a staging or restaging setting, showing good values of sensitivity and specificity [65, 68–71]. Conversely, one meta-analysis showed low sensitivity in N staging [71]. Finally, some meta-analyses showed a prognostic role of 18F-FDG PET/CT for PFS [66, 67].

#### **7.3.3 Ovarian Cancer**

Six meta-analyses about 18F-FDG PET/CT in ovarian cancer have been found [72–77]. Some meta-analyses showed high accuracy of this imaging method in a restaging setting [74–76]. Conversely, some meta-analyses showed sub-optimal sensitivity in N and M staging [72, 77]. Only one meta-analysis showed a good prognostic power of 18F-FDG PET/CT in ovarian cancer, with particular regard to OS [73].

#### **7.3.4 Peritoneal Carcinomatosis**

Three meta-analyses were focused on the role of 18F-FDG PET/CT in peritoneal carcinomatosis, showing good values of sensitivity and specificity of this method in this setting [78–80].

#### **7.3.5 PET/MRI**

Finally, recent studies evaluated the role of 18F-FDG PET/MRI in gynaecological malignancies, showing optimal diagnostic accuracy values [81, 82].


**Table 7.2** Main findings of meta-analyses about the role of PET imaging in gynaecological tumours

*HR* hazard ratio, *PFS* progression free survival, *OS* overall survival

#### **7.4 PET in Uro-genital Tumours**

Thirty-five meta-analyses on the role of PET imaging in uro-genital tumours have been selected [83–117]. In particular, pooled data about radiolabelled choline, PSMA and fluciclovine PET/CT in prostate cancer and 18F-FDG PET/CT in bladder cancer, renal cell carcinoma, testicular and penile cancer have been included (Table 7.3).

#### **7.4.1 Prostate Cancer**

#### **7.4.1.1 Radiolabelled Choline PET for Prostate Cancer**

Several meta-analyses described a very high specificity for detection of local lymph node involvement and for detection of distant metastases of prostate cancer by using radiolabelled choline PET. Radiolabelled choline PET is also widely used in patients with suspected biochemi-


**Table 7.3** Main findings of meta-analyses about the role of PET imaging in uro-genital tumours

*FDG* fluorodeoxyglucose, *PSMA* prostate-specific membrane antigen

cal relapse after initial treatments, even as a guide for salvage lymph node dissection [83–87]. Additionally, PSA kinetics was shown to be strongly related to the detection rate in patients undergoing radiolabelled choline PET [88]. Similarly, high PSA trigger was shown to be an important risk factor for positive findings of radiolabelled choline PET/CT [89]. PET with radiolabelled choline is a well-established imaging tool in clinical practice for detection of bone metastases [90, 91]. Diagnostic accuracy of radiolabelled choline PET was proven to be superior than other radiotracers as 18F-FDG and 11C-acetate [92], even if 11C-acetate PET could be considered in patients with relapse [93]. 18F-fluorocholine (FCH) PET showed higher specificity than 11C-choline PET [94]. Conversely, the choice of 18F-FCH or 11C-choline might not affect the detection of metastases in restaging patients after primary surgery and/or radiotherapy [95].

#### **7.4.1.2 Radiolabelled PSMA PET in Prostate Cancer**

Radiolabelled PSMA PET showed higher detection rate than other imaging modalities in prostate cancer [96, 97]. It was also proven to alter significantly the clinical management of these patients [98]. Diagnostic performance of PSMA PET was high for detection of node involvement in intermediate- and high-risk prostate cancer patients [99–101]. PSA kinetics may be predictor of radiolabelled PSMA PET positivity in patients with biochemical relapse [102]. PSMA detection rate ranged from 64% to 97% when PSA trigger was over 2 ng/ml at the time of scan [97].

#### **7.4.1.3 Fluciclovine PET in Prostate Cancer**

A meta-analysis demonstrated that fluciclovine ( 18F-FACBC) PET/CT had an 87% pooled sensitivity and 66% pooled specificity in detecting prostate cancer recurrence, being a useful imaging method in this setting [115].

#### **7.4.1.4 Incidental 18F-FDG Uptake in the Prostate**

A meta-analysis demonstrated that incidental 18F-FDG uptake in the prostate is observed in about 2% of 18F-FDG PET/CT scans performed in male patients carrying a significant risk of malignancy. Therefore, whenever this finding is detected further investigation is warranted to exclude malignancy [117].

#### **7.4.2 Bladder Cancer**

Several evidence-based articles were focused on the clinical usefulness of 18F-FDG PET/CT in patients with bladder cancer [103–109]. An overall sensitivity and specificity of 82% and 92% was reported, respectively [108]. Sensitivity and specificity were 90% and 100%, respectively, for primary staging, and 82% and 89%, respectively, for restaging [105]. For detection of node metastases, specificity was found high, whereas sensitivity was poor [103, 104, 106]. Additionally, detection of node involvement was assessed by other radiopharmaceuticals such as 11C-choline or 11C-acetate [109], showing low sensitivity and moderate specificity.

#### **7.4.3 Renal Cell Carcinoma**

Values of sensitivity and specificity of 18F-FDG PET/CT were 86% and 88%, respectively, for detection of recurrence [110]. If diagnostic performance of 18F-FDG PET/CT for detection of recurrent renal and extra-renal lesions was assessed separately, sensitivity and specificity of extra-renal lesions was found superior than accuracy for renal lesions [111].

#### **7.4.4 Testicular and Penile Cancer**

18F-FDG-PET sensitivity was non-optimal in the evaluation of patients with testicular cancer [112]. Similar results were drawn if PET was performed after chemotherapy treatment in patients with seminoma [113]. Clinical usefulness of 18F-FDG PET for detection of metastatic inguinal lymph nodes in patients with penile cancer is controversial [114].

#### **References**


rectal cancer to neoadjuvant chemoradiotherapy: evidence from meta-analysis and systematic review. Clin Nucl Med. 2016;41(5):371–5.


advanced rectal cancer: a meta-analysis. Int J Cancer. 2012;131(11):2604–11.


mance of positron emission tomography/computed tomography using fluorine-18 fluorodeoxyglucose in detecting locoregional nodal involvement in patients with anal canal cancer: a systematic review and meta-analysis. ScientificWorldJournal. 2014;2014:196068.


mary tumor in patients with cholangiocarcinoma: a meta-analysis. Biomed Res Int. 2014;2014:247693.


computed tomography and computed tomography in diagnosing recurrent cervical carcinoma: a systematic review and meta-analysis. Arch Med Sci. 2014;10(2):222–31.


mance of fluorine-18-fluorodeoxyglucose positron emission tomography imaging in uterine sarcomas: systematic review and meta-analysis of the literature. Int J Gynecol Cancer. 2013;23(8):1349–56.


of fluorine-18-fluorodeoxyglucose positron emission tomography in the postchemotherapy management of patients with seminoma: systematic review and meta-analysis. Biomed Res Int. 2014;2014:852681.


acid PET/CT in the diagnosis of recurrent prostate carcinoma: a meta-analysis. Acta Radiol. 2016;57(4):487–93.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Evidence-Based PET for Cutaneous, Musculoskeletal and Unknown Primary Tumours**

**8**

Luisa Knappe and Gaetano Paone

#### **8.1 Introduction**

PET/CT is extremely useful method in cutaneous and musculoskeletal tumours. In this chapter, the evidence from the literature concerning PET or PET/CT in melanoma, sarcomas, bone metastases, cancer of unknown primary (CUP) and paraneoplastic syndromes were analysed.

#### **8.2 PET in Malignant Melanoma**

#### **8.2.1 Introduction**

Melanoma is a highly malignant tumour having its origin in melanocytes from the epidermal skin layer. As prognosis is highly dependent on lymph node involvement and presence of distant metastases at the time of diagnosis, a precise staging is important for determining prognosis and choosing the fittest therapy for the patient [1–4].

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland e-mail: gaetano.paone@eoc.ch

#### **8.2.2 Staging**

18F-FDG PET/CT in staging melanoma has to be performed as a whole body protocol from head to toe to visualize the whole skin. Nonetheless, the usefulness of 18F-FDG PET/CT is limited in staging of tumour expansion and detection of satellite metastases. For the recognition of lymph nodal metastases, ultrasound and histological examination of the sentinel lymph node have a higher sensitivity and specificity than 18F-FDG PET/CT [5, 6]. Vural Topuz et al. showed that 18F-FDG PET/ CT is probably negative in the first year postsurgery if the sentinel lymph node biopsy was negative. Hence, performance of 18F-FDG PET/ CT is not recommended in early stage melanoma for not providing any significant clinical contribution [7].

In contrast, 18F-FDG PET/CT is well established for imaging of distant metastases. In a meta-analysis of nine studies, Rodriguez Rivera et al. found out usefulness of 18F-FDG PET/CT in staging of stage III melanoma having a high sensitivity (89.4%) and specificity (88.8%) [8]. Also Xing et al. valued 18F-FDG PET/CT as the most sensitive and specific method for detecting distant metastases [6]. 18F-FDG PET/CT is even superior to morphologic imaging and has replaced CT and magnetic resonance imaging (MRI) almost completely [1].

L. Knappe · G. Paone (\*)

<sup>©</sup> The Author(s) 2020 73

G. Treglia, L. Giovanella (eds.), *Evidence-based Positron Emission Tomography*, https://doi.org/10.1007/978-3-030-47701-1\_8

#### **8.2.3 Restaging and Treatment Monitoring**

The early detection of disease progression or recurrence has a huge impact on prognosis of melanoma. The usefulness of 18F-FDG PET/CT has been proven not only for the staging of advanced melanoma but also for the detection of recurrences showing a sensitivity of 96% and a specificity of 92% [7]. Accordingly, 18F-FDG PET/CT frequently leads to a change of treatment plan [6]. Due to a scarce number of prospective studies regarding use of 18F-FDG PET/ CT in melanoma, more studies are needed to find the most effective and cost-effective intervals in follow-up.

#### **8.3 PET in Sarcomas**

#### **8.3.1 Introduction**

Sarcomas are malignant tumours originating from mesenchymal cells. They are a relatively rare cancer and represent only 1% of all malignant tumours but extremely frequent in children. They can be divided in soft tissue, osseous and chondral sarcomas. Soft tissue sarcomas are a group of heterogeneous tumours as rhabdomyosarcoma, leiomyosarcoma, fibrosarcoma, liposarcoma, angiosarcoma, etc. and they are the fourth most common solid tumours in children. The bone sarcomas are the osteosarcoma and the Ewing sarcoma. Classical imaging methods for sarcomas are X-ray, CT for control of stability and MRI for the illustration of the expansion in soft tissues. Biopsy and histopathological examination ensure diagnosis [9, 10].

#### **8.3.2 Staging**

PET/CT offers the possibility of simultaneous acquisition of bone lesions and their expansion in soft tissue and is very useful for the staging of sarcomas due to its high sensitivity, specificity and accuracy. The performance of 18F-FDG PET/ CT in the initial staging provides information of the initial metabolism activity of the tumour. This is important for the follow-up and the evaluation of the therapy response [11]. About 18F-FDG PET/CT in sarcomas, according to evidencebased data, the values vary between 86 and 96% for sensitivity and from 80 to 96% for specificity [12–15]. In particular, this hybrid imaging method is very useful for detecting distant metastases, as osseous and lung metastases [12, 15]. Furthermore, 18F-FDG PET/CT might have a relevant impact on the development of treatment strategy plan [15]. Additional to the high diagnostic quality, 18F-FDG PET/CT has also a prognostic value in sarcomas. Chen et al. found out that semi-quantitative PET parameters showed a significant prognostic value for overall survival and thus are useful tools in identifying high-risk patients [16]. These findings were confirmed by other authors reporting that a high maximum standardized uptake value (SUVmax) may predict a significantly shorter overall survival period [17].

#### **8.3.3 Restaging and Treatment Monitoring**

18F-FDG PET/CT is a valuable method for detecting post-surgery recurrence in patients with sarcomas [12]. Liu et al. found a sensitivity of 92% and a specificity of 93% for the detection of recurrence in sarcoma [14]. Hongtao et al. reported that 18F-FDG PET/CT is valuable for predicting the histological response to chemotherapy as they found a response to neoadjuvant chemotherapy in osteosarcomas with a sensitivity of 73% and a specificity of 86% [18]. Muheremu et al. showed that 18F-FDG PET/CT assesses the efficacy of neoadjuvant therapy with a sensitivity and specificity of 79% and thus is a reliable imaging method not only in diagnosis but also in treatment control of osseous and soft tissue tumours [13]. Also Chen et al. valued posttreatment SUVmax as useful in monitoring therapy response [16]. Li et al. had similar results confirming that SUVmax before and after chemotherapy has effective prognostic significance for survival outcomes [19].

#### **8.4 PET for Bone Metastases**

#### **8.4.1 Introduction**

Bone metastases originate most frequently from breast cancer in women and from prostate carcinoma in men (each 60%) followed by lung carcinoma (25%), renal cell and thyroid carcinoma. Bone metastases can be differentiated in osteoblastic metastases which are typical for the prostate carcinoma, osteolytic metastases which occur in renal cell, thyroid or colon carcinomas and mixed osteoblastic and osteolytic metastases for example from breast or lung cancer. They are localized often in the spine (60%) but also in the pelvis, proximal femur and skull, rarely in distal bones. Symptoms are mainly pain, radicular symptoms if a spine metastasis causes nerve root compression and functional impairment. Furthermore, metastases can cause instability of the bone with the consequent risk of fracture [20].

#### **8.4.2 Detection of Bone Metastases**

The probably most frequently performed imaging method for osseous staging is the bone scintigraphy (BS). In the actual development, this method is being replaced by PET/CT with different tracers as 18F-FDG (which has the advantage to represent nearly all body districts) and 18F-Fluoride which is more osseous specific. In a meta-analysis, Liu et al. found a sensitivity of 93% and a specificity of 95% for 18F-fluoride PET/CT in the detection of bone metastases. This method showed significantly higher sensitivity and specificity compared to BS and thus a superior diagnostic performance in bone metastases detection and higher accuracy [21]. Shen et al. achieved similar results showing a sensitivity of 92% and a specificity of 93% for 18F-fluoride PET/CT. Compared with BS, it showed both higher sensitivity and specificity, whereas compared with 18F-FDG PET/CT it showed only a higher sensitivity but no significant difference in specificity. Consequently, the authors describe an excellent diagnostic capacity for the detection of bone metastases and advantages compared with BS and 18F-FDG PET/CT [22].

Duo et al. analysed the performance of PET/ CT with different tracers in comparison with gadolinium-enhanced MRI for detecting bone metastases: similar sensitivity and specificity values for each method were found and consequently an excellent diagnostic performance for the detection of bone metastases for both methods [23]. On the contrary, regarding only vertebral metastases, MRI showed a better performance than PET/CT both in sensitivity and specificity. This procedure outranged also all other imaging methods as CT or BS with tomographic acquisition (SPECT) [24].

Concerning the prognostic value of 18F-FDG PET/CT, Jeong et al. showed that patients with solid tumours and a lower level of 18F-FDG uptake in the bone marrow have a better event free and overall survival than patients with higher bone marrow 18F-FDG uptake and therefore propose to use the 18F-FDG uptake in the bone marrow for risk stratification of tumour progression [25].

#### **8.5 PET for Cancer of Unknown Primary (CUP) and Paraneoplastic Syndromes**

#### **8.5.1 Introduction**

Cancer of unknown primary (CUP) is a syndrome defined by the presence of a metastatic disease without identified primary tumour. In 2–5% of all malignant tumours, the localization of the primary tumour is unknown. CUP occurs in a heterogeneous group of cancers most frequently in malignant melanoma, neuroendocrine tumours, carcinoids, small cell lung carcinoma and head and neck cancer. Despite of modern imaging methods, CUP remains a challenge in clinical routine. As prognosis is rather poor, the identification of the primary tumour can be important to adjust therapy and improve survival [26].

Paraneoplastic syndromes arise from tumour secretion of hormones, peptides or cytokines or from immune cross-reactivity between malignant and normal tissues. Paraneoplastic syndromes may affect diverse organ systems, most notably the endocrine, neurologic, dermatologic, rheumatologic and hematologic systems. The most commonly associated malignancies include small cell lung cancer, breast cancer, gynaecologic tumours and hematologic malignancies. In some instances, the timely diagnosis of these conditions may lead to detection of an otherwise clinically occult tumour at an early and highly treatable stage [27].

#### **8.5.2 Impact of PET in Patients with CUP**

Since patients with CUP syndrome usually underwent a vast diagnostic procedure with negative results, the patients setting in which 18F-FDG PET/CT is performed with this question is highly selected. Consequently, the search for a primary tumour in CUP syndrome is more difficult than in other diseases. Nevertheless, Burglin et al. found a detection rate of unknown primary tumours in 41% of cases by using 18F-FDG PET/CT and they recommended an early use of 18F-FDG PET/CT to obviate too much useless diagnostic procedures [28].

#### **8.5.3 Impact of PET in Patients with Paraneoplastic Syndromes**

In patients with suspected paraneoplastic syndrome, 18F-FDG PET/CT showed a high accuracy for the detection of underlying malignancies with a sensitivity of 81% and a specificity of 88% [29]. Also in patients with paraneoplastic neurological syndrome, 18F-FDG PET/CT showed a high diagnostic performance with a detection rate of 15%, a sensitivity of 87% and specificity of 86% [30]. Generally, a heterogeneity in study design and diagnostic workup and the small number of patients in the available studies reduce interpretability of the data [29, 30].

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Evidence-Based PET for Haematological Tumours**

**9**

Francesco Bertagna, Raffaele Giubbini, and Domenico Albano

#### **9.1 Introduction**

Haematological malignancies include lymphomas such as Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL), leukaemia and multiple myeloma (MM). They can affect any organ system and positron emission tomography/ computed tomography (PET/CT) has been accepted as part of the routine management of most of them. In this chapter were only considered recent systematic reviews and meta-analyses concerning the use of PET or PET/CT with 18F-FDG in haematological malignancies dividing the results by the main areas of application.

#### **9.2 18F-FDG PET or PET/CT in Staging or Detection**

In staging HL and more aggressive NHL subtypes, 18F-FDG PET/CT was shown to be clearly more accurate than conventional radiological imaging to detect nodal and extranodal involvement. On the other hand, recent meta-analyses have addressed the diagnostic performance of this imaging method in some types of NHL, in MM and in the assessment of bone marrow involvement of HL and NHL [1–9].

#### **9.2.1 Post-transplant Lymphoproliferative Disorder**

Montes de Jesus et al. [1] evaluated the performance of advanced imaging modalities at diagnosis for post-transplant lymphoproliferative disorder (PTLD) after solid organ and haematopoietic stem cell transplantation. 18F-FDG PET/ CT was the primary imaging modality investigated. Subgroup analysis of imaging results for detection and staging in patients with PTLD indicated that 18F-FDG PET/CT identified additional lesions not detected by conventional imaging in 27.8% of cases, from which extranodal sites in 23.6%. False negative results occurred in 11.5% of cases, predominantly in physiological high background activity regions and in early PTLD lesions. False positive results occurred in 4.8% of cases, predominantly due to inflammatory conditions. They concluded that 18F-FDG PET/CT is currently the most frequently investigated imaging modality in PTLD patients with promising results in detection and staging, but available studies suffer from methodological shortcomings.

#### **9.2.2 Follicular Lymphoma**

Adams et al. [2] studied the additional value of 18F-FDG PET to CT for staging newly diagnosed follicular lymphoma (FL) in terms of Ann Arbor

F. Bertagna (\*) · R. Giubbini · D. Albano Nuclear Medicine, University of Brescia and Spedali Civili di Brescia, Brescia, Italy e-mail: francesco.bertagna@unibs.it

<sup>©</sup> The Author(s) 2020 79

G. Treglia, L. Giovanella (eds.), *Evidence-based Positron Emission Tomography*, https://doi.org/10.1007/978-3-030-47701-1\_9

staging and Follicular Lymphoma International Prognostic Index (FLIPI) risk stratification. The proportion of patients who were upstaged by 18F-FDG PET compared with CT ranged from 0 to 45.2%, with a pooled summary proportion of 18.7% (95% confidence interval (95%CI): 10.8– 30.4%). The single study that only included patients with CT-based limited non-bulky stage I to II disease reported 18F-FDG PET-induced upstaging in 40.5% of cases. No study reported data on the influence of 18F-FDG PET on FLIPI risk stratification. Although upstaging by 18F-FDG PET compared with CT occurs in a considerable proportion of patients, the available studies on this topic had numerous methodological errors. The authors concluded that future welldesigned studies are needed before 18F-FDG PET can be recommended for routine pre-treatment staging of FL.

#### **9.2.3 Marginal Zone Lymphoma of the Mucosa-Associated Lymphoid Tissue**

Treglia et al. [3] analysed the detection rate (DR) of 18F-FDG PET and PET/CT for the evaluation of patients with marginal zone lymphoma of the mucosa-associated lymphoid tissue (MALT). The pooled DR of 18F-FDG PET or PET/CT was 71% (95%CI: 61–80%). A significant difference between the DR of PET/ CT (69%; 95%CI: 61–80%) and that of PET alone (73%; 95%CI: 60–84%) was not demonstrated. A better DR of 18F-FDG PET or PET/ CT in bronchial (94%; 95%CI: 85–99%) and head-and-neck (90%; 95%CI: 78–98%) MALT lymphomas compared with gastric (62%; 95%CI: 46–77%) and ocular (49%; 95%CI: 36–63%) MALT lymphomas was found. This meta-analysis demonstrated that MALT lymphoma is an 18F-FDG-avid tumour in most of the cases, suggesting a potential clinical role in the initial evaluation of these patients. In particular, the DR of 18F-FDG PET or PET/CT is related to the primary site of the MALT lymphoma.

#### **9.2.4 Bone Marrow Involvement in Lymphoma**

Adams et al. [4] analysed the diagnostic performance of 18F-FDG PET/CT in detecting bone marrow involvement (BMI) in patients with newly diagnosed diffuse large B-cell lymphoma (DLBCL). The pooled sensitivity and specificity of 18F-FDG PET/CT for detecting BMI were 88.7% (95%CI: 82.5–93.3%) and 99.8% (95%CI: 98.8–100%), respectively. The area under the summary ROC curve was 0.9983. They concluded that 18F-FDG PET/CT is accurate and complementary to bone marrow biopsy (BMB) for detecting BMI in patients with newly diagnosed DLBCL. A negative 18F-FDG PET/CT cannot rule out the presence of BMI, but positive 18F-FDG PET/CT findings obviate the need for BMB for the detection of BMI in these patients.

The same group of authors systematically reviewed and meta-analysed published data on the diagnostic performance of 18F-FDG PET/CT in detecting BMI in newly diagnosed HL to assess whether 18F-FDG PET/CT can replace blind BMB in these patients [5]. The pooled sensitivity and specificity of 18F-FDG PET/CT for the detection of BMI range were 96.9% (95%CI: 93–99%) and 99.7% (95%CI: 98.9–100%), respectively. The area under the ROC curve was 0.986. In conclusion, although the methodological quality of studies that were included in this systematic review and meta-analysis was moderate, the meta-analysis suggests that 18F-FDG PET/CT may be an appropriate method to replace BMB in newly diagnosed HL.

Cheng et al. [6] also carried out a metaanalysis to evaluate the performance of 18F-FDG PET and PET/CT against BMB in the initial diagnosis of BMI in patients with HL. Both 18F-FDG PET and BMB had excellent specificity in detecting BMI. However, 18F-FDG PET had excellent pooled sensitivity (94.5%; 95%CI: 89.0–97.8%) in detecting BMI in the initial staging of HL patients, whereas the pooled sensitivity of iliac BMB was very poor (39.4%; 95%CI: 30.8–48.4%). The authors concluded that 18F-FDG PET significantly outperforms iliac BMB in the detection of BMI in the initial staging of HL patients and therefore should be used as a firstline study.

#### **9.2.5 Natural Killer/T-Cell Lymphoma**

Ji et al. [7] evaluated the values of 18F-FDG PET/ CT and PET in diagnosing extranodal nasal type natural killer/T-cell lymphoma (ENKTL). Pooled sensitivity, specificity and area under the curve (AUC) of 18F-FDG PET/CT for diagnosing ENKTL were 97% (95%CI: 93–99%), 97% (95%CI: 88–99%) and 0.99 (95%CI: 0.98–1.00). The same parameters for 18F-FDG PET were 81% (95%CI: 70–89%), 90% (95%CI: 66–98) and 0.86 (95%CI: 0.82–0.89), respectively. The authors concluded that in comparison with PET, 18F-FDG PET/CT had excellent diagnostic value in detecting and staging ENKTL.

Zhou et al. [8] evaluated the role of 18F-FDG PET/CT in the diagnosis and staging of natural killer/T-cell lymphoma (NKTL). On a patientbased analysis, the pooled sensitivity and specificity of 18F-FDG PET/CT in the diagnosis of NKTL were 95% (95%CI: 89–98%) and 40% (95%CI: 9–78%), respectively. For lesion-based analysis, the pooled sensitivity and specificity of 18F-FDG PET/CT in the staging of NKTL were 98% (95%CI: 96–99%) and 99% (95%CI: 99–100), respectively. The results indicated that 18F-FDG PET/CT could be used as a valuable diagnostic and staging tool for NKTL.

#### **9.2.6 Multiple Myeloma**

Lu et al. [9] conducted a systematic review and meta-analysis to evaluate the diagnostic accuracy of 18F-FDG PET or PET/CT for intramedullary and extramedullary lesions in MM. The pooled sensitivity and specificity of 18F-FDG PET or PET/CT for the detection of extramedullary lesions in MM were 96.0% (95%CI: 79.6–99.9%) and 77.8% (95%CI: 40–97.2%), respectively. The pooled sensitivity and specificity of 18F-FDG PET or PET/CT for the detection of intramedullary lesions in MM were 61.1% (95%CI: 43.5– 76.9%) and 94.1% (95%CI: 71.3–99.9%), respectively. They concluded that whole-body 18F-FDG PET or PET/CT is a valuable imaging tool for the assessment of patients with MM, especially for the appraisal of extramedullary involvement.

#### **9.3 18F-FDG PET or PET/CT in Treatment Response Evaluation (Interim and/or End of Therapy)**

#### **9.3.1 Post-transplant Lymphoproliferative Disorder**

In the meta-analysis by Montes de Jesus et al. [1] on imaging modalities in PTLD, the subgroup analysis of imaging results at treatment response evaluation indicated that 18F-FDG PET/CT findings altered or guided treatment in 29% of cases. False positive results during treatment response evaluation were reported in 20% of cases, predominantly due to inflammatory conditions. They concluded that 18F-FDG PET/CT may be promising in therapy evaluation but suffers from methodological shortcomings. Concerns remain with regard to occurrence of false negatives due to physiological high background activity and early PTLD lesions as well as false positives due to inflammatory conditions.

#### **9.3.2 Hodgkin and Non-Hodgkin Lymphomas**

Adams et al. [10] systematically reviewed and meta-analysed the proportion of false positive lesions at interim and end-of-treatment 18F-FDG PET in lymphomas (both HL and NHL) using biopsy as reference standard. The pooled proportion of false positive results among all biopsied 18F-FDG-avid lesions at PET performed during or after completion of treatment was 55.7% (95%CI: 32.6–76.6%). The pooled false positive proportions were 83% (95%CI: 72–90.2%) for interim 18F-FDG PET in NHL, 23.1% (95%CI: 4.7–64.5%) for end-of-treatment 18F-FDG PET in HL and 31.5% (95%CI: 3.9–83.9%) for end-of-treatment 18F-FDG PET in NHL. The authors concluded that both interim and end-of-treatment 18F-FDG PET in patients with lymphoma suffer from a very high number of false positive findings.

Sun et al. [11] conducted a meta-analysis to evaluate the predictive value of interim 18F-FDG PET/CT in patients with DLBCL treated with R-CHOP chemotherapy. The pooled sensitivity of interim 18F-FDG PET/CT was 52.4% and the pooled specificity was 67.8%. In conclusion, the sensitivity and specificity of interim 18F-FDG PET/CT in predicting the outcome of DLBCL patients treated with R-CHOP chemotherapy were not satisfactory. To improve this, some more work should be done to unify the response criteria and some more research to assess the prognostic value of interim 18F-FDG PET/CT with semi-quantitative analysis.

Ziakas et al. [12] assessed the diagnostic performance of interim 18F-FDG PET with regard to the final outcome of adult patients with newly diagnosed HL. The pooled sensitivity was 67% (95%CI: 57–76%) and pooled specificity was 89% (95%CI: 84–93%). The estimated negative predictive value was 93% (95%CI: 85–100%). The diagnostic performance was influenced by most covariates tested, including age, duration of follow-up, criteria used and time of interim PET. In conclusion, the use of a PET-positive study as a surrogate marker was hampered by inconsistent interpretation criteria and study populations. However, the high negative predictive value may permit treatment stratification based on a negative outcome.

#### **9.4 18F-FDG PET or PET/CT in Prognosis/Outcome Evaluation**

#### **9.4.1 Hodgkin and Non-Hodgkin Lymphomas**

Wang et al. [13] carried out a meta-analysis to detect the prognostic power of 18F-FDG PET in the evaluation of pre-stem cell transplantation (SCT) and post-SCT in HL and NHL. For the pre-SCT PET or PET/CT, the combined hazard ratios (HRs) of PET for progression-free survival (PFS) and overall survival (OS) were 2.32 and 2.64, respectively. Subgroup analysis showed that the HRs of PFS for HL and NHL were 3.28 and 2.0, respectively. For the post-SCT PET scan, the combined HR for PFS was 4.61. The authors found that 18F-FDG PET was especially effective in predicting pre-STC and post-STC prognosis. The patients with a negative PET scan had a better prognosis compared with those with a positive scan for PFS and OS. In the subgroup analysis, 18F-FDG PET had a higher value in predicting prognosis before SCT for HL patients.

Burggraaff et al. [14] aimed to assess the predictive value of visually assessed interim 18F-FDG PET on PFS or event-free survival (EFS) in DLBCL patients treated with first-line immunochemotherapy regimens. The pooled HR was 3.13 (95%CI 2.52–3.89) with a 95% prediction interval of 1.68–5.83. The negative predictive value for progression generally exceeded 80%, but sensitivity, specificity and positive predictive values ranged widely. These findings showed that interim 18F-FDG PET has predictive value in DLBCL patients. Some diagnostic test characteristics were not satisfactory, especially the positive predictive value should be improved before a successful risk stratified treatment approach can be implemented in clinical practice.

Adams et al. [15] systematically reviewed and meta-analysed the prognostic value of complete remission status at end-of-treatment 18F-FDG-PET in DLBCL patients treated with R-CHOP. The disease relapse rate among all patients with complete remission status according to end-of-treatment 18F-FDG PET ranged from 7 to 20%, with a weighted summary proportion of 13.7%. In conclusion, a non-negligible proportion of R-CHOP-treated DLBCL patients who achieve complete remission according to end-of-treatment 18F-FDG PET experiences disease relapse during follow-up.

Adams et al. [16] analysed the prognostic value of interim 18F-FDG PET in DLBCL patients treated with R-CHOP. At multivariable analysis, two studies reported interim 18F-FDG PET to have independent prognostic value in addition to the International Prognostic Index (IPI) in predicting treatment failure, whereas three studies reported that this was not the case. One study reported interim 18F-FDG PET to have independent prognostic value in addition to the IPI in predicting death, whereas two studies reported that this was not the case. In conclusion, interim 18F-FDG-PET in R-CHOP-treated DLBCL has some correlation with outcome, but its prognostic value is homogeneously sub-optimal across studies and it has not consistently proven to surpass the prognostic potential of the IPI. Therefore, at present there is no scientific base to support the clinical use of interim 18F-FDG-PET in R-CHOPtreated DLBCL.

Zhu et al. [17] analysed the prognostic value of interim 18F-FDG PET in DLBCL patients treated with rituximab-based immunochemotherapy. The pooled HR comparing PFS between patients with positive and negative results was 2.96 (95%CI = 2.25–3.89). The patients in interim 18F-FDG PET-negative group had a higher complete response (CR) rates than those in interim 18F-FDG PET-positive group (relative risk = 5.53, 95%CI = 2.59–11.8). The authors concluded that consistent evidence favouring interim 18F-FDG PET-based treatment assessment should be considered in the management of patients with DLBCL.

Pyo et al. [18] evaluated post-chemotherapy response assessment in FL. The pooled HR of end-therapy 18F-FDG PET and CT were 5.1 (95%CI: 3.7–7.2) and 2.6 (95%CI: 1.2–5.8), respectively, which implies that PET is more predictive of PFS after chemotherapy than CT. The pooled CR rates of PET- and CT-based response criteria were 75% (95%CI: 70–79%) and 63% (95%CI: 53–73%), respectively, which implies that PET is more efficient in distinguishing CR from other states with residual disease. The authors concluded that PET-based treatment assessment should be considered in the management of patients with FL.

Liao et al. [19] evaluated the prognostic value of 18F-FDG PET/CT visual interpretation in patients with aggressive NHL. PFS and OS of PET/CT-positive patients were significantly lower when determined by the visual method. In subgroup analysis, International Harmonization Project (IHP), Deauville criteria, and having no standard interpretation groups were factors able to predict PFS; IHP and having no standard interpretation group were able to predict OS. With PET/CT, IHP and Deauville 5-point criteria, the PFS of patients receiving 2–4 cycles of chemotherapy before PET/CT was significantly lower than that of PET/CT-negative patients. No significant difference in OS was observed when patients received 3 or fewer cycles of chemotherapy before PET/CT, though OS was significantly lower in patients receiving more than 3 chemotherapy cycles. They concluded that interim PET/ CT analysis after 3–4 chemotherapy cycles is capable of predicting disease prognosis in aggressive NHL.

Adams et al. [20] aimed to analyse the value of pretransplant 18F-FDG PET in predicting outcome after autologous stem cell transplantation in aggressive NHL. Pooled sensitivity and specificity of 18F-FDG PET were 54 and 73.1% in predicting treatment failure, and 54.5 and 68.7% in predicting death. They concluded that pretransplant 18F-FDG PET cannot be recommended in aggressive NHL, because available studies suffer from major methodological flaws, and reported prognostic estimates are low.

Zhu et al. [21] aimed to determine the prognostic value of interim and final 18F-FDG PET in NHL patients treated with rituximab-containing chemotherapy. The combined HRs of interim PET for PFS and OS in DLBCL were 4.4 (*p* = 0.11) and 3.99 (*p* = 0.46), respectively. The combined HRs of final PET for PFS and OS in DLBCL were 5.91 (*p* = 0.39) and 6.75 (*p* = 0.92), respectively. Regarding non-DLBCL with final PET, the combined HRs of final PET for PFS and OS were 4.05 (*p* = 0.79) and 5.1 (*p* = 0.51), respectively. In conclusion, in DLBCL, both interim and final PET can be performed for survival and progression analysis. But in other NHL, it would be necessary to perform final PET for predictive purposes.

Adams et al. [22] aimed to systematically review and meta-analyse the value of interim 18F-FDG PET in predicting treatment failure in HL. The area under the summary ROC curve was 0.877. Pooled sensitivity and specificity were 70.8% (95%CI: 64.7–76.4%) and 89.9% (95%CI: 88–91.6%). The overall prognostic value of interim PET appeared to be moderate for excluding and relatively high for identifying treatment failure in HL. However, they stated that interim PET cannot yet be implemented in routine clinical practice due to moderate-quality evidence and inter-study heterogeneity that cannot be fully explained yet.

Sickinger et al. [23, 24] assessed the effects of interim 18F-FDG PET treatment modification in individuals with HL. PFS was shorter in participants with PET-adapted therapy (without radiotherapy) than in those receiving standard treatment with radiotherapy (HR: 2.38; 95%CI: 1.62–3.5). This difference was also apparent in comparisons of participants receiving no additional radiotherapy (PET-adapted therapy) versus radiotherapy (HR: 1.86 (95%CI: 1.07–3.23) and in those receiving chemotherapy but no radiotherapy (PET-adapted therapy) versus standard radiotherapy (HR: 3.0; 95%CI: 1.75–5.14). Overall, this systematic review found moderatequality evidence that PFS was shorter in individuals with early-stage HL and a negative PET receiving chemotherapy only (PET-adapted therapy) than in those receiving additional radiotherapy (standard therapy). It was still uncertain whether PET-positive individuals benefit from PET-based treatment adaptation and the effect of such an approach in those with advanced HL.

Adams et al. [25] aimed to systematically review the prognostic value of pretransplant 18F-FDG PET in refractory/relapsed HL treated with autologous stem cell transplantation (SCT). Pooled sensitivity and specificity of pretransplant 18F-FDG PET in predicting treatment failure (i.e. either progressive, residual, or relapsed disease) were 67.2% (95%CI: 58.2–75.3%) and 70.7% (95%CI: 64.2–76.5%), respectively. Pooled sensitivity and specificity of pretransplant 18F-FDG PET in predicting death during follow-up were 74.4% (95%CI: 58.8–86.5%) and 58% (95%CI: 49.3–66.3%), respectively. In conclusion, the moderate quality of evidence suggested pretransplant 18F-FDG-PET to have value in predicting outcome in refractory/relapsed HL patients treated with autologous SCT. Nevertheless, a considerable proportion of pretransplant 18F-FDG PET-positive patients remained disease free and a considerable proportion of pretransplant 18F-FDG PET-negative patients developed disease relapse after autologous SCT.

Adams et al. [26] systematically reviewed and meta-analysed the outcome of HL patients with a post-treatment 18F-FDG PET-negative residual mass. The disease relapse rate in HL patients with a 18F-FDG PET-negative residual mass after first-line therapy was approximately 6.8%. They concluded that the presence of a non-18F-FDGavid residual mass has not been proven yet to be associated with a worse outcome than a posttreatment 18F-FDG-PET-based complete remission status without a residual mass.

The same group [27] analysed the prognostic value of complete remission status at 18F-FDG PET in HL after completion of first-line therapy. The pooled disease relapse rate during follow-up among all patients with complete remission status at end-of-treatment 18F-FDG-PET was 7.5% (95%CI: 3.9–13.8%). They concluded that, although the disease relapse rate in HL patients who achieve an 18F-FDG PET-based complete remission after first-line therapy is low, it is actually high when considering the generally favourable outcome of HL.

#### **9.4.2 Multiple Myeloma**

Caldarella et al. [28] aimed to evaluate the usefulness of 18F-FDG PET or PET/CT in monitoring response to treatment in patients with MM. Based on the findings from the literature, 18F-FDG PET or PET/CT appeared to be useful in the assessment of treatment response in patients with MM. In particular, PET or PET/CT could detect the response to treatment earlier than other imaging. Negative findings on posttreatment 18F-FDG PET or PET/CT were mostly correlated with complete clinical and histological remission or, at least, low risk of recurrences or disease progression. Persistence of metabolically active lesions was related to shorter overall and event-free survival. Therefore, post-treatment 18F-FDG PET findings could be of higher prognostic significance than standard response monitoring methods. In the near future, 18F-FDG PET or PET/CT will be used even more in the assessment of metabolic response after treatment in patients with MM, as a guidance for clinical decision and to eventually decide for alternative therapies in non-responding patients.

#### **9.5 Prognostic Role of Semiquantitative PET Parameters**

Guo et al. [29] have analysed whether baseline metabolic tumour volume (MTV) and total lesion glycolysis (TLG) measured by 18F-FDG PET/CT affect prognosis of patients with lymphoma. Patients with high baseline MTV showed a worse prognosis considering PFS and OS as well as patients with high baseline TLG. A high baseline MTV was significantly associated with worse survival in DLBCL patients treated with R-CHOP as well as a high baseline TLG. The negative effect of high baseline MTV on PFS was demonstrated in HL. A high baseline MTV was significantly associated with worse survival in ENKTL patients. A high baseline TLG was significantly associated with worse survival in ENKL patients. The authors concluded that high baseline MTV or TLG predict significantly worse PFS and OS in patients with lymphoma.

Wang et al. [30] evaluated the prognostic value of maximum standardized uptake value (SUVmax), MTV, and TLG of baseline, interim and end-of-treatment 18F-FDG PET/CT parameters in ENKTL. SUVmax, MTV and TLG on baseline PET/CT were significantly associated with PFS and with OS. For the delta SUV (DS) on interim PET/CT, the HRs for PFS and OS were 5.15 (95%CI 2.71–9.80) and 5.8 (95%CI 2.28–14.73), respectively. Similarly, the DS on end-of-treatment PET/CT was a significant predictor of PFS and OS with HRs of 3.65 (95%CI: 2.13–6.26) and 3.32 (95%CI: 1.79–6.15), respectively. They suggested that SUVmax, MTV, TLG on baseline PET/CT, DS on interim PET/CT and DS on end-of-treatment PET/CT may be significant prognostic indicators for PFS and OS in ENKTL patients. Moreover, TLG tended to be superior to SUVmax and MTV on baseline PET/ CT for predicting survival of ENKTL patients. Therefore, response monitoring and prognostication assessments based on multiple PET/CT parameters should be considered in the management of ENKTL patients.

Xie et al. [31] analysed whether SUVmax, MTV and TLG acquired from 18F-FDG PET/CT are predictors of prognosis of DLBCL. Combined results suggested a strong link between the high SUVmax, MTV and TLG values and the poor 3-year PFS with ORs of 2.59, 3.69 and 2.29, respectively. Similarly, high MTV and TLG values unfavourably influenced the 3-year OS with ORs of 5.40 and 2.19, respectively. The pooled results also showed that high SUVmax and MTV were negative predictors of PFS. The TLG value was not predictive of PFS. And for OS, only high MTV was a strong predictor of poor prognosis in DLBCL. Their results suggested that SUVmax and MTV may be significant prognostic markers for PFS and MTV may be the only predictor for OS in DLBCL.

#### **9.6 18F-FDG PET or PET/CT in Comparison with Magnetic Resonance Imaging**

Wang et al. [32] aimed to compare the performance of whole-body magnetic resonance imaging (WB-MRI) with that of 18F-FDG PET/CT for lesion detection and initial staging in patients with aggressive or indolent lymphoma. In terms of staging, the pooled accuracy of WB-MRI and 18F-FDG PET/CT for HL and aggressive NHL were 98% (95%CI: 94–100%) and 98% (95%CI: 94–100%), respectively. The pooled accuracy of 18F-FDG PET/CT dropped to 87% (95%CI: 72–97%) for staging in patients with indolent lymphoma, whereas that of WB-MRI remained 96% (95%CI: 91–100%). Subgroup analysis indicated an even lower accuracy of 18F-FDG PET/CT for staging of less 18F-FDG-avid indolent NHLs (60%; 95%CI: 23–92%), in contrast to the superior performance of WB-MRI (98%; 95%CI: 88–100%). The authors concluded that WB-MRI is a promising radiation-free imaging technique that may serve as a viable alternative to 18F-FDG PET/CT for staging of 18F-FDG-avid lymphomas, where 18F-FDG PET/CT remains the standard of care. Additionally, WB-MRI seemed a less histology-dependent functional imaging test than 18F-FDG PET/CT and may be the imaging test of choice for staging of indolent NHLs with low 18F-FDG avidity.

Regacini et al. [33] aimed to compare WB-MRI with 18F-FDG PET/CT for lymphoma staging. WB-MRI and 18F-FDG-PET/CT agreed in 90.5% of the cases. In most of the studies, when there was disagreement between the methods, WB-MRI overstaged in relation to 18F-FDG PET/CT. The sensitivity of WB-MRI and 18F-FDG PET/CT, in comparison with the clinical-radiological standard, ranged from 59 to 100% and from 63 to 100%, respectively. The authors concluded that WB-MRI has excellent agreement with 18F-FDG-PET/CT and is a great alternative for managing lymphoma patients, without using ionizing radiation or an intravenous contrast agent.

Gariani et al. [34] evaluated the diagnostic performance of WB-MRI including diffusion weighted sequences (DWI) compared to wholebody CT or 18F-FDG PET/CT in patients with MM. WB-MRI detected more lesions than 18F-FDG PET/CT (sensitivity 68–100% versus 47–100%) but was less specific (specificity 37–83% versus 62–85.7%). Despite these insights the authors concluded that, because of the heterogeneity of the studies, future prospective trials should assess the impact of WB-MRI on management of MM.

Weng et al. [35] conducted a systematic review of the published literature to evaluate the diagnostic accuracy of 18F-FDG PET, 18F-FDG PET/CT, MRI and scintigraphy for MM-related bone disease. For 18F-FDG PET and PET/CT, pooled sensitivity and specificity were 91% and 69%, respectively. Statistically significant differences were not found in the sensitivity and specificity between MRI, scintigraphy, 18F-FDG-PET and PET/CT. In conclusion, the authors suggested that 18F-FDG-PET, PET/CT, MRI and scintigraphy are all associated with high detection rate of bone disease in patients with MM. Thus, in clinical practice, it is recommended that we could choose these tests according to the condition of the patient.

#### **9.7 Conclusions**

Overall, 18F-FDG PET or PET/CT appears to be a useful and accurate diagnostic tool for haematological malignancies in clinical practice from an "evidence-based" point of view. Some topics and results need further investigations in order to overcome methodological limits and clarify the real diagnostic role of this tool and its more appropriate position in the diagnostic flow chart.

#### **References**


medullary and extramedullary lesions in multiple myeloma: a systematic review and meta-analysis. Clin Nucl Med. 2012;37:833–7.


tomography (WBCT) or 18F-fluorodeoxyglucose positron emission tomography/CT (18F-FDG PET/ CT) in patients with myeloma: systematic review of diagnostic performance. Crit Rev Oncol Hematol. 2018;124:66–72.

35. Weng WW, Dong MJ, Zhang J, Yang J, Xu Q, Zhu YJ, et al. A systematic review of MRI, scintigraphy, FDG-PET and PET/CT for diagnosis of multiple myeloma related bone disease—which is best? Asian Pac J Cancer Prev. 2014;15:9879–84.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Evidence-Based PET for Endocrine Tumours and Disorders**

**10**

Alexander Stephan Kroiss and Giorgio Treglia

#### **10.1 Introduction**

Endocrine tumours have a wide range of clinical presentations and can be found anywhere from the neck to the pelvis. Diagnostic imaging is crucial to predict the exact tumour extent, foremost in metastatic disease. Anatomical imaging as computed tomography (CT) and magnetic resonance imaging (MRI) serves as the first-line modality in the locoregional staging of these tumours. Compared with anatomical imaging, PET shows both high sensitivity and specificity. Several meta-analyses have described the diagnostic performance of positron emission tomography (PET) and hybrid imaging (PET/CT) in endocrine tumours and disorders.

#### **10.2 Adrenal Tumours and Paragangliomas**

#### **10.2.1 Characterization of Adrenal Masses**

Dinnes and colleagues reviewed the evidence on the accuracy of imaging tests for differentiating malignant from benign adrenal masses. They concluded that CT density >10 Hounsfield Unit (HU) offers high sensitivity for detection of adrenal malignancy in participants with no prior indication for adrenal imaging. With respect to a limited database and heterogeneity and low quality of included studies for meta-analysis, the authors concluded that there is insufficient evidence for the diagnostic value of individual imaging tests in distinguishing benign from malignant adrenal masses [1].

Kim and colleagues explored the role of the diagnostic accuracy of 18F-FDG PET or PET/CT for characterization of adrenal lesions [2]. The pooled sensitivity for 18F-FDG PET or PET/CT was 91% (95% confidence interval (95%CI): 88–94%) and the pooled specificity was 91% (95%CI: 87–93%). Although, at present, the literature regarding the use of 18F-FDG PET or PET/CT for the characterization of adrenal masses remains limited, 18F-FDG PET or PET/ CT demonstrated good sensitivity and specificity for the characterization of adrenal masses.

A. S. Kroiss (\*)

Department of Nuclear Medicine, Medical University Innsbruck, Innsbruck, Austria e-mail: alexander.kroiss@i-med.ac.at

G. Treglia

Clinic of Nuclear Medicine and PET/CT Center, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

Health Technology Assessment Unit, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland e-mail: giorgio.treglia@eoc.ch

#### **10.2.2 Paragangliomas**

Rufini and co-authors compared the diagnostic performance of metaiodobenzylguanidine (MIBG scintigraphy) and PET with different radiopharmaceuticals in patients with paraganglioma (PGL). The authors concluded that the diagnostic performance of PET with different radiopharmaceuticals is clearly superior to that of MIBG scintigraphy in patients with PGL, mainly for familial, extra-adrenal and metastatic diseases [3].

A review article by Treglia and co-authors investigated the diagnostic performance of 18F-DOPA PET in patients with paraganglioma (PGL). The pooled sensitivity of 18F-DOPA PET and PET/CT in detecting PGL was 91% (95%CI: 87–94%) on a per-patient-based analysis and 79% (95%CI 76–81%) on a per-lesionbased analysis. The pooled specificity of 18F-DOPA PET and PET/CT in detecting PGL was 95% (95%CI: 86–99%) on a per-patientbased analysis and 95% (95%CI: 84–99%) on a per-lesion-based analysis. The area under the receiver operating characteristic (ROC) curve was 0.95 on a per-patient- and 0.94 on a perlesion-based analysis. The authors described the possible risk of false-negative 18F-DOPA PET results in metastatic PGL, besides the fact that succinate dehydrogenase subunit B (SDHB) gene mutations could influence the diagnostic performance of 18F-DOPA PET or PET/CT [4].

Kan and colleagues performed a meta-analysis on the localization of metastatic pheochromocytoma (PHEO) and PGL with germline mutations, comparing 68Ga-somatostatin analogues and 18F-FDG PET/CT. The pooled sensitivity of 68Ga peptides and 18F-FDG PET were 95% (95%CI: 92–97) and 85% (95%CI: 78–91%), respectively. The pooled specificity of 68Ga peptides and 18F-FDG PET were 87% (95%CI: 63–96%) and 55% (95%CI:37–73%), respectively. The authors concluded that 68Ga-somatostatin analogues PET/CT demonstrated good performance in the localization of metastatic PGL, especially those with germline mutations, compared to 18F-FDG PET/CT [5].

Han and colleagues performed a systematic review and meta-analysis on the performance of 68Ga-somatostatin analogues PET in the detection of PGL. The pooled detection rate was 93% (95%CI: 91–95%), which was significantly higher than that of 18F-DOPA PET (80%; 95%CI: 69–88%), 18F-FDG PET (74%; 95%CI: 46–91%) and MIBG scintigraphy (38%; 95%CI: 20–59%). A greater prevalence of head and neck PGL was associated with higher detection rates of 68Ga-somatostatin analogues PET. The authors suggest the use of 68Ga-somatostatin analogues PET as a first-line imaging modality for the primary staging or restaging of PGL with unknown genetic status [6].

#### **10.3 Neuroblastoma**

Bleeker and colleagues described the role of MIBG scintigraphy and 18F-FDG PET for diagnosing neuroblastoma (NB). In one study, the sensitivity of 18F-FDG PET/CT compared to MIBG scintigraphy was 100% and 92%, respectively. Specificity could not be calculated for both modalities. The diagnostic accuracy of 18F-FDG PET/CT imaging in case of a negative 123I-MIBG scintigraphy could not be calculated because of very limited data. It has to be mentioned that in about 10% of the patients with histologically proven NB the tumour does not accumulate 123I-MIBG which underlines the importance of additional functional/anatomical imaging (e.g. 18F-FDG PET/CT) [7].

A review article by Xia and co-authors demonstrated a summary sensitivity for MIBG scintigraphy and 18F-FDG PET(/CT) of 79% and 89%, respectively. The summary specificity for MIBG scintigraphy and 18F-FDG PET(/CT) was 84% and 71%, respectively. The authors concluded that 18F-FDG PET(/CT) showed higher per-lesion accuracy than MIBG scintigraphy and might be the preferred modality for the staging of NB [8].

#### **10.4 Merkel Cell Carcinoma**

Treglia and co-authors investigated the diagnostic performance of 18F-FDG PET and PET/CT in patients with Merkel cell carcinoma (MCC). The meta-analysis provided the following pooled results on a per-examination-based analysis: sensitivity was 90% (95%CI: 80–96%) and specificity was 98% (95%CI: 90–100%). The area under the summary ROC curve was 0.96. No significant statistical heterogeneity between the studies was found. The authors concluded that 18F-FDG PET or PET/CT demonstrated high sensitivity and specificity, being accurate methods in the detection of MCC taking into account that literature in MCC remains limited [9].

#### **10.5 Gastroenteropancreatic and Pulmonary Neuroendocrine Tumours**

Singh and co-authors evaluated the diagnostic performance of 68Ga-somatostatin analogues PET or PET/CT on neuroendocrine tumours (NETs). For the initial diagnosis of NETs, 68Ga-somatostatin analogues PET or PET/CT had a pooled sensitivity of 91% (95%CI: 85–94%) and a pooled specificity of 94% (95%CI: 86–98%). In the setting of staging and restaging, the sensitivity of 68Ga-somatostatin analogues PET or PET/CT for detecting primary and/or metastatic lesions ranged from 78.3 to 100%, whereas specificity ranged from 83 to 100%. Change in management occurred in 45% (95%CI: 36–55%) of the cases, with majority of the changes involving surgical planning and patient selection for peptide receptor radionuclide therapy [10].

This is in line with a systematic review by Barrio and colleagues who investigated the impact of 68Ga-somatostatin analogues PET/CT in patients with NETs. A change of management occurred in 44% of cases after 68Ga-somatostatin analogues PET/CT (range: 16–71%). In some studies, 68Ga-somatostatin analogues PET/CT was performed after conventional scintigraphy ( 111In-Octreotide). In this subgroup, additional information led to a change in management in 39% of cases (range: 16–71%). The authors concluded that the management was changed in more than one-third of patients undergoing 68Ga-somatostatin analogues PET/CT even when performed after an 111In-Octreotide scintigraphy [11].

In this line, another meta-analysis was published by Deppen and co-authors who compared conventional 111In-Octreotide imaging with 68Ga-DOTATATE PET/CT in pulmonary and gastroenteropancreatic NETs, with estimated pooled sensitivity of 90.9% (95%CI: 81.4–96.4%) and pooled specificity of 90.6% (95%CI: 77.8– 96.1%) for 68Ga-DOTATATE PET/CT [12].

The high diagnostic performance of 68Ga-somatostatin analogues PET or PET/CT for thoracic and gastroenteropancreatic NETs was showed by the meta-analysis of Treglia et al. reporting a pooled sensitivity and specificity of 93% (95%CI: 91–95%) and 91% (95%CI: 82–97%), respectively. 68Ga-somatostatin analogues PET/CT should be considered as first-line diagnostic imaging method for these tumours [13].

An updated meta-analysis on this regard reported a pooled sensitivity of 93% (95%CI: 91–94%) and a pooled specificity of 96% (95%CI: 95–98%) for 68Ga-somatostatin analogues PET or PET/CT. The area under the summary ROC curve was 0.98, confirming the good diagnostic performance of 68Ga-somatostatin analogues PET or PET/CT compared to diagnostic CT and conventional scintigraphy (e.g. 111In-Octreotide) [14].

An evidence-based article compared 68Ga-DOTATOC and 68Ga-DOTATATE PET in NETs, reporting no statistically significant differences of diagnostic performance among these imaging methods on a per-patient-based analysis [15].

#### **10.6 Congenital Hyperinsulinism**

Paediatric patients with congenital hyperinsulinism (CHI) due to pancreatic disease can be evaluated by PET or PET/CT, in particular by using 18F-DOPA. A systematic review and metaanalysis by Blomberg and co-authors aimed to quantify the diagnostic performance of pancreatic venous sampling (PVS), selective pancreatic arterial calcium stimulation with hepatic venous sampling (ASVS) and 18F-DOPA PET in diagnosing and localizing focal form of CHI. 18F-DOPA PET was superior in distinguishing focal from diffuse CHI compared to PVS and ASVS. Furthermore, it localized focal CHI in the pancreas more accurately than PVS and ASVS (pooled accuracy: 82% vs. 76% and 64%, respec-

tively) [16]. Yang and colleagues performed a metaanalysis of published data on the diagnostic role of 18F-DOPA PET in patients with CHI. The pooled sensitivity of 18F-DOPA PET and PET/CT in detecting CHI was 88%. The pooled specificity of 18F-DOPA PET and PET/CT in demonstrating CHI was 79%. The area under the ROC curve was 0.92. The authors concluded that 18F-DOPA PET or PET/CT demonstrated high sensitivity and specificity in patients with CHI [17].

These findings are in line with another metaanalysis by Treglia and co-authors: the pooled sensitivity and specificity of 18F-DOPA PET or PET/CT in differentiating between focal and diffuse CHI were 89% (95%CI: 81–95%) and 98% (95%CI: 89–100%), respectively. The area under the ROC curve was 0.95. The pooled accuracy of these functional imaging methods in localizing focal CHI was 80% (95%CI: 71–88%). Although possible sources of false-negative results for focal CHI should be kept in mind, the authors concluded that 18F-DOPA PET or PET/CT are accurate methods for localizing focal CHI [18].

#### **10.7 Thyroid Diseases**

#### **10.7.1 Thyroid Incidentalomas**

Nayan and colleagues evaluated through a systematic review and meta-analysis the malignancy rates of thyroid incidentalomas identified in adults by 18FDG PET/CT. The pooled proportion of malignancy was 19.8% (95%CI: 15.3–24.7%) with most of cases being papillary thyroid cancer. The authors stated that thyroid incidentalomas identified through 18FDG PET require thorough investigation [19].

In this context, a review article by Qu and coauthors was focused on focal thyroid incidentalomas (FTI) identified on 18F-FDG PET or PET/ CT. A meta-analysis was performed to investigate whether the maximum standardized uptake value (SUVmax) could discriminate between benign and malignant FTI and to explore the cutoff value of SUVmax for the diagnosis of malignancy. The results of this article indicated that there was no statistically significant difference in the size between benign and malignant FTI, while a significantly higher SUVmax was observed in the malignant group. The authors concluded that a higher SUVmax in FTI was associated with a higher risk of thyroid malignancy, especially at a threshold of 3.3 or more [20].

Treglia and co-authors described the prevalence and malignancy risk of FTI detected by 18F-FDG PET or PET/CT. Pooled prevalence of FTI was 1.92% (95%CI: 1.87–1.99%). Considering FTI which underwent histopathology evaluation, the pooled risk of malignancy was 36.2% (95%CI: 33.8–38.6%), without significant differences among various geographic areas. The authors concluded that FTI are observed in about 2% of 18F-FDG-PET or PET/CT and they should be further investigated due to a significant risk of malignancy [21].

#### **10.7.2 Indeterminate Thyroid Nodules**

A meta-analysis by Wang and colleagues evaluated the diagnostic accuracy of 18F-FDG PET or PET/CT in discriminating between malignant and benign lesions in thyroid nodules with indeterminate fine needle aspiration biopsy (FNAB). The prevalence of malignant lesions in these patients was 26.2% (ranging from 19.6 to 40%). The pooled sensitivity and specificity of 18F-FDG PET or PET/CT for the detection of cancer were 89.0% (95%CI: 79.0–95%) and 55% (95%CI: 48–62%), respectively. Although SUVmax was higher in malignant lesions, there was still a great overlap with benign lesions. In conclusion, 18F-FDG PET or PET/CT showed a high sensitivity in detecting thyroid cancers in patients with indeterminate FNAB results [22].

#### **10.7.3 Recurrence of Differentiated Thyroid Cancer**

A meta-analysis by Haslerud and co-authors described the role of 18F-FDG PET in recurrent differentiated thyroid cancer (DTC) after total thyroidectomy and radioiodine ablative therapy. Pooled sensitivity and specificity of this method in detecting recurrent DTC were 79.4% (95%CI: 73.9–84.1%) and 79.4% (95%CI: 71.2–85.4%), respectively, with an area under the ROC curve of 0.858. The authors concluded that this method can be useful for detecting recurrent DTC in patients having undergone radioiodine ablative therapy [23].

A meta-analysis by Caetano and co-authors aimed to evaluate the accuracy of 18F-FDG PET and PET/CT for detecting recurrence of DTC, not identified by 131I whole-body scintigraphy (I-WBS). The combined sensitivity, specificity and accuracy for 18F-FDG PET were 84%, 84% and 91%, respectively; for 18F-FDG PET/CT, the combined sensitivity, specificity and accuracy were 93%, 81% and 93% respectively [24].

Another meta-analysis by Schütz and coauthors about the use of 18F-FDG PET and PET/ CT for detecting recurrent DTC demonstrated that 18F-FDG PET and PET/CT showed higher sensitivity (89.7% for PET and 94.3% for PET/ CT) compared with conventional imaging (65.4%) and comparable results for specificity [25].

Kim and colleagues investigated the diagnostic accuracy of 18F-FDG PET/CT for the detection of recurrent and/or metastatic diseases in DTC patients with progressively and/or persistently elevated thyroglobulin antibodies (TgAb) levels and negative I-WBS through a systematic review and meta-analysis. The pooled sensitivity for 18F-FDG PET or PET/CT was 84% (95%CI: 77–89%), the pooled specificity 78% (95%CI: 67–86%). The area under the ROC curve was 0.88 (95%CI: 0.85–0.90). The authors concluded that 18F-FDG PET or PET/CT demonstrated moderate sensitivity and specificity for the detection of recurrent and/or metastatic diseases in DTC patients with progressively and/or persistently elevated TgAb levels and negative I-WBS [26].

A meta-analysis by Santhanam and co-authors investigated the accuracy of 18F-FDG PET/CT in the detection of residual disease in patients with BRAFV600E mutated thyroid cancer. The authors demonstrated that presence of BRAFV600E mutation in DTC confers a higher likelihood of 18F-FDG avidity and is associated with higher SUVmax values compared to BRAFV600Emutation negative status [27].

The role of 124I-PET/CT in detecting lesions of DTC amenable to 131I-therapy was recently described. The pooled sensitivity of 124I-PET/CT in detecting DTC lesions amenable to 131I-therapy was 94.2% (95%CI: 91.3–96.4%), and the pooled specificity was 49.0% (95%CI: 34.8–63.4%). The authors concluded that 124I-PET/CT is a sensitive tool to diagnose radioiodine-avid DTC lesions, but also detects some new lesions that are not visualized on the post-treatment I-WBS [28].

#### **10.7.4 Recurrence of Medullary Thyroid Cancer**

Treglia and co-authors described the role of 18F-FDG PET or PET/CT in patients with suspected recurrent medullary thyroid cancer (MTC). A sub-analysis considering PET device used, serum calcitonin, carcino-embryonic antigen (CEA), calcitonin doubling time (CTDT) and CEA doubling time (CEADT) values was also performed. Detection rate (DR) of 18F-FDG PET or PET/CT in suspected recurrent MTC on a per-patientbased analysis was 59% (95%CI: 54–63%). DR increased in patients with serum calcitonin ≥1000 ng/L (75%), CEA ≥5 ng/mL (69%), CTDT <12 months (76%) and CEADT <24 months (91%). The authors reported that about 40% of suspected recurrent MTC remain usually unidentified by 18F-FDG PET or PET/

CT. However, 18F-FDG PET and PET/CT could modify the patient management in a certain number of recurrent MTC because these methods are often performed after negative conventional imaging studies [29].

In another meta-analysis evaluating the diagnostic performance of 18F-FDG and PET/CT for detection of recurrent or metastatic MTC, the pooled sensitivities of 18F-FDG-PET and PET/ CT were 68% (95%CI: 64–72%) and 69% (95%CI: 64–74%), respectively [30].

Other PET radiotracers beyond 18F-FDG were evaluated for detecting recurrent MTC. In a metaanalysis evaluating the DR of 18F-DOPA PET or PET/CT for recurrent MTC, the DR of 18F-DOPA PET and PET/CT on a per-patient- and a perlesion-based analysis was 66% and 71%, respectively. The DR significantly increased in patients with serum calcitonin ≥1000 ng/L (86%) and CTDT <24 months (86%). Therefore, 18F-DOPA PET/CT may be a very useful functional imaging method in detecting recurrent MTC [31].

Another meta-analysis assessed the DR of 68Ga-somatostatin analogues PET or PET/CT in patients with recurrent MTC. The DR on a perpatient-based analysis was 63.5% (95%CI: 49–77%). DR of 68Ga-somatostatin analogues PET or PET/CT increased in patients with higher serum calcitonin levels (83% for calcitonin >500 ng/L). The authors concluded that the diagnostic performance of 68Ga-somatostatin analogues PET or PET/CT in recurrent MTC was lower compared to that of the same imaging method in the majority of NETs [32].

#### **10.8 Parathyroid Diseases**

Different PET tracers may be used to detect hyperfunctioning parathyroid glands in patients with hyperparathyroidism (HPT), including 11C-methionine (11C-MET) and radiolabelled choline. 11C-MET PET has an overall good sensitivity (69%) and positive predictive value (98%) in detecting hyperfunctioning parathyroid glands in patients with HPT and it may be considered a reliable second-line imaging method to enable minimally invasive parathyroidectomy [33].

Yuan and co-authors published a metaanalysis on the diagnostic value of 11C-MET PET in detecting hyperfunctioning parathyroid glands in patients with HPT and negative 99mTc-MIBI scan. Pooled sensitivity and specificity of 11C-MET PET in patients with HPT with negative or inconclusive 99mTc-MIBI scan were 86% and 86%, respectively. The authors concluded that 11C-MET PET can be a useful functional imaging modality in patients with negative or inconclusive 99mTc-MIBI scan [34].

Caldarella and co-authors investigated the diagnostic performance of 11C-MET PET in patients with suspected parathyroid adenoma. Pooled sensitivity and DR values of 11C-MET PET in patients with suspected parathyroid adenoma were 81% (95%CI: 74–86%) and 70% (95%CI: 62–77%), respectively, on a per-patientbased analysis. The authors also concluded that 11C-MET PET could be helpful when conventional imaging techniques are negative or inconclusive in localizing parathyroid adenoma [35].

An evidence-based article by Kim and colleagues investigated the diagnostic performance of radiolabelled choline for localization of hyperfunctioning parathyroid gland in patients with HPT. The pooled sensitivity for radiolabelled choline PET/CT was 90% (95% CI: 86–94%) and the pooled specificity 94% (95%CI: 90–96%) [36].

These findings are in line with a recent metaanalysis on the diagnostic performance of radiolabelled choline PET for detecting hyperfunctioning parathyroid glands: on a per-patient analysis, the sensitivity was 95% (95%CI: 92–97%) and the positive predictive value was 97% (95%CI: 95–98%); on a per-lesion analysis, pooled sensitivity and PPV were 92% (95%CI: 88–96) and 92% (95% CI: 89–95%), respectively [37].

#### **References**

1. Dinnes J, Bancos I, Ferrante di Ruffano L, Chortis V, Davenport C, Bayliss S, et al. Management of endocrine disease: imaging for the diagnosis of malignancy in incidentally discovered adrenal masses: a systematic review and meta-analysis. Eur J Endocrinol. 2016;175:R51–64.


and gastroenteropancreatic neuroendocrine tumours: a meta-analysis. Endocrine. 2012;42:80–7.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

**Part III**

**Evidence-Based PET in Cardiology**

**11**

## **Evidence-Based PET for Cardiac Diseases**

Christel H. Kamani, Marie-Madeleine Meyer, Sarah Boughdad, Nathalie Testart, Marie Nicod Lalonde, Gilles Allenbach, Mario Jreige, Niklaus Schaefer, Giorgio Treglia, and John O. Prior

#### **11.1 Introduction**

This chapter summarizes 15 meta-analyses published in the literature on the use of PET for cardiac diseases, with the majority (*n* = 8) concerning the diagnosis of coronary artery disease (CAD) using myocardial perfusion imaging (MPI) in comparison to other modalities. Second in the number of published meta-analyses, three studies

C. H. Kamani · M.-M. Meyer · S. Boughdad N. Testart · M. Nicod Lalonde · G. Allenbach · M. Jreige Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital, Lausanne, Switzerland

N. Schaefer · J. O. Prior (\*) Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland e-mail: john.prior@chuv.ch

#### G. Treglia

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

Clinic of Nuclear Medicine and Molecular Imaging, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

Health Technology Assessment Unit, Academic Education, Research and Innovation Area, Ente Ospedaliero Cantonale, Bellinzona, Switzerland e-mail: giorgio.treglia@eoc.ch

concentrated on the prognostic value of MPI for adverse cardiovascular events. Finally, one metaanalysis was published on the use of PET for four indications among myocardial viability assessment, the presence of microvascular disease (impaired coronary vascular function in absence of obstructive, epicardial CAD), the use of cardiac hybrid imaging and the diagnostic of cardiac amyloidosis.

#### **11.2 Myocardial Blood Flow Perfusion**

**11.2.1 Performance of PET/CT in the Assessment of Myocardial Perfusion in Comparison to Other Myocardial Perfusion Imaging Modalities**

Since the years 1950s, the technology of image acquisition in nuclear cardiology has progressively evolved from initial planar to more recent rapid hybrid cadmium-zinc-telluride (CZT) single-photon emission computed tomography/ computed tomography (SPECT/CT) and digitalized 3-dimensional (3D) positron emission tomography/computed tomography (PET/CT). Contemporary to this evolution, advances in the computation of acquired quantitative data using more performant software have led to a more 100

objective, digital-based assessment of the pathophysiological processes underlying cardiovascular diseases, from the stenosis of epicardial coronary artery to microvascular dysfunction, including coronary vasomotor as well as endothelial dysfunction. In the meantime, new radiotracers for perfusion imaging have been developed, from potassium-43 (43K) to thallium-201 (201Tl) and technetium-99 metastablelabelled (99mTc) radiotracers for SPECT/CT; rubidium-82 (82Rb), oxygen-15-labelled water ([15O]-H2O), nitrogen-13-labelled ammonia ([13N]-NH3), and still under development F-18 fluorine-labelled radiotracers for PET/CT.

Apart from nuclear techniques using SPECT/ CT and PET/CT, numerous non-invasive modalities have been developed to assess the myocardial perfusion, including dobutamine stress echocardiography (DSE), cardiac magnetic resonance (CMR), computed tomographic myocardial perfusion imaging (CT-MPI), fractional flow reserve derived from computed tomography (FFR-CT), with different diagnostic performances [1–4]. The development of these non-invasive modalities aims to contribute to counteract the burden of coronary artery disease (CAD), as the latest represents one of the leading causes of death and disability in developing countries [5]. In the field of nuclear imaging, especially PET/CT, there are accumulating evidences on the importance of flow quantification to guide management of stable CAD [6, 7], suggesting the potential role of PET/CT as gatekeeper for invasive coronary angiography (ICA). Therefore, following this paradigm change in the treatment of patients with CAD, it is of great importance to accurately select the right non-invasive imaging test evaluating the hemodynamic significance of a coronary stenosis for the right patient.

Jaarsma et al. [3] performed for the first time a meta-analysis on direct comparison of the diagnostic accuracy of three imaging modalities, PET/CT, CMR and SPECT/CT, for the diagnosis of significant CAD, using ICA as reference standard. In this meta-analysis, PET/CT performed better than CMR and SPECT/CT.

Dai et al. [1] have also assessed in a metaanalysis the diagnostic performance of six different cardiac imaging modalities, including PET/CT, for the evaluation of altered myocardial perfusion, using fractional flow reserve (FFR) derived from ICA as the reference standard. Among these imaging modalities, PET/CT as well as CMR and CT-MPI demonstrate high accuracy to detect hemodynamically significant CAD, as compared with SPECT/CT, DSE and FFR-CT on a patient and coronary territory basis. From these three best performing modalities, CMR and CT-MPI perform better than PET/CT in the evaluation of perfusion deficit. However, with the advent of 3D digital PET/CT, it is expected that the improvement in the spatial resolution would add an incremental value in the evaluation of perfusion deficit [8]. One of the strengths of PET/CT and CMR over other perfusion imaging modalities is the ability to absolutely quantify myocardial blood flow, as this has been well demonstrated to be a strong predictor of coronary artery disease [9]. In this field, PET/CT performs better than CMR according to these data. Using PET/CT, different parameters are generated for the quantitative assessment of myocardial blood flow: hyperemic MBF (hMBF) defined as the myocardial blood flow during hyperemic stress test; resting MBF (rMBF) defined as the myocardial blood flow in resting conditions; relative myocardial blood flow (RBF) defined as the ratio from hMBF in the stenosed coronary territory to hMBF of a reference non-stenotic territory; and myocardial flow reserve (MFR) defined as the ratio from hMBF and rMBF [10]. Evidences have demonstrated a discrepancy in the diagnostic accuracy of different MBF parameters [1, 10]. Indeed, hMBF has been found to be more accurate than CFR and FFR in the detection of significant coronary artery disease using either FFR derived from ICA or visual evaluation of the severity of the stenosis during ICA, as well as both as reference standard. The limited performance of CFR regarding sensibility and specificity as compared to the two other parameters is linked to the clinically relevant coronary pathophysiology, as MFR evaluates the global vascular response to hyperemia [9]. Thus, epicardial coronary stenosis (from mild diffuse to focal severe) and microvascular dysfunction (from vasomotor tone to endothelial function) are determinants for the assessment of MFR. Alterations of one of these parameters in resting conditions can already lead to an activation of compensatory vasodilation of the resistance vessels in the microcirculation, resulting in an alteration of rMBF [11]. Thus, MFR could be limited in the detection of epicardial coronary stenosis in the presence of diffuse atherosclerosis. However, in contrary to hMBF, MFR has the best prediction value for major adverse cardiovascular events (MACE) when comparing both parameters [9]. FFR showed the best specificity from all three parameters. However, unlike the two other parameters, it requires a coronary territory free from relevant coronary stenosis, what can be problematic in patients with coronary 3-vessels disease. And this explains the limited sensibility of this parameter. The results of this meta-analysis are in line with the previous published meta-analysis from Takx et al. [2], with all patients presenting an intermediate epicardial coronary stenosis in ICA, further evaluated by FFR. Interestingly, in this meta-analysis, posttest probabilities following a negative respectively a positive test result in PET/CT and CMR was calculated (derived from pre-test probability and likelihood ratio), with high accuracy, ranging from 9% to 11% respectively from 84% to 85%. Indeed, over sensitivity and specificity, the clinical performance of a non-invasive perfusion modality can be best assessed by the ability of post-test probability to confidently rule-in (posttest probability ≥85%) or rule-out (post-test probability ≤15%) relevant CAD. In this line, a recent meta-analysis has investigated the ranges of pre-test probability of significant CAD, in which five imaging modalities, including PET/ CT, can highlight an accurate post-test probability [12]. For this purpose, anatomic and functional reference standard were used, derived from ICA and FFR respectively with cut-off values for determining significant CAD for a stenosis >50% on ICA or a FFR ≤0.80. Depending on the reference standard used, PET/CT as well as the other functional imaging techniques (SPECT/CT, CMR) performed different. Indeed, using anatomic reference standard, they all showed only moderate accuracy, whereas there was a significant improvement of their performance when using functional reference standard.

When comparing nuclear cardiology modalities together, PET/CT appears more advantageous than SPECT/CT in many aspects in two meta-analyses [13, 14]. PET/CT has a higher spatial resolution resulting from better count sensitivity and higher energy of their radiotracers. Moreover, PET/CT has a better image quality due to short half-life of their radiotracers leading to higher signal-to-noise ratio. This short half-life of the radiotracers enables lower patient radiation exposure as compared to SPECT, with up to tenth, respectively, half of a standard dose for 201Tl and 99mTc as compared to 82Rb. And when evaluating the efficacy of transmyocardial laser revascularization (TMR) in patients with refractory angina not amenable to conventional percutaneous or surgery revascularization, PET/CT, in contrary to SPECT/CT, demonstrates an improvement of the subendocardial to subepicardial ratio in the follow-up, where SPECT/CT showed no changes in a meta-analysis [15]. Thus, PET/CT, as compared to SPECT/CT, did have better resolution to assess the subendocardial and subepicardial perfusion, which is a relevant factor for the detection and follow-up of patients with CAD.

There are some limitations who have to be taken into consideration when interpreting the results of the most discussed meta-analysis. In the last decade, evidences from randomized controlled trials highlighted the importance of considering the hemodynamic significance of epicardial coronary lesion using fractional flow reserve (FFR) than just the traditional visual assessment of the severity of the epicardial coronary lesion in the decision of revascularization [16]. Indeed, these evidences demonstrate a discrepancy between the visual assessment of the severity and the functional significance of the lesion, with impact on the outcome when revascularizing or not. Moreover, in many patients with stable CAD and epicardial coronary stenosis, the non-inferiority of optimal medical treatment (OMT) over percutaneous coronary intervention (PCI) on top of OMT has been well demonstrated [17]. Therefore, using visual assessment of the severity of coronary stenosis as reference could have introduced significant bias of the present results.

FFR is a surrogate of the coronary perfusion, derived from measurement of the coronary pressure before and after the stenosis. Thereby, it would ideally represent the proportion of blood flow available to the myocardium distally from the stenosis, as compared to the one in the absence of coronary stenosis [18]. Using FFR as reference standard to evaluate the accuracy of coronary perfusion, PET/CT is largely questionable for many reasons. First, early evidences from the 1980s have demonstrated the predominant role of coronary flow over coronary perfusion pressure for the maintenance of an adequate myocardial function. In this animal experimental study, the significant reduction of coronary perfusion pressure to values equivalent to an FFR around 0.4 did not lead to alteration of the myocardial function as long as the coronary flow was maintained [19]. Second, perfusion imaging modalities were used to validate FFR [20], making it less suitable as reference standard to determine the accuracy of perfusion PET/CT. Third, recent data from large, multicenter, prospective randomized trial demonstrate that a significant proportion of patients (60% of the study population) with epicardial stenosis showing FFR ≤0.8 did not require coronary revascularization [21]. Following all these evidences, the results of these meta-analyses should be interpreted with care. It is not excluded that one of these imaging modalities is more accurate than what it was found, but it could not be demonstrated due to the use of FFR as reference standard. Other bias could also have an impact on the results of these different meta-analyses, such as the use of different PET/ CT scans and different PET/CT scan protocols, the difference in the prevalence of CAD in the different studies, the difference in the threshold of FFR and visual assessing of the severity of the coronary stenosis. Moreover, most of the metaanalyses have evaluated the imaging modalities as a stand-alone technique, without integrating other clinical parameters such as age and gender as well as other para-clinical parameters such as ventricular function, which could have increased the accuracy of the imaging modality. Finally, even if considerable efforts have been made to reduce the radiation exposure of patients undergoing a PET/CT, it remains a matter of concern when comparing to other imaging modalities such as CMR with no radiation exposure.

Despite the significant advances in the assessment of patients with CAD, further investigations are still needed to overcome these limitations, thus bringing a new step of comprehension in the art of performing the right exam to the right patient.

#### **11.2.2 Prognostic Value of Myocardial PET**

PET MPI is increasingly being used for an accurate assessment of myocardial ischemia in patients with known or suspected coronary artery disease (CAD). It has been proposed by the American Heart Association that PET MPI may play a role as a novel cardiovascular biomarker, allowing better risk stratification of patients with CAD.

A first meta-analysis by Siontis et al. [22] specifically evaluated the incremental prognostic value that PET MPI added to standard risk factors in patients with known or suspected CAD. They selected 20 studies (with possible overlap in 5) totaling 22,203 patients, during a long span of 20 years, with only seven prospective studies. There was considerable heterogeneity among studies as MPI acquisition protocols, image analysis and selected radiotracer were not standardized among the studies. Only five studies reported changes in model classification, discrimination and risk stratification. PET MPI improved risk classification in four out of these five studies Despite the limitations of the metaanalysis, the authors were able to show that there is a strong association between abnormal perfusion (both by quantitative and qualitative analysis) and patient's outcome. They concluded that the limited data suggest that PET MPI may improve risk stratification, but this should be confirmed by data from larger and standardized prospective randomized controlled trials.

The second systematic review and metaanalysis by Chen et al. [23] focused on evaluating the prognostic value of normal PET MPI in patients with suspected or known CAD. They selected 11 studies with a total of 20,471 patients in whom PET MPI was performed for the diagnosis of coronary artery disease and with evaluation of cardiovascular death, all-cause death and major cardiovascular events (MACE) at a followup ≥3 months. They found highly significant negative predictive values for cardiac death (98.8%; 95% confidence interval (CI), 97.64– 99.39%), all-cause death (94.8%; 95% CI: 92.99–96.30%) and MACE (90.2%; 95% CI: 78.01–96.03%), with a reasonable follow-up period of 36.9 months, 26.8 months and 35.7 months, respectively. In a subgroup analysis, there was no significant difference in negative predictive value in studies with a normal PET MPI determined by absence of perfusion abnormalities compared to those using coronary flow reserve. An important limitation is that the authors were not able to differentiate patients with known coronary artery disease from patients with only clinical suspicion.

The third wide meta-analysis carried by Smulders et al. [24] compared several noninvasive tests for depiction of coronary artery disease. The aim was to evaluate the prognostic value of negative non-invasive cardiac investigations in patients with suspected or known CAD. They compared CT angiography, stress echocardiography, cardiac magnetic resonance, exercise electrocardiographic testing, SPECT and PET. The evaluated outcome was annual event rate for cardiac death and myocardial infarction. They included 165 studies with a total of 122,721 patients. Annual event rates (AER) after a negative test varied among the imaging modalities (between 0.32 for CT angiography and 1.36 for stress echocardiography). Some modalities are preferred in populations at higher event risk or known CAD. The authors supposed the different patient population among studies could explain the differences between modalities. Indeed, when adjusting for the population risk, the AER was similar between modalities. This meta-analysis supports that a negative non-invasive cardiac study provides predictive information with good accuracy, which should influence the clinical practice by reducing the need of further cardiac investigations.

In the last years, there is growing scientific evidence describing the value of non-invasive cardiac investigations, not only as a diagnostic tool but also as a predictive prognosis biomarker. Specifically, a negative test has an excellent prognosis, which should reasonably reassure the patient.

#### **11.3 Assessment of Myocardial Viability**

The concept of hibernating myocardium has been adopted in the early 1980s following the cumulating evidences of an improvement of the myocardial regional contractile function after an aorto-coronary bypass surgery [25]. It described a condition of altered left ventricular contractile function as a consequence of prolonged reduced coronary blood supply at rest, with improving potential following coronary revascularization. Different non-invasive imaging modalities, such as dobutamine stress echocardiography, cardiac magnetic resonance imaging and nuclear imaging, focusing on different aspects of the viable myocardium, have been proposed to identify viability in dysfunctional myocardial segments, with different diagnostic accuracy [26]. Among all these modalities, PET is considered as the gold standard [27].

In the meta-analysis of Tsai et al. [28], the authors investigated the accuracy of SPECT for the diagnosis of myocardial viability in patients with CAD and left ventricular dysfunction as compared to 18F-FDG PET. Although this metaanalysis was not centred on PET, which was only used as the gold-standard comparator for SPECT, it is the only one published about viability involving PET/CT. The authors found eight prospective studies totaling 320 patients (3580 segments analysed). The criteria used for SPECT viability were diverse in these eight studies, but most frequently involved the presence of myocardial rest perfusion visible (e.g. >50% or >55%), while two studies also involved 18F-FDG in the criteria for SPECT viability (usually with SPECT vs. PET mismatch indicating viability).

SPECT as compared to PET resulted in the following pooled values: pooled sensitivity 82% (95%CI 81–84%), pooled specificity 88% (95%CI 86–90%), diagnostic odd ratio 62.6 (95%CI 19.3–203) and the area under the receiving operating characteristic curve was 0.945 ± 0.026. There was a significant heterogeneity among studies especially in the definition of myocardial viability and the largest study included 58 patients, but the QUADAS quality assessment was of excellent quality and the funnel plot indicated no publication bias.

Thus, this meta-analysis showed that SPECT can be used adequately to assess myocardial viability in patients with CAD and left ventricular dysfunction. However, nowadays SPECT appears to be an outdated modality for assessing myocardial viability, and cardiac PET would be preferred for this diagnosis.

#### **11.4 Microvascular Disease**

Coronary artery disease (CAD) has often been exclusively associated to epicardial coronary abnormalities, as visualized during invasive coronary angiography (ICA). Nowadays, there are more evidences on the involvement of the entire coronary circulation, including the microcirculation, in the development of symptoms related to CAD [29]. This issue has been highlighted in a large cohort of symptomatic patients undergoing a diagnostic ICA for suspected obstructive epicardial CAD, where about 60% of the patients did not have a significant obstructive CAD (stenosis ≤50%) [30]. Microvascular dysfunction may encounter for a significant number of symptomatic patients without relevant obstructive CAD [29, 31]. Coronary microvascular disease (CMD) englobes a group of disorders affecting the structure and function of the coronary microcirculation in relation with endothelial dysfunction or a dysregulation of the vascular smooth muscle. Numerous non-invasive techniques have been proposed to assess the coronary microvascular function, such as PET/CT, dynamic CT-MPI and CMR [32]. Among these, PET/CT is the most validated modality with good reproducibility and accuracy.

In a systematic review and meta-analysis, Brainin et al. investigated the consequences of impaired coronary vascular function in the absence of obstructive coronary arterial disease [33]. In six studies (*n* = 1192 patients) endothelial dependent epicardial dysfunction was assessed, five of them measured coronary flow reserve (CFR) during angiography using coldpressure test or acetylcholine stimulation. In one study, CFR was measured using cold-pressure testing during PET. In a pooled analysis, the relative risk (RR) of 2.49 for cardiovascular events was significantly increased. In another group of ten studies (*n* = 5134), non-endothelial dependent epicardial dysfunction was assessed measuring coronary flow velocity reserve using echocardiography. These studies showed an increased RR of 4.58. In the last group including ten studies (*n* = 3687 patients) measuring CFR by PET, non-endothelial dysfunction was associated with an increased RR of 2.44.

These results underline that coronary vascular dysfunction as assessed by PET/CT has a prognostic value on cardiovascular events in a group of patients without obstructive coronary arterial disease. RR remained significantly elevated when excluding patients with diabetes or angina, as it remained elevated when including only patients with one specific stressor. Even if the number of studies remained small with different methods applied, impaired coronary vascular function in the absence of obstructive coronary disease should be reported for its prognostic value and considered as a potential therapeutic target.

#### **11.5 Cardiac Hybrid Imaging**

Cardiac multimodality (hybrid) imaging is a technique relying on the combination (on a sideby-side or fusion mode) of imaging modalities providing in one hand cardiac morphological data, such as CT, echocardiography or CMR, and in the other hand imaging modalities providing myocardial functional data, such as PET/CT. The goal of this combination is to gain information of these different imaging modalities in a complementary fashion in order to better guide coronary revascularization [34].

Coronary CT angiography (CCTA) has high negative predictive value and sensitivity for diagnosing obstructive CAD; however, its positive predictive value and specificity are lower. In particular, CCTA may overestimate coronary artery stenosis. Therefore, a hybrid approach combining CCTA with MPI modalities—including PET, SPECT and CMR—may help diminish the falsepositive rate of CCTA by fusing anatomic data derived from CCTA with functional data obtained through MPI, thus potentially overcoming CCTA limitations. To evaluate the clinical utility of this approach, Rezvi et al. [35] conducted a systematic literature review and meta-analysis comparing the diagnostic performance of hybrid cardiac imaging modalities with stand-alone CCTA for assessing obstructive CAD, using as a reference standard invasive coronary angiography (ICA).

The results of the meta-analysis showed that, at the per-patient level, CCTA and MPI demonstrated comparable sensitivity (*p* = 0.35), but CCTA displayed lower specificity (66%) compared to MPI (83%) for predicting obstructive CAD; at the per-vessel level, specificity of CCTA and MPI was similar (*p* = 0.02), and sensitivity was higher for CCTA (89%) than for MPI alone (78%) (*p* < 0.001).

When the diagnostic performance of hybrid versus CCTA imaging method was examined on a per-patient basis, sensitivity, negative likelihood ratio (LR–) and diagnostic odds ratio (DOR) of hybrid imaging techniques compared to CCTA to detect obstructive CAD were, respectively, 91%, 0.11 and 159.00, versus 90%, 0.06 and 53.80 (*p* > 0.05, for all). Specificity and positive likelihood ratio (LR+) were higher for hybrid imaging compared to stand-alone CCTA (93% and 12.80 versus 66% and 3.39, respectively). At the per-vessel level, summary receiver-operator curves (sROC) demonstrated a statistically significant and higher area under the curve (AUC) value for hybrid imaging than for CCTA (0.97 versus 0.93, respectively; *p* = 0.047).

In conclusion, hybrid imaging techniques outperformed stand-alone CCTA for detecting obstructive CAD in patients undergoing both anatomic and functional testing, demonstrating higher specificity and LR+ on a per-patient basis. In addition, at the per-vessel level, hybrid imaging could better identify CAD based on sROC curves; however, sensitivity was comparable for hybrid versus stand-alone CCTA imaging.

#### **11.6 Cardiac Amyloidosis**

Amyloidosis encompasses a group of infiltrative disorders characterized by extracellular accumulation of fibrillary proteins, leading to functional impairment of different tissues and organs, including the heart. In patients with systemic amyloidosis, cardiac involvement could lead to significantly increased morbidity and mortality. The most common types of cardiac amyloidosis are systemic light chain (AL) and transthyretin (ATTR) amyloidosis. This differentiation is of importance as it differs in treatment and prognosis, untreated patients with AL amyloidosis having the poorest prognosis [36]. ATTR has been long time considered as a rare disease because of the lack of disease awareness, its quiescent nature at the beginning and the heterogeneity of symptoms when clinical manifest. Nowadays, accumulating evidences showed that it may be more prevalent than thought, especially in certain groups of patients such as older people or patients with aortic stenosis [37]. Moreover, with the emergence of new treatment options, early diagnosis is critical because of the better effectiveness of the treatment in the earlier course of the disease [38]. In line with this, bone scintigraphy as a non-invasive tool in the flowchart to accurately differentiate between AL and ATTR amyloidosis is well established [39]. Moreover, recent development in amyloid tracers for positron emission tomography (PET) could lead to an early diagnosis and potentially improve the prognosis of those patients. In that setting, we shortly present the results of the first systematic review and meta-analysis on the diagnostic performance of amyloid PET imaging in cardiac amyloidosis by Kim et al. [40]. The authors also reported the potential additional value of using semi-quantitative parameters for the diagnosis of cardiac amyloidosis and for distinguishing between AL and ATTR amyloidosis.

Six retrospective studies with a total of 98 patients (69 patients with systemic amyloidosis and 29 control patients) were included in this meta-analysis. Among amyloid radiotracers, 11C-PIB was the most commonly used in four studies, whereas 18F-Florbetapir and 18F-Florbetaben were each used in one study.

In the whole cohort (six studies), the pooled sensitivity was 0.95 (95% CI 0.87–0.99), and pooled specificity was 0.98 (95% CI 0.87–1.00). Positive likelihood ratio (LR) was 10.130 (95% CI 3.749–27.376), negative LR was 0.100 (95% CI 0.045–0.221), and diagnostic odds ratio was 148.83 (95% CI = 34.026–650.98). Summary receiver operating characteristic curve showed high performance with an area under curve of 0.9731 with a standard error of 0.0156. Looking only at the four 11C-PIB studies, similar results were found with a diagnostic odds ratio of 134.19 (95% CI 24.039–749.03).

The use of semi-quantitative parameters was reported in four studies (three with 11C-PIB and one with 18F-Florbetaben) and consisted of retention index (RI) and/or target-to-background ratio (TBR). RI derived from dynamic acquisitions (RI = mean myocardial SUV/integral of the arterial time–activity curve). By contrast, TBR was extracted from static acquisitions (TBR = max or mean myocardial SUV/mean SUV of the descending thoracic aorta). Patients with cardiac amyloidosis had significantly higher RI and TBR values in comparison to control patients. Using both, pooled standardized mean difference (SMD) was of 1.42 (95% CI 0.83–2.01; *p* < 0.001). The performance of RI and TBR for discriminating between AL and ATTR amyloidosis was assessed in three studies (two with 11C-PIB and one with 18F-Florbetaben) and showed significantly higher uptake in the former (pooled SMD = 0.96; 95% CI 0.13–1.79; *p* < 0.001).

To resume, amyloid PET imaging presents with strong performance to diagnose cardiac amyloidosis with high sensitivity and specificity (≥95%). The addition of semi-quantitative parameters, such as RI and TBR, could help improve the diagnosis and accurately differentiate between AL and ATTR amyloidosis.

#### **References**


2017;136(24):2325–36. https://doi.org/10.1161/ CIRCULATIONAHA.117.029992.


Med. 2010;362(10):886–95. https://doi.org/10.1056/ NEJMoa0907272.


analysis. JACC Cardiovasc Imaging. 2018;11(4):589– 99. https://doi.org/10.1016/j.jcmg.2017.05.020.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part IV**

## **Evidence-Based PET in Infection and Inflammation**

**12**

## **Evidence-Based PET for Infectious and Inflammatory Diseases**

Giorgio Treglia and Barbara Muoio

#### **12.1 Introduction**

Nuclear medicine techniques are non-invasive tools that can early detect pathophysiological changes in affected tissues in patients with inflammatory or infectious diseases. These changes usually occur before clinical onset of symptoms and before the development of anatomical changes detected by radiological techniques [1, 2]. Currently, hybrid imaging techniques as positron emission tomography/ computed tomography (PET/CT) may provide functional and morphological information for early diagnosis of infectious and inflammatory diseases [1, 2].

The ability of Fluorine-18 fluorodeoxyglucose (18F-FDG) PET/CT to identify sites of inflamma-

Health Technology Assessment Unit, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland e-mail: giorgio.treglia@eoc.ch

B. Muoio

Clinic of Oncology and Internal Medicine, San Giovanni Hospital, Oncology Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona, Switzerland

tion and infection is mainly related to the glycolytic activity of the cells involved in the inflammatory response [3, 4]. Enough evidence in the literature already exists about the diagnostic performance of 18F-FDG PET/CT in the diagnosis and management of several infectious and inflammatory diseases [5]. The results of the selected articles, including pooled values and 95% confidence interval (95%CI), are presented in Table 12.1 and summarized here below.

#### **12.2 Fever of Unknown Origin (FUO)**

Fever of unknown origin (FUO) is commonly defined as temperature ≥38.3 °C on at least two occasions, duration of illness ≥3 weeks or multiple febrile episodes in ≥3 weeks, not immunocompromised patient, and uncertain diagnosis despite thorough history-taking, physical examination, and obligatory investigations [6]. The diagnosis in patients with FUO is a challenging medical problem; the cause of FUO may be infectious diseases, non-infectious inflammatory diseases, or tumors, and 18F-FDG PET/CT detecting foci of increased glucose metabolism may be used for revealing the source of fever [6]. Several meta-analyses have estimated the diagnostic performance of 18F-FDG PET/CT in the assessment of FUO unidentified by conventional workup [7–13].

G. Treglia (\*)

Clinic of Nuclear Medicine and PET/CT Center, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

<sup>©</sup> The Author(s) 2020 111

G. Treglia, L. Giovanella (eds.), *Evidence-based Positron Emission Tomography*, https://doi.org/10.1007/978-3-030-47701-1\_12


**Table 12.1** Characteristics and main findings of included meta-analyses on the diagnostic performance of 18F-FDG PET/CT in infectious or inflammatory diseases


**Table 12.1** (continued)

*LR+* positive likelihood ratio, *LR−* negative likelihood ratio, *DOR* diagnostic odds ratio, *95%CI* 95% confidence interval, *NR* not reported, *CIED* cardiovascular implantable electronic device, *GCA* giant cell arteritis, *TA* Takayasu arteritis, *TA+* Takayasu arteritis using National Health Institute scale a Both PET and PET/CT are included

Dong et al. firstly reported that the pooled sensitivity and specificity of 18F-FDG PET/CT for the detection of FUO were 98.2% (95%CI: 93.6– 99.8) and 85.9% (95%CI: 75–93.4), respectively. Therefore, this method should be considered among the first diagnostic tools for patients with FUO in whom conventional diagnostics have been unsuccessful [7].

Hao et al. confirmed the high sensitivity of 18F-FDG PET/CT for the diagnosis of patients with FUO (pooled value: 88%; 95%CI: 81–88), but the possibility of false positive results should be kept in mind [8].

Another meta-analysis demonstrated that abnormal 18F-FDG PET/CT findings are associated with a substantially increased final diagnostic rate in FUO (pooled odds ratio: 8.94; 95%CI: 4.18–19.12, *p* < 0.00001). Consequently, 18F-FDG PET/CT could be considered for inclusion in the first-line diagnostic workup of FUO [9].

Tateuchi et al. reported that 18F-FDG PET/CT can be useful in identifying the source of fever in patients with classic FUO (immunocompetent patients). The summary sensitivity and specificity were 86% (95%CI: 81–90) and 52% (95%CI: 36–67), respectively. The contribution of 18F-FDG PET/CT may be limited in clinical settings in which infectious and neoplastic causes are less common. Indirect comparisons of test performance suggested that 18F-FDG PET/CT outperformed standalone 18F-FDG PET, Gallium-67 scintigraphy, and radiolabelled leukocyte scintigraphy in detecting causes of FUO. Studies using standardized diagnostic algorithms are needed to determine the optimal timing for testing and to assess the impact of tests on management decisions and patient-relevant outcomes [10].

Recently, Bharucha et al. reported an overall diagnostic contribution of 56% (95%CI: 50–61) of 18F-FDG PET/CT in all patients with FUO. In a subgroup analysis taking into account previous investigations the diagnostic yield/added contribution of 18F-FDG PET/CT over CT was 32% (95%CI: 22–44). The pooled proportion of abnormal 18F-FDG-PET/CT in patients with FUO was 69% (95%CI: 63–75); the higher proportion of abnormal scans was accounted for by a proportion of false positive abnormal scans with no contribution to the final diagnosis, with an overall result of 9% (95%CI: 5–14). The authors concluded that there is insufficient evidence to support the value of 18F-FDG PET/CT in investigative algorithms of FUO [11].

Conversely, in an updated meta-analysis on patients with FUO or inflammation of unknown origin (IUO), 18F-FDG PET/CT was demonstrated to be very helpful for recognizing and excluding diseases, directing further diagnostic decisions, and avoiding unnecessary invasive examinations. The pooled sensitivity and specificity were 84% (95%CI: 79–89) and 63% (95%CI: 49–75), respectively. Based on these findings, the authors recommended 18F-FDG PET/CT among the first-line diagnostic tools for patients with FUO and IUO [12].

Lastly, it has been recently demonstrated that patients with negative 18F-FDG PET/CT results were significantly more likely to present with spontaneous fever regression than those with positive 18F-FDG PET/CT results (summary relative risk = 5.6: 95%CI: 3.4–9.2; *p* < 0.001) [13].

Overall, there is not agreement among the selected meta-analyses about the added value of 18F-FDG PET/CT in patients with FUO. The main drawback of the meta-analyses evaluating the diagnostic performance of 18F-FDG PET/CT for this specific indication is that they include articles without real FUO patients or with highly variable definitions of FUO; therefore, related meta-analyses could be not accurate in this regard [14].

#### **12.3 Large Vessel Vasculitis (LVV)**

Large vessel vasculitis (LVV) is defined as an inflammatory disease mainly affecting the large arteries, with two major variants, Takayasu arteritis (TA) and giant cell arteritis (GCA). GCA often coexists with polymyalgia rheumatica (PMR) in the same patient, since both belong to the same disease spectrum [15]. 18F-FDG PET/ CT may demonstrate increased radiopharmaceutical uptake in the vascular wall of large vessels in patients with LVV; therefore, this method may be used for diagnosis, monitoring of disease activity, and evaluating disease progression in LVV [15–18], and several meta-analyses have assessed the role of this imaging method in this setting [19–25].

First meta-analyses including both 18F-FDG PET and PET/CT studies reported a valuable diagnostic performance of these methods in patients with GCA with a pooled sensitivity and specificity of 80% (95%CI: 63–91) and 89% (95%CI: 78–94), respectively [19], and a moderate value of these methods in assessing TA activity, with a pooled sensitivity and specificity of 70.1% (95%CI: 58.6–80) and 77.2% (95%CI: 64.2–87.3), respectively [20].

In a meta-analysis of Soussan et al. including both 18F-FDG PET and PET/CT studies, these imaging methods showed good performances in the diagnosis of LVV, with higher accuracy in GCA patients than in TA patients. A vascular uptake equal to or higher than the liver uptake appeared to be a good criterion for the diagnosis of vascular inflammation. 18F-FDG PET or PET/ CT showed high sensitivity and specificity for the diagnosis of LVV in GCA patients in comparison to controls, with pooled values of 90% (95%CI: 79–93) and 98% (95%CI: 94–99), respectively. 18F-FDG PET or PET/CT had a pooled sensitivity of 87% (95%CI: 78–93) and specificity of 73% (95%CI: 63–81) for the assessment of disease activity in TA, with up to 84% of specificity in studies using National Institutes of Health criteria as the disease activity assessment scale [21].

Another meta-analysis by Lee et al. confirmed that 18F-FDG PET/CT has good diagnostic accuracy for LVV with a pooled sensitivity and specificity of 83.9% (95%CI: 71.7–92.4) and 87.2% (95%CI: 72.6–95.7), respectively [22].

In a recent meta-analysis, the pooled sensitivity and specificity of 18F-FDG PET or PET/CT for detecting active disease in TA compared to clinical assessment were 81% (95%CI: 69–89) and 74% (95%CI: 55–86), respectively. Active disease by 18F-FDG PET or PET/CT was also associated with elevations of acute phase reactants, as C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) [23]. Conversely, in another meta-analysis by Gomez et al. about the association between the CRP value and 18F-FDG PET or PET/CT vascular positivity in TA, CRP concentration only moderately reflected the 18F-FDG PET vascular positivity in TA, suggesting dissociated information [24]. More prospective studies are needed to assess the value of 18F-FDG PET/CT as an independent biomarker for subtle vascular wall inflammation detection in patients with TA [24].

Lastly, an updated meta-analysis confirmed that 18F-FDG PET or PET/CT has a good performance for the detection of active disease in patients with LVV with a pooled sensitivity and specificity of 88% (95%CI: 79–93) and 81% (95%CI: 64–91), respectively. Therefore, 18F-FDG PET/CT could be suggested as a surrogate biomarker for assessment of disease activity of LVV during or after immunosuppressive therapy, but further studies are warranted to determine if PET-based treatment of LVV can improve outcomes [25].

Several factors may significantly influence the diagnostic performance of 18F-FDG PET/CT in LVV including different PET interpretation criteria, atherosclerotic vascular 18F-FDG uptake (a possible source of false positive findings), and immunosuppressive therapy (a possible source of false findings) [15].

Overall, based on the available evidence, 18F-FDG PET/CT has demonstrated high diagnostic performance for the detection of LVV. Further studies are needed to select the most clinically relevant and reproducible criteria for defining the presence of LVV with 18F-FDG PET/CT, as well as to test the clinical impact of 18F-FDG PET/CT on the management of patients with suspected LVV [15].

#### **12.4 Infectious Endocarditis and Cardiovascular Implantable Electronic Device Infections**

Infectious endocarditis (IE) is a serious and potentially life-threatening condition. The current diagnosis of IE is based on the modified Duke criteria, which has approximately 80% sensitivity for the diagnosis of native valve endocarditis (NVE), with lower sensitivity for the diagnosis of prosthetic valve endocarditis (PVE) and culture-negative endocarditis [26, 27]. Noninvasive imaging modalities may improve diagnosis of infective endocarditis (IE) [26, 27]. In particular, 18F-FDG PET/CT is currently included as diagnostic tool in the diagnostic flow chart for IE [26–29] and some meta-analyses have evaluated the diagnostic performance of this method in patients with IE or CIED infections [30–34].

A first meta-analysis published in 2016 demonstrated that the overall diagnostic performance of 18F-FDG PET/CT for the diagnosis of IE was not high due to the low sensitivity: pooled sensitivity and specificity were 61% (95%CI: 52–88) and 88% (95%CI: 80–93), respectively. However, the diagnostic performance of 18F-FDG PET/CT increased in the subgroup of patients with PVE [30].

Mahmood et al. demonstrated that 18F-FDG PET/CT may be a useful adjunctive diagnostic tool in the evaluation of diagnostically challenging cases of IE, particularly in PVE. The pooled sensitivity and specificity of 18F-FDG PET/CT for diagnosis of IE were 76.8% (95%CI: 71.8– 81.4) and 77.9% (95%CI: 71.9–83.2), respectively. Diagnostic accuracy was improved for PVE with pooled sensitivity of 80.5% (95%CI: 74.1–86) and pooled specificity of 73.1% (95%CI: 63.8–81.2). More recent studies published from 2015 to 2017 reported a higher pooled sensitivity of 81.3% (95%CI: 74.3–87) and specificity of 79% (95%CI: 71.2–85.5). The majority of the recent studies were prospective and used a specific protocol (i.e., a lowcarbohydrate fat-allowed diet for at least 24 h prior to imaging, a prolonged fasting prior to imaging, and/or an intravenous heparin bolus prior to 18F-FDG administration). 18F-FDG PET/ CT also has the potential to detect clinically relevant extra-cardiac foci of infection, malignancy, and other sources of inflammation, leading to more appropriate treatment regimens and surgical intervention. Additional extra-cardiac foci of infection were found on 17% of patients in this meta-analysis [31].

In another meta-analysis, Juneau et al. demonstrated that 18F-FDG PET/CT has a good diagnostic accuracy for the diagnosis of IE if adequate patient preparation for suppression of physiological myocardial 18F-FDG uptake was performed, including prolonged fasting at least 12 h and/or heparin injection before 18F-FDG administration, and/or high-fat carbohydrate-restricted proteinpermitted diet (minimum two meals for 24 h). Pooled sensitivity of 18F-FDG PET/CT performed with adequate cardiac preparation for the diagnosis of IE was 81% (95%CI: 73–86) and pooled specificity was 85% (95%CI: 78–91). In the subgroup of patients with PVE, the pooled sensitivity was 85% (95%CI: 77–91) but specificity was 81% (95%CI: 72–88). Therefore, 18F-FDG PET/CT may be useful in the investigation of IE, and should be considered in cases where the diagnosis is uncertain [32].

18F-FDG PET/CT may be helpful in the diagnosis of cardiovascular implantable electronic device (CIED) infections, particularly in patients with the absence of localizing signs or definitive echocardiographic findings. In a recent metaanalysis, Mahmood et al. reported a pooled sensitivity and specificity of 18F-FDG PET/CT in the diagnosis of CIED infections of 85% (95%CI: 80–89) and 90% (95%CI: 84–94), respectively. 18F-FDG PET/CT demonstrated a higher sensitivity of 96% (95%CI: 86–99) and specificity of 97% (95%CI: 86–99) for diagnosis of pocket infections. Diagnostic accuracy for lead infections or CIED-IE was lower with pooled sensitivity of 76% (95%CI: 65–85) and specificity of 83% (95%CI: 72–90). In the subgroup of studies that described the use of any myocardial suppression protocol, the pooled sensitivity was 92% (95%CI: 85–96) and the pooled specificity was 81% (95%CI: 71–89) [33].

Another recent meta-analysis confirmed the high diagnostic performance of 18F-FDG PET/ CT for the diagnosis of CIED infections with a pooled sensitivity of 87% (95%CI: 82–91) and a pooled specificity of 94% (95%CI: 88–98). Pooled sensitivity and specificity for diagnosis of pocket/generator related CIED infections were 93% (95%CI: 84–98) and 98% (95%CI: 88–100), respectively. Pooled sensitivity and specificity for diagnosis of lead or IE-related CIED infection were 65% (95%CI: 53–76) and 88% (95%CI: 77–94), respectively [34].

Overall, 18F-FDG PET/CT demonstrated a good diagnostic performance in patients with IE and CIED infections with higher diagnostic accuracy if adequate patient preparation for suppression of physiological myocardial 18F-FDG uptake was performed.

#### **12.5 Vascular Graft Infections**

Vascular graft infection (VGI), a serious complication in vascular surgery, has a high morbidity and mortality rate. The diagnosis is complicated by non-specific symptoms and challenged by the variable accuracy of different imaging techniques [35, 36]. A recent meta-analysis demonstrated a good diagnostic performance of 18F-FDG PET/ CT in patients with VGI with a pooled sensitivity and specificity of 95% (95%CI: 87–99) and 80% (95%CI: 69–89), respectively [36].

Another recent meta-analysis investigating the diagnostic accuracy of 18F-FDG PET/CT in VGI reported a pooled sensitivity and specificity for focal 18F-FDG uptake of 97% (95%CI: 89–99) and 89% (95%CI: 70–96), respectively [37].

Factors influencing the diagnostic performance of 18F-FDG PET/CT in VGI include the time at which 18F-FDG PET/CT is performed after surgery (if 18F-FDG PET/CT is performed in cases of recently implanted grafts, false positive 18F-FDG PET/CT findings for VGI are possible), the use of antibiotics prior to 18F-FDG PET/CT (causing possible false negative findings for VGI), and the PET interpretation criteria used [37].

#### **12.6 Sarcoidosis**

Sarcoidosis is a multisystem chronic inflammatory disease of unknown etiology characterized by widespread growth of non-caseating granulomas. The diagnosis of sarcoidosis is based on clinical and imaging presentation, histological confirmation, and the absence of alternative diseases. Imaging techniques may play a role in the diagnostic workup of patients with sarcoidosis to assess disease extent and activity, and treatment response evaluation [38]. The role of 18F-FDG PET/CT in patients with sarcoidosis is well established [39, 40]. Based on evidence-based data, the recommendations for use of 18F-FDG PET/CT in patients with sarcoidosis could be the following: evaluation of inflammatory active disease in patients with persistent symptoms and negative serologic markers; assessment of inflammation in radiologic stage IV sarcoidosis with lung fibrosis; evaluation of inflammatory active extrathoracic sites of sarcoidosis or assessment of cardiac sarcoidosis (especially in patients with implanted pacemakers); identification of active sites for diagnostic biopsy not revealed by other methods; evaluation of treatment response in refractory sarcoidosis [39].

The role of 18F-FDG PET/CT in cardiac sarcoidosis is currently under active investigation [41] and some meta-analyses have addressed the diagnostic performance of 18F-FDG PET/CT in this setting [42–44].

In the meta-analysis of Youssef et al., the pooled sensitivity and specificity of 18F-FDG PET or PET/CT for diagnosis of cardiac sarcoidosis were 89% (95%CI: 79–96) and 78% (95%CI: 68–86), respectively [42].

Tang et al. demonstrated that the diagnostic accuracy of 18F-FDG PET/CT for cardiac sarcoidosis depends on adequate suppression of physiological cardiac glucose uptake. Overall, 18F-FDG PET/CT had a pooled sensitivity of 75% (95%CI: 69–80) and a pooled specificity of 81% (95%CI: 76–85) for the diagnosis of cardiac sarcoidosis. This modest diagnostic accuracy was attributed to the inclusion of studies in which a short fasting duration before scanning likely influenced its sensitivity. Excluding studies without adequate myocardial suppression resulted in a pooled sensitivity of 81% (95%CI: 76–86) and a pooled specificity of 82% (95%CI: 77–86). Fasting for at least 12 h before scanning or a high-fat low-carbohydrate diet given at 3–6 h before imaging or heparin infusion before imaging has shown to improve the diagnostic accuracy of 18F-FDG PET/CT in cardiac sarcoidosis [43].

Lastly, an updated meta-analysis on the diagnostic performance of 18F-FDG PET or PET/CT in cardiac sarcoidosis demonstrated a pooled sensitivity and specificity of 84% (95%CI: 71–91) and 83% (95%CI: 74–89), respectively. The presence of combined myocardial perfusion imaging improved the diagnostic accuracy of 18F-FDG PET/CT for diagnosis of cardiac sarcoidosis. Nevertheless further large multicenter studies in this setting are needed [44].

#### **12.7 Musculoskeletal Infections**

Timely identification and precise localization of musculoskeletal infections by imaging techniques are critical for early initiation of treatment and can have a significant impact on patient outcome. In this setting, nuclear medicine and radiological imaging are complementary techniques [45]. In particular, several meta-analyses have investigated the diagnostic performance of 18F-FDG PET/CT in patients with suspicious musculoskeletal infections [46–54].

Wang et al. calculated the diagnostic performance of 18F-FDG PET or PET/CT in patients with suspicious osteomyelitis reporting a high pooled sensitivity and specificity in this setting: pooled values were 92.3% (95%CI: 86.7–96.1) and 92% (95%CI: 87–95.6), respectively [46].

A first meta-analysis focused on the diagnostic performance of 18F-FDG PET or PET/CT in osteomyelitis related to diabetic foot reported a pooled sensitivity and specificity of 74% (95%CI: 60–85) and 91% (95%CI: 85–96), respectively [47]. An updated meta-analysis on the same topic demonstrated a pooled sensitivity of 89% (95%CI: 68–97) and a pooled specificity of 92% (95%CI: 85–96) [48].

Jin et al. calculated the diagnostic performance of 18F-FDG PET or PET/CT in detecting prosthetic infection after arthroplasty. They found a pooled sensitivity and specificity of 86% (95%CI: 82–90) and 86% (95%CI: 83–89), respectively. The pooled sensitivity of 18F-FDG PET or PET/CT in demonstrating hip and knee prosthetic infection was 88% (95%CI: 83–92) and 72% (95%CI: 58–84), respectively. The pooled specificity of 18F-FDG PET or PET/CT in demonstrating hip and knee prosthetic infection was 88% (95%CI: 84–91) and 80% (95%CI: 71–88), respectively [49].

A meta-analysis focused on periprosthetic hip infection confirmed the good diagnostic accuracy of 18F-FDG PET or PET/CT in this setting with pooled sensitivity and specificity of 86% (95%CI: 80–90) and 93% (95%CI: 90–95), respectively, using increased 18F-FDG uptake in the boneprosthesis interface as the criterion for infection for the index test [50].

A meta-analysis focused on periprosthetic knee infection demonstrated a nonoptimal diagnostic accuracy of 18F-FDG PET or PET/CT in this setting with pooled sensitivity and specificity of 70% (95%CI: 56–81) and 84% (95%CI: 76–90) [51].

Some factors influencing the diagnostic performance of 18F-FDG PET/CT in patients with osteomyelitis should be underlined: first of all, several interpretation criteria of 18F-FDG PET have been used in the literature, by using visual and/or semi-quantitative criteria, leading to different diagnostic accuracy values [46–51]. Furthermore, continuous physiologic 18F-FDG activity around the prostheses may be cause of false positive 18F-FDG PET/CT findings for periprosthetic infection [49–51].

18F-FDG PET or PET/CT has an excellent diagnostic performance in detecting infectious spondylodiscitis [55]. A first meta-analysis on 18F-FDG PET or PET/CT in patients with suspicious spondylodiscitis reported a pooled sensitivity and specificity of 97% (95%CI: 83–100) and 88% (95%CI: 74–95), respectively [52]. In this setting, the diagnostic performance of 18F-FDG PET or PET/CT was higher compared with magnetic resonance imaging (MRI). Considering studies comparing 18F-FDG PET or PET/CT and MRI, pooled sensitivity and specificity of 18F-FDG PET or PET/CT were 96% (95%CI: 84–99) and 90% (95%CI: 79–96), whereas the pooled sensitivity and specificity of MRI were 76% (95%CI: 65–84) and 62% (95%CI: 45–77) [53]. Another recent meta-analysis confirmed the better diagnostic accuracy of 18F-FDG PET or PET/ CT compared to MRI for the detection of spondylodiscitis: for 18F-FDG PET or PET/CT, pooled sensitivity and specificity were 95% (95%CI: 87–98) and 88% (95%CI: 73–95), respectively; for MRI, pooled sensitivity and specificity were 85% (95%CI: 65–95) and 66% (95%CI: 48–80), respectively [54].

Overall, based on the available evidence, 18F-FDG PET/CT has demonstrated a good diagnostic performance for the detection of musculoskeletal infections.

#### **12.8 Inflammatory Rheumatic Diseases**

Molecular imaging methods, including 18F-FDG PET/CT, have been proposed for a better assessment of inflammatory rheumatic diseases [56]. 18F-FDG uptake in the shoulders or hips was often reported in PMR (pooled prevalence: 76%), especially in periarticular sites (pooled prevalence: 84%). Furthermore, interspinous 18F-FDG uptake, demonstrating interspinous bursitis, is common in PMR (pooled prevalence: 67%). However, these findings are not very specific for PMR [57].

Patients with rheumatoid arthritis (RA) may also have interspinous 18F-FDG uptake (pooled prevalence: 34%) or articular 18F-FDG uptake in shoulders or hips (pooled prevalence: 66%) or in other articular regions (pooled prevalence: 78%). Articular 18F-FDG uptake is not specific for PMR or RA, as it is common in other connective tissue diseases (pooled prevalence: 70%). Overall, 18F-FDG PET/CT is helpful in diagnostic research, but the interpretation of 18F-FDG uptake at each site is not characteristic of a specific inflammatory rheumatic disease [57].

#### **12.9 Inflammatory Bowel Diseases**

18F-FDG PET/CT may also be used to image areas of active inflammation, such as those occurring in patients with active inflammatory bowel disease (IBD) as Crohn's disease and ulcerative colitis [58]. In this setting, 18F-FDG PET or PET/CT showed a good accuracy with a pooled sensitivity and specificity of 85% (95%CI: 81–88) and 87% (95%CI 84–90), respectively [59]. These findings were confirmed by another meta-analysis including prospective studies only [60]. Nevertheless, more prospective studies evaluating the role of 18F-FDG PET/ CT for this indication are needed. Specific challenges for the use of 18F-FDG PET/CT in IBD are the physiological 18F-FDG uptake in the bowel and the movement of the bowel that may influence a correct co-registration of 18F-FDG PET and CT images [59].

#### **References**


meta-analysis, and Delphi exercise. Clin Radiol. 2017;72(9):764–71.


treatment response evaluation in a patient with aortic prosthesis. J Cardiovasc Med. 2011;12(11):814–6.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part V**

**Evidence-Based PET in Neurology**

## **Evidence-Based PET for Neurological Diseases**

**13**

Alberto Miceli, Selene Capitanio, Maria Isabella Donegani, Stefano Raffa, Anna Borra, Matteo Bauckneht, and Silvia Morbelli

#### **13.1 Introduction**

Over the past two decades, one of the major breakthroughs for the approach to neurological diseases both in the clinical and research settings has been represented by the validation of diagnostic biomarkers able to demonstrate the presence of pathological mechanisms, alteration in neurotransmission as well as to predict disease progression [1, 2]. The use of PET with different tracers as well as other imaging biomarkers support the etiological diagnosis of neurological disorders in vivo. This approach is particularly relevant in the field of neurodegenerative diseases. In fact, neurodegenerative diseases are characterized by the progressive degeneration and death of neurons. They represent a heterogeneous group of conditions characterized by different etiologies, different neuropathological and neurochemical alterations leading to different clinical pictures and courses [3]. Indeed, an early accurate diagnosis allows to tackle the disease with available or experimental intervention, lifestyle changes, or logistical arrangements, before disability has developed. Early intervention is

e-mail: silviadaniela.morbelli@hsanmartino.it

expected to have greater clinical impact, extend independent and active life, improve its quality, and decrease the burden and costs of the disease [4]. However, the validation of PET tracers in neurological disease is still ongoing, and evidence on its comparative and combined diagnostic value with respect to other biomarkers is incomplete [4, 5]. As a matter of fact, the increasing pressure for cost-effectiveness requires systematic assessment and validation of all biomarker performance in the clinical settings. Similarly only an evidence-based approach to new PET tracers can allow to select the most promising tracers for PET imaging in the research field both for pathophysiological investigations and for upcoming diagnostic approaches.

#### **13.2 Evidence-Based PET in Neurodegenerative Dementia**

Although the use of PET tracers for neurotransmission is also actively investigated, the vast majority of PET tracers recently developed for the clinical and pathophysiological evaluation of neurodegenerative dementia aim to evaluate the presence of specific pathological proteins deposition or mechanisms underlying neurodegeneration [3]. Tracers targeting neuroinflammation are also under investigation in this field but their use is still very far from the clinical

A. Miceli · S. Capitanio · M. I. Donegani · S. Raffa

A. Borra · M. Bauckneht · S. Morbelli (\*)

Nuclear Medicine Unit, IRCCS Ospedale Policlinico San Martino, Genoa, Italy

Department of Health Sciences, University of Genoa, Genoa, Italy

setting [6]. Alzheimer's disease (AD) is by far the most relevant target for brain PET clinical imaging in neurodegenerative diseases [5]. The 2011 World Alzheimer Report has underlined that only a relatively small percentage of dementia cases (ranging between 20 and 50%) are identified and correctly diagnosed in the early stages. This evidence means that at least half of the population of dementia patients does not receive a complete diagnostic workup at disease onset. This diagnostic delay gives rise to a socalled "treatment gap" between early stages of the disease and formal diagnosis, thus delaying correct care and preventing organized support which ameliorates patient's quality of life (and positively impacting also on caregivers and family members). In the field of AD, two main categories of biomarkers have been proposed to identify the prodromal stage of disease [2, 7]. On the one side, amyloidosis biomarkers are able to identify the molecular/neuropathological feature of AD and include cerebrospinal fluid (CSF) amyloid-β1-42 reduction and brain amyloid accumulation as imaged through PET technology using radiopharmaceuticals that selectively bind to the fibrillar aggregates of amyloid-β plaques (AMY-PET) [2]. On the other side, neurodegeneration biomarkers reflect neuronal injury and downstream neurodegeneration, which can be measured by the increase of tau protein in the CSF, regional atrophy on MRI or synaptic metabolic dysfunction on fluorine-18 fluorodeoxyglucose PET (18F-FDG PET) [2]. As a matter of fact, the progressive dysfunction and loss of neurons lead to distinct involvement of functional systems and major clinical symptoms are mainly determined by the anatomical regions showing neuronal and synaptic dysfunction (which however do not necessarily reflect the molecular changes in the background) [8]. In this framework, MRI has both an exclusionary and inclusionary role for the early assessment of MCI. In fact on one side it can exclude secondary etiology of cognitive symptoms (i.e., vascular damage or normal pressure hydrocephalus) and it can increase the likelihood of a neurodegenerative dementia by highlighting the presence of atrophy in specific cortical regions [9]. Similarly, 18F-FDG PET is a well-founded method for evaluation of brain function and it is useful for the early diagnosis of AD and other dementias in people with mild cognitive impairment (MCI). 18F-FDG PET is a sensitive and specific imaging modality available to support the etiological diagnosis of the underlying neurodegenerative dementia in demented patients. In particular, hypometabolism in the temporoparietal lobe, assessed by qualitative visual interpretation of the scans, represents the typical pattern found in AD [10]. However, despite its widespread use and the well-established role in the clinical settings, the quality of the available studies and thus the role of 18F-FDG PET in identifying patients affected by AD who are still at the stage of MCI are less validated. As a matter of fact, discriminating between patients who will and will not progress to dementia due to AD is critical in the context of care and it must be pursued before patients have crossed the threshold into dementia. One of the background reasons associated with this surprising lack of evidence is related to evolving definition of MCI in the last decade. Unfortunately, the characteristics of the MCI patients recruited in the available published studies are quite diverse across the literature of the last 10 years and this heterogeneity is largely reflected but the lack of a reproducible value of sensitivity and specificity of 18F-FDG PET for the identification of MCI due to AD. Moreover, the sensitivity and specificity of 18F-FDG PET (as well as of all AD biomarkers) depend largely on the method of interpretation. It is, nowadays, advisable to use a combination of visual qualitative and semiquantitative analysis. In fact, the actual general sensitivity and specificity values, although still variable, are higher than 80% for both in the centers where it is utilized computer aided visual read approach such as Statistical Parametric Mapping (SPM), three-dimensional Stereotactic Surface Projection (3-D-SSP) statistics (Neurostat) or AD t-sum implemented in other software [11, 12]. Software able to combine information from MRI and 18F-FDG PET (even in the multicenter settings) have also been developed and have been demonstrated to improve detection and differentiation of AD and FTD (with 88–100% accuracy for AD detection) [13].

In more recent years, the availability of PET biomarkers of amyloid deposition such as 11C-PiB and 18F-labeled tracers (18F-Flutemetamol, 18F-Florbetaben, and 18F-Florbetapir) has gained great attention [14]. These tracers exhibit high affinity binding for fibrillary amyloid that is a hallmark of AD and can be found in moderate to high number in cortical gray matter in all cases of AD developing many years before the onset of dementia. The vast majority of preliminary studies with amyloid PET has been carried out with 11C-PIB. It has been reported by measuring 11C-PiB distribution in MCI and control subjects, that MCI patients who converted to AD, at an estimated rate of 25% per year, had greater 11C-PiB retention in several brain area with a 93.5% sensitivity and 56% specificity [15]. Moreover, none of the 11C-PiB negative MCI patients demonstrated to be converted to AD indicating that 11C-PiB negativity had a 100% negative predictive value for progression to AD [7, 16]. However, due to the 20-min half-life of 11C, 11C-PiB can only be used in PET centers with on-site cyclotron and radiopharmacy facilities while 18F is a more suitable radioisotope for widespread clinical use. Accordingly, the availability of fluorinated amyloid PET tracers has favored a greater impact of this tool also in the clinical settings [17]. In particular, a sensitivity of amyloid 18F-labeled tracers ranging from 89 to 97%, with a specificity ranging from 63 to 93% has been reported both for visual and semiquantitative analysis and no significant differences have been highlighted among the different agents [18].

The overall impacts of AMY-PET from the reported literature are a change of diagnosis and management in 35.2% and 59.6% of cases, respectively, leading to reduction in unnecessary investigations, treatments, relief of distress of caregivers, and potential involvement in clinical trials [19].

In particular, AMY-PET (as well as CSF data) are relevant for the differential diagnosis of etiologies underlying primary progressive aphasia (PPA). In fact, it has been suggested that the current clinical classification system for PPA should aim to predict underlying pathology across different cohorts and clinical settings which has a greater specificity with respect to neurodegeneration biomarkers (MRI and 18F-FDG PET) [20]. In fact, although patients' characteristics including age and ApoE genotype should be considered when interpreting AMY-PET, it has been highlighted that AMY-PET positivity is significantly more prevalent in logopenic variant of PPA, which belongs to the AD spectrum, (86%) than in non-fluent variant of PPA (20%) or semantic variant of PPA (16%) which should be part of the spectrum of fronto-temporal degeneration [21]. It should be noted however that cognitive normal older people has a high burden of amyloid defined as incidental amyloidosis. Indeed, the presence of amyloid deposition is not sufficient to define the presence of pathophysiologic processes associated with AD which has to include the concomitant tau deposition [22]. Analyses in AMY-PET positive cognitive normal subjects have shown that the episodic memory and global cognitive function consistently had the largest, albeit still modest, effect sizes between performance and amyloid burden, whereas executive function, working memory, processing speed, visuospatial function, and semantic memory exhibited relatively smaller effect sizes [23]. Moreover, the incidence of brain amyloidosis increases with aging in all non-AD populations and has been repeatedly reported in other neurological conditions such as in dementia with Lewy bodies (DLB) and Parkinson's disease (PD). Similarly, all amyloid PET tracers have affinity to amyloid in vessel walls, and can thus result in positive scans in case of cerebral amyloid angiopathy (CAA) [7]. In this framework, it has been demonstrated that the ratio between the occipital and the whole brain uptake is higher in patients with CAA than in patients with AD, possibly suggesting a more limited regional amyloid deposition on CAA [24]. Accordingly, AMY-PET has a moderate to good diagnostic accuracy for the identification of patients with CAA; especially as a negative scan it is useful to rule out CAA in the appropriate clinical setting [25]. Finally, it should be noted that soluble Aβ oligomers and neurofibrillary tangles (NFTs), the other histopathologic cores of Alzheimer's disease are not detected by this method, possibly producing false negative results [26].

#### **13.3 Evidence-Based PET for Movement Disorders**

Movement disorders are a group of neurologic syndromes characterized by an excess of movement or a scarcity of voluntary and automatic movements [27] related to different neurodegenerative or acquired central nervous system diseases affecting structures involved in movement control and modulation such as basal ganglia, cerebellum, cortex, and different thalamic nuclei. Parkinson's disease (PD) represents the second most common neurodegenerative disorder after AD and is typically characterized by motor and non-motor manifestations. Motor dysfunctions include bradykinesia, resting tremor, and muscular rigidity [28] as the result of dopaminergic deficit due to degeneration of the dopaminergic nigrostriatal system [29]. On the other hand, nonmotor features usually associated with the damage in non-dopaminergic pathways may include depression, olfactory and autonomic dysfunction, sleep disorders, psychiatric symptoms, pain, fatigue, and cognitive impairment [30] and can, in some case, even precede the motor phase by several years [31]. Besides the most prevalent PD, there are other related atypical parkinsonian syndromes (APS) also known as Parkinson-plus syndromes. From the neuropathological point of view, on the basis of the predominant multisystemic progressive accumulation of misfolded proteins, parkinsonian disorders could be classified into α synucleinopathies (PD with and without cognitive impairment/dementia, dementia with Lewy bodies—DLB, and multiple system atrophy—MSA) and tauopathies (corticobasal degeneration—CBD and progressive supranuclear palsy—PSP). All the aforementioned conditions must be distinguished from non-degenerative causes of parkinsonism (e.g., toxic/drug-induced, psychogenic, or vascular etiologies) in which presynaptic nigrostriatal pathways are preserved. Molecular imaging has proven to be a powerful tool for improving our understanding of the pathophysiology underlying parkinsonian disorders. In particular, PET and single photon emission computed tomography (SPECT) imaging are used as surrogate of unique in vivo functional measurement of monoaminergic pathways impairment in neurodegenerative diseases since the early 1980s [32]. Actually, these techniques are able to evaluate and detect nigrostriatal degeneration from different points of view. The analogue of levodopa 18F-fluorodopa ( 18F-DOPA) is a PET tracer that estimates the activity of aromatic amino acid decarboxylase enzyme (AADC), which converts levodopa into dopamine in striatal cells, thus reflecting dopamine synthesis. After its conversion in 18F-fluorodopamine, it follows the same fate of endogenous dopamine and is thus transported to intraneuronal storage vesicles by vesicular monoamine transporter 2 (VMAT2) [33] to then be released into the synaptic cleft and interact with postsynaptic dopamine receptors.

On the other hand, presynaptic dopamine transporter (DAT) density, responsible for reuptake of dopamine (DA) from the synaptic cleft and typically reduced in PD [34], can be measured through 123I-FP-CIT (123I-ioflupane), one of the most used radiotracers for SPECT imaging, or with specific PET tracers (18F-PE2I; *N*-(3 iodoprop-2Eenyl)-2β-carbomethoxy-3β-(4 methyl-phenyl)nortropane). Finally, specific radiolabeled ligands for VMAT2 also exist and can be used in PET imaging for research purposes (11C- and 18F-dihydrotetrabenazine). The applicability of functional imaging in the evaluation of striatal dopaminergic degeneration in control subjects and PD patients is well documented. Actually, PET and SPECT studies with AADC, VMAT2, and DAT tracers are able to monitor nigrostriatal deficit over time demonstrating the negative effects of age on DA transporters and receptors. According to Karrer et al. [35], age has a significantly larger effect on D1- than D2-like receptors with an average age reduction across the DA system of about 3.7–14.0% per decade. On the contrary, DA synthesis capacity seems to be spared from this detrimental effect. This finding nicely fit with the results of a contemporary meta-analysis of Kaasinen et al. [36] including PD patients submitted to PET and SPECT studies for the evaluation of striatal presynaptic dopamine function. In this meta-analysis, it was demonstrated a higher defect in DAT and VMAT2 function compared to AADC defect in PD patients. This difference is possibly due to the effect of a compensatory upregulation in AADC function, confirming vesicular monoamine transporter 2 and dopamine transporter as the most sensitive targets to assess. In the same study, a linear correlation between disease severity and dopamine loss was demonstrated and a mean motor disease duration of 4–7 years was needed to overcome the general compensatory changes in the terminal area of the nigrostriatal tract in PD and induce a detectable loss of about 50% of putaminal presynaptic dopamine function. It should be noted, however, that 18F-DOPA PET is a valid alternative to 123I-FP-CIT SPECT. Similarly, PET VMAT2 tracers as 11Cand 18F-labeled dihydrotetrabenazine are very sensitive in detecting presynaptic dysfunction in PD without significant influence by compensatory changes [37] but are less available than other tracers. Besides dopaminergic dysfunction, serotonergic system exerts an important role in PD pathophysiology. Among PET ligands that have been developed for the evaluation of serotonergic receptors and serotonin transporter (SERT), 11C-DASB is a second-generation PET ligand with the best selectivity for the SERT [38]. In their systematic review examining the in vivo SERT changes in PD measured by 11C-DASB PET, Pagano et al. [39] demonstrated that SERT dysfunction is associated with the development of several motor (tremor and dyskinesias) and non-motor symptoms (depression, fatigue, apathy, and weight changes). In particular, they observe a highest decrease in SERT binding in the rostral raphe and caudate followed by putamen, thalamus, ventral striatum, caudal raphe, and hypothalamus, according to succession of pathological events in serotonergic neurons described by Braak's PD staging [29]. Furthermore, reduced SERT binding in putamen is associated with longer duration of the disease, whereas lower 11C-DASB binding in caudate is associated with worse cognitive function and older age. Interestingly, although SERT binding in putamen decreased with the progression of the disease, PD patients with dyskinesias have relatively preserved putaminal serotonergic function with mechanisms not fully clarified.

Moreover, PET and SPECT are able to detect pathologic changes from the earliest phase of the disease, improving diagnostic accuracy in the early stages [40]. REM sleep behavior disorder (RBD) confirmed by polysomnography is a parasomnia occurring during REM sleep characterized by the loss of physiological muscle atonia and associated with dream-enacted behaviors. It is defined "idiopathic" when appears isolated, without any other clinical sign of a neurological disorder. However, more than 80% of idiopathic RBD patients will develop a definite neurodegenerative disease, mostly a synucleinopathy. Furthermore, the presence of abnormal presynaptic dopaminergic PET or SPECT scan is considered the second most relevant risk factor for prodromal PD [41] predicting a high risk for short-term conversion into a synucleinopathy in idiopathic RBD patients [42]. A recent metaanalysis about the role of presynaptic dopaminergic imaging in RBD shows that idiopathic RBD patients exhibit decreased nigrostriatal dopaminergic functioning in comparison with healthy controls, especially at the putamen level. Furthermore, patients with idiopathic RBD and patients with PD without RBD exhibit a similar degree of nigro-caudate dopaminergic deafferentation [43].

Molecular imaging can also facilitate the differential diagnosis among PD, atypical parkinsonian syndromes (APS), essential tremors, and other degenerative conditions that represent a group of complex and heterogeneous diseases with overlapping symptomatology and variable response to dopaminergic medications. 123I-FP-CIT SPECT can distinguish degenerative forms of parkinsonism from essential tremor [44], druginduced parkinsonism [45] and could also differentiate DLB from AD. Subjects with PD and APS show indeed an early reduction of striatal dopaminergic binding [46] usually before the appearance of motor symptoms. On the contrary, patients with essential tremor, drug-induced parkinsonism, and AD are characterized by normal dopamine transporter uptake [47]. Although a differential pattern has been described at the group level, a real distinction between the various degenerative forms of parkinsonism is not possible by means of SPECT assessment only. To this purpose, 18F-FDG PET has demonstrated to be more promising. As suggested by Albrecht et al. [48], glucose hypometabolism at 18F-FDG PET can identify consistent functional brain abnormalities in PD, outperforming structural MRI. In particular, while MRI showed only focal and inconsistent alterations, in PD patients the authors found an extensive glucose hypometabolism in bilateral inferior parietal cortex and left caudate nucleus that is related to cognitive deficits (inferior parietal cortex) and motor symptoms (caudate nucleus). Disease-specific patterns of regional glucose metabolism in patients with parkinsonism are well documented [49, 50]. However, the valuable capability of 18F-FDG PET for accurate differentiation between PD and APS has been unanimously accepted only in recent years. In a preliminary meta-analysis, Meyer et al. well described the different 18F-FDG uptake pattern in PD and APS [51]. PD is characterized by a posterior temporoparietal, occipital, and sometimes frontal hypometabolism with a relative hypermetabolism of the putamen, pallidum, thalamus sensorimotor cortex, pons, and cerebellum.

MSA patients show instead a marked hypometabolism of the putamen (mainly in its posterior portion), pons, and cerebellum, which may be more pronounced in the striatum or in the pons and cerebellum, on the basis of the clinical presentation. Conversely, PSP is characterized by a regional hypometabolism preferentially involving the medial, dorsal, and ventrolateral frontal areas (i.e., the anterior cingulate gyrus, supplementary motor area, precentral gyrus, and premotor-to-posterior prefrontal areas); caudate, thalamus, and upper brain stem. Finally, CBD patients have a typically highly asymmetric hypometabolism of the frontoparietal areas, striatum, and thalamus contralateral to the most affected body side. A concomitant cortical hypometabolism may be mainly found in the parietal cortex and usually extends across the sensorimotor cortex into the cingulate gyrus and premotorto-posterior prefrontal areas.

18F-FDG PET has proved to be particularly relevant also for the diagnosis of DLB and is actually listed among the supportive biomarkers for its identification [52]. DLB is characterized by a more prominent hypometabolism affecting the primary visual cortex and occipital cortex with relative preservation of subcortical structures and primary somatomotor cortex and with a concomitant hypometabolism in posterior associative cortex. In particular, the presence of a hypometabolism in the precuneus with a relative sparing of glucose uptake in posterior cingulate gyrus is known as "cingulate island sign" and has proved to significantly increase 18F-FDG PET specificity for the differential diagnosis with respect to AD, although lower than DAT SPECT [53].

18F-FDG PET evidence is also available for Huntington's disease gene expression carriers (HDGECs). Actually, several PET imaging studies investigating the glucose metabolism in HDGECs have shown specific metabolic patterns mainly characterized by a progressive reduction of subcortical and cortical glucose metabolism in the striatum, thalamus, insula, posterior cingulate gyrus, and prefrontal and occipital cortex associated to a relative hypermetabolism in the cerebellum and pons [54]. In particular, reduction in striatal metabolism seems to be an early feature of the disease, preceding neuronal loss and thus motor onset of the disease. However, glucose metabolism deficits are only one of the many factors involved in Huntington's disease and 18F-FDG PET is not indicated for the diagnosis of this disease. Besides 18F-FDG, other striatal PET radioligands have been used in this context. In particular, a meta-analysis aiming to investigate striatal molecular changes in 158 premanifest and 191 manifest HDGECs patients [10] demonstrates a significant decrease not only in glucose metabolism in caudate and putamen but also in dopamine D2 receptors and in striatal phosphodiesterase 10A binding. This findings well reflect the different neuropathological mechanisms underlying the development of the disease.

Finally, PET and SPECT could guide the clinician in the choice of the different therapeutic modalities and in monitoring therapy response. PET has proved to be able to determine drug dosage for optimal efficacy in movement disorders as in the case of PD patients under deprenyl treatment evaluated with 11C-deprenyl PET [55]. 18F-DOPA PET has been used in several studies to evaluate the effects of potential neuroprotective agents on dopaminergic function [56]. 11C-raclopride PET was used to evaluate striatal D2 receptor status in PD patients showing a normal or raised striatal D2 binding potential in untreated patients with PD but reduced in patients with PD and a fluctuating response to L-dopa [57].

Finally, PET and SPECT studies have proved the existence of a link between impulse control disorders (ICD) and dopamine activity dysfunction across ventral and dorsal striatum in PD patients [58]. ICD are a class of psychiatric disorders including pathological gambling, hypersexuality, binge-eating, and compulsive shopping that could appear in around 30% of PD patients as a complication of D2/3 dopamine agonist treatment and, to a lesser extent, levodopa. A recent meta-analysis demonstrated the PET/ SPECT dopaminergic striatal correlates of ICD in PD. In particular, in ICD patients authors find lower DAT levels in the dorsal striatum and in its subdivisions (i.e., putamen, caudate) and reduced binding (i.e., increased dopamine release) in the ventral striatum in response to reward-related stimuli or gambling task. These lines of evidence highlight the importance of PET pharmacokinetic and pharmacodynamic studies in our understanding of the mechanisms of action, efficacy, and complications of medical interventions in patients with neurological diseases.

#### **13.4 Evidence-Based PET for Psychiatric Disorders**

While in the clinical settings either 18F-FDG or AMY-PET can be used to support the differential diagnosis between neurodegenerative dementia and depressive pseudodementia, in the specific field of psychiatric disorders, PET technology has more extensively been used for research purposes to determine pathophysiology of diseases and response to intervention [59]. In particular, the availability of tracers for imaging of neurotransmission allows the investigation of different systems (i.e., serotonin, dopamine GABA pathways) and to investigate a wide range of psychiatric diseases. Similarly, brain PET with different tracers (i.e., tracers for neuroinflammation) has been used to explore new hypotheses related to the onset of psychiatric diseases. The vast majority of PET studies carried out in the last 20 years in psychiatric patients have aimed to map functional alterations and mechanism underlying major depressive disorder (MDD), a common mental illness with high lifetime prevalence (close to 20%.). Indeed, although the presence of aberrant brain activation during cognitive and emotional processing has been extensively evaluated in MDD patients, results of available studies vary considerably. Muller and colleagues summarized the evidence derived from neuroimaging experiments (using fMRI or PET) of group comparisons between adults with unipolar depression and healthy controls and reporting significant activation differences between patients and controls during emotional or cognitive tasks [60]. Several inconsistencies across available studies emerged from this systematic review. Brain metabolism in MDD was also specifically addressed in several voxel-based PET studies which were submitted to a meta-analytical approach by Su et al. [61]. Again while decreased prefrontal, insular, and limbic cerebral glucose metabolism was repeatedly highlighted in depressed patients with respect to healthy controls, available literature has not always been consistent. In this framework, the involvement of specific regions such as insula, limbic system, basal ganglia, thalamus, and cerebellum was more frequently reported, suggesting that these regions are likely to play a key role in the pathophysiology of depression. In keeping with these results, convergent change in the limbic-cortical brain circuit in depression compared to controls was also found in multi-modal imaging studies involving both PET and MRI data. Reported specific changes include lower gray matter volumes in amygdala, dorsal frontomedian cortex, and the right paracingulate cortex, as well as relative hypermetabolism in the right subgenual and pregenual anterior cingulate cortices. Building a strong and evidence-based mapping of these alterations in MDD is relevant as these regions could serve as a specific region-of-interest-fordisease template for both in vivo imaging in individual patients and postmortem histopathologic exploration [62]. Other PET studies investigated altered function related to different pathways including 5-HT receptor and transporter dysfunction in neuropsychiatric disorders. Indeed, impairment of serotonin receptor and transporter function is increasingly recognized to play a major role in the pathophysiology of neuropsychiatric diseases including anxiety disorder, major depressive disorder, bipolar disorder, and schizophrenia. In particular, a retrospective analysis revealed that these psychiatric disorders differed in affected brain regions, affected synaptic constituents as well as direction of dysfunction in terms of either sensitization or desensitization of transporter and receptor binding sites [63].

Striatal presynaptic dopamine pathway has been the most frequent target for PET and SPECT imaging in schizophrenia [64]. As a matter of fact, the role of striatal dopaminergic neurotransmission in the onset of symptoms of schizophrenia (including psychotic symptoms) is currently targeted by dopaminergic drugs. A specific marker of the integrity of presynaptic dopamine neurons in the striatum, the density of striatal dopamine terminals, can be quantified through molecular neuroimaging of DAT. A meta-analysis of DAT density in the striatum of schizophrenic patients demonstrated that striatal DAT density was not significantly different between patients and controls [64]. Similar negative findings were regionally confirmed in putamen and caudate. There was no moderating effect for external factors such as antipsychotic medication or illness duration. Accordingly, the authors concluded that loss of integrity of striatal dopaminergic synapses is not critical for the emergence of schizophrenia and these findings are relevant for further refining dopaminergic hypotheses of schizophrenia (with possible repercussion on interventional studies aiming to identify new treatment options). By contrast, a different window on presynaptic dopaminergic function was opened by studies exploring this pathway through 11C-/18F-DOPA PET [65]. Available studies were summarized in the meta-analysis by Fusar-Poli and Meyer-Lindenberg [65]: patients with schizophrenia showed increased striatal uptake as compared with controls and this finding was regionally confirmed in both caudate and putamen. Finally, no significant effect of age, illness duration, gender, psychotic symptoms, and exposure to antipsychotics was highlighted. Of note, sensitivity analysis confirmed robustness of meta-analytic findings. Finally, data from animal models and from postmortem studies suggest that schizophrenia is associated with brain GABAergic dysfunction. However, it is still unclear the extent of this effect in vivo studies of GABA function in patients with schizophrenia [66].

PET and SPECT studies with several tracers have been used on other neuropsychiatric diseases including obsessive compulsive disorders, Tourette's syndrome (TS), and attention deficit hyperactivity disorder (ADHD) as well as to test brain function in specific conditions such as in case of ecstasy/polydrug use and disorders of consciousness. In particular, dopaminergic imaging with PET and SPECT was able to demonstrate dopaminergic alterations in TS and pathophysiology and psychostimulant treatment of attention deficit in ADHD. In fact, dopaminergic alterations in TS are in line with the current pathophysiological hypotheses of a tonic-phasic dysfunction in the dopamine system although available studies are characterized by low effect sizes due to the heterogeneity of the disease [67]. Similarly, although dopaminergic studies in ADHD yielded inconsistent results often demonstrating either high and low striatal dopamine transporter levels across different studies, a systematic review of the available evidence highlighted that striatal dopamine transporter density in ADHD is depended on previous psychostimulant exposure, with lower density in drug-naive subjects and higher density in previously medicated patients [68]. More general approaches on brain perfusion and metabolism by means of SPECT and PET allowed to demonstrate that pharmacological and psychological treatments reduce resting cortico-striato-thalamo-cortical circuit activity in obsessive compulsive disorder. Similarly markedly reduced activity within midline cortical and subcortical sites (anatomical structures linked to the default-mode network) are present in patients with disorders of consciousness [69, 70]. Finally, PET studies on postsynaptic 5HT2A receptor imaging allowed to highlight that serotonin axons with the longest projections from the raphe nuclei might be more markedly affected by ecstasy/MDMA use.

#### **13.5 Conclusions**

In conclusion, a huge body of literature has highlighted an ongoing and promising role of PET with different tracers in neurodegenerative dementia, movement and psychiatric disorders. In some clinical settings such as in case patients with mild cognitive impairment and dementia, for the differential diagnosis of underlying etiology as well as for the early and accurate identification of patients with neurodegenerative parkinsonian syndromes, PET has gained an increasing relevant clinical role. By contrast the possibility to accurately quantify neurotransmission with different tracers is increasingly supporting the use of PET technology for pathophysiological and interventional studies in patients with psychiatric disorders. In both cases (clinical use of PET in neurodegenerative dementia and parkinsonian syndromes and research use of PET in psychiatric disorders), it is crucial to proceed with robust methodology which starts with the systematic evaluation of evidence-based results of previous studies. Only this approach will allow to balance costs and clinical advancement, thus meeting the needs of both patients and health-care systems.

#### **References**

1. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging and Alzheimer's Association workgroup. Alzheimers Dement. 2011;7:270–9.


studies comparing dopamine agonists and levodopa. Neurology. 2003;61(6 Suppl 3):S43–8.


metabolism in obsessive compulsive disorder: a metaanalysis. Sci Rep. 2017;7(1):17464.

70. Hannawi Y, Lindquist MA, Caffo BS, Sair HI, Stevens RD. Resting brain activity in disorders of consciousness: a systematic review and meta-analysis. Neurology. 2017;84(12):1272–80.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

**Part VI**

**Miscellaneous**

**14**

## **Meta-Analyses on Technical Aspects of PET**

Luca Ceriani

#### **14.1 Introduction**

In literature, there are few meta-analyses that have addressed technical and methodological issues concerning positron emission tomography (PET) imaging despite their important role in determining the quality of the diagnostic results [1–19].

#### **14.2 Factors Affecting 18F-FDG Uptake**

The factors that may affect fluorine-18 fluorodeoxyglucose (18F-FDG) uptake of normal tissues/ organs and tumour lesions have been explored by three studies.

Wang et al. [1] demonstrated that the impact of time interval on standardized uptake value (SUV) in liver and mediastinal blood pool was relatively medium but clinically noticeable. Due to the rare studies, this relationship remains to be verified for other organs (cerebellum, spleen, bone marrow, muscle, bowel and adipose tissue).

Clinic of Nuclear Medicine and PET/CT Centre, Imaging Institute of Southern Switzerland, Ente Ospedaliero Cantonale, Bellinzona and Lugano, Switzerland

Institute of Oncology Research (IOR), Bellinzona, Switzerland e-mail: luca.ceriani@eoc.ch

Nevertheless, other factors such as body mass index and blood glucose level (BGL) appeared to be important in determining 18F-FDG uptake in normal organs.

The effect of BGL on 18F-FDG uptake and SUV has been more extensively explored by Eskian et al. [2] who demonstrated a correlation between increased BGLs and increased SUVmax and SUVmean values in liver and blood pool. Conversely, an increase of BGL is significantly associated to lower SUVmax and SUVmean in brain and muscle while both SUV values in tumours seemed to be affected, with significant reduction, only by BGL >200 mg/dl. The authors concluded suggesting that in patients with BGL lower than 200 mg/dl no interventions are needed for lowering BGL, unless the liver is the organ of interest. Nevertheless, new studies are warranted to evaluate sensitivity and specificity of 18F-FDG PET for diagnosis of malignant lesions in patients with hyperglycaemia.

The uptake of 18F-FDG in brown adipose tissue (BAT) is another finding that may affect the detection of tumour lesions. Hou et al. [3] demonstrated that gender, season and age are risk factors for 18F-FDG uptake in BAT. In particular, the 18F-FDG uptake rate was 2.16 times in females as that in males, 8.67 times in the minors as that in the adults and 1.94 times in winter as that in summer.

L. Ceriani (\*)

<sup>©</sup> The Author(s) 2020 139

G. Treglia, L. Giovanella (eds.), *Evidence-based Positron Emission Tomography*, https://doi.org/10.1007/978-3-030-47701-1\_14

#### **14.3 Repeatability of the Quantitative Measurements**

PET is widely used in oncology for the response assessment to treatment by the quantitative measurements of tracer uptake of the tumour lesions. For this purpose, the repeatability of these measurements in metabolic imaging is pivotal and needs to be established. Two studies dealt with the repeatability of the SUV estimation in tumour lesions [4, 5].

De Langen et al. [4] demonstrated that in 18F-FDG PET imaging SUVmean had better repeatability performance than SUVmax. For serial PET scans, a combination of 20% as well as 1.2 SUVmean units was most appropriate threshold to identify a significant metabolic change in tumoural lesions. Nevertheless, both measures showed poor repeatability for lesions with low 18F-FDG uptake since test-retest variability is affected by the level of 18F-FDG uptake while tumour volume had minimal influence on repeatability. The authors recommend to report the evaluation of biologic effects in PET by using a combination of minimal relative and absolute changes of SUV.

The same group also analysed the response evaluation using 18F-Fluorothymidine (18F-FLT) [5]. In this case, the best repeatability was obtained using SUVpeak. Differences ≥25% in 18F-FLT SUV measurements likely represented a true change in tumour uptake. Nevertheless, larger differences are required for FLT metrics comprising volume estimates when no lesion selection criteria are applied.

The partial volume effect is another factor that may hamper accurate quantification of radiopharmaceutical uptake by tumour lesions leading to underestimations of SUV values and possibly compromising the lesion detection. A metaanalysis [6] investigated the clinical impact of the partial volume effect correction factor in oncological PET studies and in particular the potential benefit in its application for diagnosis, staging, prognostication and response assessment concluding that the accumulated evidence does not support routine application of partial volume correction in standard clinical PET practice.

#### **14.4 Dual-Time-Point Imaging**

A second late acquisition after conventional 18F-FDG PET/CT imaging (dual-time-point technique) has been suggested to discriminate between inflammatory and neoplastic lesions. This approach is based on the evidence that in inflammatory lesions the 18F-FDG uptake is characterized by a progressive washout after an initial trapping, while in tumour tissues, in particular before treatment, the uptake of the tracer increases over time. Two meta-analyses addressed this issue: both showed comparable performance between standard single-time-point and dualtime-point 18F-FDG PET imaging in diagnosing pulmonary nodules [7] and in detecting lymph nodal metastases [8]. The results of the studies do not support the routine use of an additional late acquisition for these two clinical purposes.

#### **14.5 Correlation Between Proliferation Markers (Ki-67) and Tracer Uptake in Tumours**

Although 18F-FDG is not a tumour-specific agent, several studies showed that 18F-FDG uptake may be an index of biological aggressiveness of the disease. Nevertheless, whether 18F-FDG PET imaging can be a marker of tumour cell proliferation remains controversial. Deng et al. [9] analysed pooled data from clinical studies focused on this issue. The results demonstrated a moderate positive correlation between 18F-FDG uptake and tumour cell proliferation marker Ki-67 (combined correlation coefficient = 0.44) and suggested that 18F-FDG SUV may be used as an indicator of the tumour proliferation and invasiveness. A subgroup analysis based on different tumour types showed varied degrees of correlation. The correlation was highly significant in thymic epithelial tumours; significant in gastrointestinal stromal tumours (GIST); moderate in lung, breast, bone and soft tissue, pancreatic, oral, thoracic, uterine and ovary cancers; average in brain, oesophageal and colorectal cancers; and poor in head and neck, thyroid, gastric and malignant melanoma tumours.

The correlation between 18F-FLT uptake and Ki-67 was also investigated. Chalkidou A et al. [10] found sufficient data to support a strong 18F-FLT/Ki-67 correlation only for brain, lung and breast cancer. The authors highlighted the importance of the methodology used to measure Ki-67 expression: the correlation was significant and independent of cancer type only when using Ki-67 average measurements, or measuring Ki-67 maximum expression on whole surgical samples.

#### **14.6 Correlation Between 18F-FDG SUVmax and ADC Values in Tumour Tissues**

Apparent diffusion coefficient (ADC) is a parameter obtained by diffusion-weighted magnetic resonance imaging (MRI), reflecting the brownian movement of water molecules. The ADC value has been shown to link with the cell density, microvascular circulation and membrane integrity of a tumour tissue. Two meta-analyses [10, 11] examined the potential relationship between 18F-FDG SUV that characterize the metabolic activity of tumour cells and ADC. Both studies found inverse correlation between ADC and SUV in patients with cancer. This inverse correlation, which was generally weak, appeared higher in the brain tumour, cervix carcinoma and pancreas cancer. However, larger prospective studies are warranted to validate these preliminary findings in different cancer types.

#### **14.7 Diagnostic Performance of Hybrid Imaging in Oncology**

After the introduction in the last years of hybrid scanners, many experiences indicated that the integration of functional and morphological imaging (hybrid imaging) provides additional diagnostic information useful in different clinical settings and particularly in oncology.

In a meta-analysis published by Gao et al. in 2013 [13], pooled data from comparative studies revealed that integrated PET/CT has higher sensitivity (0.95 vs 0.85) and similar specificity (0.96 vs 0.95) with respect to PET alone in the detection of distant metastases. Analogous results were obtained comparing integrated PET/CT with CT alone (sensitivity 0.97 vs 0.80 and specificity 0.97 vs 0.94, respectively), confirming the additional value of the PET/CT hybrid imaging in tumour staging.

More recently published data suggested a complementary role of 18F-FDG PET/CT and MRI in oncological patients. Miles et al. [14] compared these two imaging modalities in patients with suspected residual disease or recurrent tumours. PET demonstrated greater sensitivity for detecting lymph nodal recurrence, whereas MRI was more effective than PET/CT in the detection of skeletal and hepatic recurrence. A review of studies assessing therapeutic impact of PET/MRI suggested a greater likelihood for change in clinical management when PET/MRI was used for assessment of suspected residual or recurrent disease rather than tumour staging. Supplementing the evidence-base data for 18F-FDG PET/MRI with studies that compared the components of this hybrid technology separately, 18F-FDG PET/MRI is likely to be clinically effective for the investigation of patients with suspected residual or recurrent cancers.

Xu et al. [15] demonstrated that 18F-FDG PET/CT has similar patient-based sensitivity (0.85 versus 0.85) and specificity (0.96 versus 0.97) to MRI in the detection of distant metastases. Similar lesion-based performance was also estimated (PET/CT sensitivity and specificity: 0.85 and 0.90 and MRI sensitivity and specificity: 0.88 and 0.89). The analysis of a small number of studies indicated that the combined use of these two modalities may have higher patientbased sensitivity (0.89) than PET/CT (0.82) and whole body MRI (0.81) alone, suggesting that the combined use of these two modalities may provide more benefit than PET/CT and MRI alone.

Finally, Shen et al. [16] after analysing the results of 38 studies that involved 753 patients and 4234 lesions concluded that PET/MRI has excellent diagnostic potential for the overall detection of malignancies in cancer patients. On a per-patient level, the pooled sensitivity and specificity were 0.93 and 0.92, respectively. On a per-lesion level, the corresponding estimates were 0.90 and 0.95, respectively.

#### **14.8 Varia**

The 18F-FDG PET/CT imaging has been defined by several authors as more accurate than standard radiological imaging in evaluating the response to treatment in oncological patients, in particular when a residual mass is still detectable. Kim et al. [17] compared the tumour response assessment according to the metabolic criteria developed by the European Organization for Research and Treatment of Cancer (EORTC) and morphologic criteria (RECIST1.1.) in patients with malignant solid tumours. The pooled analysis of 181 patients recruited from seven studies demonstrated a moderate agreement of tumour responses between the RECIST and EORTC criteria (*k* = 0.493). The level of agreement was not affected by the anticancer treatments (chemotherapy or targeted therapy). A disagreement was found in 66 of 181 patients (36.5%). Tumour response was upgraded in 54 patients and downgraded in 12 when adopting the EORTC criteria. The estimated overall response rates were significantly different between the two criteria (52.5% by the EORTC vs. 29.8% by the RECIST, *p* < 0.0001). The conclusions confirmed that the metabolic findings are more sensitive than the morphologic criteria to detect tumour response to the treatment.

The PET/CT imaging with radiolabelled choline is a reliable tool for the detection and localization of recurrent disease in patients with prostate carcinoma. A meta-analysis by von Eyben et al. [18] investigated whether the use of different tracers, 11C-choline (11C-Cho) and 18F-fluorocholine (18F-FCH), may provide different diagnostic performance. The detection rates of metastatic sites in studies with 11C-Cho and 18F-FCH did not differ significantly. The radiation activity of 11C-Cho and 18F-FCH injected was not significantly associated with the detection rate of extra-prostatic lesions. The authors concluded that the detection of metastatic lesions in patients with biochemical recurrence (PSA levels of 1–10 g/ml) was clinically relevant when performed by PET/CT with radiolabelled choline regardless of the radiotracer injected.

The introduction of hybrid medical imaging technology has transformed the practice of diagnostic nuclear medicine and nowadays PET/CT and single photon emission computed tomography/computed tomography (SPECT/CT) have wide acceptance for many clinical investigations. A concern with PET/CT and SPECT/CT imaging is the combined radiation doses from both radiopharmaceutical and X-ray CT components. Therefore, it is imperative to implement a radiation dose optimization process to protect patients from unwarranted high radiation burdens. Alkhybari et al. [19] systematically reviewed data published in literature to determine the variations in reported national diagnostic reference levels (NDRL) methodology and values for adult PET/CT and SPECT/CT procedures. Discrepancies were found between the methodologies applied to establish and report both PET/ CT and SPECT/CT NDRLs. In particular, the authors remarked the opportunity for hybrid imaging to report both radiation doses from the radioactivity injected and the CT dose rather than a separate NDRL. They concluded that further researches should be focused on reporting more NDRLs for hybrid examinations to collect enough data to establish a robust NDRL standard for the CT portion in PET/CT and SPECT/CT examinations.

#### **References**


18F-FDG uptake measurements in tumors: a metaanalysis. J Nucl Med. 2012;53(5):701–8.


in malignant disease. Contrast Media Mol Imaging. 2017;2017:4729547.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.