Principles and Practice of PET/CT

Home , Immunoscintigraphy, Osem (company), PET radiotracer

EANM® Eu ne rope edici an Association of Nuclear M

Publications · Brochures

Part 2 A Technologist‘s Guide

Produced with the kind Support of Editors

Giorgio Testanera Wim J.M. van den Broek Rozzano (MI), Italy Nijmegen, The Netherlands

Contributors

Ambrosini, Valentina (Italy) Kaufmann, Philipp A. (Switzerland) Beijer, Emiel (The Netherlands) Krause, Bernd J. (Germany) Bomanji, Jamshed B. (UK) Królicki, Leszek (Poland) Booij, Jan (The Netherlands) Kunikowska, Jolanta (Poland) Bozkurt, Murat Fani (Turkey) Mottaghy, Felix Manuel (Germany) Carreras Delgado, José L. (Spain) Nobili, Flavio (Italy) Chiti, Arturo (Italy) Oyen, Wim J.G. (The Netherlands) Darcourt, Jacques (France) Pagani, Marco (Italy) Decristoforo, Clemens (Austria) Papathanasiou, Nikolaos D. (Greece) Delgado-Bolton, Roberto C. (Spain) Petyt, Gregory (France) Dennan, Suzanne (Ireland) Poepperl, Gabriele (Germany) Elsinga, Philip H. (The Netherlands) Prompers, Leonne (The Netherlands) Fanti, Stefano (Italy) Schwarzenboeck, Sarah (Germany) Gaemperli, Oliver (Switzerland) Semah, Franck (France) Giammarile, Francesco (France) Souvatzoglou, Michael (Germany) Giurgola, Francesca (Italy) Tarullo, Gianluca (Italy) Grugni, Stefano (Italy) Tatsch, Klaus (Germany) Halders, Servé (The Netherlands) Testanera, Giorgio (Italy) Hanin, François-Xavier (Belgium) Urbach, Christian (The Netherlands) Houzard, Claire (France) Vaccaro, Ilaria (Italy) Jamar, François (Belgium)

2 Contents Chapter 2 – Clinical applications of PET/CT inoncology Chapter 2–Clinical applications ofPET/CT Chapter 1 – Principles of PET radiochemistry Chapter 1–Principles radiochemistry ofPET Wim vandenBroek andGiorgio Testanera Introduction Suzanne Dennan Foreword 1.5 F 1.4 [ Philip H.Elsinga Valentina Ambrosini andStefano Fanti 2.8 Neuroendocrine tumour and Leszek Królicki Kunikowska Jolanta 2.7 Digestive tumours tract L.Carreras andJosé Delgado C.Delgado-Bolton Roberto 2.6 Lymphomas J. Souvatzoglou andBernd Krause Michael Sarah Schwarzenboeck, 2.5 Prostate cancer Leonne Prompers, HaldersandFelix EmielBeijer, Mottaghy Servé Manuel Urbach, Christian 2.4 Breast cancer Francesco andClaire Giammarile Houzard 2.3 Lung cancer FaniMurat Bozkurt 2.2 Headandneckcancer Chiti Wim J.G. andArturo Oyen 2.1 In D. B.Jamshed andNikolaos Bomanji Papathanasiou 1.3 Basics of 1.3 Basics Gianluca Tarullo andStefano Grugni 1.2 Common tracer practice inPET control quality Clemens Decristoforo 1.1 In rancesca Giurgola, Ilaria rancesca Giurgola, Ilaria Vaccaro 68 18F G troduction troduction troduction to PET radiochemistry radiochemistry totroduction PET ]FDG syn a-peptide tr 18 F and thesis acers 11 C radiopharmac

and Giorg eutical chemistry eutical chemistry io Testanera 3 102 26 18 89 80 67 60 50 44 43 43 32 10 8 EANM8 6 5 Contents

Chapter 3 – Clinical applications of PET/CT in infection and in%ammation 110 François-Xavier Hanin and François Jamar

Chapter 4 – Clinical applications of PET/CT in cardiology 121 Oliver Gaemperli and Philipp A. Kaufmann

Chapter 5 – Clinical applications of PET/CT in neurology 139 5.1 Introduction to brain PET 139 Jacques Darcourt

5.2 Dementia and related disorders 141 Flavio Nobili

5.3 Psychiatry 146 Marco Pagani

5.4 Movement disorders 150 Jan Booij

5.5 Brain tumours 153 Klaus Tatsch and Gabriele Poepperl

5.6 Epilepsy 159 Franck Semah and Gregory Petyt

Imprint 163

4 eiie n t pooe ih standards high promote to and medicine nuclear of areas developing and key in tise “Technologist’s Guide” aims to develop exper- The 2004. since basis annual an Guide”on successful highly “Technologist’sa produced Technologist EANM The in%ammation andneurology. infection, cardiology,oncology, in PET/CT of radiochemistry and to the clinical applications PET to devoted series, PET/CT the of 2 part introduce to privilege great a is It care. tient pa- to control quality and assurance quality protection, radiation practical from ranging topics included and PET/CT of practice and of the series PET/CT focussed on the principles imaging.1 Part PET/CT hybrid of topic the to that the next three guides would be dedicated In 2010, the Technologist Committee decided examinations. PET/CT management and care of patients undergoing of competent and re%ective personnel in the involvement for requirement the increased rapid advancement of techniques PET/CT has the years, recent In practice. professional of Suzanne Dennan Foreword omte has Committee 5 the authors, who have ensured the outstand- of work hard and e&orts the forgrateful am I of immense help toprofessionals help all immense of working be will book PET/CT this that con*dent am I series. PET/CT the three-part for their support and generous sponsorship of ries. Special thanks are due to Siemens Medical se- annual our for time short record a within dedication in preparing and editing this guide Wim van den Broek, for their enthusiasm and indebted to the editors, Giorgio Testanera and ing quality of this book. PET/CT I am very much Committee Technologist EANM Chair, Dennan Suzanne bookin2012. PET/CT toforward thenext look I service. patient PET/CT the of quality comprehensive tool that will contribute to the nologist’s Guide” will prove to be a useful and “Tech- eighth this that hoped is It PET/CT. wouldto developlike their understanding of in departments PET/CT and also to those who

EANM Introduction Wim van den Broek and Giorgio Testanera

Owing primarily to the wide availability of lent overview of all the issues and aspects 18F-FDG, PET/CT has established its place in related to the preparation of PET tracers. the diagnosis and management of several prominent diseases, especially in the *eld t Chapter 2 presents the clinical applications of oncology. In recent years, an increasing of PET/CT in oncology, with a review of the number of alternative new PET radiopharma- strengths, weaknesses, current evidence ceuticals have become commercially avail- and future directions of this imaging tech- able, opening the way to new applications nique over a wide range of tumours. of PET/CT imaging. Since a greater variety of biological functions can be visualised by PET/ t PET/CT applications in the detection and CT with these new positron radiopharmaceu- follow-up of infection and in%ammatory ticals, the number of PET/CT applications will disease are rather new and still unrecog- increase in cardiology and neurology as well nised by some. Chapter 3 reviews the main as in oncology. indications for 18F-FDG PET/CT in this *eld and discusses the advantages and pitfalls This second PET/CT Technologist’s Guide will compared with imaging using labelled be of value not only for nuclear medicine white blood cells. technologists but also for other professionals working with PET/CT. As the *rst book cov- t The introduction of new PET tracers such ered instrumentation, protocol optimisation, as 82Rb and the increasing availability of radiation protection and patient care issues, PET devices in combination with high-end this book provides the reader with informa- CT scanners is opening the way for increas- tion on the principles of PET radiochemistry ing use of PET/CT in cardiology. Chapter 4 and the current state of clinical applications discusses the clinical applications in the of PET/CT in the *elds of oncology, cardiology, *eld and covers patient preparation and infection and in%ammation and neurology. PET/CT protocols. t The *rst chapter covers the principles of t Imaging of the brain with PET/CT is still an PET radiochemistry. In addition to present- open *eld with many possibilities. Thanks ing the basic knowledge on PET chemis- to the large number of PET tracers for brain try, it also discusses the regulatory rules imaging that are either already available or which require increasing awareness of the expected to become available. This *nal challenges involved in the production and chapter o&ers an overview of the current quality assurance of PET radiopharmaceuti- applications of PET/CT imaging in di&erent cals. This chapter o&ers the reader an excel- brain diseases.

6 posed by multimodality approaches involv- approaches multimodality by posed lished in 2012, will address the great challenge PET/CT, the third book in this series, to be pub- of applications clinical and care patient tion, protec- radiation technology, on volumes good teamwork between the professionals the between teamwork good chemists, pharmacists and technologists. Only professionals, including physicians, physicists, of number a of input expert the on pends de- imaging PET/CT of quality ultimate The radiotherapy applications of imaging.PET/CT the consider will it particular, In PET/CT. ing After the broad view o&ered by the *rst two two *rst the by o&ered view broad the After I n troduction 7 are interested inthistopic. or PET/CT of area the in work who fessionals that it will prove a valuable resource for all pro- We hope that you enjoy reading this book and tion andtheEANM Technologist Committee. In%amma- and Infection on GroupTask and Committees Neurology and Cardiovascular Oncology, Radiopharmacy, EANM the tween be- teamwork excellent the re%ects quality also applies to this booklet, in which the good disease. The importance of such collaboration of diagnosis high-quality ensure ultimately will imaging PET/CT of area the in working

EANM Chapter 1 – Principles of PET radiochemistry 1.1 Introduction to PET radiochemistry Clemens Decristoforo

PET/CT has seen tremendous growth over recently by drafting Guidelines and Guidance cent years, mainly driven by the wide use and documents on Good Radiopharmacy Prac- availability of 18F-FDG. Imaging with 18F-FDG tice for the Small-Scale Preparation of Radio- has now become the clinical standard for a pharmaceuticals. number of indications, especially in oncology. However, a major potential of molecular The role of the technologist in such an en- imaging with PET lies in the great variety of vironment requires increasing awareness biological functions that can be visualised, especially regarding challenges in the pro- including other transporters, receptors and duction and quality assurance of such highly enzymes. speci*c pharmaceuticals. Basic knowledge on the chemical processes involved and the Even though an increasing number of alter- technical challenges encountered in every- native PET radiopharmaceuticals are becom- day practice and an understanding of the ing widely commercially available based on necessity to work in an environment with supply from commercial producers – a trend high demands on quality assurance are of expected to increase in the future –, the local the utmost importance in ensuring compli- production of PET radiopharmaceuticals will ance with the highest standards, as patients remain a central pillar in the years to come. rightfully demand. Especially the use of ultra-short-lived radionuclides such as 11C, but also the increased use This technologists’ booklet provides an ex- of PET radionuclides from generators such as cellent overview of major principles of PET 68Ga and the wide interest in PET radionuclides chemistry. Section 1.2 gives an overview of with longer half-lives such as 124I, 64Cu and 89Zr, quality control procedures for PET radiophar- requires the in-house production of PET radio- maceuticals, providing an understanding of pharmaceuticals based on individual patient the daily procedures necessary to ensure needs or for use in clinical trials. the e+cacy and safety of PET radiopharmaceuticals. The following section on basic Regulatory demands concerning the prepa- principles of 18F and 11C provides information ration of radiopharmaceuticals in Europe on how major PET radiopharmaceuticals are have become a major hurdle in the estab- prepared and which challenges need to be lishment of PET radiopharmaceutical produc- overcome in these procedures. Section 1.4 tion, particularly in hospitals and academic then describes the synthesis of FDG, giving institutions. The Radiopharmacy Committee details on the principles of preparation of the of EANM has addressed these problems re- work-horse of PET. In the *nal section relat-

8 Chapter 1 – Principles of PET radiochemistry: 1.1 Introduction to PET radiochemistry

ing to PET radiochemistry, speci*c aspects of 68Ga are addressed, in particular peptide radiopharmaceuticals.

We believe that this booklet will help not only technologists but also other nuclear medicine professionals not directly involved in PET radiopharmaceutical production to understand the challenges and prospects of PET chemistry and radiopharmaceutical preparation, thereby improving their professionalism to the bene*t of patients. EANM

9 Chapter 1 – Principles of PET radiochemistry 1.2 Common practice in PET tracer quality control Gianluca Tarullo and Stefano Grugni

Introduction The radiochemical purity is the fraction of The use of radiopharmaceuticals in humans the total activity of the product in the desired requires a system of quality control tests that chemical form. Radiochemical impurities may ensure the purity, safety and e+cacy of the originate from: products. Some of these tests are normally carried out for conventional drugs while oth- - radionuclide production; ers are speci*c for radiolabelled compounds. - chemical synthesis of the product; Quality control tests of these compounds, - incomplete puri*cation; which are short-lived products, should be very - chemical changes after synthesis. fast and simple. It is possible to distinguish two categories of test: physicochemical tests and For example in the case of the most common biological tests [1]. PET tracer, [18F]FDG, di&erent impurities may be created during the synthesis process. [18F] Physicochemical tests FDM and [18F]TAG are two collateral products Physicochemical tests are used to ensure the of di&erent synthesis methods. Partially hy- purity and the integrity of radiopharmaceuti- drolysed labelled precursors are an example cals and are based on the chemical identities of incomplete puri*cation, as are 18F- free of the substances and their physical properties. ions.

Appearance and pH Other by-products may originate during stor- In normal conditions, a solution of radiophar- age of the radioactive product for radiolysis maceutical should not contain particulate. [2,3] owing to: The measure of pH is very important because both the stability of the product and - high pH values; its compatibility with human administration - high temperature; are dependent on the pH value. In order to - high concentration of activity. investigate the pH value, di&erent methods are available, but measurement with a pH Usually the minimum level of radiochemical meter is certainly more accurate than the purity required for the administration of radio- colorimetric test with pH paper. pharmaceuticals is 95% of the total activity, but the exact limit for each product is reported Radiochemical and chemical purities in the speci*c Ph. Eur monographs. Physicochemical tests also include those for the determination of radiochemical and These tests require methods based on the chemical purities. separation of the various chemical radioac-

10 Chapter 1 – Principles of PET radiochemistry: 1.2 Common practice in PET tracer quality control

tive compounds present in the product. The The value that serves to identify compounds most common method is chromatographic in both TLC and other chromatographic tech- analysis. Chromatography is a technique for niques is the Rf, the ratio of the distance trav- separation of di&erent elements in a mixture elled by the compound to the distance from of substances based on the di&erent eluo- the original point to the solvent front. To obtain tropic strengths of the mobile phase [4]. The the results of the analysis when the solvent ar- most simple chromatographic method useful rives at the required distance, the strip should for determination of radiochemical purity is be removed from the chamber with the mobile thin-layer chromatography (TLC; Fig. 1). In phase; it is then possible to evaluate the radio- TLC a small amount of the product is spot- chemical purity in two ways. The *rst method ted on one side of a paper or a strip made is to divide the strip into small segments and of several substances such as silica gel or measure the activity of each segment. Radio- EANM aluminium oxide and this side is then im- chemical purity is the ratio between the activity mersed in a mobile phase that may include of the segment where the product should be physiological solution, organic solvents and expected to run and the total activity, which bu&ers. The mobile phase runs on the strip is the sum of the activities due to the desired and the di&erent compounds distribute compound, the free isotope and any other themselves and migrate separately accord- impurities. The alternative method employs a ing to the di&erent a+nity with the mobile radiochromatographic scanner that measures and the stationary phase. the activity at each point of the strip (Fig. 2). IRCCS Istituto Clinico Humanitas, Rozzano (Milan) Courtesy of Departement of Nuclear Medicine, of Departement Courtesy Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Clinico Humanitas, Istituto IRCCS Figure 1: Thin-layer chromatography Figure 2: Radiochromatographic scanner

11 Figure 3 shows an example of an [18F]FDG TLC Normal-phase methods are based on the use report. of polar materials as the stationary phase and of non-polar solvents, usually mixed with a small amount of polar solvents. These methods are useful for the separation of polar compounds.

In reverse-phase methods, non-polar materials are used as the stationary phase. Usu- ally this kind of stationary phase is made up of hydrophobic alkyl chains with di&erent lengths, named C4, C8 or C18 depending on

Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Clinico Humanitas, Istituto IRCCS their length. Longer chains are useful for small Figure 3: [18F]FDG TLC report molecular analysis and shorter ones for proteins and peptides. An alternative method for determination of the radiochemical purity is high-performance/

pressure liquid chromatography (HPLC, Fig. 4). IRCCS Istituto Clinico Humanitas, Rozzano (Milan) These two techniques are based on the typical of Nuclear Medicine, of Departement Courtesy principles of liquid chromatography. In HPLC, mobile phases are pushed into the columns by electrical pumps, and the stationary phase is packed in a column. Working under a high pressure condition, it is possible to separate di&erent compounds with high resolution.

Generally, HPLC columns are stainless steel tubes packed with appropriate packing materials. Depending on the composition of the Figure 4: HPLC system stationary phase, it is possible to distinguish two types of chromatographic method:

- normal phase; - reverse phase.

12 Chapter 1 – Principles of PET radiochemistry: 1.2 Common practice in PET tracer quality control

In HPLC methods, a small amount of the sample is manually injected into an injection valve. After the injection, the substances present on the sample are separated, eluted from the column depending on their a+ nity with the stationary and the mobile phase and analysed using a detector. Various detectors are used in the analysis of radiopharmaceuticals for the determination of chemical and radiochemical purities, as shown in Table 1. Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Clinico Humanitas, Istituto IRCCS

Figure 5: [18F]FDG HPLC report Figure 5 shows an example of an [18F]FDG EANM HPLC report.

Radiopharmaceutical Chemical purity Radiochemical purity

[18F]Fluorodeoxyglucose UV, amperometric detector Radiochemical detector

[11C]Methionine UV detector Radiochemical detector

[11C]Choline Conductivity, UV detector Radiochemical detector

Table 1: Detectors commonly used for PET tracer quality control

Systems like those shown in Table 1, with a Several radiochemical synthesis processes radiometric and a non-radiometric detector need solvents to obtain the required product. [5,6], allow determination within the same In these cases the possible presence of a small analysis of both the radiochemical and the amount of residual solvents must be evalu- chemical purity of the product. ated. A common technique for this purpose is gas chromatography (GC; Figs. 6, 7).

13 compounds are identi*ed with a detector. The most common detector used in GC is a %ame ionisation detector (FID), which is able to work with a wide range of compounds and concentrations.

In order to investigate the chemical purity of a product, di&erent tests can be used to research the presence of a speci*c impurity. One

Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Clinico Humanitas, Istituto IRCCS example is the determination of the amount Figure 6: A GC system of Krypto*x often present in most F-18 labelled PET radiopharmaceuticals, such as [18F]FDG. The technique used for this purpose is a colorimetric test. A small aliquot of the product is placed on a test paper pretreated with an iodoplatinate reagent and the spot is compared with that of a reference solution obtained using a sample of Krypto*x at the limit concentration of Krypto*x [7].

Radionuclidic purity Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Clinico Humanitas, Istituto IRCCS Radionuclidic purity is evaluated for inves- Figure 7: An [18F]FDG GC report tigation of the isotopic forms present in a In GC, the mobile phase is an inert carrier gas radioactive substance. As reported in the “ra- and the stationary phase is usually a polymer diopharmaceutical preparations” monograph, on an inert solid support. The principles of this radionuclidic purity is the ratio, expressed as a technique are similar to those of liquid chro- percentage, of the radioactivity of the radio- matography. In GC analysis, a small amount of nuclide concerned to the total radioactivity of the sample is injected into the column, which the radiopharmaceutical preparations. Radio- is placed in an oven. Di&erent compounds are nuclidic purity can be calculated using gam- retained from the stationary phase absorbed ma spectrometry. It can be also calculated, in on the column, depending on their chemical case of short life impurities, by determination and physical properties. At column exit the of the isotope half-life.

14 Chapter 1 – Principles of PET radiochemistry: 1.2 Common practice in PET tracer quality control

Biological tests Bacterial endotoxin Biological tests are useful to ensure the sterility A test for bacterial endotoxins must be car- of radiopharmaceuticals and to detect and ried out on radiopharmaceuticals for human quantify the presence of bacterial endotoxins. administration. Various methods are useful for The aim of these tests is to ensure product this purpose: safety. - the gel clot method; Sterility - the turbidimetric method; Radiopharmaceuticals for human administra- - the chromogenic method. tion must be tested to evaluate sterility. How- ever, owing to the duration of the test (usually These tests are based on the use of amebocyte 15 days), it is not possible to obtain results lysate from the horseshoe crab (Limulus poly- EANM before the release of the product since the phemus). For short-life radiopharmaceuticals, most commonly used PET tracers are labelled like [11C]methionine, according to the speci*c with short half-life isotopes. To ensure steril- monograph it is possible to release the batch ity it is necessary to operate with methods before completion of the test. For [18F]FDG, that minimize microbial contamination, which commercial kits are available for determina- then have to be validated. A suitable test to tion of the presence of bacterial endotoxins validate the aseptic operations is the media within a suitable time. *ll test, which represents a simulation of the whole preparation process or a part of it. This All the radiopharmaceutical quality controls test is performed by replacing the solutions, should be carried out according to the speci*c diluents and materials typically used during monograph for each product. Table 2 shows the preparation or the dispensing of the ra- quality control tests, methods and limits for diopharmaceuticals with culture media, which [18F]FDG according to the European Pharma- are then incubated for a suitable time. copoeia (EuPh) and the United States Phar- macopoeia (USP).

15 Type of quality control EuPh USP Analytical method

Radio-TLC scanner/HPLC with Radiochemical purity >95% >90% radio-%ow detector HPLC with UV or amperometric Chemical purity >95% >90% detector Residual solvents Gas chromatography with an <4.1 mg/V <0.04% (acetonitrile) max FID detector

pH 4.5-8.5 4.5-7.5 Colorimetric test

Residual Krypto"x <2.2 mg <50 µg/ml Colour spot test

Sterility test Sterile Sterile Microbiological test

Bacterial endotoxin <175 IU <175 EU LAL test

Radionuclidic purity 105-115 min 105-115 min Gamma-ray spectrometry (half-life)

Table 2: Common quality control tests for [18F]FDG and limits according to di&erent pharmacopoeia

16 of [18F]FDG. Appl Radiat Isot2010;69:649-716. of [18F]FDG.ApplRadiat control forquality method J.the simple A Koziorowski 4. 2009;67:990-5. of ethanol and radioactive concentration. Appl Isot Radiat 3. Jacobson MS,Dankwart HR, Mahoney DW. Radiolysis Radiolysis DW. Mahoney HR, MS,Dankwart Jacobson 3. Isot 2007;65:1193-201. cose (FDG) and the role of reductant stabiliser. Appl Radiat 2-[ of Radiolysis RM. Fawdry 2. 2004. Springer; Saha GB. Fundamentals of nuclear pharmacy. New York: 1. tical preparation, 0125). European pharmacopoeia monographs (Radiopharmaceu- [ of control Nakao R, Ito T, Yamaguchi M, Suzuki K. Improved quality 6. Biol2006;33:441-7. trochemical NuclMed detection. control of PET radiopharmaceuticals using HPLC Quality with K. elec- Suzuki M, Yamaguchi K, Furutuka R, Nakao 5. 2-[ of References References Chapter 1.2 Suggested reading J2006;2(4):e5. Intervent Biomed Imaging 7. Yu S. Review of [ of Review YuS. 7. Biol 2005;32:907-12. 18 F]%uoro-2-deoxy-d-glucose ([ F]%uoro-2-deoxy-d-glucose 18 F]FDG by HPLC with UV detection. Nucl Med Med Nucl detection. UV HPLCwith by F]FDG 18 F]FDG synthesis and quality control. quality and synthesis F]FDG 18 F]%uoro-2-deoxy-D-glu- 18 F]FDG) and the role the and F]FDG) 17

EANM Chapter 1 – Principles of PET radiochemistry 1.3 Basics of 18F and 11C radiopharmaceutical chemistry Philip H. Elsinga

Introduction Of the many possible radiochemical prepa- This section addresses basic and practical as- ration methods, few have been extensively pects of the synthesis of 18F- and 11C-labelled developed, and these are applied by many radiopharmaceuticals and the most common PET centres since they are extremely versatile ways of producing a radiopharmaceutical with and reliable. Using two preparation methods a short-lived positron-emitting radioisotope. – nucleophilic substitution with [18F]%uoride In the pioneering phase of PET radiochemistry and [11C]methylation – a large variety of bio- (late 1970s to early 1980s), the focus was pri- logically active compounds can be prepared. marily on the development of small synthons Important PET radiopharmaceuticals have also and exploring the scope and limitations of ra- been prepared by electrophilic [18F]%uorina- diochemical reactions. During recent decades, tion. Widespread use of these speci*c meth- however, a shift occurred towards the appli- ods has been encouraged by the appearance cation of a limited number of radiochemical of commercially available synthesis modules. reactions and their optimisation to prepare Using these modules, radiochemical prepara- new PET radiopharmaceuticals. tions are relatively easily performed, including by PET centres without extensive radiochemi- Preparation of a radiopharmaceutical – cal expertise. Other radiochemical preparation general aspects methods have usually been limited to research Many synthetic routes to prepare PET radio- groups and are not frequently adopted by oth- pharmaceuticals have been developed during ers because they require speci*c expertise. recent decades. The great majority of PET tracers are labelled with the positron-emitting radioiso- Certain general aspects in PET radiochemistry topes 11C and 18F (radioactivity decay: half-lives have to be considered: Firstly, the synthesis of of 20 and 110 min, respectively). The half-lives a short-lived PET radiopharmaceutical is a *ght of 13N and 15O (10 and 2 min, respectively) are against time because of the rapid decay of the too short to perform radiochemical syntheses applied radioisotopes. Preparation times of of more than one reaction step. In most cases, three half-lives are acceptable to obtain su+- [15O]radiopharmaceuticals are limited to radio- cient amounts of radiotracer for clinical studies. labelled water, butanol and gases for inhalation 15 15 13 studies (C O, C O2). In the case of N, this radio- A second pertinent aspect is the amount of isotope is incorporated into nitrite/nitrate and reagent that is used. In Table 1, an estimate is ammonia. In addition, other positron-emitting shown of the absolute amounts that are used radionuclides such as 89Zr, 124I, 68Ga and 64Cu in a radiochemical synthesis. It is clear that the have been applied mainly in the production of non-radioactive precursor is present in a large ex- peptides and larger biomolecules [1]. cess as compared with the radioactive reagent.

18 cyclotron target or during work-up of the radio- and reagents that are used during the work-up. Nucleophilic "uorinations Stable reticalandpractical speci*c activity in Table1). todecreased speci*c activities (compare theo- ofthe produced radioisotopes, which can lead Anotherimportant aspectisotopic is dilution 2-[ The best-known example is the preparation of cyclotron. a to access without sites to ticals sibility it a&ords to distribute radiopharmaceu- activities [3,4]. An advantage of speci*c high and yields radiochemical high in prepared be can radiopharmaceuticals of Using this preparation method, a wide variety with Radiochemistry committee [2]. EANM radiopharmacy pharmaceticals have been formulated by the radio- PET for especially GMP for Guidelines assured. be to needs parameters production process. all of and Traceabilityproduct the of manufacturing the of stages all during ticals GMP is to control preparation of pharmaceu- of focus main The countries. many in datory man- is (GMP) Practice Manufacturing Good with complying conditions under ceuticals radiopharma- of production years, recent In als. Fluoride sources are the air or carbon impurities from the target materi- sourcesofcarbon are carbon dioxide from the activitypriorthetoactual synthesis. The main C hapt 18 D F]%uoro-2-deoxy- er 1 – Principles of PET radiochemistry: 1.3 Basics of 1.3Basics er 1–Principles radiochemistry: ofPET 12 C and 19 F isotopes are introduced in the 18 - glucose (FDG), which F 18 O-enriched water 18 F is the pos- 19 After the *rst steps of the synthesis procedure, and for [ [ nucleophilic for modules Di&erent suppliers o&er automated synthesis [ nucleophilic to transferred been has production FDG from [ [ of duction pro- the up scaling into put been has e&ort crease the e+ciency of FDG production, much mercially distributed in many countries. To in- glucose consumption in tumours and is com- is the work horse of PET for the investigation of liquid chromatography. high-performance or techniques extraction steps are usually employed using solid-phase puri*cation preparation, the after or During [ reactive a of Preparation 2. Direct substitution of1. an appropriate leaving oride, di&erent synthetic options are possible: i.e. cyclotron production and workup of [ 18 18 F]%uorination process. Knowledge gained Knowledge process. F]%uorination F and ation are the synthesis of [ be required.be [ Examplesa of kylation. Finally, a deprotection reaction may followed by a second reaction, such as an al- intermediatenucleophilicbysubstitution, [ arepreparationthe [ of by hydrolysis of protective groups. Examples group by [ and [ 18 F]-3-%uoro-3-deoxy-thymidine (FLT). 11 18 18 C radiopharmaceutical chemistry F]%uoroethyltyrosine (FET). F]FDG inparticular. 18 18 F]%uoride and speeding up the up speeding and F]%uoride F]%uoride, which can be followed 18 F]%uorinations in general. in F]%uorinations 18 F]FDG(Fig. and1) 18 18 F]%uorinations, F]%uorocholine 18 18 F]%uorinated F]%uoroalkyl- 18 F]%u- EANM Production of [18 F]"uoride. The cyclotron ir- Nucleophilic substitution. Once the [18F]%uo- radiation of 18O-enriched water using the ride is dried and solubilised in the presence of (p,n)-nuclear reaction is the most e&ective the phase transfer catalyst, the actual nucleo- and convenient way to produce [18F]%uoride in philic substitution can take place. The reaction large quantities up to a few hundreds of GBq. conditions that are applied are: heating for 10–30 min at temperatures varying from 80o Work-up of [18F]"uoride. After the irradiation, to 160oC in a polar, aprotic solvent. Acetonitrile [18F]%uoride is separated from 18O-enriched is the common solvent. Other possibilities are water. This separation has two purposes: re- DMSO or DMF. Acetonitrile has the advantage covery of the 18O-water and the *rst step in the that it can easily be removed by evaporation, creation of reactive %uoride in an anhydrous in contrast to DMSO and DMF. Removal of environment. Commercially available cartridg- the solvent is important in many production es, such as QMA-SepPak, are available. These methods because its presence hampers the 2- cartridges, when conditioned into the CO3 subsequent HPLC puri*cation. DMSO and form, trap >95% of the %uoride, whereas the DMF have the advantage that the pressure in 18O-water is collected. The %uoride is eluted closed reaction vessels even at higher tem- from the cartridge with a solution containing peratures is low and therefore leakage is a

K2CO3. For some reactions other potassium minor problem. salts, such as KHCO3 and potassium oxalate, are used in order to decrease the basicity of The precursor molecule should contain a the reaction medium, so avoiding side reac- leaving group at the position where the 18F tions. radioisotope is desired. One should take into account that inversion of the stereochemistry Since K18F is not soluble in organic solvents will occur. In the case of aliphatic nucleophilic suitable for nucleophilic %uorination reactions, substitutions, sulphonic esters are used as suit- a phase transfer catalyst is needed. For this able leaving groups. There are several groups purpose, tetraalkyl ammonium salts or ami- that can be used, such as tri%ates, tosylates, nopolyethers (Krypto*x 2.2.2) are used. [18F] mesylates and nosylates. The choice of a Fluoride is reactive in water-free media only. particular leaving group is determined by its Small amounts of water will generate a hy- reactivity, stability and ease of incorporation dration shield around the %uoride anion. By in a certain precursor. Aromatic nucleophilic means of a few successive evaporations with substitutions need to be carried out with ni- dry acetonitrile, the %uoride anion is stripped tro- or trimethyl ammonium groups that are from its surrounding water molecules. activated by electron-withdrawing groups in the para- or ortho-position. An example is the

20 labelling, [ The use of elemental %uorine F Electrophilic "uorinations the or MPPF 5-HT the of preparation possible to label molecules without a&ecting thereforeis it nuclide; occurring naturally a is veloped [3]. An advantage of small of variety a substitution reaction, in Whereas in with Radiochemistry technique.electrophilic %uorination the using performed also were 1970–1980 years the DOPA.in productions FDG *rst The of preparation the is example well-known A group. mercury or tin trialkyl a of substitution aromatic electrophilic an is tution. The most common reaction nowadays additions to double bonds or aromatic substi- are performed Possiblebe can that reactions is its low speci*c activity. disadvantage of using electrophilic [ critical to obtain a su+cient yield of [ passivation process is essential and sometimes target wall and transport tubing. This so-called to the target to prevent adsorption of is very reactive, it is necessary to add inactive F 0.1–0.3% F maceuticalchemistry [4]. Neon-20 containing ning of the development of the PET radiophar- C rinating agent originates from the very begin- hapt er 1 – Principles of PET radiochemistry: 1.3 Basics of 1.3Basics er 1–Principles radiochemistry: ofPET 2 18 is used as the target gas. Because F 18 F]SFB. F chemistry the *rst step is always 18 F-prosthetic groupF-prostheticforpeptide 11 C molecules have been de- been have molecules C 11 C 1A 11 antagonist C chemistry a wide 11 C is that carbon 2 as an [ L -[ 18 18 F]-%uoro- 18 F]%uoride 18 F]F 18 F to the F]%uo- p-[ 2 . The 18 F] 2 2

21 thesisof biologically interesting PET radiophar- these properties. properties. these [ ecules that can be prepared from either [ InFig. 2,examples are shown ofsmall with[ tions thatstart exclude air from synthesis modules and solu- to taken be should precautions that means This ppm. 1.6 is methane of that whereas CO of abundancy natural activity can be achieved than for [ 14 [ as target be inserted between the gas supply and the and supply gas the between inserted be ‘cold’ CO able for this purpose. To reduce the amount of ages of up to 1–2% of O [ of tion not contain carbon impurities. For the produc- and pure very be must gas target The tying. achieve a high speci*c activity and rapid emp- to possible as small as be should target The alloy. aluminium an or aluminium pure from For the production of Production of modules are available from several suppliers. radiopharmaceuticals. Commercial synthesis [ maceuticals. The majority of or[ (CH to form COform to reliable and yields many biologically interesting 11 11 N(p,a) 18 C]methyl iodide (CH C]CH F and 11 3 OTf). An C]CH 4 11 11 11 is that theoretically a higher speci*c higher theoreticallya that is 2 C nuclear reaction. The advantage of in the target gas, an ascarite trap can C radiopharmaceutical chemistry C]CO 4 and can be further used in the syn- 11 2 by hot atom chemistry. Percent-chemistry. atom hot by C]CO 11 C-methylation reaction is simple, 11 2 , traces of oxygen are needed areoxygen of traces , C 2 11 or [ or C can be produced in the in produced be can C 11 11 3 I) and [ C]CO C, the targets are made 11 2 C]CH in nitrogen are suit- 2 11 in air is 330 ppm, 330 is air in 2 C preparations use 11 . C]methyl tri%ate 4 , both via the via both , 11 C]CO 11 11 C mol-C C]CO 2 . The 2 EANM target. Another precaution is to maintain the trap into the solution containing the precursor. targets under overpressure as much as pos- Ascarite is used to remove excess of HI, whereas sible. After irradiation, the target is emptied phosphorous pentoxide removes traces of wa- through stainless steel tubing. ter. Removal of acid and water is necessary for a good 11C-methylation yield. 11C-labelled methane is produced using the 11 same target but with the target gas N2 and The ‘gas phase’ method. [ C]methyl iodide is 11 5–10% H2. Using a Porapak tube immersed directly prepared from [ C]methane in the 11 in liquid nitrogen/argon, [ C]CH4 is trapped. presence of iodine vapour. This method is more By warming to room temperature and using simple and eliminates the problems associated 11 helium, [ C]CH4 is transferred to the synthesis with the preparation of the LiAlH4 solution (spe- module. ci*c activity) and the use of HI (deterioration of tubing and valves). A quartz tube of about Preparation of [11C]methyl iodide 40 cm is prepared containing ascarite at both There are two common ways to prepare [11C] ends, iodine and an empty part. The tube is methyl iodide, i.e. the ‘wet’ method, via reduc- positioned in a module in such a way that the 11 11 o tion of [ C]CO2 to [ C]methanol by LiAlH4, empty part will be heated at 650–750 C and the followed by treatment with HI, and the ‘gas iodine part at 70–95oC. 11C-labelled methane is phase’ method using iodine vapour and [11C] passed through the iodine, which is slowly sub-

CH4. For both methods, synthesis modules are limated upon heating. This gaseous mixture is commercially available. passed through the empty part of the tube. Because of the high temperature, a part of [11C] 11 11 The ‘wet’ method. [ C]CO2 is trapped in a so- methane is converted into [ C]methyl iodide. lution of LiAlH4 in tetrahydrofuran (THF). The Alternative production methods using bromine preparation of this solution should be carried instead of iodine have been developed, thereby out with extreme caution in order to prevent avoiding the iodine evaporation step. the presence of ‘cold’ carbon dioxide. Param- eters that a&ect the amount of ‘cold’ CO2 are Because the conversion e+ciency is usually atmosphere (argon is recommended), absolute 20–30%, many PET centres use a circulation amount of LiAlH4 and volume of the solution. pump in order to pass the reaction mixture 11 11 The trapped [ C]CO2 is reduced to [ C]metha- several times through the iodine tube to in- nol as an Li salt. The THF is evaporated and HI crease the radiochemical yield. In the circula- (57%) is slowly added to the precipitate. The tion, a trap must be installed to collect the formed [11C]methyl iodide is distilled via tubing [11C]methyl iodide. Several traps are possible: containing an ascarite/phosphorous pentoxide Porapak N, glass tube in liquid argon or Car-

22 compared with the trapping conditions, [ as temperature the elevating By bosphere. 2. On solid phase support: The precursorsupport:phasesolidTheis On 2. solution: In This is mostly used in practice. 1. have intheliterature: beenreported methylations out carrying of ways Di&erent [ of preparation the are examples Well-known been labelled with N-, O- and S-nucleophiles. A large number of radiopharmaceuticals have silver tri%ate heated at 200 containingcolumnmethylthroughiodidea yl tri%ate can easily be prepared by passing [ precursor and shorter reaction times. [ lower reaction temperatures, lower amounts of reactive, methylations can be performed using iodide.Because [ actions has several advantages over [ The use of [ C Methylations free from oxygen. in the dark and the column material should be containing the silver tri%ate needs to be stored tri%ate is impregnated on a carrier. The column Preparation of[ using agentlegas%ow. trap the from released be can iodide methyl 11 hapt C]methionine and [ and C]methionine coatedon asolid support, such as an HPLC through thesolution. [ The vent, sometimes in the presence of a base. The precursor is dissolved in a suitable sol- er 1 – Principles of PET radiochemistry: 1.3 Basics of 1.3Basics er 1–Principles radiochemistry: ofPET 11 ]ehltn aet s bubbled is agent C]methylating 11 C]methyl tri%ate in methylation re- 11 11 C]meth C]methyltri%ate farmoreis 11 C]raclopride (Fig. 3). (Fig. C]raclopride yl tri#ate o C. Usually the silver 11 11 C]methyl C]Meth- 11 11 C] C] 23 Trapping of [ conditions are shown in Table 2. K needed for deprotonation. Common bases are is manybase cases, a are used. acetonitrile In and DMF,acetone DMSO, like solvents tion, solu- in out carried are methylations When pharmaceuticals. further enhance the assortment of PET radio- conditions. developmentRadiochemistry will GMP preparedunder are and available tinely rou- are radiopharmaceuticals PET Several radiopharmaceuticals. PET new of development and equipment the of operation reliable to crucial remains sta& of training Appropriate radiopharmaceuticals. 11 for modules synthesis since the availability of commercially available widespread more becoming is maceuticals of production The [ of e+ciency trapping The tive charcoal is very e&ective for this purpose. recommended for radiation safety reasons. Ac- Concluding remarks is usuallyalmostquantitative. is not quantitative, so a trap at the vent line is 2 C-methylations, covering most of the PET PET the of most covering C-methylations, 18 CO F and support by a solvent or HPLC mobile phase. solid support. The product is eluted from the [ loop or a solid phase extraction cartridge. The 11 C]methylating agent is passed through the 3 , TBOH, NaOH and NaH. Further reaction 11 C radiopharmaceutical chemistry 11 C]methyl C]methyl iodide in the solution 11 and C 18 -urnto and F-%uorination 11 C]methyl tri%ate tri%ate C]methyl 18 radiophar- F EANM References Chapter 1.3

References 1. Holland JP, Williamson MJ, Lewis JS. Unconventional 3. Miller PW, Long NJ, Vilar R, Gee AD. Synthesis of 11C, 18F, nuclides for radiopharmaceuticals. Mol Imaging 2010;9:1- 15O, and 13N radiolabels for positron emission tomography. 20. Angew Chem Int Ed Engl 2008;47:8998-9033.

2. Elsinga PH, Todde S, Penuelas I, Meyer GJ, Farstad B, 4. Coenen HH. Fluorine-18 labeling methods: features and Faivre-Chauvet A, et al. Guidance on current good radio- possibilities of basic reactions. Ernst Schering Res Found pharmacy practice (cGRPP) for the small-scale preparati- Workshop 2007;62:15-50. on of radiopharmaceuticals. Eur J Nucl Med Mol Imaging 2010;37:1049-62.

Table 1: Features of 18F and 11C labelling: isotope dilution and typical reagent amounts Theoretical SA Practical SA Amount of Amount of (TBq/mmol) (TBq/mmol) radiotracer (nmol) precursor (nmol) 11C 3.7 · 105 50 8 2500 18F 6.6 · 104 200 2 1000

SA, speci*c activity

Table 2: Typical reaction conditions of [11C]methylations

[11C]CH I [11C]CH OTf 3 3 Temperature (oC) 20–60 80–130 Reaction time (min) 1 2–10 Precursor (mg) < 1 1–10

24 © P. Elsinga © P. Elsinga Chapter 1 – Principles of PET radiochemistry: 1.3 Basics of 1.3Basics Chapter 1–Principles radiochemistry: ofPET Figures and"gure legends(seefollowing pagesforFigs. 1and2): [ common Two 2: Figure including themethylating agents[ of Production 1: Figure 11 C]raclopride (right) C]raclopride 11 11 C building blocks from cyclotron-produced [ cyclotron-produced from blocks building C C-methylation reactions: production of [ of production reactions: C-methylation 11 C]methyl iodideand[ C]methyl 25 18 F and 11 11 C radiopharmaceutical chemistry C]methyl tri%ate C]methyl 11 C]methionine (left) and (left) C]methionine 11 C]CO 2 and [ and ]CH 11 C 4 EANM, Chapter 1 – Principles of PET radiochemistry 1.4 [18F]FDG synthesis Francesca Giurgola, Ilaria Vaccaro and Giorgio Testanera

Thirty years have passed since [18F]FDG was in [18F]FDG. For more detailed information on *rst synthesised for studies in humans [1]. To- how radionuclides are produced, the reader day, it is still the most widely used radiotracer is referred to [2]. for PET scans. While oncological studies account for the majority of applications, [18F] - Fluoride, especially in its F form, is the most 2 FDG is also used electively for cardiological reactive and electronegative element in na- and neurological indications. ture, allowing many possibilities for chemical combinations. Physical features are What is 2-deoxy-2-(18F)#uoro-D-glucose? shown in Table 1. [18F]FDG is a glucose analogue in which the OH group in the C2 position of the molecule T 109.6 min is substituted by an 18F group (%uoride). It is 1/2 able to pass through the cellular membrane β+ 97% Decay by facilitated transport following the same EC 3% pathway as the glucose molecule. Inside the β+ 635 keV cell it is phosphorylated by hexokinase. Energy γ 511 keV

Since [18F]FDG, unlike glucose, lacks the OH Table 1. Physical features of [18F] group in the C2 position, it is not a substrate for subsequent enzymes of the glycolytic pathway The relatively long half-life of [18F] is very help- and cannot be further metabolised by cells. ful for chemical and medical purposes: it allows complex synthesis, long-duration scans [18F]FDG reconversion owing to glucose-6-phos- and late pharmacokinetic evaluations. phatase entails a very slow transformation and, due to the negative polarisation, the reaction The [18F] radioisotope may be produced by product ([18F]FDG-6P) is not able to pass through two di&erent types of nuclear reaction: the cellular membrane. These biochemical features mean that the molecule is unable to return - 20Ne(d, α)18F, which leads to F in a gaseous 2 to the blood circulation, allowing the tracer to form, used in electrophilic substitution re- accumulate in cells that become visible on a PET actions [3]; scan owing to their beta-plus emissivity. t 18O(p, n)18F, which leads to formation of [18F]#uoride 18F-, used in nucleophilic substitution re- [18F]%uoride is commonly produced by a cy- actions. clotron and it is the radionuclidic component

26 Evaporation [ purposes. diopharmaceutical ra- for system production used commonly most the is reactions these of second The monly used cartridges for thispurpose. monly usedcartridges 30-PS-HCO Chroma*x component water the ion exchange cartridges intended to separate module, where it needs to be trapped on an- synthesis the into then and lines transport into gas by pushed usually cyclotron, the [ with starts synthesis The Chapter 1 – Principles of PET radiochemistry: 1.4[ Chapter 1–Principles radiochemistry: ofPET T 18 - - rapping andelutionof[ F]FDG synthesis as a catalyst that increases the solubility of solubility increasesthe that catalyst a as For a correct synthesis process, the H to asananti-ionfor work [ to*x/K reactor is achieved using a solution of Kryp- The elution of [ h adto o CH of addition the through achieved is This evaporation. by cess still in the %uoride needs to be removed asKrypto*x. the samefunction to use TBA (tetrabutylammonium salt) with possible also is it that noted be should It process. synthesis the with interfering out with- reactivity %uoride increase to helps to avoid [ K binding of capable ether crown a is Krypto*x %uoride. nucleophilic 2 CO 18 3 F]KF formation, thus allowing K in CH in 18 F] from the cartridge to the 3 CN. This solution works works solution This CN. 3 18 3 N wih om an forms which CN, are the most com- most the are O.QMA Sep-Pakand 18 F] 18 18 F] coming from coming F] F]. This process + ions in order in ions 2 O ex 18 + - F]FDG synthesis 27 Substitution reaction ration a precursor is added. Mannose tri%ate tri%ate Mannose added. is precursor a ration evapo- the after reaction, nucleophilic the In types ofsubstitutionreaction: types gets for [ As stated above, use of di&erent cyclotron tar- found, whichisagood leaving group. occur only in the C2 position, where tri%ate is acetyl groups is to ensure that %uorination will increase or decrease in pH. The function of the an resultof a breakas to easy arevery bonds molecular their and positions, C6 and C4 C3, in the C2 position and acetyl groups in the C1, is similar to that of FDG: there is a tri%ate group monly used for this purpose, since its structure sulphonyl-β-D-manno methane- pyra nose) is most com- (1,3,4,6-O-acetyl-2-O-trifluoro - - - decomposition ofthecrown ethers. reduced pressure and temperature to avoid N or He (like gas inert of %ux a under and point, boiling azeotropic mixture with water, lowering its 18 cyclotronand a target in gas amanage to di+cult,technicallyveryalso is speaking, It owing to very low speci*c activity and yield. tution, but this technique is no longer used 18 Initially,synthesisFDG performed waswith Nucleophilic inS Electrophilic withgaseousF F F 2 2 in gaseous form and electrophilic substi- in gas form for the purpose of synthesis. 18 F] production leads to two di&erent 2 ). This process occurs under occurs process This ). N 2 with 18 F - 2 EANM The S 2 reaction will continue for 5 min with hydrolysed product and also formation of N a reactor temperature of 85°C and the precur- 2-chloro-2-deoxy-D-glucose (2-CIDG) as a sor concentration needs to be equimolar with side reaction, but it has the great advantage Krypto*x//K CO . of not leading to epimerisation. 2 3 - Basic hydrolysis, using NaOH: Hydrolysis oc- Hydrolysis curs with the reactor at room temperature Hydrolysis consists in removal of the protec- and takes a very short time (less than 1 min); tive acetyl groups in the C1, C3, C4 and C6 chemically speaking, it is instantaneous. positions and can be of two types: Perfect timing and NaOH concentration are - Acid hydrolysis, using HCl: Hydrolysis oc- very important for this kind of hydrolysis. Us- curs at a reactor temperature of 125°C and ing NaOH can lead to formation of epimers lasts 5–10 min. It may yield only a partially such as [18F]FDM. © Francesca Giurgola © Francesca

Figure 1: Synthesis of [18F]FDG

Puri#cation The following are the most commonly used: The subsequent reaction steps give rise to - PS-OH- and alumina to remove free %uoride many residues that need to be puri*ed: - PS-H+ to remove Krypto*x/K CO 2 3 - RP (reverse phase) cartridge to remove par- - Free [18F]%uoride tially hydrolysed [18F]FDG and precursor. - Krypto*x/K CO 2 3 - Partially hydrolysed [18F]FDG Further details about the synthesis process - Mannose tri%ate which has not reacted can be found in references [4] and [5].

This puri*cation process is achieved by means [18F]FDG synthesis today of chromatographic cartridges which also have Nowadays [18F]FDG production is achieved using the task of neutralising the hydrolysis solution. synthesis modules made of tubes and valves

28 are usually small and compact to allow adapta- also minimizes radiation exposure of sta&. They which not only ensures product protection but systems, closed being usually advantage of synthesissteps. Thesemodules have thegreat into a reactor vial and undergoes all the di&erentules, [ able to connect di&erent vials. Inside these mod- There are di&erent two ofcassette: types transfer linesandvalves. with vials di&erent connecting modules, the inside them position correctly and activate synthesis. perform Theoperator’s aretasks to to needed cartridges and component ferent dif- the all separately contain kits Single-use for every module. single-use synthesis kits or cassettes are available on the type of module used. Generally sterile and process. Practical synthesis procedures depend inertemissionsandgas duringsynthesisthe ventilationsystem permits control radioactive tion to the small spaces inside the hot cells. Their Chapter 1 – Principles of PET radiochemistry: 1.4[ Chapter 1–Principles radiochemistry: ofPET - - automatic synthesisprocedure. procedure. This is considered an almost fully standardised a with itself module the by ules. Activation and dilution are performed mod- into inserted be to only need settes *xed, reducing operator dependence. is positioning while activated be to need reagents kits, in as cassettes: ready” “Not ing is ing to“Ready use” cassettes: position- Reagent 18 F]%uoridefromthe cyclotron elutedis completely predetermined and cas- and predetermined completely 18 F]FDG synthesis 29 Thegreatthe cassettes is using in advantage sure, %ux, temperature, and leak tightness to tightness leak and temperature, %ux, sure, pres- regarding out carried be must tests of Before starting [ timeperiods. within short useful for execution of subsequent syntheses from residual radioactivity. This process is very module the of cleaning allowing synthesis, after rinsing perform to able also are settes cas- with Modules procedure. the of safety and reproducibility the increases which rors, reduction in possible operator-dependent er- able owing to the increase in standardisation. and rendered the synthesis process more reli- synthesis. It has also made maintenance easier and [ “cleaning”preparationfor time in reduction sensible a about brought has The development of fully automatic modules documentation andarchival purposes. for process production the regarding data all of report a print to common is It analysis. subsequent further for registered also are on the control PC screen. Process parameters displayed graphically and monitored are ity, temperature, pressure, %ux and radiation activ- including parameters, process synthesis The synthesis.during problems solve to intervention manual for ed, but usually there is also limited possibility control and reaction process are fully automat- ensure that the process will occur in safety. All 18 F]FDG synthesis, a sequence 18 F]FDG EANM Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Humanitas, IRCCS Figure 2: An example of an [18F]FDG module equipped with cassette: Fastlab GE Courtesy of Departement of Nuclear Medicine, (Milan) Rozzano Humanitas, IRCCS Figure 3: An example of an [18F]FDG double module: TRACERlab FX GE

30 3. Casella V, Ido T, Wolf AP et al. Anhydrous F-18 labeled F-18 Anhydrous al. et AP T,Ido WolfV, Casella 3. guide. EANM, Vienna, 2010 technologist’s A 1: Part PET/CT. of practice and Principles In : production. isotope PET 6: Chapter K. Szczepura 2. 1978;XIV:175-83. JLabelCmpdRadiopharm D-glucose. and 2-deoxy-2-%uoro-D-mannose analogs. glucose 2-deoxy-D- Labeled al. et V Casella C-N, T,Wang Ido 1. positroni. Aretrè: 1999. di emettitori farmaci radio di produzione sulla generali Concetti E. F,Bombardieri Crippa A, Bogni C, Pascali 5. dipositroni.emettitori Bologna: Patron, 2003. isotopi con marcati Radiofarmaci C. Pascali A, Bogni 4. 1980;21:750. J NuclMed preparation. radiopharmaceutical for %uorine elemental References References Chapter 1.4 18 F-labeled 2-deoxy-2-%uoro-D-glucose, 2-deoxy-2-%uoro-D-glucose, F-labeled 14 C-2-deoxy-2-%uoro- 31

EANM Chapter 1 – Principles of PET radiochemistry 1.5 68Ga-peptide tracers Jamshed B. Bomanji and Nikolaos D. Papathanasiou

Introduction Figure 1: Principles of SSTR-based imaging. Somatostatin receptor (SSTR) radionuclide trac- The radiolabelled tracer (T), with high ers are of particular interest in oncology and a+nity for SSTRs, is injected and binds their use is expanding worldwide. These trac- avidly to SSTRs expressed in high densities ers are synthetic analogues of a natural cyclic on the cell surface. The receptor–tracer neuropeptide, somatostatin (SST), which exerts complex is then internalised within the cell. inhibitory and regulatory e&ects in various parts Subsequently, the signal emitted from the of the body such as the brain, peripheral neu- tracer radionuclide maps the distribution of rons, endocrine pancreas, gut and immune sys- tracer within the body tem. The activities of SST are mediated through speci*c, G-coupled receptors (SSTRs) expressed The basis for the development of SSTR tracers on cell surfaces; to date, *ve di&erent SSTR sub- is the synthetic cyclic octapeptide octreotide. types have been recognised. Neuroendocrine Octreotide retains the pharmacological pro*le tumours (NETs) are neoplasms with distinct of SST, but unlike natural SST is metabolically biological and clinical features; in particular stable and has a relatively long biological half- they express high densities of SSTRs at their life (~2 h), which enables its use in clinical prac- cell membranes. Thus, they can be e&ectively tice. This analogue has been conjugated with targeted with SSTR radionuclide probes: the a chelator, diethylene triamine penta-acetic tracer is injected and selectively binds to these acid (DTPA), which allowed successful binding receptors, then the receptor–tracer complex is of 111In molecule to obtain the *rst commer- internalised within tumour cells and accumu- cially available agent (111In-DTPA0-octreotide, lates in high concentrations compared with the Octreoscan) in the early 1990s [1]. Whole-body concentrations in adjacent tissues. Thereafter, and SPECT studies with Octreoscan have dom- the emitted signal from the radionuclide is inated the functional imaging of NETs until detected by gamma-cameras or hybrid PET- recently. Such studies are, however, limited SPECT/CT scanners and maps the location and by inherently suboptimal spatial resolution extent of tumour (Fig. 1). for the detection of small or deeply seated deposits or lesions in organs with high physiological tracer uptake (e.g. liver). Moreover, Octreoscan shows only moderate binding a+nity for the SSTR-2 subtype (expressed by the majority of NETs), while DTPA, as a chelator, cannot form stable complexes with β-emitters (90Y or 177Lu) for the purpose of obtaining SSTR

Courtesy of Institute of Nuclear Medicine, UK London, Hospital, University College therapeutic agents.

32 SSTR-2. Applied in PET imaging they have they imaging PET in Applied SSTR-2. to a+nity binding in increase severalfold a particular in including properties, cological pharma- advantageous have tracers These imaging. PET/CT SSTR for peptides labelled novel of advent the with overcome been have Octreoscan of shortcomings The positron emitter, with 89% positron branch- positron 89% with emitter, positron min) 68 (half-life short-lived abundant an is 68 parent days) ~270 (half-life long-lived tively 68 68 isavailable. facility places where aPET and is now replacing traditional Octreoscan in NETsfor modality imaging functional art the [3]. ings *nd- equivocal or negative yields latter the when including Octreoscan, conventional with compared performance diagnostic rior and multiplied has papers 2001 [2]. Since then, the number of published in published were tumours carcinoid with 68 clinical results obtained using a Firstaccuracy. diagnostic enhancing session, information can be obtained within the same quality, combined anatomical and functional high- scanners, PET/CT hybrid newer With localisation. lesion and resolution spatial in improvements considerable about brought Chapter 1 – Principles of PET radiochemistry: 1.5 Chapter 1–Principles radiochemistry: ofPET mulated evidence shows that it o&ers supe- o&ers it that shows evidence mulated centres.accu- specialised The of number ing increas-an routinein clinical becoming is CT Ga Ga is a Ge decays by electron capture to capture electron by decays Ge patients of group small Ga-DOTA-TOC,a in Ga-peptide labelling Ga-peptide 68 68 Ge/ Ga-peptide PET/CT is the state of state the is PET/CT Ga-peptide 68 Ga Ga generator product. The rela- 68 Ga-peptide PET/ Ga-peptide 68 Ga-peptide, 68 Ga. Ga. 68 68 Ga- Ga Ga 68 Ga-p 33 eptide tracers mixed with 200–800 MBq MBq 200–800 with mixed and HEPES) or acetate sodium (commonly bu&er suitable DOTA-TATE)a in dissolved DOTA-TOC,is (e.g. peptide the of μg 10–50 and volume of HCl in the solution. Afterwards, 68 minimize to employed often is eluate the of Prepuri*cation solution. HCl using chloride 68 68 to stable decays subsequently which water), in range (β ing methods ofradiosynthesis [5]. implemented regarding worldwide centres among there variation notable experience; is and protocol department’s our account into taking and principles basic its in described been has procedure radiolabelling The nel. person- to exposure radiation reducing thus of relatively high amounts of radioactivity and min, obviating the need for manual handling 12–25 within on-site out carried is process means of semi-automated systems, the whole high-performance liquid chromatography. By and chromatography thin-layer instant by analysed is which achieved, is tracer peptide of (>95%) yield radiolabelling high A *ltration. sterile to subjected then and block heating a in 90°C at min for~10 incubated is conjugated peptides [4]. The bu&ered solution of complexes stable of tion adjust pH to 3.5–4, which facilitates the forma- the bu&er is to keep Ga Ge Ge breakthrough and to reduce the amount Ga, in ionic +3 oxidation state, is eluted as eluted is state, oxidation +3 ionic in Ga, Ge is loaded on an SnO an on loaded is Ge + max. energy 1.89 MeV; 2.9 mm mean mm 2.9 MeV; 1.89 energy max. 68 Zn. 3+ in its ionic state and 2 or TiO or 68 68 Ga. The role of role The Ga. Ga with DOTA-with Ga 2 column; 68 Ga- EANM 68Ga-peptides: tracers, properties and ed with the chelator tetraazacyclododecane advantages tetraacetic acid (DOTA) and this peptide–che- 68Ga-peptides share the same basic structure, lator complex is labelled with the positron which resembles that of “conventional” Oc- emitter 68Ga (Fig. 2). treoscan: a synthetic SST analogue is conjugat- Courtesy of Institute of Nuclear Medicine, UK London, Hospital, University College

Figure 2: Chemical structure of 68Ga-DOTA-Tyr3-octreotide. The macrocyclic chelator DOTA is conjugated with the octreotide SST analogue and this complex is labelled with 68Ga

DOTA is an excellent macrocyclic chelator tracer, having the highest a+nity among 68Ga- which forms stable complexes with metal peptides for SSTR-2 [6]. Another tracer, 68Ga- radiotracers, such as 68Ga, 90Y or 177Lu. 68Ga- DOTA0-1NaI3-octreotide (68Ga-DOTA-NOC) DOTA0-Tyr3-octreotide (68Ga-DOTA-TOC) was shows the broadest a+nity pro*le, with high one of the *rst studied tracers, with tenfold a+nities for SSTR-2 and SSTR-5 and to a lesser higher binding a+nity for SSTR-2 and an in- extent for SSTR-3; this pro*le has been sug- creased cell internalisation rate compared gested to be coupled with enhanced diagnos- with Octreoscan. By replacing the C-terminal tic performance [7]. Figure 3 shows the basic threoninol of the octreotide molecule with the structure of principal 68Ga-labelled peptides natural amino acid threonine (i.e. changing and Table 1, their a+nity pro*les. octreotide to octreotate) we get the tracer 68Ga-DOTA0-Tyr3-octreotate (68Ga-DOTA-TATE). 68Ga-DOTA-TATE is a hyperselective SSTR-2

34 Courtesy of Institute of Nuclear Medicine, University College Hospital, London, UK choice of tracer depends on the centre’sex- the on depends tracer of choice The subtypes. SSTR other the for pro*les ing bind- various exhibit and a+nities di&erent All tracers bind to SSTR-2, but they do so with NETs. of heterogeneity and rarity the given perform, to di+cult are studies Such tracers. parisons of the clinical e+cacy of these various There is a paucity of evidence regarding com- (Cys) cystein aminoacids bonds between disulphide represents -S-S- molecule. octreotide the of position C-3 the by at “inserting”NaI with the natural amino acid threonine D (Thr). The NaI B] in sequence acid amino [see octreotide of threoninol C-terminal replacingthe by obtained (-TATE) Octreotate C position. is C-3 analogue the sequence. with This Tyrat Phe replacing by Tyr octapeptide the of sequence acid Amino B Figure 3A–D: Basic structure of SST and the main SST analogues. A Amino acid sequence of SST. 1.5 Chapter 1–Principles radiochemistry: ofPET 68 G a-p 35 3 eptide tracers -octreotide (-TOC), obtained from octreotide from obtained (-TOC), -octreotide peptide. therapeutic the possible as closely as mimics used for therapy, so that the imaging peptide be to going subsequently are counterparts 68 however,either choose to perience and preference. It seems reasonable, Ga-DOTA-TATErespective their if 3 -octreotide -octreotide (-NOC) sequence is produced 68 Ga-DOTA-TOCor 90 Y or Y 177 Lu EANM 68Ga-peptides are particularly attractive for followed by adrenals, thyroid, salivary glands several reasons. The physical half-life (68 and bowel. Normal standardised uptake val- min) of 68Ga positron decay is convenient ues in various organs show a wide range [9]. for PET imaging and matches the kinetics Physiological tracer accumulation in clusters of peptides, with fast blood clearance and of pancreatic islets (expressing SSTR-2) may rapid target localisation. 68Ga is readily avail- mimic a focal lesion (most often in the pancre- able on an everyday basis, independently of atic head) and is a potential source of pitfalls. a nearby cyclotron. The long half-life (270 Non-malignant, reactive nodes may show days) of parent 68Ge enables the use of the low-grade uptake owing to SSTR expression in commercially available generator for up to activated lymphocytes [9]. Di&use, low-grade 9–12 months or even longer and allows uptake can be seen in the prostate gland and timely routine manufacture and shipment breast glandular tissue. to nuclear medicine departments. The synthesis and labelling of peptides is relatively Clinical applications of 68Ga-peptides easy, with high radiolabelling yield (>95%). These tracers have favourable pharmacoki- Neuroendocrine tumours (NETs) netic properties compared with Octreoscan, NETs are a heterogeneous group of rare and enhanced diagnostic performance and neoplasms (incidence 2.5–5 cases/100,000) clinical e+cacy. that arise from neuroendocrine cells, mainly in the gastro-entero-pancreatic tract (GEP 68Ga-peptides: protocol, biodistribution NETs) and less often in the bronchopul- and pitfalls monary system (pulmonary carcinoids). 68Ga-peptides show rapid blood clearance The incidence and prevalence of NETs has and tumoral accumulation. They are almost multiplied over the last 30 years owing to entirely excreted by the kidneys. The injected advances in endoscopic and imaging meth- activities are in the range of 100–270 MBq de- ods, either anatomical or functional, which pending on the applied tracer and the depart- have led to an increase in detection rates. ment’s protocol. Images are acquired ~60 min NETs exhibit a striking variability in clinical p.i. with a dedicated PET/CT scanner in 3-D behaviour and outcome. In a minority of acquisition mode [8]. cases (functional NETs), peptides or bio- genic amines secreted by the tumour lead The physiological distribution of 68Ga-peptides to distinct clinical syndromes, such as car- re%ects normal SSTR expression throughout cinoid or Zollinger-Ellison syndrome. Most the body. In general, there is intense tracer ac- often, tumours are non-functional and show tivity in the spleen, kidneys, pituitary and liver, a silent clinical course until late presentation

36 A is increased in 60–80% of GEP NETs, is cor- is NETs, GEP of 60–80% in increased A is Chromogranin prognosis. and grade tumour assess to utilised is Ki-67 marker proliferation The NETs. for value prognostic and agnostic di- with markers serum are There tastases. me- skeletal or hepatic and effect mass with cal, in vitro studies which demonstrated SSTR of NET cells. This has been proven by preclini- and 5 subtypes, are located on the membrane than of index >20%) showed higher uptake of tumours and a higher proliferation rate (Ki-67 high-grade with patients However, 82%). vs. with compared tivity that found al. et 18 hence showthey forlowturnover; sensitivity entiated, with low growth rate and metabolic NETs. In general, these tumours are well di&er- for well-suited less is it however, PET/CT; cal the mainstay for the vast majority of oncologi- ( %uorodeoxyglucose Fluorine-18 [10]. agents bioactive of secretion the decreasing by symptoms ameliorate to tive, pallia- or growth, tumour arrest to curative, either intent: final and tumour symptoms on burden, depending tailored, be patient should It approach. multidisciplinary a requires Treatment management. clinical proper subsequent and staging accurate for essential are modalities Imaging follow-up. therapy/ to response the assessing in later and tumours non-functioning diagnosing in useful is and burden, tumour with related Chapter 1 – Principles of PET radiochemistry: 1.5 Chapter 1–Principles radiochemistry: ofPET F-FDG. In a study of 38 NET patients, Kayani patients, Kayani NET 38 of study a In F-FDG. 68 Ga-DOTA-TATE [11]. SSTRs, mainly of 2 18 F-FDG had inferior sensi- inferior had F-FDG 68 Ga-DOTA-TATE(66% 18 F-FDG) is F-FDG) 18 F-FDG 68 Ga-p 37 eptide tracers and measured the densities of these receptor hybridisation) mRNA situ in (with expression Regarding the clinical management of NETs, of management clinical the Regarding behaviour. aggressive more and metabolism glucose increased with NETs, well-di&erentiated less tions. applica- therapeutic and imaging of number with radiolabelled peptides and an emerging molecular basis for e+cient targeting of NETs the provided density increased their as well as SSTRs of presence The nohistochemistry). immu- and autoradiography (with proteins intensity of positive lesions indicate changes indicate lesions positive of intensity tion because serial changes in the number or Such studies should be interpreted with cau- tested in monitoring response to therapy [12]. been has it burden, tumour delineates and Since are: CT of applications main the - - - peptides) or “cold” octreotide. dionuclide SST analogues ( disease who are eligible for therapy with ra- quent selection of patients with metastatic subse- and status SSTR of Determination disease. progressive or recurrent residual, detect to NETs known with patients of Restaging static disease. meta- distant and loco-regional of staging initial and tumours primary of Detection 68 18 Ga-peptide Ga-peptide imaging accurately detects F-FDG is reserved for the minority of minority the for reserved is F-FDG 68 Ga-peptide PET/ Ga-peptide 177 Lu or 90 Y-DOTA EANM in SSTR status – not tumoral metabolic activ- valuable before such a treatment approach ity as with 18F-FDG – and such changes do is considered because it can accurately deter- not necessarily correspond to a therapeutic mine tumour extent (localising the primary response. tumour and identifying small or unsuspected, loco-regional nodal metastases) and guide An increasing number of published reports subsequent surgical resection. Otherwise it show the diagnostic e+cacy of 68Ga-peptide can e&ectively rule out patients with distant PET/CT. In a prospective study of 84 proven or metastatic disease, especially in the skeleton, suspected NETs, 68Ga-DOTA-TOC PET showed from inappropriate surgical treatment (Fig. 4). superior per-patient sensitivity (97%) than CT (61%) and SSTR-SPECT (52%). 68Ga-DOTA-TOC detected more lesions than the other modalities and was very e&ective in the detection of bone lesions [13]. Putzer et al. also found that 68Ga-DOTA-TOC detected bone metastases at a signi*cantly higher rate than CT [14]. This is very important, since skeletal involvement in NETs carries a dismal prognosis and precludes extensive surgical resection. In two recent studies, with a total of 38 biopsy-proven NETs, 68Ga-peptide was superior (positive in 37/38) to 18F-DOPA (positive in 23/38 patients) PET/CT [15, 16]. The limitations of such stud-

ies include the small numbers of evaluated Courtesy of Institute Hospital, University of Nuclear Medicine, College UK London, patients (re%ecting the rarity of NETs), possible Figure 4: Patient with ileal carcinoid referral bias and the fact that imaging results tumour, with multiple soft tissue and are not validated with a standard reference bone metastases above and below the method, something which frequently is not diaphragm. This patient is unsuitable for feasible or ethical. surgical treatment; a systemic approach would be more appropriate Accurate lesion detection by 68Ga-peptide PET/CT has a signi*cant impact on thera- Frilling et al. found that 68Ga-DOTA-TOC con- peutic management. Surgery with curative tributed to clinical decision-making prior to intent is the treatment method of choice liver surgery with curative intent [17]. It could for localised NETs. 68Ga-peptide imaging is identify patients eligible for such treatment

38 deposits not detected by CT or MRI. In an- In MRI. or CT by detected not deposits of extrahepatic to owing detection screening further from excluded were transplantation the for liver evaluated In patients 7/15 series, same strategy. surgical the planning in helped and extent disease determining by compared with historical data. historical with compared bene*t survival a shows and disease static meta- progressive with in patients responses partial induces toxicity, has limited treatment of mode This e&ect”). (“bystander metabolites toxic of release associated and cells nearby to cell target the beyond damage radiation direct via enhanced is bene*t therapeutic The death. cell and damage DNA reparable ir- causes then β-emitter the and imaging, corresponding 90 the with targeted further be can imaging on uptake tracer su+cient demonstrate who Patients therapy. nuclide radio- SSTR or “cold”octreotide with ment treat- for eligible patients of selection the in lies management clinical in peptides ing imag- of contribution important most The tumour. primary of resection surgical with (10%), proportion able in a but consider- small was reported change Management NETs. pancreatic and jejunal ileal/ were majority The [18]. cases of 60% 68 primary, and an NET unknown ing metastatic hav- patients 59 with setting, clinical other Chapter 1 – Principles of PET radiochemistry: 1.5 Chapter 1–Principles radiochemistry: ofPET is internalised in a way similar to that during during that to similar way a in internalised is complex peptide receptor–therapeutic The Y- or or Y- in site primary the localised Ga-DOTA-NOC 177 Lu-labelled therapeutic analogues, analogues, therapeutic Lu-labelled 68 Ga-p 39 eptide tracers mours at extra-adrenal sites along the sym- the along sites extra-adrenal at mours tu- cell chroma+n rare are Paragangliomas Paraganglioma andphaeochromocytoma in intracellular granules or in the cytoplasm. cytoplasm. the in or granules intracellular in stored is and then mechanism uptake active an via cell chroma+n the enters It choice. of radiopharmaceutical the is MIBG analogue noradrenaline The lungs. and bones liver, nodes, to lymph metastases with malignant, are tumours these of 10% than More gland. adrenal the from originating tumours for term is the corresponding eochromocytoma pha- chain; parasympathetic and/or pathetic clide analogues is considered. is analogues clide radionu- with if treatment be applied should However, [19]. burden tumour whole the mapping in accurate be 18 with simultaneously evaluated patients MTC 18 of series small a in NETs. than other In density heterogeneous more and lower a with but surface, their on -5 and SSTR-2 express cells MTC behaviour, and origin neuroendocrine their to Owing thyroid carcinomaMedullary (MTC) β-emitters. dionuclide ra- SSTR with targeted further be can they su+cient show these If cases. in MIBG-negative especially gliomas, paragan- for depicting useful can be very CT that showing series small are There phaeochromocytomas. for than paragangliomas for sensitive less is MIBG F-FDG, none of the tracers was found to to found was tracers the of none F-FDG, 68 68 68 Ga-peptide imaging imaging Ga-peptide Ga-DOTA-TATE and and Ga-DOTA-TATE Ga-peptide uptake, uptake, Ga-peptide 68 Ga-peptide PET/ Ga-peptide EANM Meningiomas and other tumours aerosol may be developed for lung perfusion/ Meningiomas are the most frequent extra-axi- ventilation studies. 68Ga-biphosphonates for al primary brain tumours and show high SSTR- skeletal imaging has been reported [21]and 2 expression. 68Ga-peptide PET/CT has been 68Ga-DOTA-RGD has been used to image applied in meningiomas to detect potential angiogenesis. Although this statement may skull base involvement and to help design sound somewhat ambitious and premature, radiotherapy treatment planning volumes. it appears that 68Ge/68Ga generator has the potential to evolve as “the 99Mo/99mTc genera- Apart from DOTA-peptides for SSTR imaging, tor” for PET imaging. 68Ga has been used to label various other peptides, bombesin being the most important. Key points – “take-home” message Bombesin is an amino acid originally puri*ed t 68Ga-DOTA-peptide PET/CT is the con- from the skin of frog which stimulates gastrin temporary functional method of choice release from G cells. Bombesin receptors are for somatostatin receptor imaging of NETs. overexpressed in prostate and breast cancer. Bombesin-based ligands are still a research t The most widely applied tracers are 68Ga- issue at specialised centres and have not DOTA-TOC, 68Ga-DOTA-NOC and 68Ga- achieved broad use. DOTA-TATE

Potential future applications t 68Ga-peptides show excellent a+nities for For the time being 68Ga-based tracers are somatostatin receptors and are available dominated by 68Ga-peptides applied in NETs on an everyday basis from an in-house and neural crest tumours. However, other on- generator going tracer developments have notable diagnostic potential. 68Ga-citrate shares a similar t 68Ga-peptides have an impact on clinical uptake pathway to conventional 67Ga-citrate. management and are applied in initial It is a promising tracer for infection/in%amma- staging, in re-staging and, most impor- tion, and accumulates in in%ammatory lesions tantly, in the selection of eligible patients owing to increased capillary permeability, up- for targeted radionuclide therapy. take by leucocytes and bacteria and binding to lactoferrin and bacterial siderophores[20]. 68Ga-labelled macroaggregated albumin and

40 . Hofmann M, Maecke H, Borner R, Weckesser E, Scho&ski 2. 1993;20:716-31. Nucl Med Rotterdam experience with more than 1000 patients. Eur J with [111In-DTPA-D-Phe1]- and [123I-Tyr3]-octreotide: the scintigraphy receptorPP, Somatostatin Kooij HY,al. et Oei Krenning EP, Kwekkeboom DJ, Bakker WH, Breeman WA, 1. free Commun 2010;31:1025-32. patientsNuclMed 68Ga-DOTATATEof pattern al. et Distribution S, disease- in Shastry M, Kayani I, Wild D, Caplin M, Visvikis D, Gacinovic 9. 2010;37:2004-10. Imaging Mol Med DOTA-TOC,68Ga-DOTA-NOC,68Ga-DOTA-TATE Nucl J Eur 68Ga- peptides: 68Ga-DOTA-conjugated with imaging tumour PET/CT for guidelines Procedure al. et M, Gabriel RP,S, FantiBaum JB, Bomanji V,Ambrosini I, Virgolini 8. 2009;36(Suppl2):S201. Imaging Mol Med Nucl J Eur (OP231). NETs GEP of detection the in CT PJ. Comparison of Wild D, Bomanji JB, Reubi JC, Maecke HR, Caplin ME, Ell 7. 2000;27:273-82. Nucl Med J Eur use. radiotherapeutic and scintigraphic for selected somatostatinradiotracers of SST1-SST5 receptorsubtypes somatostatin human for pro*les A+nity al. et JS, A, Schmitt Heppeler S, Wenger B, Waser JC, Schar JC, Reubi 6. 2005;32:478-85. Imaging Mol Eur JNuclMed with 68Ga. DOTA-peptidesEP. Radiolabelling Krenning M, nenberg BF,Bernard E, Konij- Blois de M, Breeman Jong WA,de 5. 2010;31:753-8. (68)Ga-Dotatoc for clinical application. Nucl Med Commun bu&erof Optimal radiosynthesis A. the bruggen of choice Ver- G, Bormans H, Vanbilloen R, Chekol M, Bauwens 4. 2010;51:875-82. *ndings on 111In-DTPA-octreotide scintigraphy. J Nucl Med equivocal or negative and tumors neuroendocrine with J.68Ga-DOTATATE Bomanji of roleME, The Caplin patients J, in PET Soh AM, Quigley I, Kayani R, Srirajaskanthan 3. 2001;28:1751-7. data. Eur JNuclMed tin receptor radioligand PET (68)Ga-DOTATOC: preliminary P, Oei L, et al. and Biokinetics imaging with the somatosta- References References Chapter 1.5 68 Ga-DOTANOC and 68 Ga-DOTATATE PET/ 41 crine tumours.crine LancetOncol2008;9:61-72. WW, Thakker RV, et al. Gastroenteropancreatic neuroendo- Herder de RT, Jensen DC, Chung K, Oberg IM, Modlin 10. C, Dobrozemsky G,etal. metastasesinpatientswith Bone Uprimny D, Kendler B, Henninger M, Gabriel D,Putzer 14. receptor scintigraphy 2007;48:508-18. andCT. JNuclMed in neuroendocrine tumors: comparison with somatostatin Heute D, Uprimny C, et al. 68Ga-DOTA-Tyr3-octreotide PET G, Dobrozemsky D, Kendler C, Decristoforo M, Gabriel 13. neuroendocrine tumors. 2010;51:1349-56. JNuclMed well-di&erentiated with patients in therapy radionuclide prediction of response to somatostatin receptor-mediated 68Ga-DOTATATEal. et B, Goke C, bleis early the for PET/CT Haug AR, 12. Auernhammer CJ, Wangler B, Schmidt GP, Ue- Cancer 2008;112:2447-55. 18F-FDG. and (DOTA-DPhe1,Tyr3-octreotate) 68Ga-DOTATATE using PET/CT combined with tumors neuroendocrine of imaging Functional al. et T, Win S, Gacinovic G, Conway A, Groves JB, Bomanji I, Kayani 11. tumors. AnnSurg 2010;252:850-6. multimodal management of patients with neuroendocrine tron emission tomography/computed tomography on the ckisch A, Kuehl H, et al. The impact of 68Ga-DOTATOC posi- Bo- M, Malago A, Radtke GC, Sotiropoulos A, Frilling17. 2008;35:1431-8. Imaging Mol J NuclMed entero-pancreatic and lung neuro-endocrine tumours. Eur DOTA-NOC and 18F-DOPA PET for the detection of gastro- 68Ga- between Comparison al. et D, Rubello G, Montini V,P,TomassettiP,AmbrosiniD, 16. Castellucci Campana 2009;36:765-70. Imaging Mol Med Nucl J Eur tumours. neuroendocrine metastatic TATE and 18F-DOPA PET in patients with well-di&erentiated G, Goke B, et al. Intraindividual comparison of 68Ga-DOTA- Haug A, 15. Auernhammer CJ, Wangler B, Tiling R, Schmidt 2009;50:1214-21. Med Nucl J scintigraphy. bone and CT to comparison in 68Ga-DOTA-Tyr3-octreotide PET tumor: neuroendocrine

EANM 18. Prasad V, Ambrosini V, Hommann M, Hoersch D, Fanti S, 21. Fellner M, Baum RP, Kubicek V, Hermann P, Lukes I, Prasad Baum RP. Detection of unknown primary neuroendocrine V, et al. PET/CT imaging of osteoblastic bone metastases tumours (CUP-NET) using (68)Ga-DOTA-NOC receptor PET/ with (68)Ga-bisphosphonates: *rst human study. Eur J Nucl CT. Eur J Nucl Med Mol Imaging 2010;37:67-77. Med Mol Imaging 2010;37:834.

19. Conry BG, Papathanasiou ND, Prakash V, Kayani I, Caplin 22. Antunes P, Ginj M, Zhang H, Waser B, Baum RP, Reubi M, Mahmood S, et al. Comparison of (68)Ga-DOTATATE and JC, et al. Are radiogallium-labelled DOTA-conjugated so- (18)F-%uorodeoxyglucose PET/CT in the detection of re- matostatin analogues superior to those labelled with other current medullary thyroid carcinoma. Eur J Nucl Med Mol radiometals? Eur J Nucl Med Mol Imaging 2007;34:982-93. Imaging 2010;37:49-57.

20. Nanni C, Errani C, Boriani L, Fantini L, Ambrosini V, Bo- schi S, et al. 68Ga-citrate PET/CT for evaluating patients with infections of the bone: preliminary results. J Nucl Med 2010;51:1932-6.

Table 1: A+nity pro*les (IC values1) of peptides for SSTRs 1–5 50

SSTR-1 SSTR-2 SSTR-3 SSTR-4 SSTR-5

SST-28 3.8±0.3 2.5±0.3 5.7±0.6 4.2±0.3 3.7±0.4

Ga-DOTA-TOC >10,000 2.5±0.5 613±140 > 1,000 73±21

Ga-DOTA-TATE >10,000 0.20±0.04 >1,000 300±140 377±18

Ga-DOTA-NOC >10,000 1.9±0.4 40±5.8 260±74 7.2±1.6

In-DTPA-octreotide >10,000 22±3.6 182±13 >1,000 237±52

1 The half-maximal inhibitory concentration (IC ) is a measure of the e&ectiveness of a compound in inhibiting 50 a biological or biochemical function. This quantitative measure indicates how much of a particular drug or other substance (inhibitor) is needed to inhibit a given biological process (or component of a process, i.e. an enzyme, cell, cell receptor or microorganism) by half. The lower the IC value, the higher the a+nity of the 50 tracer. IC values are in nmol/l (mean±SE). Data are exported from Reubi et al. [6] and Antunes et al [22] 50

42 mains the main radiopharmaceutical for PET/ quently proven. Fluorodeoxyglucose (FDG) re- appr scanners PET/CT integrated of introduction the Since Chiti Wim J.G. andArturo Oyen 2.1 Introduction treatment planning based on molecular imag- radiation and monitoring response therapy towards progressing When acquisition. data and preparation patient on demand limited a places procedure qualitative and standard a Such PET/CT. whole-body using re-staging and staging beyond evolving rapidly is *eld the as discussed, are indications Upcoming patients and the management of their disease. cancer of follow-up and process diagnostic the in rolesigni*cant or decisiveestablished, an plays PET/CT which in types tumour for summarised are PET/CT of performance and for indications the sections, following the In oncology. in imaging molecular of use the celerating ac- and further even PET/CT of application the enhancing are derivatives somatostatin as such agents new of opment 18 and (FLT) %uorothymidine as such tracers well-known of revival a although CT, oncology have been investigated and subse- in technology this for indications of number Chapter 2 – Clinical applications of PET/CT in oncolo F-labelled choline derivatives and the devel- oximately 10 years ago, a large a ago,years 10 oximately 68 Ga-labelled Ga-labelled 11 C- and C-

43 EANM procedure guideline for tumour PET PET tumour for guideline procedure EANM gramme, based on the “FDG and PET PET/CT: pro- accreditation FDG-PET/CT an as up set Research initiative for Multicentre Nuclear Medicine and an as 2006 in founded (EARL), Ltd Research EANM through process this leading is EANM demand for strict standardisation, QC and QA. radiation treatment plans, there is an absolute equate incorporation of molecular imaging in To allow robust quanti*cation of data and ad- ing results, the demands dramatically increase. molecular imaging inoncology. of impact the enhance further and work our of quality the increase will medicine nuclear of *eld the within collaboration tidisciplinary mul- close the Ultimately, technologist. cine medi- nuclear the of professionalism the on demands higher place do processing image and acquisition for protocols to adherence strict and procedures agreed to according devices other and scanners of set-up equate ad- patients, of preparation and instruction Correct success. its to crucial indeed, and, technologist in this process will be important January 2010. The role of the nuclear medicine imaging: version 1.0”, published in in EJNMMI (http://earl.eanm.org). The project is gy EANM Chapter 2 – Clinical applications of PET/CT in oncology 2.2 Head and neck cancer Murat Fani Bozkurt

Introduction Role of PET/CT Head and neck cancer is the sixth most com- The role of PET/CT in the management of mon malignancy worldwide and accounts for head and neck cancer is increasing. The most 2–5% of all cancers. The great majority of head commonly used PET radiotracer is FDG, as in and neck cancers are epithelial cancers, most other tumours. The normal pattern of FDG of which are classi*ed as oropharyngeal, naso- uptake in the head and neck region is some- pharyngeal or laryngeal squamous cell cancers what complex and some normal variants may according to their anatomical location. Since mimic pathological uptake. PET/CT is a more the head and neck region is quite a limited area, useful imaging modality compared with PET the diagnosis and treatment of head and neck alone since it can provide a combination of cancers are di+cult and require a team of sur- anatomical and functional information that geons, oncologists, radiation oncologists and helps to identify normal variants more clearly. imaging specialists. The treatment options con- The indications for PET/CT in head and neck sist of surgery, radiotherapy, chemotherapy or, cancer are as follows: as employed in most patients, a combination thereof. Appropriate diagnosis which leads to 1. Staging of primary head and neck cancer use of suitable treatment options is very impor- 2. Identi*cation of sites of recurrence tant both for disease control and for the well- 3. Distinguishing residual disease from post- being of patients with regard to daily activities operative changes such as speaking and feeding [1]. 4. Assessment of the site of an unknown primary tumour Less than one-third of head and neck cancers 5. Therapy response evaluation present as stage I or II, i.e. most tumours are at 6. Evaluation of prognosis a more advanced stage at the time of presentation. The most common sites for head and neck Patient preparation and PET/CT imaging cancers are the tongue base and the tonsillar protocol fossa, which together account for approxi- Prior to oncological FDG PET/CT imaging, pa- mately 75% of all sites. Tobacco and alcohol tients should be in the fasting state for at least are causative factors for head and neck cancers 4–6 h. Blood glucose level is measured before and heavy use of both tobacco and alcohol FDG injection. The ideal glucose level is gener- often leads to an additional primary tumour ally accepted to be lower than 140 mg/dl but of the lung or oesophagus. Early diagnosis and especially for some diabetic patients, whose appropriate treatment are of great importance blood glucose levels are poorly regulated, for long-term survival, as in many other types glucose levels up to 200 mg/dl are accepted. of cancer [2]. Insulin should be avoided soon before FDG

44 Chapter 2 – Clinical applications of PET/CT in oncology: 2.2 Head and neck cancer

injection but may be applied a few hours be- scanning, immobilisation of the patient, espe- forehand whenever needed. After intravenous cially the head region, is of great importance FDG injection, patients are advised to lie still for interpretation. Arms down position is the and rest for at least 45 min, the uptake period preferred imaging position for head and neck of FDG. Since body motion and some mus- cancer patients. cular activities such as speaking and chew- ing can give rise to muscular uptake which Staging of head and neck cancer can cause confusion at scan interpretation, More than 95% of all head and neck cancers patients should be encouraged to avoid any can be detected by PET/CT. PET/CT is very suc- unnecessary body motion, eating, drinking cessful for assessment of tumour extent but and speaking (Fig. 1). it is less useful for the assessment of mucosal or submucosal extent owing to the limited EANM spatial resolution.

Lymph node metastasis has prognostic value in head and neck cancer. Given the fact that more than 300 of approximately 800 lymph nodes in the entire human body are located in

Courtesy of Başkent University Department of Nuclear Medicine, TURKEY Adana, Mehmet Aydın, the head and neck region, it is very important Figure 1: Unilateral FDG uptake at the vocal to assess whether lymph nodes harbour meta- cords (arrow), due to vocal cord paralysis static disease. Anatomical imaging modalities such as ultrasonography, CT and MRI rely on Patients should micturate following the up- lymph node size to distinguish metastatic take period and they are then taken into the nodes. However, it is known that nodal size scanning room, where they lie supine on the alone cannot di&erentiate malignant involve- imaging table. The CT part of the study is gen- ment of lymph nodes. Small lymph nodes may erally performed *rst. Intravenous contrast harbour metastasis while large lymph nodes agents are seldom used when oral agents can may only be reactive. PET/CT has been shown be administered to outline normal bowel. For to be more sensitive than MRI and CT in the head and neck cancer imaging, the scan range detection of lymph node metastasis. The often includes the whole skull to the mid- success of PET/CT is size dependent but with thighs. Although this can di&er among PET/ the help of modern PET/CT cameras with im- CT scanners, images are acquired in blocks of proved resolution capacity, small nodes of a 15 cm where *ve to six blocks (bed positions) few millimetres can be detected.Ultrasound are needed to complete the study. During the guided *ne needle aspiration biopsy of lymph

45 nodes (even when smaller than 10 mm) acts PET/CT can detect occult disease in about as a standard diagnostic tool for metastatic 7% of patients. PET/CT is more helpful in pa- lymph node involvement. PET/CT is successful tients with T1–T3 disease, among whom the in detecting occult lymph node metastasis in use of PET can reduce the probability of oc- clinically node-negative (N0) head and neck cult lymph node metastasis to below 15% [3] cancers, most of which are oral or oropha- (Figs. 2, 3). ryngeal cancers. Studies have reported that Courtesy of Başkent University Department of Nuclear Medicine, TURKEY Adana, Mehmet Aydın, A B C D Figure 2A–D: Coronal (A), sagittal (B) and transaxial (C and D) FDG PET/CT images of a 45-year-old male with the diagnosis of nasopharyngeal cancer, showing the primary tumour (wide arrows) and lymph node metastasis (arrows) Courtesy of Başkent University Department of Nuclear Medicine, TURKEY Adana, Mehmet Aydın,

A B Figure 3A, B: Whole-body maximum intensity projection (A) and transaxial (B) FDG PET/CT images of a 35-year-old male with the diagnosis of metastatic head and neck cancer to cervical lymph nodes (wide arrows) and iliac bone (arrow)

46 Chapter 2 – Clinical applications of PET/CT in oncology: 2.2 Head and neck cancer

PET/CT vs sentinel lymph node biopsy Detection of the site of an unknown Although sentinel lymph node biopsy is not primary yet a standard of care procedure for head and About 1 in 20 cases of head and neck cancer neck cancers, there are some reports which presents only as a metastatic lymph node support its use in the disease management. without any clues regarding the primary site PET/CT cannot replace sentinel lymph node on conventional imaging modalities. As al- biopsy in head and neck cancer because of ready mentioned, early detection and treat- the impact of the limited spatial resolution ment are of great importance for prognosis. on detection of micrometastases. Owing to It has been reported that PET/CT can detect this limited spatial resolution, the sensitivity the site of an unknown primary in 20–50% of PET/CT for detection of lymph node me- of cases presenting with malignant cervical tastasis is lower than that of sentinel lymph lymph nodes. The most common pitfall of PET EANM node biopsy, though the speci*city of PET/CT imaging for this purpose is physiological up- is reported to be higher. It is recommended take of FDG in the lingual and palatine tonsils. that sentinel lymph node biopsy should be Any asymmetric pattern of uptake should be performed when PET/CT is negative and neck regarded as suggestive of an unknown pri- dissection when PET/CT is positive for lymph mary site and the utmost care must be taken node metastasis. This potential use of PET/CT when interpreting the PET/CT scan [2,4]. has been reported to reduce the number of futile neck dissections. Evaluation of response to therapy PET/CT permits not only visual but, more Detection of synchronous primary cancer importantly, (semi-)quantitative evaluation. PET/CT serves as a very useful imaging modal- A quantitative approach makes it possible ity for the detection of synchronous primary for the imaging specialist to interpret more disease in patients with head and cancer. Since objectively. For therapy response evaluation, there is a common aetiological e&ect of alcohol the standardised uptake value (SUV) of lesions and tobacco for head and neck and respiratory is measured before therapy and compared tract cancers, synchronous lesions are found in with the SUV measured after chemotherapy or nearly 20% of patients at a rate of 5% per year. radiation therapy. Theoretically, it is assumed Conventional imaging modalities may have a that any signi*cant reduction in metabolic limited role compared with PET/CT, which is uptake of a malignant lesion after therapy often capable of detecting lesions that have correlates with a good therapy response and not been suspected. If PET only imaging is done is a predictor of a good outcome. There is insread of PET/CT, then combination with CT still debate regarding the exact time period of the chest becomes crucial. appropriate for evaluation of response after

47 chemotherapy and radiation therapy as well Detection of residual disease and as the exact decrease in SUV indicative of a di$erentiation of post-therapy changes good response. It is generally accepted that re- Conventional imaging modalities such as CT sponse to chemotherapy can be appropriately and MRI essentially depend on the morpho- evaluated several weeks after termination of logical features of tissues. Therefore, detection the chemotherapy. Indeed, many studies rec- of residual disease is mostly done by evalu- ommend performance of PET/CT as early as ation of any anatomical asymmetry in the 2 weeks after cessation of chemotherapy. As head and neck region. Surgery and radiation radiation therapy induces a more signi*cant therapy, however, can cause major defects in in%ammatory response, it is suggested that normal anatomy that usually reduce symme- an interval of at least 3 weeks should be al- try. For this reason, it is not always possible to lowed before evaluating treatment response. detect residual disease or to di&erentiate post- Most studies conclude that PET/CT imaging 3 therapy changes from residual or recurrent months after radiation therapy can give more disease with conventional imaging modalities. reliable results for evaluation of response to PET/CT is reported to be a more sensitive and radiation therapy. reliable tool to detect residual disease after therapy and to di&erentiate post-therapy Some studies in the literature suggest that a changes from disease if a suitable time period better long-term outcome can be predicted is allowed to elapse so that postoperative or on the basis of post-therapy FDG uptake with postradiotherapy changes can diminish. Stud- a maximum SUV of 3 or less or an 80% reduc- ies generally advocate performance of PET/ tion in SUV compared with the pretherapy CT at least 4 months after major surgery, and SUV. More data are warranted to clarify this it has been reported that by 6 months after issue [5]. therapy, negative FDG PET/CT has a very high negative predictive value in ruling out malignant disease [5,6].

48 3. Stuckensen Stuckensen T, Kovacs AF, Adams S, Baum RP. Staging of 3. Am2000;38:999-1012. neck oncology. ClinNorth Radiol Chisin R, Macapinlac HA, The indications of in FDG-PET 2. 1337-43. 2002;179: Roentgenol J Am AJR treatment. fromresulting uptake in normal anatomy, in benign lesions and in changes Nucl Med 2004;34:180-97. Nucl Med head and neck cancer including thyroid carcinoma. Semin of imaging emission PositronHW. YeungH, Schoder 6. neck andtheesophagus. 2004;45:56-68. JNuclMed and head of carcinoma with patients in response therapy of assessment the in PET SJ. Goldsmith L, Kostakoglu 5. Cancer 1998;82:1160-6. tumors cult using primary positron emission tomography. Kole AC, Niewig OE, Pruim J. Detection of unknown oc- 4. MRI. JCraniomaxillofacSurg 2000;28:319-24. mas: a prospective of comparison PET, and ultrasound, CT the neck in patients with oral cavity squamous cell carcino- sion t sion GW, Goerres Hany Von GK, Schultess TF.Positron emis- 1. References References Chapter 2.2 omography and PET/CT of the head and neck: FDG neck: and head the of PET/CT and omography 49

EANM Chapter 2 – Clinical applications of PET/CT in oncology 2.3 Lung cancer Francesco Giammarile and Claire Houzard

Introduction Initial staging Lung cancer is the most common cancer Despite many advances in the diagnosis, stag- worldwide, accounting for 12.3% of all new ing and treatment of NSCLC, the overall 5-year cases; it is also the leading cause of death survival rate of patients with resectable NSCLC globally. Histologically, lung cancer is classi*ed is less than 50%. This suboptimal survival rate into two main categories: non-small cell lung is likely due to many factors, including the cancer (NSCLC, 80% of cases) and small cell aggressiveness of the speci*c phenotype, lo- lung cancer (SCLC, 20%). NSCLC occurs in sev- cally advanced disease at presentation and eral histological variants, including squamous inaccurate pretreatment staging. It is plausible cell carcinoma (30%), adenocarcinoma (30%), that undetected locoregional and distant mi- large cell carcinoma (10%) and less frequent crometastatic disease at the time of presenta- types such as adenosquamous carcinoma, car- tion results in inappropriate treatment and, cinoid tumours, mucoepidermoid carcinoma therefore, decreased stage-speci*c survival. and adenoid cystic carcinoma. Accurate clinical staging is important in order to determine the best possible therapeutic The average survival time for untreated NSCLC option and to assess prognosis. The ana- and SCLC is only 6 months and 2 months, tomical extent of the disease at the time of respectively. If diagnosed at an early stage, diagnosis is de*ned using the TNM system. NSCLC can be cured by surgical resection. Lo- An algorithm for the staging of NSCLC with cally advanced disease is treated by preopera- the implementation of 18F-FDG PET imaging tive chemoradiotherapy followed by surgical is shown in Fig. 1 [3]. resection. In contrast, the primary therapy of SCLC is systemic chemotherapy, since this tumour type has usually metastasised at the time of diagnosis. Because of this di&erence in the treatment strategies, it is essential to correctly diagnose, stage and restage patients [1,2].

50 Chapter 2 – Clinical applications of PET/CT in oncology: 2.3 Lung cancer

NSCLC

FDG PET

Distant Metastases

YES NO EANM

Chemotherapy Lymph Node

YES NO

Histology

Surgical resection Con&rmed LN Metastases

YES NO Adapted with permission from De Wever et al. [3]. Eur Respir J 2009; 33: 201-212 33: 2009; J Respir Eur [3]. al. et Wever De from permission with Adapted

Figure 1: Algorithm for the staging of NSCLC with the implementation of 18F-FDG PET imaging

51 Tumour assessment: T-stage About 75% of pulmonary nodules are found 18F-%uorodeoxyglucose (18F-FDG) positron incidentally. A solitary pulmonary nodule emission tomography (PET) can provide ad- (SPN), or coin lesion, is a non-spiculated round ditional information. The sensitivity, speci*city lesion smaller than 3 cm in diameter without and accuracy of 18F-FDG PET in di&erentiating associated atelectasis or adenopathy. Larger benign from cancerous lesions larger than 1 lesions are likely to be cancerous and prompt cm are more than 95%, 75% and 90%, respec- pathological diagnosis; subsequent resection tively. For SPN smaller than 1 cm, only strong is usually indicated. Twenty to thirty percent of uptake of 18F-FDG may be of diagnostic value patients present with SPN, and di&erentiation (Fig. 2) [3,5]. of benign lesions from malignancy may be di+cult with conventional imaging modalities such as chest radiography and computed tomography (CT) [1,2,4].

Courtesy of - Médecine Nucléaire Hospitalier Centre France Sud, Lyon a. b. c. Figure 2a–c: SPN characterisation: 18F-FDG uptake in a suspicious nodule. a CT of a 59-year-old man showing a 23-mm speculated SPN in the right superior lobe. b 18F-FDG PET image with intense uptake in the nodule. c 18F-FDG PET/CT fused image (true positive)

False-positive results are due to conditions demonstrate increased 18F-FDG uptake. Since where 18F-FDG accumulation occurs in meta- generally malignant nodules show increased bolically active tissue that is not cancerous. 18F-FDG uptake over time, while pulmonary Most benign pulmonary nodules do not ac- nodules of benign origin have a declining pat- cumulate 18F-FDG. However, active granu- tern of uptake with time, some authors have lomatous in%ammations or infections like proposed performing delayed 18F-FDG PET im- tuberculosis, sarcoidosis (Fig. 3), aspergillosis, ages or dual-time-point imaging [6]. histoplasmosis and coccidioidomycosis can

52 Chapter 2 – Clinical applications of PET/CT in oncology: 2.3 Lung cancer

Courtesy of - Centre Médecine Nucléaire France Sud, Hospitalier Lyon a. b. EANM Courtesy of - Centre Médecine Nucléaire France Sud, Hospitalier Lyon c. Figure 3a–c: Tissue characterisation: 18F-FDG uptake in sarcoidosis. a CT of a 55-year-old woman showing mediastinal nodes (red arrows). b 18F-FDG PET image with intense uptake in the nodes. c 18F-FDG PET/CT fused image (false positive)

False-negative PET images can occur for low- Nodal assessment: N-stage metabolism tumours such as bronchioloalve- The involvement of regional lymph nodes has olar carcinomas or squamous cell carcinomas a signi*cant impact on prognosis and in%u- (up to 60%) and bronchial carcinoid tumours ences treatment in NSCLC. The size, number (up to 85%). In addition, very small lesions (less and nodal levels involved in%uence survival. than 5 mm) can be missed because of the par- Only patients with limited mediastinal lymph tial volume e&ect and motion artefacts [5,7]. node metastases (stage IIIa) should proceed to surgical resection. CT provides good anatomi- Overall, 18F-FDG PET is more accurate than CT cal de*nition but signi*cant over- and under- for correct prediction of the T-stage of malig- staging. 18F-FDG PET has been shown to be nant tumours; however, di&erences are not more accurate than CT for staging mediastinal signi*cant, and CT is fundamental to assess nodes since detection is dependent not on lesion size, used for the determination of T- size but on metabolic activity of nodes (Fig. 4). stage [8].

53 Courtesy of Hospi- - Centre Médecine Nucléaire France Sud, talier Lyon a. b. c. Figure 4a–c: N staging: downstaging with 18F-FDG. a CT of a 62-year-old man showing a pathological node in the mediastinum (green circle). b 18F-FDG PET image with no uptake in the node. c 18F-FDG PET/CT fused image (true negative)

As in primary tumours, false positive uptake Assessment of distant metastases: M-stage in nodes occurs in tuberculosis, sarcoidosis, The detection of distant metastases implies histoplasmosis, Wegener’s granulomatosis, palliative treatment. At presentation, the com- amyloidosis, anthracosis and organising pneu- monest sites for metastatic disease in NSCLC monias. The clinical importance of 18F-FDG PET are, in decreasing order, brain, bone, liver and lies in the high negative predictive value in adrenals. 18F-FDG PET is a whole-body imaging lymph node staging. False-negative results system that will identify unsuspected metas- can occur in micrometastases. Therefore, a tases (Fig. 5). positive 18F-FDG PET for mediastinal nodes requires pathological con*rmation (to ensure that no potentially curable patient is refused appropriate treatment) while a negative 18F- FDG PET/CT may negate the requirement for mediastinoscopy [9].

54 Chapter 2 – Clinical applications of PET/CT in oncology: 2.3 Lung cancer EANM

Courtesy of Médecine Nucléaire - Centre Hospitalier Lyon Sud, France Sud, Hospitalier Lyon Courtesy - Centre of Médecine Nucléaire a. c. d.

Courtesy - of Médecine Nucléaire France Sud, Hospitalier Lyon Centre b. e. Figure 5a–e. M staging: upstaging with 18F-FDG. a,b CT of a 60-year-old man showing lung cancer and normal left adrenal (red arrow). c 18F-FDG PET image with necrotic primitive neoplastic pulmonary tumour (yellow arrow) and adrenal metastasis (red arrow). d,e 18F-FDG PET/CT fused image (true positive)

55 It is signi*cantly better than CT in detect- Response to treatment ing extrathoracic metastases, but is limited Neoadjuvant chemoradiation therapy fol- in assessing brain metastases because of the lowed by curative resection is an acceptable high rate of glucose uptake by brain cells (in treatment for resectable stage IIIa NSCLC. this case, additional CT or MRI is performed) Owing to a lack in survival bene*t, in stage [10,11]. IIIb and unresectable stage IIIa patients only chemoradiation is proposed. Follow-up Assessment of response to these treatments Prognosis may result in alteration of the course of man- There are numerous reports on the prognos- agement and provide prognostic information. tic value of 18F-FDG PET using the SUV of the Since changes in metabolism always precede primary tumour although the values used are changes in morphology, 18F-FDG PET can de- very variable [12]. termine early response to therapy (Fig. 6).

a. b. c.

Courtesy of Médecine Nucléaire - Centre Hospitalier Lyon Sud, France Sud, Hospitalier Lyon Courtesy - Centre of Médecine Nucléaire d. e. f. Figure 6a–f. Response assessment with 18F-FDG: axial 18F-FDG PET views (a,d), sagittal 18F-FDG PET views (b,e) and coronal 18F-FDG PET views (c,f) in a stage IIIb NSCLC (red arrows: primary tumour and mediastinal lymph node) before (a–c) and after (d–f) radiochemotherapy (good response with lower intensity metabolism)

56 Chapter 2 – Clinical applications of PET/CT in oncology: 2.3 Lung cancer

A reduction in tumour 18F-FDG uptake by mediastinal node involvement not seen on more than 20%, as assessed by SUV, is used CT scans, decrease the treatment *eld size as a criterion for a metabolic response in owing to better localisation of the tumour in NSCLC [13]. areas of associated atelectasis or postobstruc- tive pneumonitis and decrease inter-observer Recurrent disease variation in target de*nition [11,15]. After curative treatment of NSCLC, most of the conventional imaging techniques are sensitive The addition of 4D 18F-FDG PET (i.e. images for structural changes but have limited abil- acquired with respiratory gating) will improve ity to distinguish scars from viable tumours. image quality since 3D PET images su&er from 18F-FDG PET is both sensitive and speci*c for motion blurring due to breathing. Whether recurrent disease, with a high negative pre- the resulting modi*cations in treatment plan EANM dictive value, and will alter treatment plans in will signi*cantly improve tumour control and up to 63% of patients. However, the clinical quality of life has still to be demonstrated signi*cance of this potential superiority is still [11,15,16]. unknown [14]. New radiopharmaceuticals for PET imaging Future applications and developments Proliferation and metabolic activity could be analysed with 18F-%uorothymidine and Radiation treatment planning 18F-%uorocholine. Apoptosis is studied with In localised NSCLC, 18F-FDG PET has been 18F-annexin V and hypoxia with 18F-%uoromi- widely explored and has been recommended sonidazole. However, to date, none of these as a standard means of evaluation for patients markers under development has proved su- undergoing radiotherapy or surgery [2,3]. It perior to 18F-FDG in published studies and permits assessment of stage with a higher no tumour-speci*c radiopharmaceutical for sensitivity and speci*city than standard mor- NSCLC has been identi*ed [17]. phological imaging and is therefore of major importance for therapeutic strategy [4-7]. The Economic aspects potential impact of 18F-FDG PET on radiother- Several studies have evaluated the cost-e&ec- apy patient management may be particularly tiveness of PET/CT in NSCLC. All the authors important as it may upstage the extent of have concluded that 18F-FDG PET is cost-ef- disease, change the treatment intent from fective in the management of SPN and in the curative to palliative, increase the radiation pretreatment staging of NSCLC [18]. *eld size owing to the *nding of unsuspected

57 Conclusion 18F-FDG PET is an imaging technology with evolving potential. The advantage of 18F-FDG PET lies in its ability to detect metabolic changes in cancer cells before anatomical changes (commonly identi*ed by conventional imaging modalities). This advantage may help to achieve more accurate staging than is possible with conventional imaging. It may also identify tumours at an earlier stage, assess their response to neoadjuvant therapy and help with follow-up surveillance. The potential capability of PET in assessing the tumour responsiveness to chemotherapy can be used as a prognostic factor, thereby in%uencing the direction of further management. Furthermore, 18F-FDG PET can be used in conformal radiotherapy planning in order to reduce the radiation dose to normal tissue and deliver higher tumour dose.

58 3. De Wever W, Stroobants S, Coolen J, Verschakelen JA. Verschakelen J, Coolen S, Stroobants W,Wever De 3. nodule. 2003;348:2535-42. NEnglJMed pulmonary solitary The SH. Feinsilver AM, Fein D, Ost 2. Oncol 2004;22:330-53. Clin J 2003. update guideline: cancer lung cell non-small- unresectable of treatment Oncology Clinical malities. Chest2001;120:1791-7. of malignancy in patients presenting with new lung abnor- xyglucose positon emisson tomography for the evaluation Lee J, 10. Aronchick JM, Alavi A. Accuracy of 18-%uorodeo- lung cancer. AmJClinOncol2000;23:47-52. resectable potentially in imaging FDG-PET with staging al. Accuracy and clinical impact of mediastinal lymph node Wenget JuillardG, E, A, Sadeghi A, Safa TranS, Rege L, 9. of non-smallcelllungcancer. AnnSurg 1999;229:286-91. computed tomographic scanning in preoperative staging tive e+cacy of positron emission tomography with FDG and Gupta NC, Graeber GM, Rogers JS, Bishop HA. Compara- 8. 2006;47:426-31. value of2.5.JNuclMed lesions with 18-FDG uptake below the standardized uptake forpulmonary diagnosisof PET of al.Accuracy et A, Nagai Hashimoto Y, Tsujikawa T, Kondo C, Maki 7. M, Momose M, JAMA 2001;285:936-7. sis of pulmonary nodules and mass lesions: a meta-analysis. DK. Accuracy of positron emission tomography for diagno- Gould MK, Maclean CC, Kuschner WG, Rydzac CE, Owens 6. *ndings, AmJRoentgenol 1998;170:935-9. PET FDG carcinoma: bronchioloalveolar of form Localized al. BT, Kim Kim Y,LeeKS, YoonOJ,Kwonet EM, SB,Cheon 5. nodule. Chest2003;123;89-96. pulmonary solitary MD.The Ianettoni EA, Kazerooni KR, Flaherty BB, Tan 4. 2009;33:201-12. J Respir Eur integration. clinical and aspects technical cer: Integrated in PET/CT the staging of nonsmall cell lung can- . P*ster DG, Johnson DH, Azzoli CG. American Society of 1. References References Chapter 2.3 59 tomography/computed tomography and positron emis- positron and tomography tomography/computed Hillner BE, 14. Siegel BA, Liu D. Impact of positron emission study. JClinOncol2001;19:111-8. prospective a cancer: lung non-small-cell with patients in tomography emission positron %uorodeoxyglucose F (18) of MP. impact MacManus Clinical RJ, Hicks V, Kal& 13. 2007;34:54-9. vors. Imaging Mol Eur JNucl Med survi- potential identi*es PET cancer: lung cell non-small III stage with patients of outcome on 18F-FDG-PET with staging of F. PaulsenImpact G, Friedel SM, Eschmann 12. 2001;50:287-93. Phys. Biol Oncol Radiat J Int therapy. radiation radical for apparent implications stage lung III cancer: non-small-cell in PET by metastases distant unsuspected of detection of rate High JP. Matthews RJ, Hicks MP, MacManus 11. J Technol Assess HealthCare 2008;24:212-20. mography on radiotherapy treatment decision making. Int stage early of innovation: The use of positron emission to- an at costs and e&ects short-term Assessing MO. Carrère F, Giammarile P, Pommier M, Morelle R, Remonnay 18. scan for patient. Lung theright Cancer 2008;59:48-56. PET right the Ensuring G. Cook J, Buscombe SG, Spiro 17. other Oncol2000;55:317-24. cancer.lung cell non-small volumeswith patients in Radi- lymph node (FDG-PET) staging on the radiation treatment tomographyemission positron %uoro-2-deoxy-D-glucose (18)F- of impact The al. et EK, W,WeverVerbeken De PR, LJ,LeynVanuytselJF, Vansteenkiste De 16. SG, Stroobants treatment. Cancer 2001;92:886-95. on impact high and survival with correlation powerful diotherapy candidates with nonsmall cell lung carcinoma: ra- radical in staging tomography emission positron cose %uorodeoxyglu- F-18 DL. Ball MP,RJ, MacManus Hicks 15. Registry. JClinOncol2008;26:2155-61. Oncologic PET National the from results initial cancer: with patients of management expected on alone (PET) tomography sion

EANM Chapter 2 – Clinical applications of PET/CT in oncology 2.4 Breast cancer Leonne Prompers, Emiel Beijer, Christian Urbach, Servé Halders and Felix Manuel Mottaghy

Introduction loco-regional surgery and systemic treatments Breast cancer is the most common malignan- have a curative aim. cy in women [1]. In 2008, 429,000 new cases of breast cancer were diagnosed in Europe, and Various imaging tools, including mammog- 129,000 women died from breast cancer in the raphy, MRI, bone scans, CT, PET and PET/CT, same year [2]. It is a cancer originating from are widely available for breast cancer imaging. breast tissue, most commonly either from The role of 2-[18F]%uoro-2-deoxy-D-glucose milk ducts or from the lobules that supply the (18F-FDG) PET/CT in breast cancer imaging will ducts with milk. Cancers originating from milk be the topic of this chapter. ducts are called ductal carcinomas, while cancers originating from the lobules are known as lobular carcinomas. Some breast cancers have Medicine, Maastricht University Medical Center Medical University Medicine, Maastricht receptors that are sensitive to hormones such as oestrogen and progesterone. Some are Courtesy of Departement of Nuclear of Departement Courtesy characterised by an overexpression of human epidermal growth factor receptor 2 (HER2). The advent of targeted antibody therapy with trastuzumab (Herceptin) targeting this HER2 receptor has been a major breakthrough in breast cancer treatment. Breast cancers that do not express the genes for oestrogen receptor, progesterone receptor or HER2 are called triple-negative tumours. These triple-negative 18F-FDG avidity tumours have a poorer prognosis owing to In*ltrating ductal carcinomas are 18F-FDG avid their lack of susceptibility to targeted therapy. and 18F-FDG uptake is signi*cantly higher com- In addition to the histological subtype of the pared with lobular carcinomas [3-5]. The lim- tumour, the most essential prognostic factor ited uptake in lobular carcinomas restricts the in breast cancer is the presence and extent indication for 18F-FDG PET/CT in these breast of axillary or other (lymph node) metastases. cancers. 18F-FDG uptake also increases with The survival rate drops sharply if malignant histological grade and depends on hormone axillary nodes are present and it drops even receptor status: oestrogen receptor-negative, more if distant metastases are found. Treat- progesterone receptor-negative and triple- ment has developed considerably in the past negative tumours show higher 18F-FDG avid- decade and this has had an impact on sur- ity than hormone receptor-positive tumours. vival. Certainly in patients with no metastases, In addition, a high maximum standardised

60 Chapter 2 – Clinical applications of PET/CT in oncology: 2.4 Breast cancer

uptake value is associated with a higher TNM to that for the MRI breast coil used in clinical stage [3-6]. The TNM classi*cation is a cancer routine. However, also in this study the MRI staging system that describes the extent of seemed more accurate when assessing the cancer: T describes the size of the tumour, N T-stage. Taking into account the rather high whether lymph nodes are involved and M sensitivity of MRI in the detection of breast tu- whether metastases are present. mours, the accuracy of T-staging, the relatively low cost and the absence of ionising radiation, Procedure MRI will probably remain the cornerstone in Technical protocols concerning 18F-FDG ad- the initial detection and T-staging of breast ministration, patient preparation and scan tumours. procedure in breast cancer imaging are largely equivalent to other (standard) whole-body EANM tumour imaging protocols. Of special note is that breast cancer patients should be injected Center Medical University Medicine, Maastricht in the arm contralateral to the side of interest.

In the case of bilateral breast lesions, a foot of Nuclear of Departement Courtesy injection should be administered. The PET/ CT scan should be acquired with the patient’s arms raised above the head.

Indications

Detection of primary breast tumours Previous data show that the sensitivity of PET for detecting small breast tumours (<1 cm) and in situ lesions is rather low [7, 8]. The rea- Assessment of axillary lymph node son for this may be related to tumour biology status since in situ cancers may be associated with The assessment of regional lymph node in- decreased vascularity and glycolytic activity volvement by 18F-FDG PET/CT is quite good, [3]. Nevertheless, more recently Heusner et al. although micrometastases are often missed, showed a similar accuracy of PET/CT to MRI for which diminishes the utility of PET/CT in de- the detection of breast cancer lesions using a *ning the axillary lymph node status [10-12]. PET/CT mammography protocol [9]. This pro- It seems clear that 18F-FDG PET/CT cannot re- tocol utilises a special breast-positioning aid place sentinel lymph node biopsy or axillary that allows a pendant breast position similar lymph node dissection [12].

61 Staging of breast cancer of the liver), whereas 18F-FDG PET/CT is per- Adequate staging is essential in breast cancer formed in a single examination. In addition, patients in order to select the appropriate ther- given the relatively low speci*city of conven- apeutic strategy and thereby enhance the likeli- tional imaging, supplementary examinations hood and duration of survival. Conventional are often needed to con*rm abnormalities, staging of breast cancer consists of chest x-ray, causing anxiety and delaying the start of liver ultrasonography or CT and bone scintigra- treatment [32]. One of the disadvantages of phy. In recent years, several studies have high- 18F-FDG PET/CT is that the costs of PET/CT are lighted the superiority of 18F-FDG PET(/CT) over still about 3-4 times higher than those of con- such conventional modalities in the staging ventional imaging. Whether the use of PET/ of initial and recurrent breast cancer [13-27]. CT in breast cancer staging is cost-e&ective The higher diagnostic value of 18F-FDG PET(/ is not clear at present [33]. However, several CT) resides in the more accurate detection of renowned clinics now use PET/CT for staging distant metastases. Its superiority particularly of patients with locally advanced or recurrent relates to small sites of nodal disease. Further- breast cancer because of its superiority over more, numerous studies have pointed out the conventional staging. additional value of 18F-FDG PET(/CT) in altering treatment management. Up to 42% of patients Therapy management have been reported to be up- or downstaged PET/CT has a role in assessing response to sys- by 18F-FDG PET(/CT) [18, 20, 23, 24, 26], with temic therapy [34]. In the neo-adjuvant set- consequences for treatment in up to 30% of ting, PET/CT can early identify non-responders patients [18, 19, 23]. PET/(CT) is clearly superior or chemotherapy-resistant patients, thus de- to bone scanning in demonstrating osteolytic termining timing of surgery [35]. The optimal metastases [28, 29]. Bone scanning seems to date for an interim PET/CT to evaluate che- have a better sensitivity in the detection of os- motherapy response is probably immediately teoblastic metastases, and it has therefore been before the third cycle of chemotherapy [36, suggested that the two scans should have a 37]. PET/CT might also be useful in predict- complementary role [30, 31]. ing response to therapy as it detects metabolic changes that usually precede anatomical Another advantage of 18F-FDG PET/CT over changes [38]. conventional staging is the higher patient comfort compared with conventional multi- Detection of suspected recurrences step examinations. Conventional imaging In patients who have had previous lumpec- takes at least three examinations (bone scin- tomies, mammography assessment and CT tigraphy, chest x-ray and ultrasound or CT assessment can be very di+cult owing to sur-

62 Chapter 2 – Clinical applications of PET/CT in oncology: 2.4 Breast cancer

gical scars and the e&ect of radiation therapy In the United States, approximately 1 million [39]. In these cases, PET/CT demonstrates an surgical and needle breast biopsies performed excellent ability to detect local recurrences. each year yield benign *ndings [46]. As stated before, the use of whole-body 18F-FDG PET/CT In asymptomatic breast cancer patients with is restricted in the primary diagnosis of breast rising serum tumour markers or in patients cancer. High-resolution PET, also known as with clinical symptoms, PET/CT is superior to positron emission mammography (PEM), is conventional imaging in detecting metastatic a small organ-speci*c PET device with a re- disease [25, 39-44]. ported spatial resolution of 2 mm. In addition it has the capability of providing PET biopsy guidance that facilitates percutaneous sam- pling of identi*ed lesions [47, 48]. Recent data EANM show that PEM and breast MRI have the same overall sensitivity for depiction of index cancerous lesions, including invasive cancer and DCIS [49, 50]. Furthermore, these studies show a trend towards higher speci*city of PEM. PEM is therefore less likely to prompt unnecessary biopsies, which warrants future investigations.

Conclusions The lack of recommendations to guide appropriate use of PET/CT imaging is one of the factors currently limiting its application in breast cancer. It seems clear that PET/CT is su- Courtesy of Departement of Nuclear Medicine, Maastricht University Medical Center perior to conventional imaging in the primary Positron emission mammography staging of breast cancer patients. PET/CT also MRI has been successfully implemented has a role in assessing response to systemic in many medical centres as a valuable tool therapy and in detecting local recurrences or to diagnose breast cancer and is recom- metastatic disease. The role of PET/CT in the mended as a supplemental screening tool detection of primary breast tumours is rather to mammography in women considered to limited. However, PEM could be an alterna- be at high risk for developing breast cancer tive for women who cannot tolerate MRI and [45]. One of the important issues with MRI, may play a more important role in the future however, is the rather high false-positive rate. because of its high speci*city.

63 References Chapter 2.4

References 1. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Gha- 10. Wahl RL, Siegel BA, Coleman RE, Gatsonis CG. Prospec- foor A, et al. Cancer statistics, 2005. CA: Cancer J Clin tive multicenter study of axillary nodal staging by positron 2005;55:10-30. emission tomography in breast cancer: a report of the staging breast cancer with PET Study Group. J Clin Oncol 2. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. 2004;22:277-85. Estimates of worldwide burden of cancer in 2008: GLOBO- CAN 2008. Int J Cancer 2010;127:2893-917. 11. Veronesi U, De Cicco C, Galimberti VE, Fernandez JR, Rotmensz N, Viale G, et al. A comparative study on the value 3. Heudel P, Cimarelli S, Montella A, Bouteille C, Mognet- of FDG-PET and sentinel node biopsy to identify occult ti T. Value of PET-FDG in primary breast cancer based on axillary metastases. Ann Oncol 2007;18:473-8. histopathological and immunohistochemical prognostic factors. Int J Clin Oncol 2010;15:588-93. 12. Heusner TA, Kuemmel S, Hahn S, Koeninger A, Otterbach F, Hamami ME, et al. Diagnostic value of full-dose FDG PET/ 4. Inoue T, Yutani K, Taguchi T, Tamaki Y, Shiba E, Noguchi CT for axillary lymph node staging in breast cancer patients. S. Preoperative evaluation of prognosis in breast cancer pa- Eur J Nucl Med Mol Imaging 2009;36:1543-50. tients by [(18)F]2-Deoxy-2-%uoro-D-glucose-positron emission tomography. J Cancer Res Clin Oncol 2004;130:273-8. 13. van der Hoeven JJ, Krak NC, Hoekstra OS, Comans EF, Boom RP, van Geldere D, et al. 18F-2-%uoro-2-deoxy-d-glu- 5. Crowe JP Jr, Adler LP, Shenk RR, Sunshine J. Positron cose positron emission tomography in staging of locally emission tomography and breast masses: comparison with advanced breast cancer. J Clin Oncol 2004;22:1253-9. clinical, mammographic, and pathological *ndings. Ann Surg Oncol 1994;1:132-40. 14. van Oost FJ, van der Hoeven JJ, Hoekstra OS, Voogd AC, Coebergh JW, van de Poll-Franse LV. Staging in patients with 6. Basu S, Chen W, Tchou J, Mavi A, Cermik T, Czerniecki B, et locoregionally recurrent breast cancer: current practice and al. Comparison of triple-negative and estrogen receptor-po- prospects for positron emission tomography. Eur J Cancer sitive/progesterone receptor-positive/HER2-negative breast 2004;40:1545-53. carcinoma using quantitative %uorine-18 %uorodeoxyglucose/positron emission tomography imaging parameters: 15. Port ER, Yeung H, Gonen M, Liberman L, Caravelli J, a potentially useful method for disease characterization. Borgen P, et al. 18F-2-%uoro-2-deoxy-D-glucose positron Cancer 2008;112:995-1000. emission tomography scanning a&ects surgical management in selected patients with high-risk, operable breast 7. Avril N, Rose CA, Schelling M, Dose J, Kuhn W, Bense S, carcinoma. Ann Surg Oncol 2006;13:677-84. et al. Breast imaging with positron emission tomography and %uorine-18 %uorodeoxyglucose: use and limitations. J 16. Mahner S, Schirrmacher S, Brenner W, Jenicke L, Haber- Clin Oncol 2000;18:3495-502. mann CR, Avril N, et al. Comparison between positron emission tomography using 2-[%uorine-18]%uoro-2-deoxy-D- 8. Uematsu T, Kasami M, Yuen S. Comparison of FDG PET glucose, conventional imaging and computed tomography and MRI for evaluating the tumor extent of breast cancer for staging of breast cancer. Ann Oncol 2008;19:1249-54. and the impact of FDG PET on the systemic staging and prognosis of patients who are candidates for breast-conserving 17. Eubank WB, Manko& DA, Takasugi J, Vesselle H, Eary JF, therapy. Breast Cancer 2009;16:97-104. Shanley TJ, et al. 18%uorodeoxyglucose positron emission tomography to detect mediastinal or internal mammary 9. Heusner TA, Kuemmel S, Umutlu L, Koeninger A, Freu- metastases in breast cancer. J Clin Oncol 2001;19:3516-23. denberg LS, Hauth EA, et al. Breast cancer staging in a single session: whole-body PET/CT mammography. J Nucl Med. 18. Dizendorf EV, Baumert BG, von Schulthess GK, Lutolf 2008;49:1215-22. UM, Steinert HC. Impact of whole-body 18F-FDG PET on staging and managing patients for radiation therapy. J Nucl Med 2003;44:24-9.

64 Chapter 2 – Clinical applications of PET/CT in oncology: 2.4 Breast cancer

19. Groheux D, Moretti JL, Baillet G, Espie M, Giacchetti S, 28. Cook GJ, Houston S, Rubens R, Maisey MN, Fogelman I. Hindie E, et al. E&ect of (18)F-FDG PET/CT imaging in pati- Detection of bone metastases in breast cancer by 18FDG ents with clinical stage II and III breast cancer. Int J Radiat PET: di&ering metabolic activity in osteoblastic and osteo- Oncol Biol Phys 2008;71:695-704. lytic lesions. J Clin Oncol 1998;16:3375-9.

20. Fuster D, Duch J, Paredes P, Velasco M, Munoz M, Santa- 29. Du Y, Cullum I, Illidge TM, Ell PJ. Fusion of metabolic func- maria G, et al. Preoperative staging of large primary breast tion and morphology: sequential [18F]%uorodeoxyglucose cancer with [18F]%uorodeoxyglucose positron emission to- positron-emission tomography/computed tomography mography/computed tomography compared with conven- studies yield new insights into the natural history of bone tional imaging procedures. J Clin Oncol 2008;26:4746-51. metastases in breast cancer. J Clin Oncol 2007;25:3440-7.

21. Dose J, Bleckmann C, Bachmann S, Bohuslavizki KH, 30. Ben-Haim S, Israel O. Breast cancer: role of SPECT Berger J, Jenicke L, et al. Comparison of %uorodeoxyglucose and PET in imaging bone metastases. Semin Nucl Med positron emission tomography and „conventional diagnos- 2009;39:408-15. tic procedures“ for the detection of distant metastases in breast cancer patients. Nucl Med Commun 2002;23:857-64. 31. Costelloe CM, Rohren EM, Madewell JE, Hamaoka T, The-

riault RL, Yu TK, et al. Imaging bone metastases in breast EANM 22. Isasi CR, Moadel RM, Blaufox MD. A meta-analysis of cancer: techniques and recommendations for diagnosis. FDG-PET for the evaluation of breast cancer recurrence Lancet Oncol 2009;10:606-14. and metastases. Breast Cancer Res Treat 2005;90:105-12. 32. Gerber B, Seitz E, Muller H, Krause A, Reimer T, Kundt 23. Eubank WB, Manko& D, Bhattacharya M, Gralow J, Lin- G, et al. Perioperative screening for metastatic disease is den H, Ellis G, et al. Impact of FDG PET on de*ning the not indicated in patients with primary breast cancer and extent of disease and on the treatment of patients with no clinical signs of tumor spread. Breast Cancer Res Treat recurrent or metastatic breast cancer. AJR Am J Roentgenol 2003;82:29-37. 2004;183:479-86. 33. Pennant M, Takwoingi Y, Pennant L, Davenport C, Fry- 24. Jager JJ, Keymeulen K, Beets-Tan RG, Hupperets P, Smith A, Eisinga A, et al. A systematic review of positron van Kroonenburgh M, Houben R, et al. FDG-PET-CT for emission tomography (PET) and positron emission to- staging of high-risk breast cancer patients reduces the mography/computed tomography (PET/CT) for the dia- number of further examinations: A pilot study. Acta Oncol gnosis of breast cancer recurrence. Health Technol Assess 2010;49:185-91. 2010;14:1-103.

25. Radan L, Ben-Haim S, Bar-Shalom R, Guralnik L, Israel O. 34. Avril N, Sassen S, Roylance R. Response to therapy in The role of FDG-PET/CT in suspected recurrence of breast breast cancer. J Nucl Med 2009;50 Suppl 1:55S-63S. cancer. Cancer 2006;107:2545-51. 35. Krak NC, Boellaard R, Hoekstra OS, Twisk JW, Hoekstra CJ, 26. Dirisamer A, Halpern BS, Flory D, Wolf F, Beheshti M, Lammertsma AA. E&ects of ROI de*nition and reconstruc- Mayerhoefer ME, et al. Performance of integrated FDG-PET/ tion method on quantitative outcome and applicability in contrast-enhanced CT in the staging and restaging of colo- a response monitoring trial. Eur J Nucl Med Mol Imaging rectal cancer: comparison with PET and enhanced CT. Eur 2005;32:294-301. J Radiol 2010;73:324-8. 36. Rousseau C, Devillers A, Sagan C, Ferrer L, Bridji B, Cam- 27. Segaert I, Mottaghy F, Ceyssens S, De Wever W, Stroo- pion L, et al. Monitoring of early response to neoadjuvant bants S, Van Ongeval C, et al. Additional value of PET-CT chemotherapy in stage II and III breast cancer by [18F] in staging of clinical stage IIB and III breast cancer. Breast J %uorodeoxyglucose positron emission tomography. J Clin 2010;16:617-24. Oncol 2006;24:5366-72.

65 37. Martoni AA, Zamagni C, Quercia S, Rosati M, Cacciari 44. Zangheri B, Messa C, Picchio M, Gianolli L, Landoni C, N, Bernardi A, et al. Early (18)F-2-%uoro-2-deoxy-d-glucose Fazio F. PET/CT and breast cancer. Eur J Nucl Med Mol Ima- positron emission tomography may identify a subset of ging 2004;31 Suppl 1:S135-42. patients with estrogen receptor-positive breast cancer who will not respond optimally to preoperative chemotherapy. 45. Enriquez L, Listinsky J. Role of MRI in breast cancer ma- Cancer 2010;116:805-13. nagement. Cleve Clin J Med 2009;76:525-32.

38. Duch J, Fuster D, Munoz M, Fernandez PL, Paredes P, 46. Tafra L, Fine R, Whitworth P, Berry M, Woods J, Ekbom Fontanillas M, et al. 18F-FDG PET/CT for early prediction of G, et al. Prospective randomized study comparing cryo- response to neoadjuvant chemotherapy in breast cancer. assisted and needle-wire localization of ultrasound-visible Eur J Nucl Med Mol Imaging 2009;36:1551-7. breast tumors. Am J Surg 2006;192:462-70.

39. Siggelkow W, Rath W, Buell U, Zimny M. FDG PET and 47. Weinberg IN, Beylin D, Zavarzin V, Yarnall S, Stepanov PY, tumour markers in the diagnosis of recurrent and metastatic Anashkin E, et al. Positron emission mammography: high- breast cancer. Eur J Nucl Med Mol Imaging 2004;31 Suppl resolution biochemical breast imaging. Technol Cancer Res 1:S118-24. Treat 2005;4:55-60.

40. Lind P, Igerc I, Beyer T, Reinprecht P, Hausegger K. Ad- 48. Weinberg IN. Applications for positron emission mam- vantages and limitations of FDG PET in the follow-up of mography. Phys Med 2006;21 Suppl 1:132-7. breast cancer. Eur J Nucl Med Mol Imaging 2004;31 Suppl 49. Schilling K, Narayanan D, Kalinyak JE, The J, Velasquez 1:S125-34. MV, Kahn S, et al. Positron emission mammography in 41. Suarez M, Perez-Castejon MJ, Jimenez A, Domper M, breast cancer presurgical planning: comparisons with ma- Ruiz G, Montz R, et al. Early diagnosis of recurrent breast gnetic resonance imaging. Eur J Nucl Med Mol Imaging cancer with FDG-PET in patients with progressive elevation 2011;38:23-36. of serum tumor markers. Q J Nucl Med 2002;46:113-21. 50. Berg WA, Madsen KS, Schilling K, Tartar M, Pisano ED, 42. Gallowitsch HJ, Kresnik E, Gasser J, Kumnig G, Igerc I, Larsen LH, et al. Breast cancer: comparative e&ectiveness Mikosch P, et al. F-18 %uorodeoxyglucose positron-emission of positron emission mammography and MR imaging in tomography in the diagnosis of tumor recurrence and me- presurgical planning for the ipsilateral breast. Radiology tastases in the follow-up of patients with breast carcino- 2011;258:59-72. ma: a comparison to conventional imaging. Invest Radiol 2003;38:250-6.

43. Tatsumi M, Cohade C, Mourtzikos KA, Fishman EK, Wahl RL. Initial experience with FDG-PET/CT in the evaluation of breast cancer. Eur J Nucl Med Mol Imaging 2006;33:254-62.

66 cases), and it was one of the leading causes leading the of one was it and cases), all of (24.1% 2006 in cancer prostate of cases incident 301,500 were there Union, ropean the highest prevalence in men. Within the Eu- is cancer Prostate Introduction Sarah Schwarzenboeck, Michael Souvatzoglou and Bernd J 2.5 Prostate cancer prostate cancer is based on increased phos- increased on based is cancer prostate image to choline of use The phospholipids. an essential constituent of the cell membrane is Choline [12]. al. et Krause see overview an 11 with PET/CT and PET using obtained been have results Promising [5–11]. derivatives and methionine %uorodihydrotestosterone, are carcinoma prostate of diagnosis the tracers that PET/CT have been introduced for metastasised prostate cancer. Further PET and [3–5]. cancer prostate recurrent of imaging and nomas carci- prostate di&erentiated of detection the for studies various in sensitivity limited 18 emission tomography are (PET, used. PET/CT) positron and (MRI) imaging resonance netic mag- (CT), tomography computed as such methods imaging functional and phological mor- methods, diagnostic these Besides [2]. biopsies ultrasound-guided and ultrasound transrectal (PSA), antigen prostate-speci*c of levels serum of measurement examination, evaluation of prostate cancer are digital rectal the in tools diagnostic used Commonly [1]. of carcinoma-associated death in men (10.4%) Chapter 2 – Clinical applications of PET/CT in oncolo observed in dedi&erentiated, aggressive and dedi&erentiated,aggressive in observed F-%uorodeoxyglucose ( F-%uorodeoxyglucose C- and C- 18 F-labelled choline derivatives; for derivatives; choline F-labelled 18 F-FDG accumulation is mostly is accumulation F-FDG among the carcinomas with 11 C- or C- 18 18 F-FDG) has shown has F-FDG) F-labelled choline F-labelled 11 C-acetate, 11 18 C- F- 67 views views the using use radio- of and PET PET/CT [16–19]. detected havebeen re- chapterThis rate transportation choline elevated an and transporters choline of expression increased Additionally, [15,16]. cells cancer prostate in lism, such as choline kinase, are up-regulated metabo- choline the of enzymes Key brane. mem- phosphatidyl the into incorporated is is the *rst step in the Kennedy pathway – and which – kinase choline by phosphorylated is via a high-a+nity system,transporter choline cells [13,14]. After uptake into the tumour cell cancer prostate in turnover phatidylcholine phos- elevated an and levels phorylcholine (Table 1). The detection rate of primary pros- primary of rate 1). detection (Table The reported lower detection rates [23, 24, 29, 31] prostate cancer [26, 30, 36], while others have high sensitivities for the detection of primary groupspatient selected havewith ies shown than lesion-based analyses. Some of the stud- (98–100%) results better shown have which ties have been based on per patient analyses, sensitivi- reported of spectively. majority The re- 73% and 91% are cancer prostate local of detection the for speci*city and sensitivity 20–37]. mean 11, The [8, controversialresults partially with studies several in examined been has cancer prostate primary of nosis 18 using PET/CT and PET of value The prostate cancer ofprimary Imaging biochemical recurrence. cancer, with special emphasis on patients with actively labelled choline derivatives in prostate F-labelled choline derivatives for the diag- the for derivatives choline F-labelled . Krause 11 C- and C- gy EANM tate cancer using 11C-choline PET and PET/CT ten not possible. Furthermore, di&erentiation may be in%uenced by the tumour con*gura- between benign prostatic hyperplasia, pros- tion: The detection of small and partly ‘rind-like’ tatitis and high-grade intraepithelial neoplasia carcinomas (onion ring form of growth) is of- is not always possible [23, 24, 29, 33, 35–37]. Courtesy of Department of Nuclear UniversityMedicine, Hospital Rostock, University of Rostock Figure 1. 65 year old patient with prostate cancer and status after radiation therapy 11/03, referred for 11C-choline PET/CT due to increasing PSA up to 3,34 ng/ml. 11C-choline PET/CT at 11/10 revealed local recurrence (A) CT scan, (B) PET scan, (C) PET/CT fused images

With respect to lymph node staging, only a choline PET/CT *ndings with nomograms, no few studies have reported results for sensitiv- statistically signi*cant di&erence was found in ity and speci*city. De Jong et al. showed that sensitivity and speci*city (Briganti nomogram: 11C-choline PET identi*ed metastatic pelvic 60% and 74%; Kattan nomogram: 60% and lymph nodes with a size between 0.5 and 3 cm 64%). Poulsen et al. examined 25 consecutive [38]. The authors reported 19 correct negative patients with primary prostate cancer using *ndings in 19 patients without lymph node 18F-choline PET/CT. The sensitivity and speci- metastases and one instance of false positive *city of 18F-choline PET/CT for patient-based 11C-choline enhancement in a lymph node lymph node staging of prostate cancer were with in%ammatory changes. Schiavina et al. 100% and 95% respectively [39]. examined 57 patients with biopsy-proven prostate cancer and an intermediate or high 11C-choline PET/CT has also been successfully risk for lymph node metastases using 11C- used in patients with advanced prostate can- choline PET/CT prior to prostatectomy and cer [40]. In these patients, PET/CT allowed a extended pelvic lymph node dissection [32]. more precise assessment of disease localisa- 11C-choline PET/CT showed a sensitivity of tion and extent within a single examination 60% and a speci*city of 98% for the detection (lymph nodes, bones and other organs, e.g. of lymph node metastases. Comparing 11C- lung) and demonstrated additional value for

68 Chapter 2 – Clinical applications of PET/CT in oncology: 2.5 Prostate cancer

the detection of bone metastases, thereby tives for the diagnosis of recurrent prostate in%uencing individualised therapy planning. cancer, showing an increasingly valuable potential of this hybrid imaging technique for Imaging of recurrent prostate cancer the detection of local, regional and systemic A number of studies have been performed on disease [22, 25, 33, 41–59]. the use of 11C- and 18F-labelled choline deriva- EANM Courtesy of Department of Nuclear UniversityMedicine, Hospital Rostock, University of Rostock

Figure 2. 72 year old patient with prostate cancer and status after salvage prostatectomy and radiation therapy 01/99, status after local recurrence and anti-androgen therapy 11/05, referred for 11C-choline PET/CT due to increasing PSA up to 3 ng/ml. 11C-choline PET/CT at 12/10 revealed lymphnode metastasis. (A) CT scan, (B) PET scan, (C) PET/CT fused images

Table 2 shows a summary of relevant publica- lated as 69.1% and 69.2% for the detection of tions regarding the diagnostic e+cacy of PET local recurrence, lymph node metastases and and PET/CT using 11C- and 18F-labelled choline bone metastases. Only a few patients were derivatives in patients with recurrent prostate under anti-androgen therapy during PET/CT cancer after primary therapy. Overall, the re- scanning (Table 2). sults of 22 groups and 1731 patients were analysed. The majority of patients were examined Recurrence of prostate cancer can be found after radical prostatectomy, while the others in 20%–50% of patients within 10 years after were initially treated with conformal radiation radical prostatectomy as primary treatment therapy and/or anti-androgen therapy. The [60,61] and in up to 53% within 10 years af- mean interval between *rst local treatment ter conformal radiation therapy [62]. In most and PET/CT at the time of biochemical recur- of the patients, recurrence is suspected as a rence was 42 months. The mean serum PSA result of an increase in the serum PSA level, value at the time of PET or PET/CT was 8.0 ng/ which is a sensitive biomarker for recurrence ml. Mean sensitivity and speci*city were calcu- of prostate cancer and often precedes clinical

69 manifestation by months or even years [63]. <4 ng/ml [44]. Rinnab et al. [54] performed 11C- However, increase in serum PSA does not al- choline PET/CT in 50 patients who had previ- low di&erentiation between local, regional and ously undergone primary therapy. Most of the systemic disease. Lymph nodes and bone are pathological 11C-choline PET/CT *ndings were the most common sites of metastases of pros- compared with histopathology as the gold tate cancer. About 50% of patients develop standard. The investigators reported that the local recurrence while the remainder su&er sensitivity of PET/CT at PSA values of less than from metastatic disease with or without local 2.5 ng/ml was 91%, though with a lower speci- recurrence [64]. It is crucial to distinguish be- *city of 50%. Scattoni et al. examined 25 pa- tween local, regional and systemic recurrence tients with biochemical recurrence and lymph for the purpose of individualised treatment. node metastases with 11C-choline PET/CT and At present, conventional imaging methods reported a high positive predictive value even such as transrectal ultrasound, CT and MRI are at low PSA concentrations. The negative pre- not su+ciently accurate in the localisation of dictive value, on the other hand, was low; this recurrence of prostate cancer. This clinical was most likely related not to lymph node situation demands a diagnostic procedure metastases but to microscopic lesions [57]. with a high sensitivity and speci*city for early Reske et al. [52] performed 11C-choline PET/CT and exact detection of disease localisation. in 49 patients with biochemical recurrence This is a key challenge that needs to be met and suspicion of local recurrence of prostate in order to improve diagnostic performance cancer. Local recurrence was biopsy proven and accuracy. in 36 of the 49 patients (74%), and 11C-choline PET/CT correctly predicted local recurrence in Picchio et al. compared 18F-FDG and 71% of these 36 patients. Husarik et al. evalu- 11C-choline for re-staging in 100 prostate can- ated the value of 18F-%uorocholine PET/CT in cer patients with biochemical recurrence and 111 patients with prostate cancer, 68 of whom a mean serum PSA level of 6.6 ng/ml. In 47% of were restaged because of biochemical recur- patients, lesions with pathological 11C-choline rence. The authors showed a sensitivity of accumulation were found [51]. Cimitan et al. 86% for the diagnosis of recurrent prostate performed an 18F-%uorocholine PET/CT study cancer [25]. Krause et al. [48] examined 63 pain 100 patients with biochemical recurrence tients with biochemical failure after primary after primary therapy. A positive 18F-%uorocho- therapy. Pathological areas were detected in line PET/CT scan was obtained in 54 of the pa- 35 of the patients (56%). The detection rate for tients (54%). Negative 18F-%uorocholine PET/ PSA values <1 ng/ml was 36%. Furthermore, CT imaging without pathological uptake was the authors demonstrated a linear correlation demonstrated in 89% of patients with a PSA between 11C-choline PET/CT detection rate

70 Chapter 2 – Clinical applications of PET/CT in oncology: 2.5 Prostate cancer

and serum PSA value. Castellucci et al. [42] lymph nodes might and micrometastases will assessed the e&ect of total PSA, PSA veloc- not be detectable by PET and PET/CT owing ity and PSA doubling time on the 11C-choline to the partial volume e&ect. In the future, ad- PET/CT detection rate in 190 patients with bio- equately designed, prospective studies with chemical recurrence of prostate cancer. The exact de*nition of patient groups and com- investigators con*rmed the linear correlation parison of imaging results with histology, as between PSA value and 11C-choline PET/CT well as multicentre trials, will be needed to detection rate. A multivariate analysis showed corroborate the data and de*ne the clinical that PSA doubling time and PSA velocity were e+cacy of PET/CT in the re-staging of pros- additional independent predictive factors for tate cancer. PET positivity. Giovacchini et al. [45] examined 358 patients with biochemical recurrence af- Imaging results using 11C- and 18F-labelled EANM ter radical prostatectomy. The percentage of choline derivatives might in%uence the thera- positive scans was 19% in those with a PSA peutic management. Distant metastases have concentration between 0.2 and 1 ng/ml, 46% to be excluded before choosing the treatment in those with a PSA of 1–3 ng/ml and 82% in modality, e.g. salvage radiation therapy, sal- those with a PSA >3 ng/ml. The authors con- vage radical prostatectomy or lymphadenec- cluded that a PSA value of 1 ng/ml might be tomy. Therapy could also be optimised using a reasonable threshold when considering a choline PET/CT. For example, a single-site local choline PET/CT scan. Castellucci et al. evalu- lymph node metastasis could be treated by ated the potential of 11C-choline PET/CT in radiation therapy with an extended *eld or re-staging of prostate cancer patients after by salvage lymphadenectomy, while systemic radical prostatectomy who presented a mild therapy could be initiated in the case of dis- increase in PSA (<1.5 ng/ml) during follow-up. seminated disease. The authors showed positive *ndings in 29 of 102 patients (28%) [43]. Therapy response assessment Systemic therapy of advanced prostate cancer Nevertheless, study results regarding the value comprises androgen deprivation by LHRH ag- of choline PET/CT in patients with recurrence onists/antagonists and anti-androgen therapy, have to be interpreted carefully owing to the chemotherapy, bisphosphonates, and – within low numbers of patients in some studies, the clinical studies – anti-neovascular agents and lack of histopathology in most studies, the use tyrosine kinase inhibitors. Methods of con- of retrospective data analysis in most studies, ventional follow-up such as clinical response, and the use of di&erent tracers and di&erent serum PSA level and conventional imaging PET and PET/CT systems. Moreover, small modalities have been insu+cient for assessing

71 the therapeutic e+cacy of any treatment regi- of 4 months (range 3–12 months). The authors men. The use of PET and PET/CT with radioac- found a signi*cant reduction in choline uptake tively labelled choline derivatives may o&er a and a high inter-patient variability. The imag- new method for early and precise prediction ing biomarker signal decrease was paralleled of therapy response and outcome of a chosen by a decrease in PSA values of 78% [24]. therapy regimen. Preclinical studies show that 11C-choline PET can be used to monitor ef- Conclusion/Outlook fects of a cytotoxic therapy in prostate cancer Choline PET/CT is a clinically valuable tool for xenograft models [65]. In an in vivo study of re-staging of patients with increasing PSA se- prostate cancer mouse xenograft models with rum levels after de*nitive local therapy. At the xenotransplanted PC3 cells in NMRI (nu/nu) present time, however, choline PET/CT can- mice under docetaxel treatment which were not be recommended as a *rst-line screening imaged with 11C-choline small animal PET/CT, procedure for primary prostate cancer in men the authors showed that docetaxel therapy at risk, though it may be of value for the de- led to a signi*cant reduction in mean 11C-cho- tection of clinically suspected prostate cancer line uptake of 45%, paralleled by signi*cant in patients with multiple negative prostate tumour growth inhibition. Prospective clinical biopsies. Preliminary studies evaluating the studies are necessary to evaluate the useful- use of choline PET/CT for therapy response ness of PET and PET/CT imaging with 11C- and assessment after anti-hormonal therapy have 18F-labelled choline for therapy response as- revealed partially promising results, and clini- sessment. These clinical studies will have to cal studies of the use of choline PET/CT for elucidate the best tracer, time of scanning and therapy monitoring in patients with metasta- clinical signi*cance of the magnitude of signal ses under docetaxel chemotherapy are being changes [66]. Currently, few data have been launched. published regarding use of PET/CT for therapy monitoring of prostate cancer patients among Although considerable progress has been whom a relevant number are under anti-an- made in recent years with PET/CT using 11C- drogen treatment. Preliminary results with 18F- and 18F-labelled choline derivatives, diagnostic %uorocholine PET [9] and 11C-choline PET/CT performance still needs to be improved. Many [24] showed a signi*cant reduction in choline strategies for prostate cancer imaging with uptake after initiation of anti-androgen ther- novel radiotracers such as bombesin-based apy. Giovacchini et al. used 11C-choline PET/ tracers or tracers for androgen receptor imag- CT to monitor a group of six patients with ing are being assessed in preclinical studies localised prostate cancer under bicalutamide and in some cases in initial clinical feasibility treatment (150 mg/day) for a median duration studies. In the future more tumour-speci*c

72 Chapter 2 – Clinical applications of PET/CT in oncology: 2.5 Prostate cancer

tracers may increase the sensitivity and speci- combining the molecular imaging properties *city of multimodality imaging for the diag- of PET with the high resolution and soft tis- nosis of prostate cancer. Advances in hybrid sue contrast of MRI and the functional and imaging, especially PET/MRI, could also lead molecular MRI techniques. to improvements in diagnostic accuracy by

Table 1 Pts. Tracer Ref. Author Year Modality Local tumour Lymph nodes (n) Sensitivity Speci&city Sensitivity Speci&city (%) (%) (%) (%) 18F-choline [26] Kwee 2005 PET 17 100 - - - EANM [33] Schmid 2005 PET/CT 19 100 - - - [27] Kwee 2006 PET 26 100 - - - [25] Husarik 2008 PET/CT 43 98 - 33 100 [34] Steuber 2010 PET/CT 20 - - 0 100 [20] Beheshti 2010 PET/CT 130 n.c. n.c. 45 96 [39] Poulsen 2010 PET/CT 25 - - 100 95 11C-choline [8] Kotzerke 2000 PET 23 100 - 50 90 [21] de Jong 2002 PET 25 100 - 80 95 [38] de Jong 2003 PET 67 - - 80 96 [35] Sutinen 2004 PET 14 100 - - - [36] Yamaguchi 2005 PET 20 100 - - - [37] Yoshida 2005 PET 13 - - - - [23] Farsad1 2005 PET/CT 36 66 81 - - [30] Reske1 2006 PET/CT 26 100 - - - [31] Scher 2007 PET/CT 58 86 70 - - [29] Martorana1 2006 PET/CT 43 66 84 - - [24] Giovacchini1 2008 PET/CT 19 72 43 - - [32] Schiavina 2008 PET/CT 57 - - 60 98 [28] Li2 2008 PET/CT 49 90 86 - -

73 Pts. Tracer Ref. Author Year Modality Local tumour Lymph nodes (n) Sensitivity Speci&city Sensitivity Speci&city (%) (%) (%) (%) [11] Watanabe 2010 PET 43 73 59 - - [67] Souvatzoglou 2011 PET/CT 43 79 - - - Sum 816 Mean 89.4 70.5 56 96.2 Median 99 75.5 55 96 Table 1: Diagnostic e+cacy of 18F-choline and 11C-choline PET and PET/CT in patients with primary prostate cancer (modi*ed and updated from [12])

1 Sextant-based comparison with histology 2 Uptake ratio of lesion to muscle was compared with histology

Table 2 Pts. Pts. Pts. Pts. Tracer Ref. Author Year Modus Time (Mo) PSA Sens. Spec. (n) (n) (n) (n)

all RP RT ADT Tx-PET/CT (ng/mL) (%) (%)

18F-choline [47] Heinisch 2006 PET/CT 34 31 3 4 - 17.1 41 - [33] Schmid 2005 PET/CT 9 8 1 - 49 14.1 100 - [44] Cimitan 2006 PET/CT 100 58 21 21 - 48.3 54 - [25] Husarik 2008 PET/CT 68 68 - 13 - 10.8 86 - [59] Vees 2007 PET 20 20 - - 35 0.4 50 - [50] Pelosi 2008 PET/CT 56 56 - - - - 43 - [58] Steiner 2009 PET/CT 47 17 30 - 67 3.3 81 - 11C-choline [51] Picchio1 2003 PET 100 77 23 - - 6.6 47 -

74 Chapter 2 – Clinical applications of PET/CT in oncology: 2.5 Prostate cancer

Pts. Pts. Pts. Pts. Tracer Ref. Author Year Modus Time (Mo) PSA Sens. Spec. (n) (n) (n) (n)

all RP RT ADT Tx-PET/CT (ng/mL) (%) (%)

[22] de Jong1 2003 PET 36 20 16 - - 12.0 55 100 [49] Ohlmann 2007 PET/CT 45 0 45 - - 7.8 65 - [54] Rinnab 2007 PET/CT 50 40 10 4 22 3.6 95 40 [57] Scattoni 2007 PET/CT 25 25 - - - 4.0 100 66 [41] Breeuwsma1 2010 PET 80 0 70 - - 12.3 81 100 [48] Krause 2008 PET/CT 63 42 21 17 47 5.9 56 - [55] Rinnab 2008 PET/CT 15 15 - - 24 2.3 100 0 [52] Reske 2008 PET/CT 49 49 - 9 59 2.0 73 88 EANM [56] Rinnab 2009 PET/CT 41 41 - - 24 2.8 93 36 [42] Castellucci 2009 PET/CT 190 190 - - 46 4.2 39 - [45] Giovacchini 2010 PET/CT 358 358 155 3.8 85 93 [46] Giovacchini 2010 PET/CT 170 170 45 3.2 87 89 [53] Richter 1 2010 PET 73 49 24 15 2.7 61 80 [43] Castellucci 2011 PET/CT 102 102 - 16 0.9 28 - Sum 1731 1436 264 254 Mean 42 8.0 69.1 69.2 Median 45.5 4.0 69 88

Table 2: Diagnostic e+cacy of 18F-choline and 11C-choline PET and PET/CT in patients with recurrent prostate cancer (modi*ed and updated from [12])

RP, radical prostatectomy; RT, radiotherapy; ADT, androgen deprivation therapy; Tx, therapy 1 PET only

75 References Chapter 2.5

References 1. Ferlay J, Autier P, Boniol M, Heanue M, Colombet M, 11. Watanabe H, Kanematsu M, Kondo H, Kako N, Yama- Boyle P. Estimates of the cancer incidence and mortality in moto N, Yamada T, et al. Preoperative detection of prostate Europe in 2006. Ann Oncol 2007;18:581-92. cancer: a comparison with 11C-choline PET, 18F-%uorodeoxyglucose PET and MR imaging. J Magn Reson Imaging 2. Smith RA, Cokkinides V, Eyre HJ. Cancer screening in the 2010;31:1151-6. United States, 2007: a review of current guidelines, practices, and prospects. CA Cancer J Clin 2007;57:90-104. 12. Krause BJ, Souvatzoglou M, Treiber U. Imaging of prostate cancer with PET/CT and radioactively labeled choline 3. Hofer C, Laubenbacher C, Block T, Breul J, Hartung R, derivatives. Urol Oncol 2011;in press. Schwaiger M. Fluorine-18-%uorodeoxyglucose positron emission tomography is useless for the detection of local 13. Ackersta& E, Glunde K, Bhujwalla ZM. Choline phospho- recurrence after radical prostatectomy. Eur Urol 1999;36:31- lipid metabolism: a target in cancer cells? J Cell Biochem 5. 2003;90:525-33.

4. Morris MJ, Akhurst T, Osman I, Nunez R, Macapinlac H, 14. Casciani E, Gualdi GF. Prostate cancer: value of magnetic Siedlecki K, et al. Fluorinated deoxyglucose positron emissi- resonance spectroscopy 3D chemical shift imaging. Abdom on tomography imaging in progressive metastatic prostate Imaging 2006;31:490-9. cancer. Urology 2002;59:913-8. 15. Ramirez de Molina A, Penalva V, Lucas L, Lacal JC. Re- 5. Nunez R, Macapinlac HA, Yeung HW, Akhurst T, Cai S, gulation of choline kinase activity by Ras proteins involves Osman I, et al. Combined 18F-FDG and 11C-methionine Ral-GDS and PI3K. Oncogene 2002;21:937-46. PET scans in patients with newly progressive metastatic prostate cancer. J Nucl Med 2002;43:46-55. 16. Ratnam S, Kent C. Early increase in choline kinase activity upon induction of the H-ras oncogene in mouse *broblast 6. Albrecht S, Buchegger F, Soloviev D, Zaidi H, Vees H, cell lines. Arch Biochem Biophys 1995;323:313-22. Khan HG, et al. (11)C-acetate PET in the early evaluation of prostate cancer recurrence. Eur J Nucl Med Mol Imaging 17. Ackersta& E, P%ug BR, Nelson JB, Bhujwalla ZM. Detec- 2007;34:185-96. tion of increased choline compounds with proton nuclear magnetic resonance spectroscopy subsequent to mali- 7. Dehdashti F, Picus J, Michalski JM, Dence CS, Siegel BA, gnant transformation of human prostatic epithelial cells. Katzenellenbogen JA, et al. Positron tomographic assess- Cancer Res 2001;61:3599-603. ment of androgen receptors in prostatic carcinoma. Eur J Nucl Med Mol Imaging 2005;32:344-50. 18. Hara T, Bansal A, DeGrado TR. Choline transporter as a novel target for molecular imaging of cancer. Mol Imaging 8. Kotzerke J, Prang J, Neumaier B, Volkmer B, Guhlmann 2006;5:498-509. A, Kleinschmidt K, et al. Experience with carbon-11 choline positron emission tomography in prostate carcinoma. Eur 19. Katz-Brull R, Degani H. Kinetics of choline transport J Nucl Med 2000;27:1415-9. and phosphorylation in human breast cancer cells; NMR application of the zero trans method. Anticancer Res 9. Larson SM, Morris M, Gunther I, Beattie B, Humm JL, 1996;16(3B):1375-80. Akhurst TA, et al. Tumor localization of 16beta-18F-%uoro- 5alpha-dihydrotestosterone versus 18F-FDG in patients 20. Beheshti M, Imamovic L, Broinger G, Vali R, Waldenber- with progressive, metastatic prostate cancer. J Nucl Med ger P, Stoiber F, et al. 18F choline PET/CT in the preoperative 2004;45:366-73. staging of prostate cancer in patients with intermediate or high risk of extracapsular disease: a prospective study of 130 10. Toth G, Lengyel Z, Balkay L, Salah MA, Tron L, Toth C. patients. Radiology 2010;254:925-33. Detection of prostate cancer with 11C-methionine positron emission tomography. J Urol 2005;173:66-9; discussion 9. 21. de Jong IJ, Pruim J, Elsinga PH, Vaalburg W, Mensink HJ. Visualization of prostate cancer with 11C-choline positron emission tomography. Eur Urol 2002;42:18-23.

76 Chapter 2 – Clinical applications of PET/CT in oncology: 2.5 Prostate cancer

22. de Jong IJ, Pruim J, Elsinga PH, Vaalburg W, Mensink HJ. 32. Schiavina R, Scattoni V, Castellucci P, Picchio M, Corti B, 11C-choline positron emission tomography for the evalu- Briganti A, et al. 11C-choline positron emission tomogra- ation after treatment of localized prostate cancer. Eur Urol phy/computerized tomography for preoperative lymph- 2003;44:32-8; discussion 8-9. node staging in intermediate-risk and high-risk prostate cancer: comparison with clinical staging nomograms. Eur 23. Farsad M, Schiavina R, Castellucci P, Nanni C, Corti B, Urol 2008;54:392-401. Martorana G, et al. Detection and localization of prostate cancer: correlation of (11)C-choline PET/CT with histopa- 33. Schmid DT, John H, Zweifel R, Cservenyak T, Westera G, thologic step-section analysis. J Nucl Med 2005;46:1642-9. Goerres GW, et al. Fluorocholine PET/CT in patients with prostate cancer: initial experience. Radiology 2005;235:623-8. 24. Giovacchini G, Picchio M, Coradeschi E, Scattoni V, Bet- tinardi V, Cozzarini C, et al. [(11)C]choline uptake with PET/ 34. Steuber T, Schlomm T, Heinzer H, Zacharias M, Ahyai S, CT for the initial diagnosis of prostate cancer: relation to Chun KF, et al. [F(18)]-%uoroethylcholine combined in-line PSA levels, tumour stage and anti-androgenic therapy. Eur PET-CT scan for detection of lymph-node metastasis in high J Nucl Med Mol Imaging 2008;35:1065-73. risk prostate cancer patients prior to radical prostatectomy: preliminary results from a prospective histology-based stu- 25. Husarik DB, Miralbell R, Dubs M, John H, Giger OT, Gelet dy. Eur J Cancer 2010;46:449-55. EANM A, et al. Evaluation of [(18)F]-choline PET/CT for staging and restaging of prostate cancer. Eur J Nucl Med Mol Imaging 35. Sutinen E, Nurmi M, Roivainen A, Varpula M, Tolvanen 2008;35:253-63. T, Lehikoinen P, et al. Kinetics of [(11)C]choline uptake in prostate cancer: a PET study. Eur J Nucl Med Mol Imaging 26. Kwee SA, Coel MN, Lim J, Ko JP. Prostate cancer loca- 2004;31:317-24. lization with 18%uorine %uorocholine positron emission tomography. J Urol 2005;173:252-5. 36. Yamaguchi T, Lee J, Uemura H, Sasaki T, Takahashi N, Oka T, et al. Prostate cancer: a comparative study of 11C- 27. Kwee SA, Wei H, Sesterhenn I, Yun D, Coel MN. Loca- choline PET and MR imaging combined with proton MR lization of primary prostate cancer with dual-phase 18F- spectroscopy. Eur J Nucl Med Mol Imaging 2005;32:742-8. %uorocholine PET. J Nucl Med 2006;47:262-9. 37. Yoshida S, Nakagomi K, Goto S, Futatsubashi M, Torizuka 28. Li X, Liu Q, Wang M, Jin X, Yao S, Liu S, et al. C-11 choline T. 11C-choline positron emission tomography in prostate PET/CT imaging for di&erentiating malignant from benign cancer: primary staging and recurrent site staging. Urol Int prostate lesions. Clin Nucl Med 2008;33:671-6. 2005;74:214-20.

29. Martorana G, Schiavina R, Corti B, Farsad M, Salizzoni E, 38. de Jong IJ, Pruim J, Elsinga PH, Vaalburg W, Mensink Brunocilla E, et al. 11C-choline positron emission tomogra- HJ. Preoperative staging of pelvic lymph nodes in prostate phy/computerized tomography for tumor localization of cancer by 11C-choline PET. J Nucl Med 2003;44:331-335. primary prostate cancer in comparison with 12-core biopsy. J Urol 2006;176:954-60; discussion 60. 39. Poulsen MH, Bouchelouche K, Gerke O, Petersen H, Svolgaard B, Marcussen N, et al. [18F]-%uorocholine posi- 30. Reske SN, Blumstein NM, Neumaier B, Gottfried HW, tron-emission/computed tomography for lymph node sta- Finsterbusch F, Kocot D, et al. Imaging prostate cancer with ging of patients with prostate cancer: preliminary results of 11C-choline PET/CT. J Nucl Med 2006;47:1249-54. a prospective study. BJU Int 2010;106:639-43; discussion 44. 31. Scher B, Seitz M, Albinger W, Tiling R, Scherr M, Becker HC, et al. Value of 11C-choline PET and PET/CT in patients with suspected prostate cancer. Eur J Nucl Med Mol Ima- ging 2007;34:45-53.

77 40. Tuncel M, Souvatzoglou M, Herrmann K, Stollfuss J, 48. Krause BJ, Souvatzoglou M, Tuncel M, Herrmann K, Buck Schuster T, Weirich G, et al. [(11)C]Choline positron emis- AK, Praus C, et al. The detection rate of [11C]choline-PET/ sion tomography/computed tomography for staging and CT depends on the serum PSA-value in patients with bio- restaging of patients with advanced prostate cancer. Nucl chemical recurrence of prostate cancer. Eur J Nucl Med Mol Med Biol 2008;35:689-95. Imaging 2008;35:18-23.

41. Breeuwsma AJ, Pruim J, van den Bergh AC, Leliveld AM, 49. Ohlmann C, P*ster D, Thüer D, Wille S, Engelmann UH, Nijman RJ, Dierckx RA, et al. Detection of local, regional, Heidenreich A. 11C-choline-positron emission tomogra- and distant recurrence in patients with psa relapse after phy/computerized tomography (11C-PET/CT) for tumour external-beam radiotherapy using (11)C-choline posi- localization of locally recurrent prostate cancer after radia- tron emission tomography. Int J Radiat Oncol Biol Phys tion therapy. Eur Urol Suppl 2007;6:229. 2010;77:160-4. 50. Pelosi E, Arena V, Skanjeti A, Pirro V, Douroukas A, Pupi 42. Castellucci P, Fuccio C, Nanni C, Santi I, Rizzello A, A, et al. Role of whole-body 18F-choline PET/CT in disease Lodi F, et al. In%uence of trigger PSA and PSA kinetics on detection in patients with biochemical relapse after radical 11C-choline PET/CT detection rate in patients with bio- treatment for prostate cancer. Radiol Med 2008;113:895-904. chemical relapse after radical prostatectomy. J Nucl Med 2009;50:1394-400. 51. Picchio M, Messa C, Landoni C, Gianolli L, Sironi S, Brioschi M, et al. Value of [11C]choline-positron emission 43. Castellucci P, Fuccio C, Rubello D, Schiavina R, Santi I, tomography for re-staging prostate cancer: a comparison Nanni C, et al. Is there a role for (11)C-choline PET/CT in the with [18F]%uorodeoxyglucose-positron emission tomogra- early detection of metastatic disease in surgically treated phy. J Urol 2003;169:1337-40. prostate cancer patients with a mild PSA increase <1.5 ng/ ml? Eur J Nucl Med Mol Imaging 2011;38:55-63. 52. Reske SN, Blumstein NM, Glatting G. [11C]choline PET/ CT imaging in occult local relapse of prostate cancer af- 44. Cimitan M, Bortolus R, Morassut S, Canzonieri V, Garbe- ter radical prostatectomy. Eur J Nucl Med Mol Imaging glio A, Baresic T, et al. [18F]%uorocholine PET/CT imaging for 2008;35:9-17. the detection of recurrent prostate cancer at PSA relapse: experience in 100 consecutive patients. Eur J Nucl Med Mol 53. Richter JA, Rodriguez M, Rioja J, Penuelas I, Marti-Climent Imaging 2006;33:1387-98. J, Garrastachu P, et al. Dual tracer 11C-choline and FDG-PET in the diagnosis of biochemical prostate cancer relapse after 45. Giovacchini G, Picchio M, Coradeschi E, Bettinardi V, Gi- radical treatment. Mol Imaging Biol 2010;12:210-7. anolli L, Scattoni V, et al. Predictive factors of [(11)C]choline PET/CT in patients with biochemical failure after radical 54. Rinnab L, Mottaghy FM, Blumstein NM, Reske SN, prostatectomy. Eur J Nucl Med Mol Imaging 2010;37:301-9. Hautmann RE, Hohl K, et al. Evaluation of [11C]-choline positron-emission/computed tomography in patients with 46. Giovacchini G, Picchio M, Scattoni V, Garcia Parra R, Bri- increasing prostate-speci*c antigen levels after primary ganti A, Gianolli L, et al. PSA doubling time for prediction of treatment for prostate cancer. BJU Int 2007;100:786-93. [(11)C]choline PET/CT *ndings in prostate cancer patients with biochemical failure after radical prostatectomy. Eur J 55. Rinnab L, Mottaghy FM, Simon J, Volkmer BG, de Petrico- Nucl Med Mol Imaging 2010;37:1106-16. ni R, Hautmann RE, et al. [11C]Choline PET/CT for targeted salvage lymph node dissection in patients with biochemi- 47. Heinisch M, Dirisamer A, Loidl W, Stoiber F, Gruy B, Haim cal recurrence after primary curative therapy for prostate S, et al. Positron emission tomography/computed tomogra- cancer. Preliminary results of a prospective study. Urol Int phy with F-18-%uorocholine for restaging of prostate cancer 2008;81:191-7. patients: meaningful at PSA <5 ng/ml? Mol Imaging Biol 2006;8:43-8.

78 Chapter 2 – Clinical applications of PET/CT in oncology: 2.5 Prostate cancer

56. Rinnab L, Simon J, Hautmann RE, Cronauer MV, Hohl 62. Chism DB, Hanlon AL, Horwitz EM, Feigenberg SJ, Pollack K, Buck AK, et al. [(11)C]choline PET/CT in prostate cancer A. A comparison of the single and double factor high-risk patients with biochemical recurrence after radical prosta- models for risk assignment of prostate cancer treated with tectomy. World J Urol 2009;27:619-25. 3D conformal radiotherapy. Int J Radiat Oncol Biol Phys 2004;59:380-5. 57. Scattoni V, Picchio M, Suardi N, Messa C, Freschi M, Ros- cigno M, et al. Detection of lymph-node metastases with 63. Roberts SG, Blute ML, Bergstralh EJ, Slezak JM, Zincke integrated [11C]choline PET/CT in patients with PSA failure H. PSA doubling time as a predictor of clinical progression after radical retropubic prostatectomy: results con*rmed by after biochemical failure following radical prostatectomy open pelvic-retroperitoneal lymphadenectomy. Eur Urol for prostate cancer. Mayo Clin Proc 2001;76:576-81. 2007;52:423-9. 64. Bott SR. Management of recurrent disease after radical 58. Steiner C, Vees H, Zaidi H, Wissmeyer M, Berrebi O, prostatectomy. Prostate Cancer Prostatic Dis 2004;7:211-6. Kossovsky MP, et al. Three-phase 18F-%uorocholine PET/ CT in the evaluation of prostate cancer recurrence. Nukle- 65. Krause BJ, Souvatzoglou M, Herrmann K, Weber AW, armedizin 2009;48:1-9; quiz N2-3. Schuster T, Buck AK, et al. [11C]Choline as pharmacodyna- mic marker for therapy response assessment in a prostate

59. Vees H, Buchegger F, Albrecht S, Khan H, Husarik D, Zaidi cancer xenograft model. Eur J Nucl Med Mol Imaging EANM H, et al. 18F-choline and/or 11C-acetate positron emission 2010;37:1861-8. tomography: detection of residual or progressive subclinical disease at very low prostate-speci*c antigen values (<1 ng/ 66. Contractor KB, Aboagye EO. Monitoring predominantly mL) after radical prostatectomy. BJU Int 2007;99:1415-20. cytostatic treatment response with 18F-FDG PET. J Nucl Med 2009;50 Suppl 1:97-105. 60. Freedland SJ, Presti JC Jr, Amling CL, Kane CJ, Aronson WJ, Dorey F, et al. Time trends in biochemical recurrence 67. Souvatzoglou M, Weirich G, Schwarzenboeck S, et al. after radical prostatectomy: results of the SEARCH database. The sensitivity of [11C]choline PET/CT to localize prostate Urology 2003;61:736-41. cancer depends on the tumour con*guration. Clin. Cancer Res. 2011 61. Han M, Partin AW, Zahurak M, Piantadosi S, Epstein JI, Walsh PC. Biochemical (prostate speci*c antigen) recurrence probability following radical prostatectomy for clinically localized prostate cancer. J Urol 2003;169:517-23.

79 Chapter 2 – Clinical applications of PET/CT in oncology 2.6 Lymphomas Roberto C. Delgado-Bolton and José L. Carreras Delgado

Introduction with conventional imaging methods [1,2]. 18F-FDG PET has a main role in the manage- Therefore, 18F-FDG PET has a potential im- ment of patients with lymphoma. It has dem- pact on management decisions as therapy onstrated a higher e+cacy than conventional is based on the stage of the disease. In ad- imaging methods such as CT, MRI and gallium dition, the advent of multimodality imaging scans, providing a more accurate distinction with 18F-FDG PET/CT has further improved between scar or *brosis and active tumour. upon the performance of 18F-FDG PET alone, Studies have evaluated the performance of as is re%ected by the higher speci*city of the 18F-FDG PET in pretreatment staging, monitor- combined technique in recent studies [1]. ing response to therapy and post-therapy surveillance, among other applications. 18F-FDG An interesting issue is whether intravenous PET has a major role in monitoring response contrast must be used in PET/CT. Preliminary to therapy and has allowed the introduction data suggest that intravenous contrast allows of the concept of risk-adapted therapy. This a more accurate assessment of the liver and is discussed more fully in this section, which spleen compared with unenhanced CT [3]. A reviews the role of PET in the management review of the published data concluded that of lymphoma. intravenous contrast-enhanced low-dose PET/CT may represent a reasonable choice Initial staging as a single imaging modality for staging Based on the fact that most lymphomas FDG-avid lymphomas [4]. show high levels of 18F-FDG uptake, many studies analysing the performance of PET in E+cacy studies have evidenced the supe- the staging of lymphomas have shown very rior performance of 18F-FDG PET in the initial high sensitivity in patients with Hodgkin’s staging of lymphomas, but there is very little lymphoma (HL) and high-grade non-Hodg- evidence regarding its impact on patient out- kin’s lymphoma (NHL). 18F-FDG PET improved comes. Even so, 18F-FDG PET/CT has become the performance of conventional imaging the cornerstone of the staging imaging pro- methods (including CT) in the detection of cedures in the state-of-the-art management nodal and extranodal disease as well as in in FDG-avid lymphomas (HL and aggressive the characterisation of lesions considered NHL) [1]. In low-grade or indolent lymphomas, equivocal on other imaging methods [1]. The where 18F-FDG PET has shown a variable diag- additional information supplied by 18F-FDG nostic performance and impact on manage- PET improves the accuracy of the staging ment, 18F-FDG PET/CT is used sporadically or of the disease, resulting in the upstaging or in clinical trials with frequently encouraging downstaging of the disease when compared results [1].

80 Chapter 2 – Clinical applications of PET/CT in oncology: 2.6 Lymphomas

Therapy response evaluation Risk strati#cation and evaluation of response to therapy: prognostic indexes and imaging Concept of risk-adapted therapy procedures Without doubt the most promising clinical For both HL and NHL there are well-de*ned application of 18F-FDG PET is in the early moni- pretreatment prognostic indexes that have toring of response to therapy: new possibilities demonstrated a prognostic value and the for risk strati*cation and individualisation of prediction of survival in prospective stud- therapy have emerged based on the results ies [1,7–10]. The therapeutic strategy is de- of interim 18F-FDG PET [1,5,6]. The concept of termined mainly by the clinical stage and risk-adapted therapy in lymphoma is increas- these prognostic indexes. However, tumour ingly accepted as a way to achieve higher cure response to therapy is an important surrogate rates with a lower or equal risk of treatment- for other assessments of clinical bene*t, in- EANM related morbidity and mortality. Tailoring and cluding progression-free survival and overall individualising therapy according to the need survival [1,5,6]. of the patient is a therapeutic option which could perhaps soon become the standard of Several prognostic indexes have been devel- care. However, it is still not proven that modify- oped for HL and NHL. The indexes developed ing therapy on the basis of interim PET can im- for advanced HL have demonstrated a limited prove patient outcomes. Therefore, this issue clinical utility and their predictive value has must be analysed in appropriately designed been questioned [11–13]. All these indexes clinical trials. 18F-FDG PET enables evaluation take into account di&erent risk factors (age, of the early metabolic changes rather than stage, etc.) to establish the prognosis and risk the morphological changes which occur lat- of recurrence. However, the values obtained er during therapy. In lymphoma these early are based on a group of patients who had metabolic changes are highly predictive of been analysed retrospectively. Therefore, they the *nal treatment response. PET performed do not necessarily re%ect the risk in the indi- after a few courses of standard chemotherapy vidual patient [6]. Recently, the concept of risk- is a reliable prognostic tool to identify poor adapted therapy has been proposed, based responders to therapy. Interim PET is a pow- on the fact that the chemosensitivity of the erful prognostic tool when compared with tumour or early response to therapy in the in- other well-established clinical parameters in dividual patient would allow risk strati*cation lymphoma [1,5,6]. and, possibly, help with the therapeutic decisions as it predicts the probability of achieving control of the disease [1,5,6,11].

81 Until recently, CT and MRI have had a main role cytokines that assure the immortality of the in diagnostic imaging in oncology. In recent Reed-Sternberg cells and work as an ampli- years, however, there has been a change in *er of the capacity of PET to detect response this paradigm, as 18F-FDG PET has demonstrat- [5,29–31]. This non-neoplastic compartment is ed a superior e+cacy compared with conven- disconnected very early by chemotherapy, giv- tional imaging methods in the evaluation of ing rise to a situation denominated “complete the oncological disease, especially in monitor- metabolic response” [5,28,31–33]; this phenom- ing response to therapy, where the functional enon explains the impressive prognostic value information supplied by 18F-FDG PET allows of PET in HL. This is the reason why, in around assessment of the early metabolic changes in- 80% of patients with HL, normalisation of the stead of the morphological changes which oc- PET scan is achieved after two courses of ABVD cur later during therapy [1,5]. Many studies in [24,26,31]. On the other hand, in DLBCL che- which 18F-FDG PET has been performed after motherapy progressively destroys a fraction one to three courses of chemotherapy in ag- of neoplastic cells, the percentage of cellular gressive NHL [14–22] and HL [11,23–26] have destruction being predictive of the response evidenced that these early metabolic changes at the end of the chemotherapy [28]. In DLBCL have a high predictive value for the response the accompanying non-neoplastic lympho- at the end of the treatment and progression- mononuclear cells do not seem to have a role in free survival [1]. A meta-analysis by Terasawa 18F-FDG uptake [28,34]. Instead, 18F-FDG uptake et al. concludes that 18F-FDG PET is a reliable re%ects the metabolic situation and represents prognostic tool to identify patients with a bad the balance between the cellular destruction response to therapy in HL [27]. of the chemosensitive components and the growth of the resistant components [35]. It Di$erences between HL and NHL: is characteristic in DLBCL that the metabolic repercussions on therapy response activity decreases in a continuous way, as a evaluation with PET fraction of neoplastic cells dies progressively. The histopathological and physiopathological Owing to this progressive behaviour, the sig- characteristics of the neoplastic tissue are di&er- ni*cance and predictive value of the early as- ent for HL and NHL. In HL, the Reed-Sternberg sessment of response probably di&er depend- cells represent less than 1% of the neoplastic ing on the timing of the PET scan [35]. At the tissue, whereas in di&use large B cell lymphoma beginning of the treatment (after one or two (DLBCL) they represent more than 90% of the courses) PET assesses chemosensitivity as the total cell population [28]. In HL, the accompa- prevailing phenomenon is the destruction of nying non-neoplastic lymphomononuclear the most chemosensitive cells, whereas after cells produce an interconnected network by three or four courses PET evaluates resistance

82 Chapter 2 – Clinical applications of PET/CT in oncology: 2.6 Lymphomas

to therapy as FDG uptake depends more on for HL in 2005 [23], although it was later ap- the growth of the resistant components of the plied for both HL and NHL. The de*nition of tumour [5,35]. MRU has changed since then, a consensus being achieved in 2009. Gallamini et al. [31] Evaluation of 18F-FDG PET-CT: visual or reviewed the evolution of the concept of MRU, semiquantitative? which can be summarised in three points: Based on the histopathological and physiopathological di&erences between HL and NHL 1. In 2005 Hutchings et al. considered MRU to discussed in the previous section, it seems that re%ect unspeci*c in%ammatory processes a visual analysis would be preferable in HL (be- related to cell destruction [23,31,37]. cause of its dichotomous response), whereas 2. In 2007 Gallamini et al. [11] de*ned MRU in DLBCL a semiquantitative analysis (using as low 18F-FDG uptake similar to the medi- EANM the maximum standardised uptake value or astinal blood pool [31]. SUV ) would be preferable (because of its 3. In 2008 Barrington et al. [36,38] de*ned max progressive response) [5,28]. However, when MRU as a residual 18F-FDG uptake with an applying a semiquantitative analysis it must intensity lower than or equal to the liver be taken into account that SUV depends on uptake [31]. many factors (patient preparation and blood glucose levels, 18F-FDG administration proto- The evolution of the de*nition of MRU has ex- col, 18F-FDG uptake time, PET scanner, image tended the limits of what is considered MRU. reconstruction, etc.). Therefore, it is necessary These modi*cations have been aimed at in- to standardise the 18F-FDG PET procedure in creasing speci*city and reducing the number of order to obtain PET images which will be com- false positives when PET is used to predict the reparable between di&erent centres and will sponse to therapy and patient outcome [5,31,39]. also permit comparison of successive studies within each centre, especially when using PET However, these modi*cations in the de*nition for response assessment [1,36]. of MRU have had consequences, especially in the design of clinical trials, as several of those Considerations regarding the concept of currently under way are using di&erent crite- minimal residual disease ria for the interpretation of interim PET [31]. The concept of minimal residual uptake Therefore, in April 2009 an international meet- (MRU) has been de*ned as the presence of ing took place in Deauville (France), where low-grade 18F-FDG uptake (just above back- uniform criteria were established for interim ground) in a location where previously there PET scan interpretation. The conclusions of was disease. This concept was *rst described this meeting are presented in the next section.

83 Evaluation of response to therapy: Is there a recently [28]. Gallamini et al. [31] revised the consensus? conclusions of the meeting and indicated that The response evaluation criteria with 18F-FDG the criteria for the interpretation of interim PET PET di&er depending on the timing of the PET can be summarised in three main statements: scan, i.e. whether it is performed early during (a) visual analysis is preferable, although in therapy or after its completion. some cases SUV can be used; (b) interim max PET must be analysed comparing the focal Interim PET. When applying PET in the evalua- 18F-FDG uptakes with those described in the tion of response to therapy, the main problem pretherapy study; and (c) the intensity of 18F- is the absence of uniform criteria in studies FDG uptake must be categorised using a *ve- available in the literature. The use of di&erent point scale where the reference organ is the criteria has led to a great variability in the re- liver [5,31]. The proposed *ve-point scale is: ported sensitivity and speci*city [28]. In some studies the International Harmonization Proj- 1. Absence of uptake ect criteria were applied, although these were 2. Uptake ≤ mediastinum designed to evaluate response alter comple- 3. Uptake > mediastinum but ≤ liver tion of therapy [3,40]. Another issue to take 4. Uptake moderately higher than the liver into account is the concept of MRU and its uptake, in any location limits [23]. Recently, with many new clinical 5. Marked increased uptake in any location trials dedicated to this subject, the need for and new focal lesions standard criteria for PET interpretation has become even more important, in order to obtain During this meeting, international validation reproducible results all over the world [28]. studies were designed for HL and NHL to validate these interpretation criteria [28,31]. In this context an international meeting was A follow-up meeting, entitled the Second In- organised with experts in nuclear medicine ternational Workshop on Interim PET in Lym- and haematology. The First International phoma, took place in Menton (France) in April Workshop on Interim PET in Lymphoma took 2010. The conclusions of this second meeting place in Deauville (France) in April 2009, with had not been published at the time of submis- the aim of establishing uniform criteria for in- sion of this manuscript. terim PET scan interpretation and launching international validation studies, as had been Evaluation of response after completion of done previously in the International Harmoni- therapy. zation Project in 2007 [3,5,28,40]. The conclu- The International Harmonization Project estab- sions of the Deauville meeting were published lished standard criteria to evaluate response

84 Chapter 2 – Clinical applications of PET/CT in oncology: 2.6 Lymphomas

after completion of therapy. The conclusions to be evidence that PET was positive at of the International Harmonization Project all disease sites with a diameter ≥1.5 cm were published in 2007 [3,40] and can be sum- detected by CT [3]. marised in the following recommendations: 3. Timing of 18F-FDG PET performed for response assessment at the conclusion of 1. Use of 18F-FDG PET for response assess- therapy: To minimize the frequency of ment at the conclusion of therapy: Many post-therapy in%ammatory changes, which studies have demonstrated the value of potentially confuse the interpretation of PET for response assessment of HL and PET scans, PET should not be performed DLBCL at the conclusion of front-line, earlier than 3 weeks after chemotherapy salvage or high-dose therapy. The role of and preferably 8–12 weeks after comple- PET for response assessment of aggressive tion of radiotherapy [3]. EANM NHL subtypes other than DLBCL and of 4. Interpretation of 18F-FDG PET scans per- indolent and mantle cell lymphomas is formed for response assessment at the less clear [3]. conclusion of therapy: Visual assessment 2. Requirement for pretherapy 18F-FDG PET alone appears to be su+cient to decide scan for response assessment of lympho- whether PET is positive or negative at the ma at the conclusion of therapy: Prethera- conclusion of therapy, and semiquanti- py PET is not compulsory for assessment of tative analysis (using SUV ) does not max response after treatment of patients with seem necessary. The generally accepted HL, DLBCL, follicular lymphoma or mantle de*nition of a positive or abnormal PET cell lymphoma because these lymphomas *nding when using visual assessment is are routinely FDG avid. However, prether- focal or di&use FDG uptake above back- apy PET is de*nitely encouraged for these ground in a location incompatible with subtypes because it can facilitate the inter- normal anatomy or physiology [3]. This pretation of post-therapy PET. In contrast, de*nition looks as if it is suitable in most pretherapy PET is obligatory for variably instances, although there are some excep- FDG-avid lymphomas if PET is to be used to tions. assess their response to treatment. These include aggressive NHL subtypes other These recommendations established by the than DLBCL, such as T-cell lymphomas, and International Harmonization Project were all subtypes of indolent NHL other than designed to standardise the interpretation of follicular lymphoma. If PET is to be used PET in the evaluation of response to therapy for response assessment of patients with in the clinical setting as well as in clinical tri- these histological subtypes, there needs als [3,5,40].

85 Follow-up and early detection of recurrent 18F-FDG PET in the initial staging of lympho- disease mas, but there is very little evidence regard- The available literature is very limited regarding its impact on patient outcomes. Even so, ing the value of 18F-FDG PET in the follow-up 18F-FDG PET/CT has become the cornerstone setting. Therefore the role of 18F-FDG PET in of the staging imaging procedures in state-of- the detection of preclinical relapse is not clear, the art management of FDG-avid lymphomas. although it could be discussed that 18F-FDG PET could perhaps allow patients to be treated 18F-FDG PET has a main role in monitoring with salvage therapy when they present mini- response to therapy and has allowed the mal disease instead of an overt relapse. Further introduction of the concept of risk-adapted studies are needed to assess the e+cacy of therapy. The concept of risk-adapted therapy 18F-FDG PET in the follow-up setting [1]. in lymphoma is increasingly accepted as a way to achieve higher cure rates with a lower Response prediction before high-dose or equal risk of treatment-related morbidity chemotherapy and autologous stem cell and mortality. Tailoring and individualising transplantation therapy according to the need of the patient Available evidence indicates that 18F-FDG PET is a therapeutic option which could perhaps performed after induction chemotherapy and soon become the standard of care. However, it just before autologous stem cell transplanta- is still not proven that modi*cation of therapy tion can predict which patients will achieve based on interim PET can improve patient out- long-term remission after the high-dose sal- comes. Therefore, this issue must be analysed vage regimen. Patients with persistent disease in appropriately designed clinical trials. on 18F-FDG PET present short progression-free survival, although the published studies have Recent meetings have developed recommen- also reported a higher rate of false-positive dations for the assessment of PET for moni- results compared with when PET is performed toring response to therapy in lymphomas. In early during front-line therapy. The role of 18F- 2009 an international meeting in Deauville FDG PET in this setting is unclear, and available established uniform criteria for interim PET data do not support a change in treatment in- scan interpretation. In 2007 the International tent despite a suboptimal metabolic response Harmonization Project published recommen- to induction therapy [1]. dations and standard criteria for the evaluation of response after completion of therapy. Conclusions 18F-FDG PET has a key role in the management of patients with lymphoma. E+cacy studies have evidenced the superior performance of

86 . Juweid ME, Stroobants S, Hoekstra OS, et al. Use of po- 3. 2005;104:1066–74. stagingrestagingand Cancerlymphoma. with patients of positronxy-2-%uoro-D-glucose emission tomography in the Isasi CR, Lu P, Blaufox MD. A metaanalysis of 18F-2-deo- 2. 2009;50:21S–30S. assessment inlymphoma.JNuclMed Hutchings M, Barrington SF. for PET/CT therapy response 1. Oncol 1989;7:1630–6. patients with Hodgkin’s disease: Cotswolds meeting. J Clin stagingof and mittee convenedevaluation the todiscuss Lister 10. TA, Crowther D, Sutcli&e SB, et al. Report of a com- 1998;339:1506–14. Med J Engl N disease. Hodgkin’s advanced on project factors prognostic international disease: Hodgkin’s ced advan- for score prognostic A V.Diehl D, Hasenclever 9. Blood index. 2004;104:1258–65. Follicu- prognostic al. international et P, lymphoma Colombat lar P, Roy P, Solal-Celigny 8. Hodgkin’s 1993;329:987–94. lymphoma.NEnglJMed non- aggressive for model predictive A Project. Factors The International Non-Hodgkin’s Lymphoma Prognostic 7. 2009;36:2089–90. Imaging Mol Med Nucl J Eur management? clinical in standard-of-care the will this replace the established prognostic indices and be tion and further therapeutic decision-making in lymphoma: Basu S. Early FDG-PET response–adapted risk strati*ca- 6. response?] Nucléaire Médecine 2011;35:29–37. in lymphoma: Is there a consensus regarding evaluation of réponse? [18F-FDG PET-CT for therapy response evaluation la de l’évaluation dans consensus un existe-t-il 18F-FDG: TEP la avec lymphome du traitement au réponse la de Évaluation JL. Delgado Carreras RC, Delgado-Bolton 5. scans in patients with lymphoma. Blood 2007;110:3507-16. FDG-PET of role BD.The Cheson P,ME, Juweid Seam 4. Oncol 2007;25:571-8. Clin J Lymphoma. in Project Harmonization International of Subcommittee Imaging the of consensus lymphoma: of assessment response for tomography emission sitron References References Chapter 2.6 87 lymphoma. Haematologica 2000;85:613–8. therapy is predictive of treatment failure in non-Hodgkin’s polychemo- of cycles few a after uptake 18F-FDG tumor Persistent al. et MF,FassotteY, Beguin G, Jerusalem 14. in Hodgkin’s disease. AnnOncol2002;13(Suppl):75-8. Hasenclever D. 13. The disappearance of prognostic factors systems. Cancer 2001;91:1467-78. sease: promising results from integration of the best three prognostic models in patients with advanced di- Hodgkin of Comparison al. et C, Broglia PL, Zinzani PG, Gobbi 12. 2007;25:3746–52. ma: a fromreport a joint study.Italian–Danish J Clin Oncol Hodgkin’slympho- advanced-stage in score prognostic international to superior prognostically is tomography emission positron 2-[18F]%uoro-2-deoxy-D-glucose rim inte- Early al. et L, Rigacci M, Hutchings A, Gallamini 11. Imaging 2004;31:22–8. Imaging lymphoma and Hodgkin’s Mol lymphoma. Eur JNuclMed non-Hodgkin’s aggressive in FDG-PET with monitoring therapy Early al. et T, F, Kanno Nakamura T,Torizuka19. 2002;43:1018–27. J NuclMed motherapy in aggressive lymphoma and Hodgkin’s disease. che- of cycle 1 after prognosis predicts PET SJ. Goldsmith H, Zoe I, JP, Kuji Leonard M, Coleman L, Kostakoglu 18. 1993;34:1706–10. Nucl Med J scintigraphy. %uorine-18-%uorodeoxyglucose planar by treatment response in malignant lymphoma, as determined Early al. et R, Golding GJ, Ossenkoppele OS, Hoekstra 17. Hodgkin’s lymphoma.AnnOncol2002;13:1356–63. non- aggressive with patients in outcome predicts cose 18F-%uorodeoxyglu- with tomography emission positron Spaepen K, 16. Stroobants S, Dupont P, et al. Early restaging CT. Leuk Lymphoma 2000;39:543–53. of aggressive non-Hodgkin’s lymphoma: comparison with MN. 18-FDG PET as a prognostic indicator in the treatment Mikhaeel NG, 15. Timothy AR, MJ,O’Doherty Hain S, Maisey

EANM 20. Mikhaeel NG, Hutchings M, Fields PA, O’Doherty MJ, 30. Skinnider BF, Mak TW. The role of cytokines in classical Timothy AR. FDG-PET after two to three cycles of che- Hodgkin lymphoma. Blood 2002;99:4283–97. motherapy predicts progression-free and overall survival in high-grade non-Hodgkin lymphoma. Ann Oncol 31. Gallamini A, Fiore F, Sorasio R, Meignan M. Interim po- 2005;16:1514–23. sitron emission tomography scan in Hodgkin lymphoma: de*nitions, interpretation rules, and clinical validation. Leuk 21. Haioun C, Itti E, Rahmouni A, et al. [18F]%uoro-2-deoxy- Lymphoma 2009;50:1761–4. D-glucose positron emission tomography (FDG-PET) in aggressive lymphoma: an early prognostic tool for predicting 32. Kostakoglu L. Early prediction of response to therapy: patient outcome. Blood 2005;106:1376–81. the clinical implications in Hodgkin’s and non-Hodgkin’s lymphoma. Eur J Nucl Med Mol Imaging 2008;35:1413–20. 22. Kostakoglu L, Goldsmith SJ, Leonard JP, et al. FDG-PET after 1 cycle of therapy predicts outcome in di&use lar- 33. MacManus MP, Seymour JF, Hicks RJ. Overview of early ge cell lymphoma and classic Hodgkin disease. Cancer response assessment in lymphoma with FDG-PET. Cancer 2006;107:2678–87. Imaging 2007;7:10–8.

23. Hutchings M, Mikhaeel NG, Fields PA, Nunan T, Timothy 34. Gallamini A. The prognostic role of positron emission AR. Prognostic value of interim FDG-PET after two or three tomography scan in Hodgkin’s lymphoma. Haematologica cycles of chemotherapy in Hodgkin lymphoma. Ann Oncol 2009;3:144–50. 2005;16:1160–8. 35. Meignan M, Itti E, Gallamini A, Haioun C. Interim 18F- 24. Hutchings M, Loft A, Hansen M, et al. FDG-PET after %uorodeoxyglucose positron emission tomography in two cycles of chemotherapy predicts treatment failure di&use large B-cell lymphoma: qualitative or quantitative and progression-free survival in Hodgkin lymphoma. Blood interpretation – where do we stand? Leuk Lymphoma 2006;107:52–9. 2009;50:1753–6.

25. Zinzani PL, Tani M, Fanti S, et al. Early positron emission 36. Barrington SF, Qian W, Somer EJ, et al. Concordance tomography (PET) restaging: a predictive *nal response in between four European centres of PET reporting criteria Hodgkin’s disease patients. Ann Oncol 2006;17:1296–300. designed for use in multicentre trials in Hodgkin lymphoma. Eur J Nucl Med Mol Imaging 2010;37:1824–33. 26. Gallamini A, Rigacci L, Merli F, et al. The predictive value of positron emission tomography scanning performed after 37. Spaepen K, Stroobants S, Dupont P, et al. [18F]FDG PET two courses of standard therapy on treatment outcome in monitoring of tumour response to chemotherapy: does advanced stage Hodgkin’s disease. Haematologica 2006; [18F]FDG uptake correlate with the viable tumour cell frac- 91:475–81. tion? Eur J Nucl Med Mol Imaging 2003;30:682–8.

27. Terasawa T, Lau J, Bardet S, et al. Fluorine-18-%uoro- 38. Barrington SF, Qian W, Somer EJ, et al. Concordance deoxyglucose positron emission tomography for interim between four European centres of PET reporting criteria response assessment of advanced-stage Hodgkin’s lym- designed for use in multicentre trials in Hodgkin lymphoma. phoma and di&use large B-cell lymphoma: a systematic Eur J Nucl Med Mol Imaging 2009;36 (Suppl.2):S252. review. J Clin Oncol 2009;27:1906–14. 39. Gallamini A, Hutchings M, Avigdor A, Polliack A. Early in- 28. Meignan M, Gallamini A, Haioun C. Report on the First terim PET scan in Hodgkin lymphoma: where do we stand? International Workshop on Interim-PET-Scan in Lymphoma. Leuk Lymphoma 2008;49:659–62. Leuk Lymphoma 2009;50:1257–60. 40. Cheson BD, P*stner B, Juweid ME, et al. Revised re- 29. Canellos GP. Residual mass in lymphoma may not be sponse criteria for malignant lymphoma. J Clin Oncol residual disease. J Clin Oncol 1988;6:931–933. 2007;25:579–86.

88 by n nrclua tasot ad t s phos- is it and transport, intracellular in used for PET is %uorodeoxyglucose (FDG), (FDG), %uorodeoxyglucose is PET for used tracer major The pathway. anaerobic an by mainly is metabolised cells in cancer glucose phenotype: by a particular characterised are Warburg 1924, In as as 18 acids, amino labelled including introduced, been have overexpression receptor of knowledge or pathways metabolic other using diotracers using performed are techniques 90% of research almost Currently, is marker. not a speci*c radiopharmaceutical and muscles. is This why tissue the reason this adipose brown in%ammation, instance, for in, seen also is it for malignancy; speci*c not is GLUT1 of Overexpression GLUT1. protein transport glucose the of expression the to Cellular activity). hexokinase and activity transport of glucose mechanisms the of indicator an rather, (or glucose for pathways metabolic of indicator of radioactivity measured the Thus, cell. the in accumulated is structure, in changes to owing glucose-phosphate), (6-deoxy- phosphorylation of product the However, hexokinase). of result a (as cycle Krebs the of reaction *rst the in phorylated ( emitter positron a %uorine, radioactive with labelled easily be can Deoxyglucose glucose. of analogue an Introduction Chapter 2 – Clinical applications of PET/CT in oncolo Jolanta Kunikowska andLeszek Królicki Kunikowska Jolanta tumours 2.7 Digestive tract F-choline and somatostatin analogues such analogues and somatostatin F-choline the same mechanisms as are involved involved are as mechanisms same the 68 Ga-DOTATATE. 18 F-%uorothymidine ( F-%uorothymidine showed that cancer cells cells cancer that showed 18 F-FDG uptake is related related is uptake F-FDG 18 18 F). It is transported transported is It F). F-FDG PET. Novel ra- Novel PET. F-FDG 18 F-FDG is an an is F-FDG 18 F-FLT), F-FLT), 89 Oesophagealcancerappears principaltwo in cancerOesophageal tional testing is usually performed to estimate Addi- cancer. oesophageal of diagnosis the Endoscopy with biopsy plays the main role in geal tumour symptoms are dysphagia and pain. sophagusstomach.and Thetypicaloesopha- cells that are present at the junction of the oe- adenocarcinomawhile glandularfromarises cells that line the upper part of the oesophagus, cinoma. Squamous cell cancer arises from the subtypes: squamous cell cancer and adenocar- reduces thesensitivity. which mm), 6 (approximately resolution tial to the primary tumour is its relatively poor spain the detection of nodal metastases adjacent [1,2,5]. The major limitation of 100% and 82% between is metastases nodal of accuracy of CT is 40–73% [1–5]. The speci*city the while 24–90%, is nodes lymph regional of curacy ac- reported in%ammation. Theto due occur 91% and 100% [1–3]. False positive uptake can tection, with reported sensitivities of between can play a role PET/CT tumour in de- primary torestricted those larger than1 cm. fact that its ability to detect masses is generally The sensitivity of CT is, however, limited by the tant organs (especially regional lymph nodes). dis- or tissues adjacent to spread has cancer the whether evaluate to used is pelvis and abdomen chest, the of CT stage. tumour the 18 F-FDG PET/CT F-FDG in PET/CT the detection of regional 18 F-FDG PET/CT in the staging of staging the in PET/CT F-FDG 18 F-FDG PET/CT F-FDG PET/CT gy EANM PET/CT is strictly recommended to improve metastases [6]. 18F-FDG PET/CT scans can also the accuracy of detection of distant metas- exclude metastatic disease at sites considered tases, thereby facilitating treatment planning abnormal on conventional imaging. (Fig. 1). For the diagnosis of regional and distant recurrences 18F-FDG PET/CT has a sensitivity of 94%, a speci*city of 82% and an accuracy of 87% (compared with 81%, 82% and 81%, respectively, for conventional imaging) [7]. Other tracers such as 18F-FLT could be used, but data are still limited.

Gastric cancer Stomach cancer is often asymptomatic or in its early stage causes only non-speci*c symptoms like loss of appetite and abdominal discomfort. The cancer has often reached an advanced stage by the time that further symptoms occur, including abdominal pain, loss of weight and bleeding resulting in anaemia. If clinical symptoms suggest gastric cancer, endoscopy is the best imaging

Courtesy of Nuclear Medicine Department, Warsaw Medical University of method. Additionally, CT scanning of the Figure 1: Oesophageal cancer with metastasis abdomen may reveal gastric cancer, but it to paratracheal lymph node is more useful in determining invasion into adjacent tissues or the presence of spread to Distant metastatic disease has a signi*cant im- local lymph nodes. pact on patient management because these patients are no longer eligible for surgical In gastric cancer, 18F-FDG PET/CT may be used resection. For the detection of distant metas- for staging, diagnosis of recurrent disease, tases, 18F-FDG PET has a reported sensitivity prognosis and therapy monitoring. However, of 69–100%, a speci*city of 84–90% and an di&erent types of cancer are characterised by accuracy of 84–91%, while the sensitivity is a variety of types of 18F-FDG uptake, and the lower [4, 6]. 18F-FDG PET has been reported to clinical role of 18F-FDG PET/CT in this setting reveal more than 20% of unsuspected distant is currently not well de*ned.

90 Chapter 2 – Clinical applications of PET/CT in oncology: 2.7 Digestive tract tumours

The sensitivity of 18F-FDG PET for the detec- ing on tumour size, depth of invasion and cell tion of primary tumours is 58–94% and the type (Fig. 2). speci*city, 78–100% [8–12], the rates depend- EANM

Courtesy of Nuclear Medicine Department, Warsaw Medical University of a. b. Figure 2a,b: Gastric cancer. There is increased uptake of 18F-FDG (a) and thickening of the gastric wall (b)

Its reported detection rate for tumours mea- that the sensitivity and speci*city of 18F-FDG suring less than 30 mm is 17% while for tu- PET were, respectively, 85% and 74% for liver mours of more than 30 mm the rate rises to metastases, 67% and 88% for lung metasta- 77% [9]. In the di&use type of gastric cancer, ses, 24% and 76% for ascites, 30% and 82% for according to Lauren’s classi*cation, the de- bone metastases and 4% and 100% for pleural tection rate with 18F-FDG PET is 41%, which metastases [12]. is signi*cantly lower than that for the intestinal type (83%) [13]. Mucinous carcinoma and 18F-FDG PET/CT could play a role in the diag- signet ring cell carcinoma typically show low nosis of distant lesions, but the low metabolic uptake of 18F-FDG. activity seen in most primary tumours could also be anticipated in metastases. 18F-FDG PET has a lower sensitivity than CT in the detection of regional lymph node Because of the limited usefulness of 18F-FDG metastases, mainly owing to its poor spatial in gastric cancer owing to the relatively high resolution [14], while the reported sensitivity number of primary tumours that are not is 17–46% [9,10,15]. Yoshioka et al. reported glucose avid, other tracers have been inves-

91 tigated. Measurement of tumour growth and carcinomas, colloid carcinomas, undi&erenti- DNA synthesis in vivo might be superior for ated carcinomas, and undi&erentiated carci- imaging malignancies of the gastrointestinal nomas with osteoclast-like giant cells. Exocrine tract. In a study by Herrmann et al. using 18F- pancreatic tumours are far more common FLT, the sensitivity for the detection of primary than pancreatic endocrine tumours, which tumours was 100% whereas that of 18F-FDG account for only 1% of all cases. Pancreatic was 69% [16]. cancer is sometimes called a “silent killer” because in its early form it is symptomless, and Pancreatic cancer later symptoms are usually non-speci*c and Pancreatic cancer has a very poor prognosis, varied. Therefore, pancreatic cancer is often with the survival rate being only 5% at 5 years not diagnosed until it is advanced. following diagnosis. About 95% of exocrine pancreatic cancers are adenocarcinomas; the The sensitivity of PET in the diagnosis of pan- remaining 5% include adenosquamous carci- creatic adenocarcinoma is 67–97%, and the nomas, signet ring cell carcinomas, hepatoid speci*city, 61–97% [17–27] (Fig. 3).

Courtesy of Nuclear Medicine Department, Warsaw Medical University of a. b. Figure 3a,b: Pancreatic cancer. There is increased uptake of 18F-FDG in the head of the pancreas (a), in the visualised CT lesion (b) Hypoxia accompanying tumours and in%am- considered a supplementary examination if matory processes limits the value of examina- the results of other studies do not allow for tion. As its sensitivity and speci*city are similar unequivocal identi*cation. to those of other methods, PET imaging is now

92 Chapter 2 – Clinical applications of PET/CT in oncology: 2.7 Digestive tract tumours

Since neuroendocrine tumours of the pan- metastatic disease in patients with pancreatic creas are known to be 18F-FDG negative, a cancer, and its use has been reported to lead di&erent tracer should be used when such to treatment changes in 12–69% of such pa- a lesion is suspected; for example, 68Ga-DOT- tients [21,22,25,28]. ATATE may be used owing to the expression of somatostatin receptors on such tumours, Another issue is the characterisation of cys- or 18F-DOPA may be employed to target cat- tic lesions in the pancreas (di&erentiation echolamine pathways. between benign and proliferative changes) (Fig. 4). 18F-FDG PET/CT is signi*cantly more sensitive than conventional CT imaging in detecting EANM

Courtesy of Nuclear Medicine Department, Warsaw Medical University of a. b. Figure 4a,b: Cystic lesion of the pancreas. Decreased uptake of 18F-FDG is present in the tail of the pancreas (a), in the visualised CT lesion (b)

In this application, the additional functional as- In conclusion, 18F-FDG PET/CT is a comple- sessment is very useful. The sensitivity of PET is mentary study to assess the nature of focal higher than that of CT; however, the speci*city lesions in the pancreas. Moreover, it is a par- of the two methods is comparable. ticularly useful modality in the diagnosis of metastases.

93 Hepatocellular carcinoma Recently, 11C- or 18F-choline and 11C-acetate Hepatocellular carcinoma (HCC) is a primary have been introduced as new tumour-seeking malignancy of the liver. Most cases of HCC ap- agents for the evaluation of a variety of malig- pear in patients with chronic viral hepatitis nant tumours. 11C-acetate, a tracer of lipid me- infection (hepatitis B or C, ca. 20% of cases) tabolism, is a promising radiopharmaceutical or cirrhosis (ca. 80% of cases, alcoholism be- for the diagnosis of HCC. The reported sensi- ing the most common cause). HCC may pres- tivity in the detection of low- or intermediate- ent with jaundice, bloating from ascites, easy grade HCC is 87% [35]. bruising from blood clotting abnormalities, loss of appetite, unintentional weight loss, The accumulation of 18F-FDG and 11C-acetate abdominal pain, especially in the upper right depends on the biological activity of the part, nausea, emesis or fatigue. Owing to its cancer. High-grade tumours exhibit uptake cost-e&ectiveness, ultrasound (US) is usually of 11C-acetate, but not of 18F-FDG, whereas the *rst examination. In patients with a high low-grade tumours accumulate 18F-FDG but suspicion of HCC but negative US, the best show no or low accumulation of 11C-acetate. method of diagnosis is a CT scan of the abdo- Tumours of intermediate grade exhibit vari- men using an intravenous contrast agent and able accumulation of the two tracers. With three-phase scanning. respect to the use of PET/CT for detection of primary tumours, Park et al. reported an overall Owing to the high background of 18F-FDG sensitivity of 61% with 11C-acetate and 75% uptake by the normal liver and the low meta- with 18F-FDG as single tracers, rising to 83% bolic activity of HCC, only 50–55% of HCCs can for dual-tracer imaging [33]. Accumulation of be visualised by 18F-FDG PET [29–32]. For this 11C-acetate was signi*cantly greater than that reason, 18F-FDG PET/CT is not recommended of 18F-FDG in well-di&erentiated HCCs, with in the diagnosis of HCC. 18F-FDG PET can, positive lesions in 71.4% and 50% of cases, however, play a role in prognostic assessment respectively [33]. Owing to the short half-life since poorly di&erentiated tumours are able to of 11C (t½=20 min), 11C-acetate is only available accumulate 18F-FDG. Overall survival is lower in nuclear medicine departments equipped among patients with positive 18F-FDG PET/CT with a cyclotron. *ndings for all indexed lesions than among those with 18F-FDG negative or partially posi- Choline is the main component of phos- tive *ndings [33]. 18F-FDG may show extra- phatidylcholines, an important constituent hepatic metastases or residual activity after of cell membrane phospholipids. Malignant interventional therapy [34]. tumours may show high proliferation and increased metabolism of cell membrane

94 Chapter 2 – Clinical applications of PET/CT in oncology: 2.7 Digestive tract tumours

components, which will lead to increased world [37]. It arises from adenomatous pol- uptake of choline. yps in the colon. These mushroom-shaped growths are usually benign, but some develop As demonstrated in spectroscopic studies, into cancer over time. Localised colon cancer HCC is characterised by signi*cantly higher is usually diagnosed through colonoscopy. concentrations of choline than normal liver cells. It seems that 18F-choline is as good a PET is recommended in the staging and re- marker as 11C-acetate in the diagnosis of HCC, staging of colorectal cancer. The sensitivity of and it is much easier to use. Currently, how- 18F-FDG PET/CT in staging is 96–100%, and ever, no su+ciently large studies have been higher than that of other diagnostic tech- performed to identify the role of this marker niques, but speci*city is limited to 58–95% in clinical practice. Talbot et al. reported a sen- [38,39]. The sensitivity depends on the histo- EANM sitivity of 88% for 18F-%uorocholine and 68% for logical type of the carcinoma: lower sensitivity, 18F-FDG [36] in a study of 81 patients. Of 11 58%, is noted for the detection of mucinous patients with well-di&erentiated HCC, six had cancer, while for non-mucinous cancer sensi- a positive result with 18F-%uorocholine alone, tivity reaches 92% [40]. whereas 18F-FDG was never positive alone. In non-malignant lesions, 18F-%uorocholine ap- For the detection of lymph node involvement, pears less speci*c than 18F-FDG (62% vs 91%) 18F-FDG PET/CT has low sensitivity but high because of uptake by focal nodular hyperpla- speci*city. In a study by Abdel-Nabi et al., sia [36]. sensitivity was 29% and speci*city was 96% [38]; Mukai et al. [39] found a similar sensitivity In conclusion, patients with suspected HCC (22.2%). Owing to the e&ectiveness of other should be examined using 18F-choline or diagnostic methods (particularly endoscopy) 11C-acetate, while an additional study using and limitations in the assessment of micro- 18F-FDG is recommended when a negative metastases in regional lymph nodes, PET is or ambiguous result is obtained. Owing to of limited signi*cance in the initial staging of the resolution of the images, changes with the disease. a diameter of less than 7 mm cannot always be assessed. PET has a high sensitivity in the evaluation of distant metastases. The reported sensitivity Colorectal cancer of PET in the detection of liver metastasis is Colorectal cancer is the most common cancer between 85% and 100% (mean 88%) while of the digestive tract and the third leading that of CT is 82%; similarly, the speci*city of cause of cancer-related death in the Western PET is 96% and that of CT, 84% [38, 41–48].

95 The likelihood of false-negative *ndings de- The results of 18F-FDG PET change the treat- pends on the size of metastases (for lesions ment plan in 31% (range 20–58%) of patients <1 cm, sensitivity is only 25%) [41] and the [42,45,46]. histological type. Other distant metastases may be recognised with 18F-FDG PET/CT. 18F-FDG PET/CT is a useful tool in the recog- Overall, PET has 91% sensitivity and 76–95% nition of local recurrences: di&erentiation be- speci*city for metastatic disease, and is thus tween postoperative or post-radiation therapy superior to CT (61% sensitivity and 91% spec- scar/*brotic tissue and vital tumour in rectal i*city) [45]. cancer is di+cult or impossible when using morphology-based imaging modalities (Fig. 5).

Courtesy of Nuclear Medicine Department, Warsaw Medical University of a. b. Figure 5a,b: Colon cancer after surgery and radiotherapy. There is absence of increased uptake of 18F-FDG (a) in the visualised CT lesion (b). Typical scar is present after radiotherapy

In a meta-analysis, the sensitivity of 18F-FDG 94% and that of CT, 79% (2,244 patient studies); PET/CT in detecting local recurrence was speci*city is 87% for PET and 73% for CT (2,244 found to be 98% and the speci*city, 76% [49]. patient studies), and PET leads to a change in A review of the literature for evaluation of re- treatment in 32% (915 patient studies) (Fig. 6). currence shows that the sensitivity of PET is

96 Chapter 2 – Clinical applications of PET/CT in oncology: 2.7 Digestive tract tumours

Courtesy of Nuclear Medicine Department, Warsaw Medical University of

a. b. EANM Figure 6a,b: Colon cancer after surgery. There is increased uptake of 18F-FDG at the site of surgery and metastasis to pelvic lymph nodes (a), in the visualised CT lesions (b) 18F-FDG PET/CT has been found to be success- Usually, like most malignancies, cholangiocar- ful in localising the site of relapse in patients cinoma cells have high glucose uptake, and with elevated CEA, with a reported sensitivity the sensitivity of 18F-FDG PET/CT in carcinoma of nearly 100% and a speciﬁcity of 71% [50]. of the gallbladder and cholangiocarcinoma is quite high, although it seems to be dependent Gallbladder carcinoma and on the tumour subtype. The sensitivity of 18F- cholangiocarcinoma FDG PET/CT for the detection of gallbladder Cholangiocarcinoma is a malignant primary carcinoma is 78%, and in nodular cholangio- tumour of the biliary duct system that may carcinoma it rises to 85%; however, sensitiv- originate in the intrahepatic/extrahepatic bile ity is only 18% in inﬁltrating cholangiocarci- ducts and terminate at the ampulla of Vater. noma [51]. Detection of cholangiocarcinoma This is the reason it can involve three regions depends on localisation. The sensitivity and of the biliary system: intrahepatic, perihilar speci*city of 18F-FDG PET/CT for intrahepatic and distal extrahepatic. The most common cholangiocarcinoma are 93% and 80%, respec- localisation is perihilar; such tumours are also tively, and better than the values for CT. For called Klatskin tumours. Intrahepatic tumours extrahepatic cholangiocarcinoma, the sensi- are the least common. More than 95% of these tivity of 18F-FDG PET/CT is 55% and speci*city tumours are ductal adenocarcinomas; many is 33%, which is comparable to CT [52]. patients present with unresectable or metastatic disease.

97 18F-FDG PET/CT can identify distant metasta- in comparison to only 44% with CT [56,57]; ses of peripheral cholangiocarcinoma that are 18F-FDG PET/CT could predict response as ear- not detected with MRI and CT. The sensitiv- ly as 24 h after the start of treatment, whereas ity of 18F-FDG PET/CT for detection of distant CT only did so several weeks later [56]. This metastases has been reported to be 94–100%, ability to predict the long-term treatment re- with 100% speci*city; however, the sensitivity sponse is useful [57, 58]. An absolute SUV max for detection of regional lymph nodes me- threshold of 3.4 and a 40% reduction in SUV max tastases was only 12%, with 96% speci*city have been found to be more predictive of time [52,53]. The ability of PET to detect distant to treatment failure than the SUV threshold max metastases may alter the surgical manage- of 2.5 and 25% reduction reported in a previ- ment plan in up to 30% of cases [51,54]. ous trial [59,60].

Gastrointestinal stroma tumours Unfortunately, access to PET is still limited for Gastrointestinal stroma tumours (GISTs) are patients with GISTs because the glucose up- the most common mesenchymal tumours of take before treatment is not su+cient to be the gastrointestinal tract (1–3% of all gastro- detected by 18F-FDG PET/CT in some lesions. intestinal malignancies). The tumours most often arise in the stomach (70%), followed by Conclusion the small intestine (20%), large intestine (5%) PET/CT is a very useful technique in the and oesophagus (5%). The biology of the tu- imaging of oncological disease. The com- mour is de*ned by mutations in the c-kit gene. monly available radiotracer 18F-FDG is well In 1998, Hirota et al. reported a mutation of the established as a diagnostic tool for a variety c-kit proto-oncogene that activates tyrosine of gastrointestinal cancers. It is not a speci*c kinase in the absence of a stem cell factor, radiotracer and further developments of spe- leading to uncontrolled cell proliferation [55]. ci*c radiotracers will improve the modality. Labelled acetate and choline are now well de- 18F-FDG PET has been shown to be highly sen- veloped tracers in the diagnosis of hepatocel- sitive in detecting early response to treatment lular carcinoma. Novel reconstruction software with the tyrosine kinase inhibitor imatinib: 1 and new technical ideas such as time of %ight month after initiation of therapy, the response with new radiotracers will be the future of on- was accurately predicted in 95% with PET/CT cological imaging.

98 mography and positron emission tomography in the pre- tomographythe emission in positronmography and 2000;120:1085–92. Surg Cardiovasc Thoraccancer.J esophageal recurrent of sitron emission tomography for the diagnosis and staging 2. Rankin SC, Taylor H, Cook GJ, Mason R. Computed to- Computed R. Mason GJ, Cook TaylorSC, H, Rankin 2. 1997;168:417–24. Roentgenol J Am 18F-%uorodeoxyglucose AJR tomography. emission with positron cancer esophageal Staging al. et of BA, Siegel F, Dehdashti FL, Flanagan 1. recurrent cancer. gastric 2003;44:690–9. JNuclMed or metastatic, advanced, with patients in PET 18F-FDG of Evaluation al. et K, Kubota YoshiokaT,12.K, Yamaguchi Positron 1998;4:213–21. Imaging Clin experience. Preliminary cancer. gastric in FDG-PET of Accuracy al. et M, Karpehet H, Yeung HW,Macapinlac 11. for cancer. gastric World JSurg 2004;28:247–53. positron 2-deoxy-2-%uoro-d-glucose emission tomography E, H, Kuwano H, et Katoh al.Mochiki Evaluation of 18F- 10. Cancer 2006;9:192–6. Gastric cancer. gastric of detection for FDG-PET perative preo- of Usefulness al. et K, Ishida Y, Okajima K, Mukai 9. 2005;46:1582–8. with CT. JNuclMed study comparison a PET: (18)F-FDG using cancer gastric of staging node Lymph al. et SH Noh JS, Lim M, Yun 8. Flamen P, Lerut A, Van Cutsem E, et al. The of utility po- 7. tomography in thoracic oncology. Thorax 1998;53:703–12. Lowe VJ, Naunheim KS. Current role of positron emission 6. Ann Thorac Surg 1997;64:765–9. cancer.esophageal staging in tomography emission tron Luketich JD, PR, Schauer Meltzer CC, et al. of Role posi- 5. PET. 2000;20:713–23. Radiographics FDG at cancer esophageal recurrent and primary of tures fea- Imaging al. et AL, BrownSJ,M, Thompson Skehan 4. troesophageal malignancy. BrJCancer 1998;78:521–27. emission tomography for staging of oesophageal and Positrongas- VaalburgW. OE, Nieweg JT, Plukker AC, Kole 3. 1998;53:659–65. Radiol Clin carcinoma. oesophageal of staging operative References References Chapter 2.7 99 18. Nishiyama Y, Nishiyama Evaluation 18. Yamamoto al. Y, et T, Monden Nucl2004;23:90–4. Esp Med ma camera coincidence imaging in pancreatic cancer. Rev Giorgi 17. MC, Cunha RM, Soares J Jr, et al. Dual-head gam- comparative analysis. 2007;48:1945–50. JNuclMed 18F-FLTa radiotracers 18F-FDG: the and and PET with cer can- al.et gastric Imaging AK, Buck K, Ott K, Herrmann 16. emission tomography. Cancer 2005;103:2383–90. operative staging of gastric adenocarcinoma with positron J,Chen JH, Cheong 15. YunMJ, al.et Improvementpre- in 1999;72:525–9. phageal carcinoma. BrJRadiol oeso- and gastric in tomography emission positron FDG 18F- of D,Evaluation al.McAteer F,Wallis et 14. G, Couper 2003;30:288–95. Imaging Mol Med Nucl J Eur *ndings. histopathological and doscopic en- with correlation carcinomas: gastric advanced locally of imaging PET FDG al. et WA,Weber K, Ott A, Stahl 13. with 104patients. 2004;45:1279–86. JNuclMed study prospective a -- lesions pancreatic in value clinical digital image fusion of multidetector and CT 18F-FDG PET: Retrospective al. et N, Hosten SM, Niehues AJ, Lemke 23. cost-e&ectiveness. AnnSurg 2005;242:235–43. its and cancer pancreatic resectable of management the on in%uences tomography tomography/computed sion Heinrich S, Goerres GW,22. Schafer M, et al. Positron emis- J ClinGastroenterol 2006;40:923–9. after concurrent chemoradiotherapy for pancreatic cancer. the di&erential diagnosis, staging, and response evaluation of 18-%uorodeoxyglucose positron emission tomography in Bang S, 21. Chung HW, Park SW, et al. The clinical usefulness 2005;32:399–404. Imaging Mol Eur JNuclMed able to detect pancreatic carcinoma in chronic pancreatitis. van Kouwen 20. MC, Jansen JB, van Goor H, et al. FDG PET is JSurgatic masses?Scand 2004;93:191–7. 11C-acetate valuable in diagnosing indeterminate pancre- and 18F-%uorodeoxyglucose using tomography emission I, Rasmussen Sorensen J, Langstrom 19. B, et al. Is positron pancreatic tumour. Commun 2005;26:895–901. NuclMed with patients in imaging PET FDG additional delayed of

EANM 24. Lytras D, Connor S, Bosonnet L, et al. Positron emission 34. Torizuka T, Tamaki N, Inokuma T, et al. Value of %uorine- tomography does not add to computed tomography for 18-FDG-PET to monitor hepatocellular carcinoma after the diagnosis and staging of pancreatic cancer. Dig Surg interventional therapy. J Nucl Med 1994;35: 1965–9. 2005;22:55–62. 35. Ho CL, Yu SC, Yeung DW. 11C-acetate PET imaging in 25. Maemura K, Takao S, Shinchi H, et al. Role of positron hepatocellular carcinoma and other liver masses. J Nucl emission tomography in decisions on treatment strate- Med 2003;44:213–21. gies for pancreatic cancer. J Hepatobil Pancreat Surg 2006;13:435–41. 36. Talbot J, Fartoux L, Balogova S, et al. Detection of hepatocellular carcinoma with PET/CT: a prospective compari- 26. Sperti C, Bissoli S, Pasquali C, et al. 18-Fluorodeoxy- son of 18F-%uorocholine and 18F-FDG in patients with cirrho- glucose positron emission tomography enhances com- sis or chronic liver disease. J Nucl Med 2010;51:1699–706. puted tomography diagnosis of malignant intraductal papillary mucinous neoplasms of the pancreas. Ann Surg 37. Cancer. National Cancer Institute. 2009. http://www. 2007;246:932–9. cancer.gov/cancertopics/commoncancers.

27. Casneuf V, Delrue L, Kelles A, et al. Is combined [18] 38. Abdel-Nabi H, Doerr RJ, Lamonica DM, et al. Staging of F%uorodeoxyglucose-positron emission tomography/com- primary colorectal carcinomas with %uorine-18 %uorodeo- puted tomography superior to positron emission tomogra- xyglucose whole-body PET: correlation with histopatholo- phy or computed tomography alone for diagnosis, staging gic and CT *ndings. Radiology 1998;206:755–60. and restaging of pancreatic lesions? Acta Gastro-Ent Belg 39. Mukai M, Sadahiro S, Yasuda S, et al. Preoperative eva- 2007;70:331–8. luation by whole-body 18F-%uorodeoxyglucose positron 28. Nishiyama Y, Yamamoto Y, Yokoe K, et al. Contribution of emission tomography in patients with primary colorectal whole body FDG-PET to the detection of distant metastasis cancer. Oncol Rep 2000;7:85–7. in pancreatic cancer. Ann Nucl Med 2005;19:491–7. 40. Whiteford MH, Whiteford HM, Yee LF, et al. Usefulness of 29. Iwata Y, Shiomi S, Sasaki N, et al. Clinical usefulness of FDG-PET scan in the assessment of suspected metastatic positron emission tomography with %uorine-18-%uorode- or recurrent adenocarcinoma of the colon and rectum. Dis oxyglucose in the diagnosis of liver tumors. Ann Nucl Med Colon Rectum 2000 ;43 :759–67. 2000;14:121–6. 41. Fong Y, Saldinger PF, Akhurst T, et al. Utility of 18F-FDG 30. Khan M, Combs C, Brunt E, et al. Positron emission to- positron emission tomography scanning on selection of mography scanning in the evaluation of hepatocellular patients for resection of hepatic colorectal metastases. Am carcinoma. J Hepatol 2000;32:792–7. J Surg 1998;178:282–7.

31. Delbeke D, Martin WH, Sadler MP, et al. Evaluation of 42. Ogunbiyi OA, Flanagan FL, Dehdashti F, et al. Detection benign vs malignant hepatic lesions with positron emission of recurrent and metastatic colorectal cancer: comparison tomography. Arch Surg 1998;133:510–6. of positron emission tomography and computed tomography. Ann Surg Oncol 1997;4:613–20. 32. Trojan J, Schroeder O, Raedle J, et al. Fluorine-18 FDG positron emission tomography for imaging of hepatocel- 43. Rohren EM, Paulson EK, Hagge R, et al. The role of F-18 lular carcinoma. Am J Gastroenterol 1999;94:3314–9. FDG positron emission tomography in preoperative assessment of the liver in patients being considered for curative 33. Park JW, Kim JH, Kim SK, et al. A prospective evaluation resection of hepatic metastases from colorectal cancer. Clin of 18FDG and 11C acetate PET/CT for detection of prima- Nucl Med 2002;27:550–5. ry and metastatic hepatocellular carcinoma. J Nucl Med 2008;49:1912–21. 44. Boykin KN, Zibari GB, Lilien DL, et al. The use of FDG- positron emission tomography for the evaluation of colorectal metastases of the liver. Am Surg 1999;65:1183–5.

100 Chapter 2 – Clinical applications of PET/CT in oncology: 2.7 Digestive tract tumours

45. Wiering B, Krabbe PF, Jager GJ, et al. The impact of 54. Fritscher-Ravens A, Bohuslavizki KH, Broering DC, et al. %uor-18-deoxyglucose-positron emission tomography in FDG PET in the diagnosis of hilar cholangiocarcinoma. Nucl the management of colorectal liver metastases. Cancer Med Commun 2001;22:1277–85. 2005;104:2658–70. 55. Hirota S, Isozaki K, Moriyama Y, et al. Gain-of-function 46. Delbeke D, Vitola JV, Sandler MP, et al. Staging recur- mutations of C-kit in human gastrointestinal stromal tu- rent metastatic colorectal carcinoma with PET. J Nucl Med mors. Science 1998;279:577–80. 1997;38:1196–201. 56. Stroobants S, Goeminne J, Seegers M, et al. 18FDG- 47. Lai DT, Fulham M, Stephen MS, et al. The role of whole- positron emission tomography for the early prediction body positron emission tomography with [18F]%uorodeo- of response in advanced soft tissue sarcoma treated with xyglucose in identifying operable colorectal cancer meta- imatinib mesylate (Gleevec). Eur J Cancer 2003;39:2012–20. stases the liver. Arch Surg 1996;131:703–7. 57. Antoch G, Kanja J, Bauer S, et al. Comparison of PET, 48. Beets G, Pennickx F, Schiepers C, et al. Clinical value CT, and dual-modality PET/CT imaging for monitoring of of whole-body positron emission tomography with imatinib (STI-571) therapy in patients with gastrointestinal [18F]%uorodeoxyglucose in recurrent colorectal cancer. stromal tumors. J Nucl Med 2004;45:357–65.

Br J Surg 1994;81:1666–70. EANM 58. Therasse P, Arbuck SG, Eisenhauer EA, et al. New 49. Huebner RH, Park JE, Shepard JE, et al. Meta-analysis of guidelines to evaluate the response to treatment in solid the literature for whole body FDG-PET detection of recur- tumors: European Organization for Research and Treatment rent colorectal cancer. J Nucl Med 2000;41:1177–89. of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 50. Tutt ANJ, Plunkett TA, Barrington SF, et al. The role of 2000;92:205–16. positron emission tomography in the management of colorectal cancer [review]. Colorectal Dis 2004;6:2–9. 59. Van den Abbeele AD. The lessons of GIST — PET and PET/CT: a new paradigm for imaging. The Oncologist 51. Anderson CD, Rice MH, Pinson CW, et al. Fluorodeo- 2008;13:8–13. xyglucose PET imaging in the evaluation of gallbladder carcinoma and cholangiocarcinoma. J Gastrointest Surg 60. Holdsworth CH, Manola J, Badawi RD, et al. Use of com- 2004;8:90–7. puterized tomography (CT) as an early prognostic indicator of response to imatinib mesylate (IM) in patients with gas- 52. Petrowsky H, Wildbrett P, Husarik DB, et al. Impact of trointestinal stromal tumors (GIST). Proc Am Soc Clin Oncol integrated positron emission tomography and computed 2004;22:197. tomography on staging and management of gallbladder cancer and cholangiocarcinoma. J Hepatol 2006;45:43–50.

53. Jadvar H, Henderson RW, Conti PS. F-18]%uorodeoxyglucose positron emission tomography and positron emission tomography: computed tomography in recurrent and metastatic cholangiocarcinoma. J Comput Assist Tomogr 2007;31:223–8.

101 Chapter 2 – Clinical applications of PET/CT in oncology 2.8 Neuroendocrine tumour Valentina Ambrosini and Stefano Fanti

Introduction typical feature of NE cells is the expression of Over past decades, nuclear medicine pro- several receptors in high quantities [3]; soma- cedures have acquired a relevant role in the tostatin receptor (SST) subtypes are the ones detection of neuroendocrine tumours (NET), mostly represented. and the recent introduction of speci*c PET tracers for NET studies has further increased As regards disease staging, a TNM staging sys- nuclear physicians’ diagnostic and therapeutic tem has been proposed for tumours of the (peptide receptor radionuclide therapy, PRRT) fore-gut , mid-gut and hind-gut [4,5] by the interest in this subset of tumours. European Neuroendocrine Tumour Society (ENETS), taking into account the most recent Knowledge of the biological features of NET WHO classi*cations and the need for a bet- is of crucial importance in understanding ter prognostic assessment of NET. Histologi- the mechanism of action and the potential cal grading identi*es three categories based clinical relevance of the radiotracers currently on the Ki-67 levels (G1: Ki-67 <2%; G2: Ki-67 used in NET imaging, since nuclear medicine between 3% and 20%; G3: Ki-67 >20%) [4,5]. compounds have been designed taking into Although traditionally the TNM staging classi- account the speci*c characteristics of NET *cation has not been applied to lung NET, the cells. TNM has been demonstrated to be useful in these patients; therefore the International As- NET are heterogeneous slow-growing neo- sociation for the Study of Lung Cancer (IASLC) plasms that occur in 1-4/100,000 people per recently recommended that the TNM be ap- year [1,2]. Derived from endocrine cells origi- plied to pulmonary NET [6]. nating from the neural crest [3], NET are characterised by an endocrine metabolism and a From a diagnostic point of view, NET have typical pathological pattern. Moreover, NET represented a diagnostic challenge since they can produce a large variety of substances, and can present as small lesions with variable ana- as they belong to the APUD cell system (amine tomical localisation. The diagnostic %ow-chart precursor uptake and decarboxylation), NET of NET has traditionally included ultrasound, can take up, accumulate and decarboxylate computed tomography (CT), magnetic reso- amine precursors such as dihydroxyphenyl- nance imaging and somatostatin receptor alanine (DOPA) and hydroxyltryptophan. The scintigraphy (SRS) [7,8]. SRS has played a major presence of hormone syndromes related to role in the functional assessment of NET, since amine/hormone production allows the dif- it has been reported to have a higher accu- ferentiation of NET in functional (33–50% of racy than conventional imaging procedures cases) and non-functional tumours. Another [9]. However, more recently, the introduction

102 Chapter 2 – Clinical applications of PET/CT in oncology: 2.8 Neuroendocrine tumour

of speci*c PET tracers for NET assessment Currently the most relevant PET tracers for completely revolutionised the diagnostic ap- well-di&erentiated NET tumours (Fig. 1) are proach to this subset of tumours owing to the 68Ga-DOTA peptides (DOTATOC, DOTANOC, better spatial resolution of PET as compared DOTATATE) and 18F-DOPA while 18F-FDG is em- to scintigraphy. ployed in selected cases with undi&erentiated lesions or variable expression of SST [10]. EANM

Courtesy of Nuclear Medicine Unit, S.Orsola-MalpighiAzienda Policlinico Ospedaliero-Universitaria di Bologna, Figure 1a–c: PET coronal images showing the physiological di&erences in tracer biodistribution. a 68Ga-DOTA peptides (DOTANOC); b 18F-DOPA; c 18F-FDG

68Ga-DOTA peptides SST (TOC, NOC, TATE), a chelant (DOTA) and 68Ga-DOTA peptides are a group of positron- the beta-emitting isotope (68Ga). These com- emitting tracers that speci*cally bind to SST pounds show variable binding a+nity to SST over-expressed on NET cells. Six di&erent SST receptor subtypes [11]: all can bind to SST2 have been identi*ed (SST1, SST2A, SST2B, and SST5, but only DOTANOC also presents SST3, SST4, SST5) in humans. All tracers be- good a+nity for SST3. There is nevertheless longing to this group have a common struc- at present no indication that these di&erences ture that includes the active part binding to in SST binding a+nity are associated with a

103 direct clinical correlate and/or advantages Universitaria di Bologna, Policlinico S.Orsola-Malpighi Ospedaliero- Unit, Azienda of Nuclear Medicine Courtesy in clinical employment. The uptake of 68Ga- DOTA peptides is an indirect measure of the presence of SST on NET cells, which are overexpressed in well-di&erentiated tumours, and is not dependent on cell metabolism (unlike, for example, uptake of 18F-DOPA or 18F-FDG). Therefore, 68Ga-DOTA peptide PET/CT not only provides data on disease extension but also allows non-invasive assessment of the presence of SST on NET cells; this has direct therapeutic implications, since NET are treated with cold or ‘hot’ (PRRT) somatostatin analogues. Antunes et al. also reported how the labelling with 68Ga presents advantages over other employed iso- Figure 2: 68Ga-DOTANOC PET/CT images of a topes [11]. patient with biopsy-proven NET metastatic lesions and unknown primary tumour. PET/ Indications for 68Ga-DOTA peptide PET/CT in- CT allowed the detection of a focal area of clude staging, re-staging after therapy, iden- tracer uptake in the ileum wall suspicious for ti*cation of the site of an unknown primary tumour primary site, one abdominal node, tumour (in patients with proven NET second- one mesenteric nodule and a focal area at ary lesions) and selection of cases eligible for the liver level (*rst segment). 68Ga-DOTANOC therapy with either cold or hot (PRRT) soma- PET/CT showed that all the detected lesions tostatin analogues. presented an increased expression of SST

At present, 68Ga-DOTA peptides are employed In the past few years, the employment of 68Ga- in specialised centres as part of clinical trials DOTA peptides in well-di&erentiated NET has for the diagnostic work-up of NET and they increased exponentially, and the published have not yet been registered for routine clini- results con*rm their higher accuracy as com- cal use. The most commonly employed tracers pared with CT and SRS for the identi*cation are 68Ga-DOTATOC, 68Ga-DOTANOC (Fig. 2) and of both the primary tumour and metastatic 68Ga-DOTATATE. sites [12,13]. The study with the largest patient population with NET (84 patients) reported a superior sensitivity of DOTATOC (97%) in comparison to CT (61%) and SRS (52%) [13]. 68Ga-

104 Chapter 2 – Clinical applications of PET/CT in oncology: 2.8 Neuroendocrine tumour

DOTA peptides have been reported [14–17] to a positron emitter with 89% positron emission accurately visualise well-di&erentiated NET le- and negligible gamma emission (1077 keV) sions at lymph nodes, bone and liver and to be of 3.2% only. For labelling, the 68Ga eluate is of value in detection of the unknown primary *rst concentrated and puri*ed using a micro- tumour. A recent paper [18] also described chromatography method [20]. Radiolabelling how the maximum standardised uptake yields of >95% can usually be achieved within value (SUV ) may have prognostic value in 15 min. max patients with NET: in a population of 44 patients, the SUV was signi*cantly higher in 68Ga-DOTA peptide PET/CT image acquisi- max cases with stable disease or partial response at tion (skull base to the middle of the thigh) follow-up. These data re%ect the observation should be performed following currently that well-di&erentiated lesions (characterised available guidelines [21]. Intravenous injected EANM by higher SST expression and therefore tracer doses (68Ga-DOTATOC, 68Ga-DOTANOC, 68Ga- uptake) are associated with a slower growth DOTATATE) vary between 100 and 200 MBq. rate. Recently, an evaluation of the clinical The amount of injected radioactivity strictly impact of 68Ga-DOTANOC PET/CT on patient depends on the daily production of the gen- management was performed in a population erator for each single elution (usually ranging of 90 subjects: 68Ga-DOTANOC PET/CT a&ected from 300 to 700 MBq) and, of course, the num- either the staging or the therapy in half the ber of patients scanned per day. Uptake time patients, with modi*cation of the therapeutic varies between 45 and 90 min after injection; approach being the most frequently observed although there is no generally accepted acqui- impact [19]. sition time in the literature, best results at most centres are achieved acquiring the images at From a technical point of view, 68Ga-DOTA 60 min. Considering the prognostic relevance peptides present several advantages. The of SUV measures, it is important that up- max labelling with 68Ga is relatively easy and in- take time is standardised among patients. expensive: 68Ga can be easily eluted from a commercially available 68Ge/68Ga generator Images should be reconstructed using the it- and does not require an on-site cyclotron. erative reconstruction algorithm implement- However it should be noticed that all gen- ed in the system and with the system settings. erators available so far are not GMP compliant. The long half-life of the mother radionuclide Sites of physiological SST receptor expression 68Ge (270.8 days) makes it possible to use the include the liver, spleen, pituitary, thyroid, kid- generator for approximately 9–12 months de- neys, adrenal glands, salivary glands, stomach pending on requirement. 68Ga (t1/2=68 min) is wall and bowel.

105 Potential false positive *ndings are repre- age acquisition is performed after an uptake sented by the presence of in%ammation (due time of 60 min from injection (intravenous in- to the expression of SST on activated lym- jection of 370 MBq). Oral premedication with phocytes), accessory spleen or physiologi- carbidopa has been reported to increase im- cal activity at the adrenal level. The head of age quality [25]. Regions of physiological 18F- the pancreas may show variable 68Ga-DOTA DOPA uptake include the striatum and pan- peptide uptake and represents a potential creas with subsequent elimination though the pitfall in image interpretation since it is not biliary, digestive and urinary tracts. necessarily associated with the presence of active disease. Its clinical signi*cance and the 18F-FDG mechanism of uptake in the pancreatic head It is generally accepted that 18F-FDG is not the are still debated. tracer of choice to visualise well-di&erentiated NET. However, the evidence that NET lesions 18F-DOPA may present areas with variable tracer uptake Originally employed for the diagnosis of within the same lesion supported the hypoth- movement disorders and in cases of hypo- esis of employing 18F-FDG in undi&erentiated glycaemia of the newborn, 18F-DOPA also NET forms. In fact, less di&erentiated NET le- may be used in patients with NET. In fact, as sions present a higher growth rate and there- they belong to the APUD cell system, NET fore higher glucose consumption and can be cells are easily visualised on 18F-DOPA PET/ visualised on 18F-FDG PET/CT. From a clinical CT images. point of view, the presence of 18F-FDG-avid NET lesions is associated with a worse prog- 18F-DOPA has been reported to show a higher nosis and a signi*cantly higher risk of death accuracy for the detection of NET lesions [26]. In addition, 18F-FDG may contribute to the compared with CT and SRS [9]. Currently, identi*cation of NET histological subtypes that there are only two studies comparing 68Ga- are characterised by variable expression of SST DOTA peptides (DOTANOC and DOTATATE) (e.g. medullary thyroid carcinoma). with 18F-DOPA in NET: in both papers a higher sensitivity was reported for 68Ga-DOTA pep- Conclusions tides [22,23]. Several PET tracers are currently available to study NET. 68Ga-DOTA peptides seem to be From a technical point of view, the relatively the most promising owing to their practical di+cult synthesis process [24] and higher advantages (easy synthesis process, poten- costs have limited widespread use of 18F-DOPA tially lower costs, patient friendliness) and in favour of 68Ga-DOTA peptides. 18F-DOPA im- higher sensitivity for the detection of well-

106 Chapter 2 – Clinical applications of PET/CT in oncology: 2.8 Neuroendocrine tumour

di&erentiated lesions. Moreover, 68Ga-DOTA peptides non-invasively provide information on SST expression on NET cells, with a direct impact on treatment decisions (selection of eligible candidates for either ‘hot’ or ‘cold’ somatostatin analogue therapy). 18F-DOPA also has a relevant role, being more accurate than SRS and CT; the main advantage as compared to 68Ga-DOTA peptides is related to commercial availability of the tracer while limitations are relatively high cost and incomplete di&usion throughout Europe. EANM The role of 18F-FDG in NET imaging is mainly limited to undi&erentiated NET forms and cases presenting a variable SST expression (e.g. medullary thyroid carcinoma).

107 References Chapter 2.8

References 1. Modlin IM, Kidd M, Latich I, Zikusoka MN, Shapiro MD. 11. Antunes P, Ginj M, Zhang H, Waser B, Baum RP, Reubi Current status of gastrointestinal carcinoids. Gastroentero- JC, et al. Are radiogallium-labelled DOTA-conjugated so- logy 2005;128:1717–51. matostatin analogues superior to those labelled with other radiometals? Eur J Nucl Med Mol Imaging 2007;34:982–93. 2. Taal BG, Visser O. Epidemiology of neuroendocrine tumours. Neuroendocrinology 2004;80 Suppl 1:3–7. 12. Kowalski J, Henze M, Schuhmacher J, Mäcke HR, Hof- mann M, Haberkorn U. Evaluation of positron emission 3. Reubi JC. Neuropeptide receptors in health and di- tomography imaging using [68Ga]-DOTA-D Phe(1)-Tyr(3)- sease: the molecular basis for in vivo imaging. J Nucl Med octreotide in comparison to [111In]-DTPAOC SPECT. First 1995;36:1825–35. results in patients with neuroendocrine tumors. Mol Ima- 4. Rindi G, Klöppel G, Alhman H, Caplin M, Couvelard A, de ging Biol 2003;5:42–8. Herder WW, et al. TNM staging of foregut (neuro)endocrine 13. Gabriel M, Decristoforo C, Kendler D, Dobrozemsky G, tumors: a consensus proposal including a grading system. Heute D, Uprimny C, et al. [68Ga]DOTA-Tyr3-octreotide PET Virchows Arch 2006;449:395–401. in neuroendocrine tumors: comparison with somatostatin 5. Rindi G, Klöppel G, Couvelard A, Komminoth P, Körner receptor scintigraphy and CT. J Nucl Med 2007;48:508–18. M, Lopes JM, et al. TNM staging of midgut and hindgut 14. Putzer D, Gabriel M, Henninger B, Kendler D, Uprimny (neuro) endocrine tumors: a consensus proposal including C, Dobrozemsky G, et al. Bone metastases in patients with a grading system. Virchows Arch 2007;451:757–62. neuroendocrine tumor: 68Ga-DOTA-Tyr3-octreotide PET 6. Travis WD; IASLC Staging Committee. Reporting lung in comparison to CT and bone scintigraphy. J Nucl Med cancer pathology specimens. Impact of the anticipated 7th 2009;50:1214–21. Edition TNM classi*cation based on recommendations of 15. Ambrosini V, Tomassetti P, Castellucci P, Campana D, the IASLC Staging Committee. Histopathology 2009;54:3– Montini G, Rubello D, et al. Comparison between [68Ga] 11. DOTA-NOC and [18F]DOPA PET for the detection of gastro- 7. Sundin A, Garske U, Orlefors H. Nuclear imaging of neu- entero-pancreatic and lung neuro-endocrine tumours. Eur roendocrine tumours. Best Pract Res Clin Endocrinol Metab J Nucl Med Mol Imaging 2008;35:1431–8. 2007;21:69–85. 16. Fanti S, Ambrosini V, Tomassetti P, Castellucci P, Montini 8. Ramage JK, Davies AH, Ardill J, Bax N, Caplin M, Gross- G, Allegri V, et al. Evaluation of unusual neuroendocrine man A, et al. UKNETwork for neuroendocrine tumours. tumours by means of [68Ga]DOTA-NOC PET. Biomed Phar- Guidelines for the management of gastroenteropancreatic macother 2008;62:667–71. neuroendocrine (including carcinoid) tumours. Gut 2005;54 17. Prasad V, Ambrosini V, Hommann M, Hoersch D, Fanti S, Suppl 4:iv1–16. Baum RP. Detection of unknown primary neuroendocrine 9. Krenning EP, Kwekkeboom DJ, Bakker WH, Breeman WA, tumours (CUP-NET) using (68)Ga-DOTA-NOC receptor PET/ Kooij PP, Oei HY, et al. Somatostatin receptor scintigraphy CT. Eur J Nucl Med Mol Imaging 2010;37:67–77. with [111In-DTPA-D-Phe1]- and [123I-Tyr3]-octreotide: the 18. Campana D, Ambrosini V, Pezzilli R, Fanti S, Labate AM, Rotterdam experience with more than 1000 patients. Eur J Santini D, et al. Standardized uptake values of (68)Ga-DO- Nucl Med 1993;20:716–31. TANOC PET: a promising prognostic tool in neuroendocrine 10. Adams S, Baum R, Rink T, Schumm-Dräger PM, Usadel tumors. J Nucl Med. 2010;51:353–9. KH, Hör G. Limited value of %uorine-18 %uorodeoxyglucose 19. Ambrosini V, Campana D, Bodei L, Nanni C, Castellucci positron emission tomography for the imaging of neuro- P, Allegri V, et al. 68Ga-DOTANOC PET/CT clinical impact endocrine tumors. Eur J Nucl Med 1998;25:79–83. in patients with neuroendocrine tumors. J Nucl Med 2010;51:669–73.

108 Chapter 2 – Clinical applications of PET/CT in oncology: 2.8 Neuroendocrine tumour

20. Zhernosekov KP, Filosofov DV, Baum RP, Ascho& P, Bihl H, Razbash AA, et al. Processing of generator-produced 68Ga for medical application. J Nucl Med 2007;48:1741–8.

21. Virgolini I, Ambrosini V, Bomanji JB, Baum RP, Fanti S, Gabriel M, et al. Procedure guidelines for PET/CT tumour imaging with 68Ga-DOTA-conjugated peptides: 68Ga- DOTA-TOC, 68Ga-DOTA-NOC, 68Ga-DOTA-TATE. Eur J Nucl Med Mol Imaging 2010;37:2004–10.

22. Ambrosini V, Tomassetti P, Castellucci P, Campana D, Montini G, Rubello D, et al. Comparison between [68Ga] DOTA-NOC and [18F]DOPA PET for the detection of gastroentero-pancreatic and lung neuro-endocrine tumours. Eur J Nucl Med Mol Imaging 2008;35:1431–8.

23. Haug A, Auernhammer CJ, Wängler B, Tiling R, Schmidt EANM G, Göke B, et al. Intraindividual comparison of [68Ga]DOTA- TATE and [18F]DOPA PET in patients with well-di&erentiated metastatic neuroendocrine tumours. Eur J Nucl Med Mol Imaging 2009;36:765–70.

24. DeVries EFJ, Luurtsema G, Brussermann M, Elsinga PH, Vaalburg W. Fully automated synthesis module for the high yield one-pot preparation of 6-[18F]%uoro-l-DOPA. Appl Radiat Isot 1999;51:389–94.

25. Orlefors H, Sundin A, Lu L, Oberg K, Langstrom B, Eriks- son B, et al. Carbidopa pretreatment improves image interpretation and visualisation of carcinoid tumours with 11C-5-hydroxytryptophan positron emission tomography. Eur J Nucl Med Mol Imaging 2006;33:60–5.

26. Binderup T, Knigge U, Loft A, Federspiel B, Kjaer A. 18F- %uorodeoxyglucose positron emission tomography predicts survival of patients with neuroendocrine tumors. Clin Cancer Res 2010;16:978–85.

109 Chapter 3 – Clinical applications of PET/CT in infection and in%ammation François-Xavier Hanin and François Jamar

Introduction this field and discusses its advantages and 18F-FDG PET and PET/CT now have proven pitfalls compared with WBC scintigraphy. high diagnostic accuracy in disease assessment in oncology, neurology and cardiology. Fever of unknown origin In the field of inflammation and infection, Fever or pyrexia of unknown origin (PUO) nuclear medicine offers several techniques poses a difficult challenge. True PUO is allowing detection and follow-up of disease, defined as a temperature of more than such as labelled white blood cell (WBC) 38.3°C that lasts for 3 weeks or more and scintigraphy, anti-granulocyte monoclonal remains undiagnosed after 3 days in hos- antibody (MoAb) immunoscintigraphy and pital or 1 week of outpatient evaluation 18F-FDG PET and PET/CT. During the past (Petersdorf’s criteria adapted according to decade, PET/CT has become a commonly Durack et al. [1]). There are more than 200 accepted tool in the detection of infection possible diagnoses. They are usually classi- and inflammation, but WBC scintigraphy fied as infections, non-infectious inflamma- remains a more accurate diagnostic proce- tory diseases or tumours, each of these ac- dure in specific cases. This chapter reviews counting for one-third of the final diagnoses the main indications for 18F-FDG PET/CT in (Fig. 1: cancer in PUO). © F. Hanin, F. Jamar Hanin, F. © F.

110 right lower lobe with intenselowerwith lobe right slice CT (C) showing a pulmonary mass in the hilar and mediastinal lymph nodes. C,D: Axial masses. Note the two foci of uptake in the right adrenallarge two in uptake intense showing (B) image PET coronal and (A) image PET/CT for fever of unknown origin. A,B: Coronal fused Figure 1A–D: An 84-year old male was referred by many centres when there is a diagnostic a is there when centres many by preferred remains technique this infection, labelled diagnosingroleof in standing WBCs long- the Considering compare. to hard are sion criteria, results from di&erent publications inclu- in di&erences and heterogeneity wide formed in less than 3 days [3]. Because of this on whether a diagnostic work-up can be per- instance, for depending,another, to hospital one from varies diagnosis *nal a reach to ity abil- the de*nition, the separately.Whatever explored been not have that entities further constitute fever postoperative and patients immunocompromised with associated tions condi- Other infection. (HIV) virus de*ciency immuno- human the with patients in PUO et al.: nosocomial PUO, neutropenic PUO and Durack by de*ned were entities other PUO, classic Besides [2]. obtained *nally is agnosis di- no however,cases, of one-third to up In lymphoma lungcancer mimicking B-cell large a revealed *nally however,Pathology,cancer. lung metastatic of suggestive was masses adrenal large and foci mediastinal and pleural nodule, lung a C on the fusion image (D). The conjunction of conjunction The (D). image fusion the on hapt er 3 – Clinical applications of PET/CT in infection andin"ammation ininfection applicationser 3–Clinical ofPET/CT 18 F-FDG uptake uptake F-FDG 111 out CT gave a of pooled 83% sensitivity (95% out CT for imaging. All but one of those studies have since reported been have studies few a only and using performed were PUO classic with dealing studies initial The patients [4]. of number limited a only in performed been with comparisons and CT for available not are scanning WBC with comparisons direct since data comparative by supported however, not, is preference This infection. of suggestive clue slightly less sensitive in infection and non-in- and infection sensitivein less slightly was and cases) (23 neoplasia of cases in well 18 Overall, accuracy. anatomical improved the number of cases,false-positive rather than to the in reduction a to related mainly was CT with bene*t the Interestingly, respectively. (95% CI: 94–100%) and 86% (95% CI: 75–93%), the pooled sensitivity and speci*city were 98% where studies, four last the in added was CT when improved was This 49–66%). CI: (95% 58% of speci*city pooled a and 73–90%) CI: *ve the of analysis considered helpful in 177 (46%). The statistical and (62%) them of 242 in reached was diagnosis *nal A total. in patients 388 covering PUO, classic for criteria the met CT) with four ing (includ- studies nine [5], meta-analysis cent re- a to According retrospective. been have them of half about and single-centre been F-FDG PET seemed to perform particularly particularly perform to seemed PET F-FDG 18 F-FDG PET was positive in 240 (62%) and 18 F-FDG PET/CT became the standard the became PET/CT F-FDG 18 F-FDG PET studies with- studies PET F-FDG 18 -D PT alone PET F-FDG 18 F-FDG PET have PET F-FDG 18 F-FDG PET/ F-FDG EANM fectious in%ammation. In the only prospective 18F-FDG-avid lymph nodes, starting from the multicentre study published so far, with a very upper part of the torso and then involving rigorous diagnostic algorithm [6], the analysis the lower part of the body [9]. In addition, the of 70 patients, among whom a *nal diagnosis degree of uptake is related to viral load [10] was obtained in 35, showed that 18F-FDG PET and increases in the event of treatment with- had a positive and negative predictive value drawal [2]. In addition, 18F-FDG PET can play of 70% and 92%, respectively, and was useful a role in the management of infectious and in 33% of the entire population. Although the oncological complications of AIDS, such as number of evaluable studies remains limited, opportunistic infections, lymphoma, Kaposi’s there is now evidence that 18F-FDG PET/CT sarcoma and AIDS-related dementia complex should be considered as one of the *rst di- [9]. Nevertheless, the involvement of lymph agnostic tools in patients with PUO in whom nodes related to viral load may complicate conventional diagnostic procedures have not the interpretation of results [11]. been successful. The additional value of combining the 18F-FDG PET study with a whole- Granulomatous and in#ammatory diseases body CT study with iodinated contrast injec- 18F-FDG PET has not been widely used in tution remains to be prospectively evaluated, berculosis. An interesting pilot study [12] de- although retrospective data reported by Ferda veloped a two-time point method to better et al. [7] are convincing. A prospective study characterise tuberculosis involvement in HIV- of 18F-FDG PET combined with full CT scan in negative patients. 18F-FDG PET has emerged intensive care unit patients with fever dem- as a very valuable tool in evaluating sarcoid- onstrated a high diagnostic yield as a *rst-line osis. The whole-body tomographic capability tool [8]. 18F-FDG PET has also proved useful in together with the physical properties of 18F- the other types of PUO although the number FDG PET as compared with 67Ga scintigraphy of reported studies remains small. explains why the latter has been almost completely abandoned in this indication [13]. The In AIDS, there are typical and well-de*ned method is well suited for whole-body evalua- anatomical steps for the appearance of the tion of this very polymorphic disorder (Fig. 2).

112 © F. Hanin, F. Jamar be required in patients with enlarged lymph enlarged with patients in required be will diagnosis pathological a therefore, and, certitude with impossible is lymphoma with thology: in particular, the di&erential diagnosis that mind in keep to however,essential is, It [16]. steroids with especially treatment, of e&ect 15] and seems to be helpful for assessing the [14, sites anatomical of number a in disease It is useful for the diagnosis and staging of the complete disappearance of abnormal uptake repeat the and sarcoidosisthereforefor treated was lymph nodes without malignant cells. Patient which revealed granulomatous in%ammatory biopsy, a to led and nodes lymph hilar mediastinal right in uptake intense of Foci melanoma. for treated previously woman 34-year-old a of images (right) PET coronal and (left) PET/CT fused Coronal 2: Figure C hapt 18 er 3 – Clinical applications of PET/CT in infection andin"ammation ininfection applicationser 3–Clinical ofPET/CT F-FDG PET/CT F-FDG will PET/CT not substitute for pa- 18 F-FDG PET 6 months later showed later months 6 PET F-FDG 113 identify the right andeasiestnodeto biopsy.identify theright nodes. (Fig. 3). patients with GCA and in%ammatory markers (89–100%) were found to be high in untreated speci*city and (77–92%) sensitivity The [17]. and in subsets of studies in patients with PUO studies dedicated in observed been has This ant-cell (GCA)arteritis and Takayasu’s disease. gi- as such arteritides, large-vessel assessing conditions, immune 18 or in%ammatory In large vessels, consistent withvasculitis and aorta the in uptake intense reveals CT C-reactive protein and asthenia. increased loss, weight for referred woman maximal intensity projections of a 59-year-old (right) sagittal and (left) Coronal 3: Figure F-FDG PET has proved very promising for promising very proved has PET F-FDG 18 F-FDG PET/CT may,F-FDG PET/CT however, help to 18 F-FDG PET/

© F. Hanin, F. Jamar F. Hanin, F. © EANM A lower sensitivity was observed under im- often severe and leads to high mortality rates munosuppressive therapy (e.g. steroids). Fur- (20–70%) if diagnosis is delayed [19]. These ther, in GCA, 18F-FDG PET has proved useful infections can appear up to several years after in assessing the extent of the disease and in placement and are challenging to diagnose monitoring the disease activity during ther- early as the clinical course is often indolent. apy. It must be noted that 18F-FDG PET does not allow reliable diagnosis and monitoring The role of WBC scintigraphy has been well of localised temporal arteritis [18]. Use of 18F- documented in the past, showing a sensitiv- FDG PET in other in%ammatory or immune ity from 53% to 100% and a speci*city from conditions (e.g. rheumatoid arthritis) has been 50% to 100% [20]. 18F-FDG PET/CT is a prom- shown to be feasible but its relevance remains ising tool for the early detection of vascular to be determined in daily practice. graft infection by virtue of the combination of morphological and functional imaging in- Infection of vascular grafts formation (Fig. 4). The incidence of vascular graft infection ranges from 0.25% to 6%, but such infection is © F. Hanin, F. Jamar Hanin, F. © F.

Figure 4A–D: A 37-year-old patient with multiple vascular grafts showing multiple septic foci due to Staphylococcus aureus. Coronal (A, B) and axial (C, D) views of PET/CT fused images (A, D) and PET images (B, C) revealing focal uptake of FDG in the right axillary region, allowing accurate localization on the infected vascular graft

114 was described to show better accuracy than accuracy better show to described was gest that gest CT alone [20, 21], while more recent data sug- past, the In uptake. tense in- of localisation correct the for mandatory is imaging PET and CT of fusion addition, In dications. in- speci*c two in value limited very of is but [20]), 90% about of accuracy (overallmyelitis shown good very results for detecting osteo- or either (using scintigraphy WBC infection Bone infection. graft vascular for value any has not shown predictive of uptake focal analysis To infection. semi-quantitative date, without body, foreign a around %ammation in- to related be to likely is graft the along uptake and regular smooth di&use, contrast, In [23]. infection for criteria predictive were CT on boundary graft irregular an and take up- FDG irregular an that showed patients 96 in al. et Spacek by study interesting An body. foreign a to reaction in%ammatory from graft infected discriminate better to quantitative, as well as qualitative criteria, de*ne to authors some led has graft fected non-in- on FDG of uptake physiological The or thrombosis. bleeding haematomas, as especially period, has been described in the early postoperative uptake positive False [22]. accuracy global therefore and speci*city in increase an in C hapt 111 In-oxine [or tropolonate] labelling) has labelling) tropolonate] [or In-oxine er 3 – Clinical applications of PET/CT in infection andin"ammation ininfection applicationser 3–Clinical ofPET/CT 18 F-FDG PET/CT furthermore results furthermore PET/CT F-FDG 18 F-FDG PET alone PET F-FDG 99m Tc-HMP AO 115 First, it is poorly e&ective for detecting spinal detecting fore&ective poorly is First,it curacy, ranging from 53% to 76%, for chronic ac- poor shows scintigraphy WBC Secondly, show [24]. asimilarpattern can infarction and disease Paget’stumours, ity of this photopenic picture is very low since the infection and vascular spasm. The speci*c- cyte migration related to an encapsulation of leuco- of absence the to owing scintigraphy WBC on lesions photopenic as appear cases spondylodiscitis. Typically, these pathological and/or osteomyelitis vertebral i.e. infection, osteomyelitis andspondylodiscitis(Fig. 5). vertebral in reported been have 87–100% of while a sensitivity of 94–100% and a speci*city [25], reported been have 99% to 75% from ranging speci*city a and 98% of sensitivity a osteomyelitis, acute for *eld: this in interest raphy, As a result of these limitations of WBC scintig- (21% and60%,respectively) [26]. for chronic osteomyelitis of the axial skeleton than respectively) 80%, and imaging(84% tis osteomyeli- chronic peripheral for speci*city and sensitivity better shows it addition, In [25]. recurrence suspected or osteomyelitis 18 F-FDG PET/CT F-FDG has PET/CT attracted increasing

Figure 5: A 75-year-old patient hospitalized for back pain. Sagittal (A, B) and axial (C, D) views of fused PET/CT images (A, D) and PET images (B, C) revealed intense uptake in the lumbar spine, further diagnosed as Mycobacterium avium-related spondylodiscitis

This excellent accuracy of 18F-FDG PET/CT has in spondylodiscitis; in this latter indication, a major impact on the clinical management 18F-FDG PET/CT is especially to be preferred of patients with (disco)-vertebral infection, by when MRI is hampered by the presence of modifying treatment in 52% of cases [27]. A metallic hardware. meta-analysis of chronic osteomyelitis imaging in 2005 by Termaat et al. reported aggre- Orthopaedic prosthesis infection: hip and gated sensitivity and speci*city of 96% and knee 91%, respectively [26]. Hip and knee joint replacement is subject to two major potential complications: infec- In conclusion, WBC scanning remains the *rst- tion and loosening. In both cases, the clinical line technique for acute infection in peripheral symptoms will usually be limited to a painful osteomyelitis, whereas 18F-FDG PET/CT has arthroplasty. 18F-FDG PET is a relatively new become the preferred tool, along with MRI, tool in these indications, while three-phase

116 be related to the exudation of immunologi- of exudation the to related be can prosthesis a around FDG of uptake The Patterns andquantitative approach the traditionalinvestigative tools. are scintigraphy WBC and (TPBS) scan bone arthroplasty arthroplasty using qualitative criteria are 85%, of racy accu- and speci*city sensitivity, overall The and non-pathological arthroplasties. a signi*cant di&erence between pathological infectionnoted,and was therealthough was di&erence in SUV aseptic between loosening a group of 92 arthroplasties [29], no signi*cant ue as (SUV) a potential discriminative factor. In titative analysis of the standardised uptake val- The second approach is the use of semi-quan- periprosthetic soft tissues (pattern 5). from no uptake (pattern 1) to FDG uptake in the severalpatternsuptake,withthe tribution of dis-reliesessentiallyscalethethis [29]:on al. A scale of uptake was proposed by Reinartz et correlationtheclinicalingthe outcome.with of FDG uptake around a hip prosthesis, evaluat- A *rst approach is to analyse the characteristics causes ofpainfularthroplasty. discriminate better to order in sought been Therefore, criteria allowing di&erentiation have [28]. well as uptake FDG induce to likely are loosening, metallic and polyethylene particles aseptic of case the in but [25], macrophages and leucocytes especially cells, active cally C hapt er 3 – Clinical applications of PET/CT in infection andin"ammation ininfection applicationser 3–Clinical ofPET/CT 18 F-FDG PET in patients with a painful a with patients in PET F-FDG 117 series 18F-FDG PET shows a higher accuracy accuracy higher a shows PET 18F-FDG series and 81% forAlthough from knee). combined hip for (80% TPBS or knee) for 83% and hip for (89% PET either than knee) for 84% and hip for (91% accuracies higher showing [20], infection prosthetic of detection for nique tech- best the remains scintigraphy WBC Role of chanical looseningfrom infection [30]. me- toimproveof discrimination mandatory are criteria precise more Nevertheless, [29]. respectively,83%, and 75% 98%, for knee the and hip, the for respectively, 89%, and 90% tive asitisaccessible, fastandnon-invasive. alterna- PET,valuable or a remainsning TPBS where there is little or no access to WBC scan- departments in that undisputed is it Finally, of value added the on tool. To date, there are no data in the literature patients with painful arthroplasty as a *rst-line 18 loosening.from infection discriminate better more speci*c criteria need to be developed to ing complications of joint replacement, while diagnos- for e&ective highly considered be time-consuming). Therefore, (while WBC scintigraphy is more complex and procedure one-day a is it that fact the and TPBS than accuracy better a are advantages sible technique compared with TPBS. Its main acces- less and costly a remains it loosening, and infection prosthetic of diagnosis the in pared with F-FDG PET should not be used routinely in routinely used be not should PET F-FDG 18 F-FDG PETvsWBCandTPBS 18 F-FDG PET alone. F-FDG PET 18 F-FDG PET/CT com- PET/CT F-FDG 18 F-FDG PET F-FDG can PET EANM The infected diabetic foot uninfected Charcot’s neuroarthropathy [32]. 18F- The clinical diagnosis, treatment and follow- FDG PET/CT was initially found in a series of 14 up of diabetic foot infections remain di+cult patients to be accurate in assessing diabetic owing to several diabetic-speci*c causes: neu- foot [33]. Others, however, have found 18F-FDG ropathy, higher susceptibility to infection and PET/CT to have a much lower accuracy. In ad- vascular impairments. It is often of major clinical dition, whether combination with CT has an importance to distinguish between soft tissue added diagnostic value remains undetermined. and underlying bone infection since osteomy- Therefore, 18F-FDG PET/CT cannot fully replace elitis leads to a more aggressive treatment. WBC scintigraphy, especially when the latter is acquired in combination with bone marrow As stated in the section on bone infection, scintigraphy and using SPECT-CT [20]. WBC scintigraphy is the modality of choice in the diagnosis of peripheral infection, the Technical issues1 reported sensitivity and speci*city being be- t The clinical indication must be reviewed tween 72% and 100% and between 67% and with care in order to choose the correct 98%, respectively [31]. The main disadvantage imaging protocol of WBC scintigraphy is the poor image quality, which leads to di+culties in assessing whether t Whole-body imaging is necessary in PUO abnormal activity in the foot is attributable to a but a limited *eld of view can be su+cient bone or a soft tissue infection. More recently it in answering speci*c questions has been possible to address this pitfall by using SPECT/CT detection cameras, which allow t Diabetic patients must be checked care- better localisation of the hotspots. fully for hyperglycaemia, although this may not preclude valid imaging in infection and Recent data, however, show high sensitivity, in%ammation as it does in tumours [34] speci*city and accuracy (81%, 93% and 90%, respectively) for 18F-FDG PET. Interestingly, com- t Careful examination of possible pregnancy parisons of these values with MRI in the same is always necessary indication have shown MRI to have a better sensitivity (91%) but a lower speci*city (78%) and a t Interpretation should consider potential global accuracy of 81%. The authors concluded false positive results related to physiologi- that the two techniques are complementary cal or tumour uptake in that the high sensitivity of MRI is combined 18 with the higher speci*city of F-FDG PET, which 1 Detailed EANM guidelines will be available shortly better discriminates between osteomyelitis and and will cover all issues

118 3. Habibzadeh F, Yadollahie M. Time for a change in the in change a for F,Time Habibzadeh M. Yadollahie 3. 2008;52:17-29. Imaging Mol Q JNuclMed of infectious diseases using [18F] %uorodeoxyglucose PET. Bleeker-Rovers CP, Vos FJ, Corstens FH, 2. Oyen WJ. Imaging mined and rede*ned. Curr Clin Top Infect Dis 1991;11:35-51. Mol Imaging 2010;54:698-703. Imaging Mol losis patients: implications for cancer staging. Q J Nucl Med Wiele C. FDG uptake in lymph-nodes of HIV+ and tubercu- Sathekge M, 11. Maes A, Kgomo M, Pottel H, Stolz A, Van De 2010;31:137-40. Commun Med Nucl count. cell CD4 to relatedinversely is patients HIV of nodes lymph by uptake rodeoxyglucose Fluo- C. Wiele de VanM, Kgomo A, Maes M, Sathekge 10. 2009;36:1176-84. Imaging infection. Mol Eur JNuclMed tron emission tomography in Posi-patients su&eringC. Wielefrom de HIV-1 van A, Maes I, Goethals M, Sathekge 9. infection. Care Intensive suspected 2010;36:504-11. Med tomography combined with CT in critically ill patients with der Hoeven JG. F-18-%uorodeoxyglucose positron emission P, Pickkers KS, Simons CP,Bleeker-Rovers Oyen WJ,van 8. 2010;73:518-25. integrated full diagnostic isotropic CT imaging. Eur J Radiol Fever of unknown origin: a value of with (18)F-FDG-PET/CT Ferda J, Ferdova E, Zahlava J, Matejovic M, Kreuzberg B. 7. 2007;34:694-703. Imaging Mol origin. Eur JNuclMed tured diagnostic protocol in patients with fever of unknown multi-centre study of the value of FDG-PET as part of a struc- Bleeker-Rovers CP, Vos FJ, Mudde AH, 6. et al. A prospective 3.[Epubaheadofprint] 2010;Dec Radiol J Eur origin. unknown feverof of evaluation the in PET-CT %uorodeoxyglucose-PET/ of value the of meta-analysis A GJ. ZF,SY,YangZhou GL, Liu Wang K, Zhao MJ, Dong 5. 2004;31:622-6. Imaging Mol Eur JNuclMed value of 18F-FDG PET and scintigraphy.111In-granulocyte diagnostic of comparison prospective origin: unknown of Fever L. Hojgaard A, Eigtved AM, Lebech A, Kjaer 4. de*nition of fever of unknown origin. J Infect 2008;57:166-7. 1. Durack DT, Street AC. Fever of unknown origin--reexa-DT, Durack unknown FeverAC.Streetof 1. References References Chapter 3 119 14. Bonardel G, Carmoi T, Gontier E, et al. [Use of positron of [Use al. et E, Carmoi T, Gontier G, Bonardel 14. patients withsarcoidosis. 2006;47:1571-6. JNucl Med in scintigraphy 67Ga and PET 18F-FDG of evaluation tive Nishiyama Y,13. Yamamoto Y, Fukunaga et al. Compara-K, treatment. Apilotstudy. 2010;49:35-40. Nuklearmedizin de Wiele C. Impact of FDG PET on the management of TBC H, VanPottel A, Stoltz M, Kgomo A, Maes M, Sathekge 12. 16. Keijsers Keijsers RG, 16. Verzijlbergen FJ, Oyen WJ, et al. 18F-FDG PET, sarcoidosis. Chest2007;132:1949-53. with patients 137 in tomographyscans emission positron %uorodeoxyglucose whole-body 188 of Results MC. nuzzi Teirstein15. J, O,Almeida Machac AS, Lu P, Padilla Ian- ML, 2011;32:101-8. Interne Med Rev sarcoidosis]. in tomography emission Eur J Nucl Med Mol Imaging 2009;36:850-8. Imaging Mol Eur JNuclMed using 18F-FDG PET/CT: our experience with 96 prostheses. P.of infection “non-acute”Diagnosticsprosthesis vascular Spacek M, 23. Belohlavek O, Votrubova J, Sebesta P, Stadler 2009;53:35-40. Imaging vascular grafts. Mol QJNuclMed Keidar Z, Nitecki 22. S. FDG-PET for the detection of infected 2010;40:348-54. Surg Endovasc Vasc J Eur infection. graft prosthetic lar vascu- of work-up diagnostic the in FDG-PET-CT of racy Accu- al. et BR, AW,Saleem Glaudemans JL, Bruggink 21. 2010;37:1986-91. the imaging method of choice? Eur J Nucl Med Mol Imaging infections: in AW, Glaudemans FDG-PET/CT 20. SignoreA. 2009;39:103-14. NuclMed tion andin%ammation.Semin Ben-Haim S, Gacinovic S, Israel O.19. Cardiovascular infec- Spine2010;doi:10.1016/j.jbspin.2010.09.013Bone Aortic involvement in giant cell arteritis: Current data. Joint D. Wendling A, Lohse JC, Balblanc C, Prati M, Bossert 18. 2009;53:51-63. Imaging Mol portance of diagnosing large vessel vasculitis. Q J Nucl Med in patients FDG-PET with fever origin:of the unknown im- B. Meller T,Liersch O, Gurocak CO, Sahlmann J, Meller 17. sarcoidosis. 2009;36:1131-7. Imaging Mol Eur JNuclMed diagnosed newly in sIL-2R and ACE genotype-corrected

EANM 24. Palestro CJ, Love C, Bhargava KK. Labeled leukocyte ima- 30. van der Bruggen W, Bleeker-Rovers CP, Boerman OC, ging: current status and future directions. Q J Nucl Med Mol Gotthardt M, Oyen WJ. PET and SPECT in osteomyelitis and Imaging 2009;53:105-23. prosthetic bone and joint infections: a systematic review. Semin Nucl Med 2010;40:3-15. 25. Stumpe KD, Strobel K. 18F FDG-PET imaging in musculoskeletal infection. Q J Nucl Med Mol Imaging 31. Palestro CJ, Love C. Nuclear medicine and diabetic foot 2006;50:131-42. infections. Semin Nucl Med 2009;39:52-65.

26. Termaat MF, Raijmakers PG, Scholten HJ, Bakker FC, Patka 32. Nawaz A, Torigian DA, Siegelman ES, Basu S, Chryssikos T, P, Haarman HJ. The accuracy of diagnostic imaging for the Alavi A. Diagnostic performance of FDG-PET, MRI, and plain assessment of chronic osteomyelitis: a systematic review *lm radiography (PFR) for the diagnosis of osteomyelitis in and meta-analysis. J Bone Joint Surg Am 2005;87:2464-71. the diabetic foot. Mol Imaging Biol 2010;12:335-42.

27. Ito K, Kubota K, Morooka M, Hasuo K, Kuroki H, Mimori 33. Keidar Z, Militianu D, Melamed E, Bar-Shalom R, Israel A. Clinical impact of (18)F-FDG PET/CT on the management O. The diabetic foot: initial experience with 18F-FDG PET/ and diagnosis of infectious spondylitis. Nucl Med Commun CT. J Nucl Med 2005;46:444-9. 2010;31:691-8. 34. Rabkin Z, Israel O, Keidar Z. Do hyperglycemia and dia- 28. DeHeer DH, Engels JA, DeVries AS, Knapp RH, Beebe JD. betes a&ect the incidence of false-negative 18F-FDG PET/ In situ complement activation by polyethylene wear debris. CT studies in patients evaluated for infection or in%am- J Biomed Mater Res 2001;54:12-9. mation and cancer? A comparative analysis. J Nucl Med 2010;51:1015-20. 29. Reinartz P, Mumme T, Hermanns B, et al. Radionuclide imaging of the painful hip arthroplasty: positron-emission tomography versus triple-phase bone scanning. J Bone Joint Surg Br 2005;87:465-70.

120 nacd u ptohsooia under- pathophysiological our enhanced greatly has and conditions, cardiac of variety a of assessment the for tool non-invasive a as interest great generated has (PET) raphy the accrued number of studies document- studies of number accrued the [e.g. rubidium-82 ( compounds radiolabelled novel of duction intro- and development The services. nostic diag- cardiac among acceptance gaining is PET Nonetheless, imaging. PET of costs high the limited availability of PET scanners and the tracers, perfusion most of production the for gands, the need for an on-site cyclotron facility physical half-lives of positron-emitting radioli- short and production of complexity the by constrained been has services diagnostic cal spread dissemination of cardiac PET into clini- Despite its undisputed advantages, the wide- [1,2]. heart the of utilisation substrate and health bolic meta- and (MBF) %ow blood myocardial of regulation the into insights groundbreaking enabled have radiotracers for sensitivity tion ventional nuclear methods and its high detec- and temporal resolution compared with con- spatial superior Its kinetics. tracer of models using dynamic acquisition and mathematical PET allows quanti*cation of radiotracer uptake unique physical and technical characteristics, toits cardiac of pathologies.standing Owing Since its *rst days, *rst its Since Introduction Chapter 4 – Clinical applications of PET/CT in cardio Oliver Gaemperli 82 positron emission tomog- Rb) Rb) or

and Philipp A.K 18 F-BMS-747158-02], aufmann 121 cardiac applica Technical relevant ofPET to characteristics cular research [3]. andclinicalpractice cardiovas- in PET of use the promote further will and applications future for avenues new ners into hybrid scanners which have opened scan- CT high-end with PET of combination the include developments Recent imaging. have contributed to a steady growth in cardiac oncology in PET of success clinical great the by fosteredmainly devices PET of availability ever-increasing the and use clinical their ing to itsincreasing useinnuclearcardiology. technology and have signi*cantly contributed portant milestones in the development of PET than two-dimensional acquisitions [5] are im- rather three- of use increasing the as well as systems with reduced detection dead-time [4] detector and oxyorthosilicate) lutetium (e.g. recent introduction of new detector materials The movement. wall cardiac and respiratory by degraded slightly is resolution imaging cardiac in although [4], (SPECT) tomography computed emission single-photon for mm 10–12 with compared view of *eld the of centre the in mm 4 reaches devices PET rent detector design, the spatial resolution of cur- PET and detection coincidence to Owing applications. cardiac for technique accurate highly a as operate to PET allow instrumental properties and technical of number A tions logy EANM Despite the high photon energy released Radiopharmaceuticals in annihilation of positrons (511 keV), PET is The most common radiolabelled compounds more susceptible to photon attenuation than available for cardiac PET applications are sum- SPECT, where photon energies from cardiac marised in Table 1. All compounds listed in applications range from 70 to 140 keV. The Table 1 [except for 18F-%uorodeoxyglucose reason for this seemingly paradoxical *nd- (18F-FDG)] are used for myocardial perfusion ing is that coincidence registration with PET imaging, the most common imaging ap- requires two photons to travel through the plication in nuclear cardiology. Radiokinetic whole cross-section of the body before a true properties of these compounds should ideally event is recorded. This, however, makes pho- ful*l a number of requirements to allow their ton attenuation more predictable than with use as perfusion tracers (Table 2). It should SPECT as it is uniform along a given line of be noted, however, that all currently available response between two detectors and inde- perfusion tracers listed in Table 1 have rela- pendent from the origin of the positron along tive advantages and disadvantages: none of this line, and therefore allows for more robust them meets all the radiokinetic requirements means of attenuation correction. As a result, to be considered the ideal perfusion tracer. attenuation correction is now widely accepted However, promising new agents are currently for PET imaging and has proven indispensable in preclinical and clinical evaluation with the for quantitative assessment of MBF [6]. potential to overcome most of the limitations of currently used compounds [8]. Image acquisition in list mode is widely available for clinical use and allows images to be Radiolabelled water reconstructed retrospectively using di&erent Radiolabelled water (15O-H O) is a metaboli- 2 rebinning criteria [7]. This provides high %ex- cally inert and freely di&usible perfusion tracer. ibility in data processing and reconstruction Consequently, myocardial uptake is linearly of images using ECG or respiratory gating, dy- related to MBF, which allows accurate quan- namic image frames for quantitative analysis ti*cation of MBF over a wide range of values or simple addition of information into static (Fig. 1) [9–11]. images. In cardiology, this approach facilitates quanti*cation of blood %ow by applying dynamic tracer kinetic modelling or assessment of left ventricular volumes and function from ECG-gated data.

122 Adapted from Knnuti J et al. J Nucl Cardiol 2009; 16:497 For research purposes, however, this tracer tracer this however, purposes, research For 15 of use ample the limited have features netic radioki- These [12,13]. ventricle left the of images anatomical create and pool blood and myocardial separate to processing cal with MBF of assessment Quantitative images. perfusion appropriate generate to contrast myocardial enough without pool, blood and myocardium between equilibrium an reaches rapidly it compartment, myocardial However, since 99m include to plateau at increasing %ow values. andSPECTperfusion tracers) extraction tends extraction and MBF; with all other tracers (PET a quasi-linear relationship between myocardial of MBF. myocardial radiotracer extraction as a function Figure 1: Simpli*ed schematic representation of C hapt O-H Tc-tetrofosmin er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT 15 2 O for clinical PET perfusion imaging. imaging. perfusion PET clinical for O O-H 15 O-H 99m 2 Tc-methoxyisobutylisonitrile and O requires complex mathemati- complex requires O 2 O is the only perfusion tracer with 15 O-H 2 O is not retained O in the is not retained 99m Tc tracers 123 the intracellular compartment after metabolic internalisation, Upon [14–16]. 1) (Fig. rates %ow high very at cept is linear overuptake a wide range of MBF, ex- myocardial and 80% to up of rate extraction 13 13 session. same the in conditions biological di&erent under MBF of measurements repeated allows min) the of half-life physical short the very Indeed, popularity. wide has gained perfusion defect reversibleof the laterobasal partially wall a show (B) maps polar perfusionand (A)imagesperfusion rest and occluded left circum%ex artery. The paired stress perfusionPET in a 77-year-old patient with an stress/restof Example 2A,B: Figure ing highmyocardial contrastimages(Fig. 2). conversion into N-ammonia ( N-ammonia N-ammonia 13 13 N-NH3) has a high *rst-pass high a has N-NH3) N-glutamine, N-glutamine, thereby creat- 13 N-NH3 is trapped in trapped is N-NH3 15 O isotope (2 (2 isotope O 13

N-NH3 University Hospital Zurich, Switzerland Zurich, Hospital University Courtesy of the Cardiac Imaging Departement, Departement, Imaging Cardiac the of Courtesy EANM However, its short physical half-life (10 min) Figure 3A: Dynamic image acquisition (18 necessitates an on-site cyclotron for produc- frames) after 13N-NH3 injection showing tion and immediate administration. A similar sequential activity in the right ventricle [dark performance of 13N-NH3 compared to 15O-H O blue regions of interest (ROI), frames 2–4], the 2 for quanti*cation of MBF has been demon- left ventricle (light blue ROI, frame 5) and the strated in experimental animal [10] and human left ventricular myocardium (white ROI, frames studies [17]. Its ability to provide high-quality 6–18). perfusion images and allow quanti*cation of Figure 3B: Time-activity curves for the right regional MBF simultaneously have advanced ventricular cavity (RV), left ventricular cavity this compound as an excellent clinical per- (LV) and left ventricular myocardial segments fusion tracer in diagnostic centres with an (myocardium). on-site cyclotron (Fig. 3), and an increasing Figure 3C: Segmental values for coronary %ow number of publications underline its clinical reserve (CFR) calculated from time-activity diagnostic and prognostic value [18–21]. curves after pixelwise tracer kinetic modelling using a three-compartment model [14]

Rubidium-82 Rubidium-82 (82Rb) is an analogue of the potassium cation and undergoes cellular uptake by the Na/K-ATPase. Unlike other positron-emitting compounds which are cyclotron products, 82Rb can be obtained from an 82Sr/82Rb generator by decay from 82Sr attached to an elution column (Table 1). Downsides of 82Rb are the lower myocardial extraction rate, its non-linear relationship between myocardial tracer uptake and MBF (Fig. 1) [22,23], the high positron energy (which allows the positron to travel for several millimetres before annihilation and thereby lowers spatial resolution) (Table 1) and the ultrashort half-life (leading to detector oversaturation at

Courtesy of the Cardiac ImagingCourtesy Departement, of the Cardiac University Hospital Zurich, Switzerland the beginning of the acquisition (particularly with three-dimensional acquisition protocols) and low count statistics and increased image

124 The relative drawbacks ofavailable perfusion 18 the cell[24]. of state metabolic prevailing the with action inter- potential a regarding concerns raises Na/K-ATPase energy-dependent the by take uptake, a surrogate for myocardial viability. myocardial for surrogate a uptake, glucose myocardial assess to used clinically myocyte. the in trapped remains therefore and metabolisation, further dergo 6-phosphate, but unlike glucose does not un- internalisation, FDG is phosphorylated to FDG- (GLUT-1GLUT-4). and Upon transporter cose cytes via facilitated di&usion through the glu- cal analogue of glucose – is taken up by myo- 18 18 cyclotron. on-site an without centres in imaging PET perfusion allows half-life long the ditionally, tracer extraction at variable %ow rates [8]. Ad- myocardialstable longed high retentionand pro- data, biodistribution excellent indicates studies animal in evidence available The ed. phase II and III clinical trials are eagerly await- current of results the and PET, with imaging perfusion myocardial for compound novel promising a as introduced been recently has The limitations. mentioned previously the come over- to ability the with compounds novel for search intensive an fuelled have tracers noise in later phases. Additionally, cellular up- C hapt F-%uorodeoxyglucose( F-labelled BMS-747158-02 F-"uorodeoxyglucose 18 er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT F-labelled compound BMS-747158-02 compound F-labelled 18 F-FDG) – a biologi- a – F-FDG) 18 F-FDG is F-FDG 125 scars pathways (i.e. anaerobic glycolysis), whereas in FDG uptake due to upregulation of anaerobic dium characteristically shows an increased Chronically underperfused but viable myocar- preparation includes obtaining the patient’s in- tise in advanced life support techniques. Patient able pharmacological stress agents and exper- quali*ed physician with knowledge of the avail- All stress procedures should be supervised by a is preferred. dobutamine) or vasodilators (with stress cal pharmacologi- Therefore,acquisition. image the temporal proximity of tracer injection and to owing perform to di+cult are procedures Inherently with PET, however, physical exercise of choice in patients who are able to exercise. or treadmill) is generally the stress procedure perfusion imaging. Dynamic exercise (bicycle stress techniques are available for myocardial Several rest. at present be not may that sion perfu- myocardial in changes elicit to stress cardiac of state a during images entails acquiring study perfusion myocardial full A lifethe of half- physical long rather The [25,26]. absent Myocardial perfusion PET Myocardial perfusion Patient preparation andprotocols an on-site 1). (Table cyclotron forneed basis,the obviating day-to-day a on centresnearby PET to distributed and facility cyclotron a in produced centrally be to FDG 18 F-FDG uptake is markedly reduced or reduced markedly is uptake F-FDG 18 F-nuclide (110 min) allowsmin) for(110 F-nuclide 18 18 F- EANMF- formed consent and a clinical history covering direct near-maximal myocardial hyperaemia, indications for the test, symptoms, risk factors, thereby uncoupling MBF from myocardial medication and prior diagnostic or therapeutic oxygen demand. In myocardium served by a procedures, and *nally establishing a secure stenosed coronary artery, hyperaemic blood intravenous line for radiopharmaceutical and %ow is reduced; this leads to heterogeneities stress agent administration. Haemodynamic of perfusion during vasodilation, demarcating non-invasive monitoring (including 12-lead ischaemic myocardial territories. ECG, heart rate and blood pressure) should be ensured at all times during the stress procedure. Dosing schemes for adenosine and dipyri- Cardiac medications which could interfere with damole are given in Table 3. Owing to its the pharmacological stress agent should be non-selective action on all adenosine recep- withheld if possible. This includes nitrates for tor subtypes (A1, A2A, A2B, A3), vasodilators at least 24 h and beta-blockers and calcium are contraindicated in patients with a history antagonists for at least 48 h prior to the test. If of bronchospasm, high-grade AV block of sick a vasodilator stress agent (adenosine or dipyri- sinus syndrome, symptomatic aortic stenosis damole) is used, ca&eine-containing beverages and hypertrophic obstructive cardiomyopa- (co&ee, tea, cola etc.), foods (chocolate) and thy, and hypotension [27]. Newer agents with medications (some pain relievers, stimulants selectivity for A2A receptors are under clinical and weight-control drugs) and methylxan- and preclinical evaluation. Regadenoson is the thine-containing medications should be dis- *rst selective A2A agonist approved for use in continued for at least 12 h prior to the test [2]. the United States and has recently received market approval in Europe in 2010. It has a Vasodilator stress (adenosine, dipyridamole). longer half-life than adenosine, can be admin- Adenosine mediates direct coronary arterio- istered as a bolus and is better tolerated than lar dilation via the A2A adenosine receptor adenosine or dipyridamole [28]. subtype, and in normal coronary arteries results in a three- to fourfold increase in MBF. Dobutamine. Dipyridamole is an indirect coronary arteriolar Dobutamine is a sympathomimetic agent dilator that increases the tissue levels of ad- with high a+nity to adrenergic β1-receptors. enosine through inhibition of the intracellu- It causes secondary coronary vasodilation lar reuptake and deamination of adenosine. through increased myocardial oxygen demand Adenosine and dipyridamole tend to induce as a result of β1-mediated dose-dependent in- a modest increase in heart rate and a mod- creases in heart rate, blood pressure and myo- est decrease in both systolic and diastolic cardial contractility. Dobutamine is adminis- blood pressure. Thus, both agents induce tered as a continuous intravenous infusion that

126 uniform throughout the left ventricle, as the as ventricle, left the throughout uniform (and hence (and Under fasting conditions, myocardial glucose 18 and uncontrolled cardiac arrhythmias. severehypertension,dissectionaortic acute include acute coronary syndrome, uncontrolled 3).Absolute contraindications todobutamine account of its relative convenience. Oral ad- Oral convenience. relative its of on account uptake glucose myocardial dardising stan- of method common most the is This Oral glucose loading. areuptake discussedbelow: glucose myocardial standardise to methods common most three The utilisation. glucose myocardialof interpretableimages obtain to order in environment metabolic the dardise stan- to thereforeis necessary It [29]. acids ty prevailing myocardial energy source is free fat- the precludes performing 13 ever, with study is performed if the perfusion on the logistics of each individual centre; how- depends study perfusion the after or before 12 h. Whether the least at for discontinued be should smoking fasted overnight (at least 6h) and alcohol be and should patient The SPECT). or (PET ies stud- perfusion with conjunction in formed per- often are studies viability Myocardial u istitrated according to the patient’s heart rate C hapt p F-FDG PET N-NH3 PET, the longer half-life of half-life longer the PET, N-NH3 to a maximaltoa dose of40μg/kg/min (Table er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT 18 F-FDG) uptake is low and non- and low is uptake F-FDG) 18 F-FDG study is performed 18 F-FDG study*rst. 18 F-FDG 127 intravenous administration of short-acting in- oral glucose loading protocol with additional substantial proportion of patients. A modi*ed a in quality image reduced for account may which failure, heart congestive with patients and subjects diabetic 2 type in common is handling glucose abnormal therefore and resistance insulin myocardial Increased [30]. optimal image quality in up to 25% of patients However, take. tosub- maylead method this increases myocardial glucose and the release of pancreatic insulin, which in turn ministration of 25–100 g of glucose results in reducing circulating free fatty acid concen- acid fatty free circulating reducing thereby lipolysis, peripheral of inhibitors are acipimox as such derivatives acid Nicotinic Nicotinic acidderivatives. reader isreferred to othersources [1,30,31]. interested the protocol, this of description For moreexpertise. detailed a particular with institutes of number a for reserved therefore col is cumbersome and and time-consuming 2 diabetic patients. However, the clamp proto- and yields images of high quality even in type subject individual each of resistance insulin glucose uptake. This protocol accounts for the in order to achieve a steady state of myocardial taneous administration of glucose and insulin consists of an intravenous protocol for simul- clamp euglycaemic hyperinsulinaemic The euglycaemicHyperinsulinaemic clamp. in patientswithinsulinresistance [1]. maysulin improve quality,image particularly 18 F-FDG up- EANM trations. They are e&ective for improving Detector corrections 18F-FDG image quality, and when combined Detector normalisation and correction for with subcutaneous or intravenous injection dead-time loss are crystal-speci*c calibrations of small amounts of short-acting insulin, performed prior to imaging. Detector normali- good image quality can also be obtained sation re%ects the individual sensitivity of each in diabetic subjects. With the exception of scintillation crystal in the detector array and %ushing, no side-e&ects of acipimox have ensures uniform detector sensitivity. Detector been observed. dead-time loss is the fraction of counts lost to pulse pile-up and can be assessed by the Image processing and reconstruction count rate characteristics of the scanner. Ap- In cardiovascular applications, PET needs to plying dead-time loss factors ensures a linear achieve a high level of quantitative accuracy relationship between injected activity and to allow for a reliable comparison of myocar- actual count rate. dial perfusion in all left ventricular territories. This is accomplished by applying a number of Randoms and scatter correction corrections to the raw image datasets (Fig. 4): Correction for random events recorded by PET detectors can be achieved by direct measurement using the delayed coincidence technique. This technique is based on the principle that events occurring at time intervals much longer than the detector response time are due solely to random coincidences and are therefore subtracted from the number of prompt events along a given line of response. Scatter correction can be implemented by applying a model of contribution of scattered counts based on the theory underlying Compton scatter of gamma rays. Attenuation maps can be used to provide an estimate of the expected scatter in a given subject. Major factors af- fecting the amount of scatter are the energy windows of acceptance set according to the Adapted from DiCarli MF and Lipton MJ (eds). Cardiac PET Cardiac MJ (eds). DiCarli and MF and Lipton from Adapted PET/CTYork. 1st edition. Springer Imaging. Science+Business Media, New energy of the radiotracer and the energy reso- Figure 4: Flowchart for PET image processing lution of the detector and whether axial septa and reconstruction are used to block axially oblique scatter events.

128 of PET is that attenuation is constant along a along constant is attenuation that is PET of property interesting An 2%. as low as be can escape the body without absorption or scatter that photons of fraction the subjects, heavy In counts. of degradation signi*cant in result As mentioned above, photon attenuation can hood expectation-maximisation (MLEM) and (MLEM) expectation-maximisation hood maximum-likeli- (FBP), tered back-projection *l- include algorithms reconstruction Image Image reconstruction algorithms artefacts. misregistration avoid to scans emission and be given to correct alignment of transmission should attention Careful [6]. correction tion attenua- reproducible and reliable provide to proven been has and scans transmission source external time-consuming often the than faster considerably being of advantage data using the intrinsic CT system. This has the slow helical scan CT to generate transmission es are hybrid scanners PET/CT which employ a However, nowadays, the majority of PET devic- ( caesium-137 germanium-68 ( as such sources transmission with equipped maps. systemsA number of areavailable PET source which is used to compute attenuation gamma-ray external an with body the of *le tice by performing a transmission shadow pro- This characteristic feature is exploited in prac- even if the source is placed outside the body. along this line positron annihilation occurred, where matter no coincidence, of line given Attenuation correction C hapt er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT 137 68 Cs, 662 keV) point sources. point keV) 662 Cs, Ge, 511 keV) rod sources or Ge, 511 keV) 129 coronary angiography as the standard or ref-angiographystandardor the as coronary invasive with performance diagnostic high a dynamic cardiac PET imaging using tracer tracer using imaging PET cardiac dynamic is low. However, for density or count present is activity extracardiac which in situations in particularly a choice, are better usually rithms algo- these available, are techniques OSEM If images. quality best the produce not may and artefacts to prone is it however, time; computational little requires and step) tion backprojec- single a involves only (it exact is FBP protocols. most in su+cient often is density count myocardial since imaging PET cardiac for well performs often FBP (OSEM). expectation-maximisation ordered-subsets showed a higher diagnostic accuracy of PET, of accuracy diagnostic higher a showed head comparisons of PET versus SPECT which head-to- small three in con*rmed was This favourably with data published for SPECT [32]. compare and respectively) 89%, and 90% of speci*city and sensitivity (weighted erence with studies PET able Avail- (CAD). disease artery coronary of tion detec- the for accuracy diagnostic superior a have to considered is PET technique, SPECT available widely more the with Compared cor valueDiagnostic andriskstrati#cation in Clinical applications of cardiac PET imaging poorer. be can quality image though even nature, exact its of because algorithm best the considered is FBP MBF, estimate to modelling kinetic onary artery disease artery onary 13 N-NH3 or N-NH3 82 Rb report report Rb EANM which had an overall sensitivity and speci*city crease at diameter stenoses greater than 40% of 93% and 82%, respectively, versus 85% and and is virtually nulli*ed with stenoses greater 67%, respectively, for SPECT [33–35]. than 80% [41]. There are insu+cient data to support an added diagnostic value of quan- Besides its diagnostic value for the detection titative MBF measurements over perfusion of CAD, perfusion imaging provides relevant imaging alone [42]. However, the potential data for patient risk strati*cation and predic- clinical utility of measuring MBF is supported tion of cardiac events [36,37]. Compared with by a number of prognostic studies in patients SPECT, PET is a newer technique and thus with cardiovascular risk factors [43], in patients prospective studies documenting the pre- with hypertrophic [20], dilated [21] or isch- dictive value of PET data are just starting to aemic cardiomyopathy [44], and in patients accrue [38,39]. These reports have consistently with suspected or known CAD [18]. The last- demonstrated a graded and independent re- mentioned trial documented an independent lationship between the extent of perfusion prognostic value of myocardial perfusion ab- abnormalities on PET and cardiovascular out- normalities and coronary %ow reserve (CFR) comes during a limited follow-up of 2.7–3.5 assessed with dynamic 13N-NH3 PET over a years. Most importantly, event rates are low follow-up of 5.4 years. In patients with normal with normal perfusion on PET (annualised perfusion, abnormal CFR was independently event rates 0.4%– 2.4%) but increase by a associated with a higher annual event rate factor ranging from approximately 3 to 17 in over 3 years compared with normal CFR (1.4% patients with moderate to severe perfusion vs 6.3%). Similarly, in abnormal perfusion, CFR defects. Furthermore, functional information remained predictive throughout the 10-year from gated PET has proven to have an added follow-up period. independent prognostic value [38, 40]. Assessment of myocardial viability Added value of quantitative MBF and "ow In patients with ischaemic cardiomyopathy, reserve dysfunctional but viable myocardium can ex- The ability to provide non-invasive regional ist in a state of hibernation (where glucose absolute quanti*cation of MBF is a distinct and uptake is preserved or even increased but useful advantage of PET over other imaging contractility impaired), and can recover con- modalities. In the presence of epicardial coro- tractility upon adequate restoration of blood nary artery stenoses, the hyperaemic response %ow (Fig. 5) [45]. to adenosine or dipyridamole tends to de-

130 Courtesy of the Cardiac Imaging Departement, University Hospital Zurich, Switzerland ycrim Te eae enhancement delayed The myocardium. ) indicating D), segments oftheinferolateral wall. remaining the in rim viable preserved a with (arrows) wall inferolateral the of midportion the areain small scar showa transmural F) of and (E images resonance magnetic cardiac preserved with C) and (A seen be can rest at defect perfusion inferolateral large A mismatch. metabolism demonstrate the typical pattern of perfusion– hibernating 13 of with PET perfusion Myocardial myocardium. Example 5A–F. Figure C hapt N-NH3 er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT (A and C) and C) and (A the presence of hibernating hibernating of presence the 18 -D utk ( and (B uptake F-FDG 18 F-FDG PET (B and D) and (B PET F-FDG 131 *city *city of 92% and 63%, respectively, to predict speci- and sensitivity weighted a yielded has A recent meta-analysis of should be noted that only very limited pro- limited very only that noted be should it Nonetheless, [46]. (17%) therapy medical receiving viability with patients in observed was rate mortality annualised highest the whereas 4%), mortality (annual survival best the had revascularisation underwent who myocardium viable with Patients [46]. sation revasculari- after function ventricular left of recovery global predict to respectively, 64%, and 83% and recovery, functional regional (95%CI 0.42–0.93). could be observed, with a hazard ratio of 0.62 cant di&erence favouring the PET guided arm signi*- a included, were revascularisation for only patients who met PET recommendations and the standard therapy arms. Nonetheless, if PET-guided the between noted were comes out- cardiovascular in di&erences signi*cant to guide treatment strategy) [47]. At 1 year, no ed therapy arm (where based on clinical judgement) and a PET-guid- strategy treatment regarding (decision arm therapy standard a to randomised were 35% than less of fraction ejection ventricular left and CAD with patients 430 PARR-2trial, the In prognosis. on impact its and PET of value clinical the on available are data spective 18 F-FDG PET was used 18 F-FDG PE T studies EANM Hybrid PET/CT imaging: combination of The *rst pioneering report of hybrid PET with morphology and function 13N-NH3 and CT angiography was presented The rapid advances in multislice CT technol- by Namdar and colleagues [3]. Sensitivity, ogy have facilitated hybrid PET/CT imaging, speci*city and positive and negative predic- which consists of fusion of both imaging mo- tive values of hybrid PET/CT for the detection dalities into one hybrid image. By exploiting of %ow-limiting stenoses were 90%, 98%, the relative merits of each technique, i.e. ability 82% and 99%, respectively, compared with of CT to reveal the burden of anatomical CAD the clinical gold standard of PET and invasive and ability of PET to evaluate its physiological coronary angiography. A number of similar relevance, hybrid imaging can non-invasively studies using SPECT/CT hybrid imaging have provide unique information to guide manage- followed, given the wider availability of the ment decisions in CAD patients (Fig. 6). SPECT technique, con*rming the superiority of the hybrid approach over the single modalities for the detection of CAD [48–51]. Through spatial co-localisation of perfusion defects with their subtending coronary arteries, the hybrid approach allows targeted revascularisation of haemodynamically relevant stenoses and avoids overuse of angioplasty in non-obstructive lesions [52]. First prospective data document a prognostic value of hybrid imaging [53]. A match of perfusion defect with the corresponding coronary stenosis

Courtesy of the Cardiac ImagingCourtesy Departement, of the Cardiac University Hospital Zurich, Switzerland identi*es a group of patients with a very high Figure 6A–C. Hybrid three-dimensional annual event rate (6.0%) compared to those volume-rendered PET/CT images in an with unmatched (2.8%) or normal *ndings 80-year-old gentleman. The stress and rest (1.3%). The clinical use of hybrid imaging has perfusion images (A) document inferior further been facilitated by the development ischaemia. His CT coronary angiogram (B) and introduction of commercially available reveals intracoronary stents encompassing the dedicated cardiac fusion software [54]. whole right coronary artery. Hybrid imaging (C) shows superposition of the right coronary A recent joint position statement by the artery and the inferior perfusion defect, raising European Association of Nuclear Medicine the strong suspicion of in-stent restenosis of (EANM), the European Society of Cardiac the right coronary artery Radiology (ESCR) and the European Council

132 patients should undergo such hybrid evalua- novel technique and that evidence on which a is imaging hybrid that noted be should It CAD. of probability pretest intermediate an imaging. They support its use in patients with hybrid of use clinical the on ommendations the non-invasive ofCAD. evaluation in protocols hybrid of use widespread more a to contribute will [55], considerably posure ex- radiation lower can which angiography, CT for techniques acquisition novel of tion implementa- the and devices CT high-end with equipped scanners PET of availability increasing the that however,foreseeable, is It investigations. multicentre in and cohorts ment strategies need to be evaluated in larger of hybrid imaging and its impact on manage- tions is therefore scarce. The incremental value of Nuclear Cardiology (ECNC) has given rec- given has (ECNC) Cardiology Nuclear of C hapt er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT 133

EANM References Chapter 4

References 1. Dilsizian V, Bacharach SL, Beanlands RS, Bergmann SR, 11. Bergmann SR, Fox KA, Rand AL, McElvany KD, Welch Delbeke D, Gropler RJ, et al. PET myocardial perfusion and MJ, Markham J, et al. Quanti*cation of regional myocardial metabolism imaging. http://www.asnc.org/section_73.cfm blood %ow in vivo with H215O. Circulation 1984;70:724-33. (accessed march 12th, 2011). 2009. 12. Araujo LI, Lammertsma AA, Rhodes CG, McFalls EO, Iida 2. Hesse B, Tagil K, Cuocolo A, Anagnostopoulos C, Bardies H, Rechavia E, et al. Noninvasive quanti*cation of regio- M, Bax J, et al. EANM/ESC procedural guidelines for myo- nal myocardial blood %ow in coronary artery disease with cardial perfusion imaging in nuclear cardiology. Eur J Nucl oxygen-15-labeled carbon dioxide inhalation and positron Med Mol Imaging 2005;32:855-97. emission tomography. Circulation 1991;83:875-85.

3. Namdar M, Hany TF, Koep%i P, Siegrist PT, Burger C, Wyss 13. Hermansen F, Ashburner J, Spinks TJ, Kooner JS, Cami- CA, et al. Integrated PET/CT for the assessment of coronary ci PG, Lammertsma AA. Generation of myocardial factor artery disease: a feasibility study. J Nucl Med 2005;46:930-5. images directly from the dynamic oxygen-15-water scan without use of an oxygen-15-carbon monoxide blood-pool 4. Pichler BJ, Wehrl HF, Judenhofer MS. Latest advances in scan. J Nucl Med 1998;39:1696-702. molecular imaging instrumentation. J Nucl Med 2008;49 Suppl 2:5S-23S. 14. Muzik O, Beanlands RS, Hutchins GD, Mangner TJ, Ngu- yen N, Schwaiger M. Validation of nitrogen-13-ammonia 5. Schepis T, Gaemperli O, Treyer V, Valenta I, Burger C, Ko- tracer kinetic model for quanti*cation of myocardial blood ep%i P, et al. Absolute quanti*cation of myocardial blood %ow using PET. J Nucl Med 1993;34:83-91. %ow with 13N-ammonia and 3-dimensional PET. J Nucl Med 2007;48:1783-9. 15. Bellina CR, Parodi O, Camici P, Salvadori PA, Taddei L, Fusani L, et al. Simultaneous in vitro and in vivo validation 6. Koep%i P, Hany TF, Wyss CA, Namdar M, Burger C, Kon- of nitrogen-13-ammonia for the assessment of regional stantinidis AV, et al. CT attenuation correction for myocardial myocardial blood %ow. J Nucl Med 1990;31:1335-43. perfusion quanti*cation using a PET/CT hybrid scanner. J Nucl Med 2004;45:537-42. 16. Choi Y, Huang SC, Hawkins RA, Kim JY, Kim BT, Hoh CK, et al. Quanti*cation of myocardial blood %ow using 13N- 7. Bengel FM, Higuchi T, Javadi MS, Lautamaki R. Cardiac po- ammonia and PET: comparison of tracer models. J Nucl Med sitron emission tomography. J Am Coll Cardiol 2009;54:1-15. 1999;40:1045-55.

8. Nekolla SG, Reder S, Saraste A, Higuchi T, Dzewas 17. Nitzsche EU, Choi Y, Czernin J, Hoh CK, Huang SC, Schel- G, Preissel A, et al. Evaluation of the novel myocardial bert HR. Noninvasive quanti*cation of myocardial blood perfusion positron-emission tomography tracer 18F- %ow in humans. A direct comparison of the [13N]ammonia BMS-747158-02:comparison to 13N-ammonia and va- and the [15O]water techniques. Circulation 1996;93:2000-6. lidation with microspheres in a pig model. Circulation 2009;119:2333-42. 18. Herzog BA, Husmann L, Valenta I, Gaemperli O, Siegrist PT, Tay FM, et al. Long-term prognostic value of 13N-am- 9. Schafers KP, Spinks TJ, Camici PG, Bloom*eld PM, Rhodes monia myocardial perfusion positron emission tomography CG, Law MP, et al. Absolute quanti*cation of myocardial added value of coronary %ow reserve. J Am Coll Cardiol blood %ow with H(2)(15)O and 3-dimensional PET: an ex- 2009;54:150-6. perimental validation. J Nucl Med 2002;43:1031-40. 19. Koep%i P, Wyss CA, Namdar M, Klainguti M, von Schul- 10. Bol A, Melin JA, Vanoverschelde JL, Baudhuin T, Vogela- thess GK, Luscher TF, et al. Beta-adrenergic blockade and ers D, De Pauw M, et al. Direct comparison of [13N]ammonia myocardial perfusion in coronary artery disease: di&erential and [15O]water estimates of perfusion with quanti*cation e&ects in stenotic versus remote myocardial segments. J of regional myocardial blood %ow by microspheres. Circu- Nucl Med 2004;45:1626-31. lation 1993;87:512-25.

134 Circulation 2002;105:186-93. ventricular %ow dysfunction. in idiopathic left impairment Prognostical. et myocardialPrataliL, of A, roleblood getti 21. Neglia D, Michelassi C, Trivieri MG, Sambuceti G, Gior- G, Sambuceti MG, TrivieriC, Michelassi D, Neglia 21. pertrophic cardiomyopathy. N Engl J Med 2003;349:1027-35. PG. Coronary microvascular dysfunction and prognosis in hy- humans. 1990;31:1749-56. JNuclMed fasted normal in %uorine-18-%uorodeoxyglucose of lation Bergmann SR, et al. Nonuniformity in myocardial accumu- DJ, Perry SM, Moerlein KJ, Lee BA, Siegel RJ, Gropler29. trial.ternational JNuclCardiol 2007;14:645-58. in- multicenter 3 phase ADVANCE the of results imaging: myocardialperfusion in evaluation comparative denoson rega- versus Adenosine al. et RC, MD,CerqueiraHendel B, Blackburn L, Belardinelli TM, Bateman AE, Iskandrian 28. Coll Cardiol 1994;23:384-9. results from the Adenoscan Multicenter Trial Registry. J Am imaging: perfusion stress adenosine of pro*le Safety AS. Cerqueira MD,27. Verani MS, Schwaiger M, Heo J, Iskandrian 1998;339:173-81. N EnglJMed Wijns 26. W, Vatner SF, Camici PG. Hibernating myocardium. ovasc Dis1989;32:217-38. toimaging bypositron tomography. emission Prog Cardi- disease: basic and principles in application ischemic heart Camici P,25. Ferrannini E, Opie LH. Myocardial metabolism 1983;24:907-15. Nucl Med II. E&ects of metabolic and pharmacologic J interventions. rubidium-82. with perfusion Jr.Myocardial HA O’Brien KL, Gould DJ, Fisher SK, Marani NA, Mullani RA, Goldstein 24. 1990;82:1377-86. Circulation model. mathematical a of graphy.Exploration tomo- emission positron and rubidium-82 with perfusion mann SR. Noninvasive quanti*cation of regional myocardial Herrero P,23. Markham J, Shelton ME, Weinheimer CJ, Berg- 2009;36:576-86. Imaging Mol Med Nucl J Eur stenosis. artery coronary of ment of myocardial blood %ow: validation in a canine model Voicu for C, et al.quantitative assess- Rubidium-82 PET-CT Lautamaki R, 22. George RT, Kitagawa K, Higuchi T, Merrill J, 0 Cecchi20. F, Olivotto I, Gistri R, Lorenzoni R, Chi C hapt er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT r iatti G, Camici 135 DR, Underwood DA, et al. A prospective comparison of rubi- Neumann GB, Saha WJ, MacIntyre TH, Marwick RT, Go 33. 2007;115:1464-80. ninvasive evaluation of coronary disease.artery Circulation no- for technology New MF,R. Hachamovitch Carli Di 32. 2002;29:1257-66. Imaging Mol Med Nucl J Eur dysfunction. ventricular left ischaemic chronic with patients in viability myocardial of evaluation the in imaging PET FDG cardiac of disation standar- for need The JJ. Bax HR, Schelbert J, Knuuti 31. 1992;33:1255-62. J NuclMed tomography.emission positron during utilization glucose myocardial stimulating in load glucose oral and clamp hyperinsulinemic Euglycemic al. et A, Ahonen R, konen Har- M, Saraste U,P, Ruotsalainen Nuutila MJ, Knuuti 30. Cardiol 2008;15:745-53. Nucl J disease. artery coronary suspected or known with mined by stress gated rubidium 82 PET imaging in patients prognostic value of left ventricular ejection fraction deter- incremental and Independent al. et SJ, Cullom L, Volker D, Katten TM, Bateman AW, Ahlberg K, Lertsburapa 38. sestamibi imaging. JAmColl Cardiol 1998;32:57-62. photon emission computed tomographic technetium-99m Iskander S, 37. Iskandrian AE. Risk assessment using single- sion computed tomography. Circulation 2003;107:2900-7. dergoing stress myocardial single photon emis- perfusion therapy in patients with no prior coronary disease artery un- medical with compared revascularization with associated bene*t survival short-term the of Comparison DS. man Hachamovitch R, Hayes36. SW, Friedman JD, Cohen I, Ber- ammonia positron tomography. J Nucl Med 1988;29:1181-8.emission computed tomography: comparisonet al. with Value nitrogen-13and limitation of stress thallium-201 Tamaki35.single photon N, Yonekura Y, Senda M, Yamashita K, Koide H,1991;67:1303-10. Cardiol Saji J H, Am disease. artery coronary of detection for imaging SPECT thallium-201 and tomography emission positron rubidium-82 of Comparison al. et M, Kalus GM, Gacioch J, Popma E, Molina M, Schwaiger RE, Stewart 34. sis of coronary disease.artery J Nucl Med 1990;31:1899-905. imaging utilizing a single dipyridamole stress in the diagno- dium-82 PET and thallium-201 SPECT myocardial perfusion

EANM 39. Yoshinaga K, Chow BJ, Williams K, Chen L, deKemp RA, 48. Rispler S, Keidar Z, Ghersin E, Roguin A, Soil A, Dragu R, et Garrard L, et al. What is the prognostic value of myocardial al. Integrated single-photon emission computed tomogra- perfusion imaging using rubidium-82 positron emission phy and computed tomography coronary angiography for tomography? J Am Coll Cardiol 2006;48:1029-39. the assessment of hemodynamically signi*cant coronary artery lesions. J Am Coll Cardiol 2007;49:1059-67. 40. Dorbala S, Hachamovitch R, Curillova Z, Thomas D, Vangala D, Kwong RY, et al. Incremental prognostic value 49. Gaemperli O, Schepis T, Valenta I, Husmann L, Sche&el H, of gated Rb-82 positron emission tomography myocardial Duerst V, et al. Cardiac image fusion from stand-alone SPECT perfusion imaging over clinical variables and rest LVEF. JACC and CT: clinical experience. J Nucl Med 2007;48:696-703. Cardiovasc Imaging 2009;2:846-54. 50. Santana CA, Garcia EV, Faber TL, Sirineni GK, Esteves FP, 41. Uren NG, Melin JA, De Bruyne B, Wijns W, Baudhuin T, Sanyal R, et al. Diagnostic performance of fusion of myocar- Camici PG. Relation between myocardial blood %ow and dial perfusion imaging (MPI) and computed tomography the severity of coronary-artery stenosis. N Engl J Med coronary angiography. J Nucl Cardiol 2009;16:201-11. 1994;330:1782-8. 51. Slomka PJ, Cheng VY, Dey D, Woo J, Ramesh A, Van 42. Kajander S, Joutsiniemi E, Saraste M, Pietila M, Ukkonen Kriekinge S, et al. Quantitative analysis of myocardial per- H, Saraste A, et al. Cardiac positron emission tomography/ fusion SPECT anatomically guided by coregistered 64-slice computed tomography imaging accurately detects anato- coronary CT angiography. J Nucl Med 2009;50:1621-30. mically and functionally signi*cant coronary artery disease. Circulation 2010;122:603-13. 52. Gaemperli O, Husmann L, Schepis T, Koep%i P, Valenta I, Jenni W, et al. Coronary CT angiography and myocardial 43. Schindler TH, Nitzsche EU, Schelbert HR, Olschewski perfusion imaging to detect %ow-limiting stenoses: a po- M, Sayre J, Mix M, et al. Positron emission tomography- tential gatekeeper for coronary revascularization? Eur Heart measured abnormal responses of myocardial blood %ow J. 2009;30:2921-9. to sympathetic stimulation are associated with the risk of developing cardiovascular events. J Am Coll Cardiol 53. Pazhenkottil AP, Nkoulou RN, Ghadri JR, Herzog BA, Bue- 2005;45:1505-12. chel RR, Kuest SM, et al. Prognostic value of cardiac hybrid imaging integrating single-photon emission computed 44. Tio RA, Dabeshlim A, Siebelink HM, de Sutter J, Hillege tomography with coronary computed tomography angio- HL, Zeebregts CJ, et al. Comparison between the prognostic graphy. Eur Heart J. 2011 Feb 14 [Epub ahead of print].54. value of left ventricular function and myocardial perfusion Gaemperli O, Schepis T, Kal& V, Namdar M, Valenta I, Stefani reserve in patients with ischemic heart disease. J Nucl Med L, et al. Validation of a new cardiac image fusion software 2009;50:214-9. for three-dimensional integration of myocardial perfusion SPECT and stand-alone 64-slice CT angiography. Eur J Nucl 45. Camici PG, Prasad SK, Rimoldi OE. Stunning, hiberna- Med Mol Imaging 2007;34:1097-106. tion, and assessment of myocardial viability. Circulation 2008;117:103-14. 55. Husmann L, Valenta I, Gaemperli O, Adda O, Treyer V, Wyss CA, et al. Feasibility of low-dose coronary CT angio- 46. Schinkel AF, Bax JJ, Poldermans D, Elhendy A, Ferrari R, graphy: *rst experience with prospective ECG-gating. Eur Rahimtoola SH. Hibernating myocardium: diagnosis and Heart J 2008;29:191-7. patient outcomes. Curr Probl Cardiol 2007;32:375-410. 56. McNeill AJ, Fioretti PM, el-Said SM, Salustri A, Forster T, 47. Beanlands RS, Nichol G, Huszti E, Humen D, Racine N, Roelandt JR. Enhanced sensitivity for detection of coronary Freeman M, et al. F-18-%uorodeoxyglucose positron emissi- artery disease by addition of atropine to dobutamine stress on tomography imaging-assisted management of patients echocardiography. Am J Cardiol 1992;70:41-6. with severe left ventricular dysfunction and suspected coronary disease: a randomized, controlled trial (PARR-2). J Am Coll Cardiol 2007;50:2002-12.

136 Table ofthe properties 2:Radiokinetic “ideal” tracer perfusion PET Table tracersfor cardiac commonPET applications 1:Most TABLES C Tracer a E+cient, fast,reliableE+cient, andfully automated radiochemicalsynthesis inertness Metabolic andLVgating for thesimultaneousassessmentofperfusion function Prolonged to give ECG retention andhighcountstatisticspermit astabledistribution intheheart Linear correlation ofmyocardial withMBFrate traceruptake andimpr tion intheheart atbothbaselineandhigh%ow accumula- rates fraction for maximalactivity *rst-pass extraction High 18 13 82 15 hapt Calculated for tissueusingthefullwidthat20%ofmaximumamplitude(FW20M)method soft -D 1 i 0 5 .40.54 0.64 200 – 350 110min F-FDG N-NH Rb O-H er 4 – Clinical applications of PET/CT incardiology applicationser 4–Clinical ofPET/CT 2 O 3 half-life Ph 0mn30–7011 1.26 1.19 370 – 740 10 min ysical i 0 ,0 .01.87 1.70 700 – 1,500 2 min 8s110–15031 4.10 3.15 1,100 – 1,500 78 s (MBq) activity mended Recom- oved imagequality (MeV) energy positron Maximum 137 (mm) range positron Tissue a Uptake mechanism hexo-kinase (viability) hexo-kinase Glucose transport/ (perfusion) Na/K-ATPase (perfusion) meta-bolic trapping Di&usion/ (perfusion) Free di&usion production Radionuclide generator Cyclotron Cyclotron Cyclotron 82 Sr/ 82 Rb Rb EANM Table 3: Dosing regimen for pharmacological stress agents in myocardial perfusion PET

Adenosine

Continuous intravenous infusion at 140 μg/kg/min over 6–7 min a

PET tracer injection 3 min into adenosine infusion

Dipyridamole

Continuous intravenous infusion at 140 μg/kg/min over 4 min

PET tracer injection 3–5 min after completion of dipyridamole infusion

Dobutamine

Continuous intravenous infusion starting at 5–10 μg/kg/min

Increase by 10 μg/kg/min increments to reach target heart rate (at least 85% of the age-predicted maximal heart rate) or maximal dose of 40 μg/kg/min

If necessary, increase heart rate with intravenous atropine (0.25 mg, repeated every 1–2 min up to a maximal dose of 1 mg) [56]

Continue dobutamine infusion for at least 2 min after tracer injection a For patients at risk of complications (AV conduction disturbances, borderline hypotension, inadequately controlled asthma), adenosine infusion can be started at a lower dose (50 μg/kg/min) and increased at 1-min intervals to 75, 100 and 140 μg/kg/min if tolerated. PET tracer injection can be performed 1 min into the 140 μg/kg/min dose. Shorter adenosine infusion protocols are not recommended for PET perfusion imaging, particularly if dynamic image acquisition for MBF quanti*cation is envisaged

138 in or will become available. The brain “targets” brain The available. become will or are which radiopharmaceuticals of number the to as well as diseases neurodegenerative of incidence high the to due is This cine. Nobili reviews this large *eld, pointing out out pointing *eld, large this reviews Nobili Flavio years. many for dementia in PET brain of use the supported has data of rent force” “driving body A large PET. brain of cur- the certainly is which disorders, related and dementia addresses chapter *rst The imaging. PET of brain applications the cover current which chapters *ve in Guide Technologist’s this of section this organised has Committee aging Neuroim- the EANM background, this Against process. this of beginning very the at are We tice. prac- daily and trials clinical in techniques these of usefulness the demonstrate and protocols the standardise to together work should physicians medicine nuclear and technologists recognition, this achieve to order In neuro-oncology. and geriatrics try, psychia- neurology, in guidelines clinical by recognised usefulness their and physicians referring by more understood and be known widely to need procedures medicine nuclear neuroimaging However, etc. tion, activa- microglial tumours, plaques, amyloid neurotransmission, metabolism, perfusion, rich: exceptionally are imaging molecular for Positron emission tomography emission Positron Chapter 5 – Clinical applications of PET/CT in neurol Jacques Darcourt Jacques to brainPET 5.1 Introduction g is a fast-growing field in nuclear medi- nuclear in field fast-growing a is g neuroimag- 18 F-FDG F-FDG 139 in dementias for example. for dementias in as indications in these as mature being from far is imaging PET of role the However, tions. indica- promising these of aware be should we Therefore autism. and disorders de*cit attention disorders, stress post-traumatic as such depression, applications future major of *eld a is It situation. current the explains clearly Pagani Marco psychiatry. with deals and di&erent very is chapter second The future. near the in available become should which disease, Alzheimer’s for imaging plaque amyloid addresses also He indications. accepted currently the major and Gregory Petyt, reviews the use of brain brain of use the reviews Petyt, Gregory and Semah Franck by written chapter, last The metabolism. tumour speci*c or acid amino as such glucose features logical of bio- advantage taking data, morphological to MRI information molecular valuable adds It of stands. PET where brain representative larly is particu- indication This management. mour tu- brain for imaging PET of capabilities the Tatsch describes Klaus chapter, fourth the In diagnosis. the for pharmaceuticals radio- di&erent use the elegantly lustrates il- it and imaging neurotransmission of use well-established *rst the is it since portance im- major of is This disorders. movement of diagnosis for imaging PET of applications the reviews Booij Jan chapter third the In ogy EANM PET in epilepsy. The authors explain in particular how 18F-FDG PET images can detect metabolic abnormalities which draw attention to speci*c regions which can be missed on MRI.

The EANM Neuroimaging Committee along with the Technologist Committee hopes that these chapters will contribute to the further development of brain PET imaging, to increased standardisation of the techniques and to recognition of the techniques in clinical strategies for patient management.

140 unit, where it undergoes phosphorylation via energy acid pathway, requiring oxygen and leading and oxygen requiring pathway, acid tricarboxylic the via terminals synaptic the at place takes production Energy hexokinase. ues. As compared to SPECT tracers, this topo- this tracers, to ues. As SPECT compared val- the show lowest cortex temporal medial and the cerebellum while cortex, and mesial polar and occipital ganglia basal the thalami, in found are values highest matter. The grey the in that of 25–30% about is rate the ter mat- white the in adult; normal the in rest at tissue/min brain of g glucose/100 mmol 60 to 40 about from ranges (CMRglc) cose glu- for rate metabolic cerebral matter grey reference The level. synaptic the at happens utilisation of glucose as part the unit, greater functional neuron–astrocyte its in tabolism me- terminal synaptic of measurement tive or semi-quantita- quantitative allows %uorine [ with glucose Labelling activation. tional func- during and rest at function neuronal to coupled closely is metabolism Glucose more economicenergy production [1]. for pyruvate to transformed is and neurons end-product is transferred from astrocytes to lactate rapidly.The more so doing but ergy anaerobic glycolysis, providing much less en- utilise mainly astrocytes turn, In activation. quickly meet the need for energy during brain cannot but productive very is pathway This glycolysis). (aerobic availability ATP high to The brain mainly brain The ofbrainBasic principles glucose utilisation Chapter 5 – Clinical applications of PET/CT in neurol Flavio Nobili andrelated5.2 Dementia disorders . Glucose enters the neuron–astrocyte utilises glucose to produceto glucose utilises 18 F] 141 uptake distribution is mainly driven by basal basal by driven mainly is distribution uptake cortex, cortex, and the lateral posterior parietal, both the following: precuneus, posterior cingulate of more or one in found is hypometabolism obtained with with obtained Alzheimer’s disease cognitiveMild impairment and for brain guidelines provides The activity. EANM metabolic synaptic duced or re- of synapses in the number a reduction either represents essentially uptake glucose reduced unit, Fortissue a integrity. neuronal general represents and activity neuronal with obtained that resembles closely distribution graphic and the mesial temporal lobe, which is indeed hippocampus the segmenting in adequate more is slices transaxial of analysis [3]), plane 30° upward with respect to the bicommissural transaxial plane (i.e. with the nose tilted about Ohnishi so-called the on re-orientation ing extent, lateral frontal association Us- cortices. lesser a to and, temporal mesial and lateral with suspicion of AD undergoes develop dementia. When a typical MCI patient tests. However, not all individuals with MCI will cognitive standardised by con*rmed de*cit memory a as de*ned been has and (MCI) stage as is mild known cognitive impairment her autonomy occupations). in everyday This his/ maintains subject the (i.e. present yet not is dementia when disease, the of stages early to moved has (AD) disease Alzheimer’s Recently, the diagnostic e&ort in patients with 99m 99m Tc-ECD but is less than that Tc-ECD that than is but less Tc-HMPAO. glucose rest, At 18 F-FDG PET [2]. PET F-FDG 18 F-FDG PET, ogy EANM not parallel to the bicommissural plane. The In a study including 284 patients, of whom 138 primary sensorimotor cortex, visual cortex, had a neuropathological diagnosis, a typical thalami, basal ganglia and cerebellum are AD pattern (i.e. parietal and temporal hypome- relatively spared. tabolism, with or without frontal involvement) showed a sensitivity of 94% and a speci*city of A very frequent feature especially evident in 73% in comparison to the neuropathological the early stages of disease is the strong asym- diagnosis [4]. In another study, the sensitivity metric involvement. As a consequence, visual of PET for Alzheimer’s disease was 86% (95% analysis of scans strongly relies on asymmetric CI: 76–93%) and the speci*city was 86% (95% evaluation of abnormalities. Early-onset AD CI: 72–93%) [5]. Its high accuracy has posi- (i.e. before the age of 60) is more often char- tioned 18F-FDG PET as one of the biomarkers acterised by posterior parieto-temporal hy- for the diagnosis of AD, even before the onset pometabolism as compared to late-onset AD, of dementia, the others being MRI and tau in which medial temporal hypometabolism protein and Ab 1-42 assays in cerebrospinal prevails. An example in a patient with early- %uid [6]. onset AD is shown in Fig. 1. 18F-FDG PET has been reported to be highly sensitive to hypometabolism in AD patients over time, in follow-up studies lasting 1 year. This means that it is a suitable marker to follow the disease evolution and to evaluate the potential e&ect of both symptomatic and neu- roprotective agents. It has been computed that metabolism reduction in critical regions is in the order of 16–19% over 3 years, while in healthy subjects reduction is virtually absent over such a time span. In the next few years, some radiopharmaceuticals able to image the amyloid brain burden will become commercially available. Virtually all patients with AD show amyloid brain deposition (very high sensitivity), but patients with dementia with Lewy

Courtesy of Clinical Neurophysiology and Nuclear MedicineCourtesy Dept., of Clinical Neurophysiology University Hospital San Martino, Genoa, Italy bodies (DLB) or Parkinson’s disease dementia Figure 1: 18F-FDG PET *ndings in a patient with and some elderly healthy subjects are also early-onset AD ‘amyloid positive’ (moderate speci*city) [7].

142 Chapter 5 – Clinical applications of PET/CT in neurology: 5.2 Dementia and related disorders

Frontotemporal dementia Frontotemporal dementia accounts for about

10% of all dementias and is mainly charac- of Clinical Neurophysiology Dept., Courtesy and Nuclear Medicine terised by early behavioural disturbances, including apathy, dysinhibition, altered social behaviour and stereotyped behaviour, variously mixed according to the parts of the frontal lobes maximally involved. In the left- Italy Genoa, Hospital San Martino, University sided variants (progressive aphasia), language complaints are found. Both the frontal lobes may be variably involved, often in an asymmetric fashion, together with the pole and EANM the anterolateral aspect of the temporal lobes, the insula and the anterior cingulate gyrus. When the frontal lobe is severely a&ected, hypometabolism in the contralateral cerebellar hemisphere is found (‘crossed cerebellar dias- Figure 2: 18F-FDG PET *ndings in a patient chisis’) as a consequence of dea&erentation with FLD via the fronto-ponto-cerebellar *bres. In the variant ‘frontal lobe degeneration’ (FLD), the two frontal lobes are more symmetrically af- fected. Figure 2 shows the 18F-FDG PET of a patient with FLD.

143 Dementia with Lewy bodies Pitfalls and limitations Dementia with Lewy bodies accounts for Since most equipment is nowadays PET/CT about 10–15% of cases of dementia. The and attenuation correction is based on CT clinical symptoms may include spontaneous scan, movement artefacts between CT and (not drug-induced) parkinsonism, early visual PET acquisitions should be carefully checked hallucinations, %uctuating attention and REM because they can generate false asymme- sleep behaviour disorders. Neuroleptic hy- tries. A fast of at least 4 h is essential because persensitivity (severe parkinsonism following a standard dose of about 150–185 MBq of neuroleptic administration) is a key sign. The 18F-FDG (3D acquisition) without fasting may metabolic picture is similar to that observed be insu+cient to achieve a good image qual- in AD, but occipital hypometabolism is more ity. Another limitation of 18F-FDG PET is that frequent and more severe, while medial tem- no standard analytical procedure has as yet poral lobe hypometabolism is milder. Presyn- been established. Scans are analysed in very aptic dopaminergic PET radiopharmaceuticals, di&erent ways, ranging from simple visual such as 11C-FE-CIT and 18F-%uorodopa, and of evaluation to sophisticated computer-assisted course the regular DaT SPECT (FP-CIT SPECT) comparisons with a normative database, ei- highlight reduced uptake at the basal ganglia ther already available in working stations and level. supplied by the manufacturer or locally established. While simple visual analysis may be too Vascular dementia sensitive even to trivial or physiological asym- Cognitive impairment up to dementia can metries and impairment in median structure be found in a large family of cerebrovascular hypometabolism can be missed, automatic diseases, the most representative of which are analysis su&ers from opposite issues because it subcortical ischaemic vascular disease, large is generally very speci*c but scarcely sensitive, brain infarctions and infarctions a&ecting although automatic analysis will help relative ‘strategic’ areas. MRI is the cornerstone of the inexperienced Nuclear Medicine Physicians in diagnosis, while 18F-FDG PET may highlight better interpretation of the images. hypometabolism in the basal ganglia, in the thalami and, patchily, in several cortical areas, more often the temporal lobes.

144 References Chapter 5.2 References . Ohnishi T, Hoshi H, Nagamachi S, Jinnouchi S, Flores LG 3. 2009;36:2103-10. Imaging version Mol 2.Eur JNuclMed cedure guidelines forbrain imaging PET using [(18)F]FDG, Varrone A, Asenbaum S, Vander Borght T, 2. et al. EANM pro- 2000;886:108-12. Res Brain coupling. metabolic neural-glia from insight ging: ima- brain functional of basis PJ.Cellular Magistretti 1. deposition in Lewy body diseases. Neurology 2008;71:903-10. Gomperts SN, Rentz DM, Moran E, et7. al. Imaging amyloid NINCDS-ADRDA LancetNeurol criteria. 2007;6:734-6. the revising disease: Alzheimer’s of diagnosis the for teria cri- Research al., et C, Jacova HH, Feldman B, Dubois 6. stics ofPET—a 2004;231:73-80. metaanalysis. Radiology characteri- operating disease: OT.Alzheimer Rutschmann GP,Samsa DB, Matchar DC, McCrorg MB, Patwardhan 5. metabolism and long-term outcome. JAMA 2001;286:2120-7. siontomography evaluationin dementia:of regional brain Silverman DH, Small GW, Chang CY, et al. Positron4. emis- 1995;36:1163-9. Med hippocampal perfusion in neuropsychiatric diseases. J Nucl 2nd, Futami S, Watanabe K. High-resolution SPECT to assess 145

EANM Chapter 5 – Clinical applications of PET/CT in neurology 5.3 Psychiatry Marco Pagani

Introduction and suicidality not due to a medical condition, Neuroimaging in psychiatry has been used medication, abused substance or psychosis. to assess functional, neurochemical and struc- MD is estimated to rival virtually every other tural patterns related to speci*c disorders, to known medical illness in burden of disease evaluate new molecules as therapeutic agents morbidity early in this millennium since it is and to foresee individual response to treat- estimated to a&ect, on a life-time basis, up to ment. In psychiatric studies, comorbidity is a quarter (26%) of the population. The neural a critical issue since the co-occurrence of networks modulating emotional expression di&erent pathological dimensions might im- seem to be implicated in the pathophysiol- pact on the neurobiological patterns, causing ogy of MD, and this is in particularly true of the neuroimaging *ndings to be unspeci*c. the medial prefrontal cortex, amygdala, hip- Furthermore, macroscopic changes have sel- pocampus and basal ganglia, as reported by dom been described since most psychiatric several PET studies. Serotonin 1A receptor diseases show changes at the molecular level. (5-HT1AR) function appears to be abnormal All this results, for all forms of anxiety disor- in MD but disagreement still exists regarding der, in a diagnostic accuracy far below that the presence and direction of binding abnor- reported in neurodegenerative disorders. In malities [1]. An important use of PET in MD is this respect the diagnostic approach to the the study and follow-up of antidepressant ef- various mental disorders included in DSM-IV fects on monoamine receptor occupancy [2]. Axis I (major mental and learning disorders) needs speci*c knowledge in order to adopt Post-traumatic stress disorder the correct strategies for patient selection as Post-traumatic stress disorder (PTSD) is an well as advanced radiochemical and statis- anxiety disorder following major psychologi- tical methodologies. In most of the mental cal trauma reported to a&ect between 4% and disorders included in DSM-IV Axis I, CBF, me- 8% of the general population. It is de*ned by tabolism and transporter/receptor patterns the coexistence of three clusters of trauma- have been investigated but only in some of related symptoms: re-experiencing, avoidance them has a general consensus been reached. and hyperarousal. Neural correlates of PTSD have been investigated by means of di&erent Major depression research paradigms such as symptom provo- Major depression (MD) is a common primary cation, cognitive activation and functional idiopathic condition characterised by one or connectivity. Functional studies by PET have more major depressive episodes of persis- consistently shown during emotional stimuli tently low mood, dysregulated sleep, appetite a concomitance of amygdala hyper-reactivity and weight, anhedonia, cognitive impairment, and a correspondingly reduced prefrontal

146 Chapter 5 – Clinical applications of PET/CT in neurology: 5.3 Psychiatry

cortex and anterior cingulate cortex control Attention de"cit hyperactivity disorder over amygdalae. However, other structures Attention de*cit hyperactivity disorder (ADHD) have been consistently found to be involved is a multifactorial and clinically heterogeneous in PTSD, i.e. anterior and posterior insular cor- neurobehavioural disorder characterised by tex, posterior cingulate cortex and certain por- symptoms of hyperactivity, inattention and tions of the temporal lobe and orbitofrontal impulsivity. It is suggested to be the most cortex ([3]; for review see [4]). Figure 1 illus- common chronic undiagnosed psychiatric trates increased CBF in PTSD as demonstrated disorder in adults and its prevalence in adults by 99mTc-HMPAO SPECT imaging analysed by is estimated to be about 4%. Comorbidity with SPM2. at least one other psychiatric disorder occurs in 87% of adult ADHD subjects. Neuroanatom- ical and neurobiological data underscore alter- EANM ations in dopamine/noradrenaline-modulated fronto-striatal-cerebellar circuits. Owing to the high catecholamine concentration in these sensitive regions, both dopamine transporters and receptor ligands have often been used in ADHD studies [5] and the e&ect of stimulants, mostly methylphenidate, investigated. Moreover de*cits in CBF and metabolism have been reported in both adolescents and adults with ADHD (for review see [6]).

Autism spectrum disorders Autism spectrum disorders (ASDs) are de- *ned on the clinical basis by impairment in social interaction, impairment in verbal and non-verbal communication and repetitive or Courtesy of Department Karolinska of Nuclear Medicine, Hospital, Stockholm and of Nuclear Medicine Communication stereotypical behaviours. They a&ect almost Figure 1: Increased CBF in PTSD. 99mTc-HMPAO 1% of the general population and have shown SPECT images analysed by SPM2. Three- an impressive increase in prevalence since the dimensional rendering of voxels re%ecting early 1990s, probably attributable to the better higher tracer distribution in patients (n=15) as awareness of the disease. Structural *ndings compared to controls (n=27). The signi*cant support the hypothesis of anomalous perina- statistical di&erences are highlighted tal processes of neurogenesis and neuronal

147 maturation [7]. Regional CBF or metabolism abnormalities have been found in the cerebellum, limbic system, frontal and temporal cortices, corpus callosum and basal ganglia even though speci*c patterns for ASD have not yet been identi*ed [8]. Furthermore, a growing body of evidence suggests weaker functional connectivity, especially in the default network, often associated with symptom severity.

Conclusion Neuroimaging in psychiatric disorders will help to provide a better understanding of the underlying pathophysiology, to assess the neurobiology of treatment outcome and to demonstrate the neuroanatomical, functional and chemical pro*les associated with speci*c endophenotypes. It will also contribute in disclosing the neurobiology of the speci*c symptoms underlying psychiatric disorders and in progressing towards a more suitable dimensional approach.

148 D, Hood K, et al. Occupancy of serotonin transporters by transporters serotonin of Occupancy al. et K, Hood D, Hussey V, Goulding N, Ginovart AA, Wilson JH, Meyer 2. 2007;34:865-77. depression: replication and literature review. Nucl Med Biol Kupfer DJ, et al. Serotonin-1A receptor imaging in recurrent disorders. Ment Retard Dev Disabil Res Rev 2000;6:171-9. disorders. Rev DisabilRes Retard Dev Ment Rumsey JM, Ernst M. Functional neuroimaging of autistic 8. of autism. Trends Neurosci 2008;31:137-45. Amaral DG, Schumann CM, Nordahl CW. Neuroanatomy 7. 2009;57:601-7. pharmachology for a better understanding of the biology of ADHD. Neuro- Zimmer L. Positron emission tomography neuroimaging 6. hyperactivity. 2005;57:229-38. BiolPsychiatry association between striatal dopamine markers and motor disorder: attention-de*cit/hyperactivity with adolescents male in binding transporter dopamine midbrain duced ForssbergC, FernellFardeHalldin A, H, JucaiteE, Re- L. 5. 2007;24:202-18. Anxiety Depress *ndings. and methods current of view re- disorder: stress posttraumatic in studies roimaging neu- Functional JD. Bremner E, V, FrancatiVermetten 4. stress disorder. Commun 2007;28:757-65. NuclMed HMPAOpost-traumaticoccupation-related in distribution Jonsson C, et al. E&ects of EMDR psychotherapy O, on Sundin 99mTc-D, Nardo D, Salmaso G, Högberg M, Pagani 3. 1843-9. a [(11)C]DASB PET imaging study. Am J Psychiatry 2001;158: paroxetine and citalopram during treatment of depression: . Drevets WC, Thase EL, ME, PriceMoses-Kolko J, Frank E, 1. References References Chapter 5.3 149

EANM Chapter 5 – Clinical applications of PET/CT in neurology 5.4 Movement disorders Jan Booij

The movement disorder Parkinson’s disease Radiopharmaceuticals to assess central do- (PD) is a clinical syndrome characterised by pamine transporters in vivo have been devel- parkinsonism: bradykinesia, hypokinesia/aki- oped and validated for both SPECT and PET nesia, rigidity, tremor and postural instability. imaging in humans. Loss of striatal dopamine This syndrome is characterised neuropatho- transporters in PD patients has been reported logically by a severe loss of dopaminergic consistently, even in preclinical (premotor) nigrostriatal neurons. The integrity of these phases of the disease (for a review see Booij dopaminergic neurons can be measured and Knol [2]). Radiotracers for the vesicular using radiopharmaceuticals for the dopa- monamine transporter are available only for mine transporter (located exclusively in the PET [1]. Use of such tracers also permits the membrane of dopaminergic neurons) or the accurate detection of loss of nigrostriatal neu- vesicular monamine transporter (located in rons. Although in Europe the majority of cen- the membrane of vesicles in dopaminergic tres use dopamine transporter radioligands terminals) (Fig. 1). and SPECT to assess the integrity of nigrostriatal neurons in clinical routine studies, some centres use 18F-DOPA PET [3].

In the majority of PD patients, the dopamine transporter binding or DOPA uptake in the striatum is asymmetrical. In addition, the reduced uptake/binding is more pronounced in the putamen than in the caudate nucleus (Fig. 2). In hemiparkinsonian patients (in the © J. Booij © J. vast majority of PD patients, motor signs start Figure 1: Schematic representation of a on one side of the body) even bilateral loss of terminal of a dopaminergic neuron at the striatal DOPA uptake or dopamine transporter level of the striatum binding has been documented. Moreover, there is a correlation between striatal uptake/ Another method to measure the integrity of binding and disease severity [1]. nigrostriatal neurons is PET imaging following injection of 18F-DOPA. The uptake rate con- Loss of dopamine neurons is not speci*c for stant of DOPA is determined by its transfer PD. Also in certain other movement disorders, across the blood–brain barrier, its decarboxyl- such as multiple system atrophy (MSA) or pro- ation to dopamine and the retention in nerve gressive supranuclear palsy (PSP), there is a terminals (for a review see Booij et al. [1]). severe loss of nigrostriatal neurons. MSA is a

150 Chapter 5 – Clinical applications of PET/CT in neurology: 5.4 Movement disorders

movement disorder characterised by a variable of Neurology,Department Center Medical University combination of parkinsonism, cerebellar dys-

function, autonomic failure and involvement of Groningen, Groningen, The Netherlands Courtesy: L.K. L.K. TeuneCourtesy: Leenders, and K.L. the pyramidal tract. PSP is characterised by parkinsonism (frequently symmetrical and with an axial rigidity), vertical supranuclear gaze palsy, dysphagia, dysarthria and cognitive decline. Generally speaking, in both MSA and PSP the response to dopaminergic drugs is poor. On the other hand, drug-induced parkinsonism, essential tremor and psychogenic parkinsonism may sometimes be hard to di&erentiate clini- Figure 2: Transverse PET slices at the level of the cally from PD; however, these disorders are not striatum. Upper panel: left: normal 18F-DOPA characterised by loss of dopaminergic neurons uptake in the striatum in a patient su&ering (Fig. 2). Therefore, imaging of the nigrostriatal from essential tremor; right: reduced striatal dopaminergic neurons may be helpful in con- uptake (particularly in posterior putamen) in *rming or excluding an underlying dopaminer- a hemiparkinsonian PD patient (with motor gic de*cit in uncertain cases. signs on left side of body). Lower panel: note the lower striatal 18F-FDG uptake in the Additionally, MSA and PSP are characterised striatum of an MSA patient (right side) than by loss of postsynaptic striatal neurons bear- in a patient su&ering from PD (left). Images ing dopamine D2 receptors. Particularly in the courtesy of L.K. Teune and K.L. Leenders, earlier clinical phases, it may be hard to di&er- Department of Neurology, University entiate these syndromes from PD clinically. To Medical Center Groningen, Groningen, The help di&erentiate these syndromes from PD, Netherlands some centres use PET and radiotracers for dopamine D2 receptors (such as 11C-raclopride) Additional reduced uptake may be detected [1]. However, since several studies have ob- in MSA in the cerebellar hemispheres [4]. In served a certain degree of overlap between PSP, reduced 18F-FDG uptake may not be re- data obtained in PD patients and in MSA/PSP stricted to the striatum, but may also involve patients, centres are starting to use 18F-FDG the prefrontal lobe, thalamus and mesen- PET to di&erentiate MSA/PSP from PD. 18F-FDG cephalon [4]. PET studies have demonstrated loss of 18F-FDG uptake in the striatum of MSA patients but not in PD (Fig. 2).

151 References Chapter 5.4

References 1. Booij J, Tissingh G, Winogrodzka A, van Royen EA. Ima- ging of the dopaminergic neurotransmission system using single-photon emission tomography and positron emission tomography in patients with parkinsonism. Eur J Nucl Med 1999;26:171-82.

2. Booij J, Knol RJJ. SPECT imaging of the dopaminergic system in (premotor) Parkinson’s disease. Parkinsonism Re- lated Disord 2007;13:S425-8.

3. Eshuis SA, Jager PL, Maguire RP, Jonkman S, Dierckx RA, Leenders KL. Direct comparison of FP-CIT SPECT and F-DOPA PET in patients with Parkinson’s disease and healthy controls. Eur J Nucl Med Mol Imag 2009;36:454-62.

4. Teune LK, Bartels AL, de Jong BM, Willemsen AT, Eshuis SA, de Vries JJ, et al. Typical cerebral metabolic patterns in neurodegenerative brain diseases. Mov Disord 2010;25:2395-404.

152 ceiain peprtv eauto and, evaluation preoperative acterisation, char- classi*cation, diagnosis, for helpful are which glioma of features metabolic several addressing by role important an plays PET in recent ischaemic infarcts, in some benign some in infarcts, ischaemic recent in lesions, in%ammatory active in observed be tion, addi- In there. located when missed be may matter,white normal of that to close uptake with present usually which mours, tu- low-grade and masked be may tumours matter, high-grade grey normal of sumption a minor role. toOwing the high glucose con- Regarding this question, or [ O-(2-[ MET), tracers such as [methyl- acid amino using transport/(incorporation) ( 2-deoxy-D-glucose 2-[ using metabolism glucose are two main functions extensively studied so far The monitoring. post-therapeutic especially, In the diagnostic work-up diagnostic the In Introduction tumours andbenign lesions Di$erential diagnosis between brain covered.of thesetechniques willbebrie%y since elevated glucose metabolism may also may metabolism glucose elevated since Chapter 5 – Clinical applications of PET/CT in neurol Klaus Klaus Tatsch Poepperl andGabriele 5.5 Braintumours 18 F]%uorodopa. Major clinical applications 18 F-FDG lacks speci*city for malignancy malignancy for speci*city lacks F-FDG 18 F]%uoroethyl)-l-tyrosine ( 18 F-FDG) and amino acid 11 18 C]-L-methionine ( F-FDG PET may play of brain tumours, brain of 18 F]%uoro- 18 18 F-FDG F-FET) F-FET) 11 C- 153 trast, increased amino acid uptake is present is uptake acid amino increased trast, con- In seizures. epileptic during or tumours should be as complete as possible but at the at completebut as possible be as should tumours brain of resection surgical Since oftumourbordersDe"nition do not exhibit increased amino acid transport. generally scar post-traumatic or cavernoma haematoma, dysplasia, gliosis, reactive sions, ischaemic demyelinatingor in%ammatory le- subacute or chronic as such lesions tumoral those with low-grade types. In addition, non- in even gliomas, with patients all almost in acid tracers seem superior to acid tracersseemsuperior amino radiolabelled context, this In therapy. tory to de*ne the optimal volume for radiation termination of tumour borders is also manda- is needed for treatment planning. Precise de- normal brain tissue, exact tumour delineation intact functionally spare largely time same 18 F-FDG (Fig. 1). ogy EANM Courtesy of Department of Nuclear Medicine, Municipal Hospital Karlsruhe Inc, Germany

Figure 1: Glioblastoma: Major parts of the tumour in the left parietal lobe show 18F-FDG metabolism similar to or even below that of the normal cortex. There is reduced 18F-FDG metabolism in the surrounding oedema zone. Better delineation of the tumour borders and extent is achieved with the amino acid tracer 18F-FET. This may favour this type of tracer for biopsy and therapy planning

Especially brain tumours which appear hypo- take correlates well with the histological grade or isometabolic compared with cortical up- of gliomas, 18F-FDG PET has been considered take and which are located in or near to the as a valuable diagnostic tool for this purpose. grey matter may hardly be identi*ed with Cut-o& levels of 1.5 and 0.6, respectively, for tu- 18F-FDG, thus hampering clear de*nition of mour-to-white matter and tumour-to-cortex tumour boundaries. On the other hand, reuptake ratios have been proven to di&erenti- sults regarding tumour delineation by means ate accurately between low- and high-grade of amino acid tracers are encouraging. As tumours [1]. The role of radiolabelled amino documented by stereotactic serial biopsies, acids or analogues for tumour grading is cur- these tracers may even better re%ect the real rently changing. Earlier studies using stan- tumour size than morphological imaging with dard ratio methods taking into account the CT or MRI. late uptake phase (>30 min) showed a slight but statistically not signi*cant increase from Grading/malignant progression low- to high-grade tumours [2,3]. The clini- According to the World Health Organisation cally most important di&erentiation between (WHO) system, tumour grading is based on low- and high-grade, and especially between histological criteria. Accurate pathological WHO grade II and III, tumours was not reliably grading is essential because it determines achieved. More recently, however, assessment treatment and prognosis. Since 18F-FDG up- of uptake kinetics with a simple dynamic ac-

154 Chapter 5 – Clinical applications of PET/CT in neurology: 5.5 Brain tumours

quisition protocol has been reported to have Biopsy planning an excellent capability to discriminate be- Brain tumours are histologically often het- tween low- and high-grade gliomas [4]. Char- erogeneous; therefore, stereotactic biopsies acteristically, steadily increasing uptake until should be guided, taking into account the the end of the acquisition period suggests metabolic PET information. Nowadays PET reactive (benign) post-therapeutic changes imaging routines have been successfully im- or low-grade tumours, whereas decreasing plemented in planning systems of stereotactic curves following an early peak indicate re- brain biopsies. 18F-FDG PET-guided biopsies current glioblastoma or dedi&erentiation of have previously shown an improvement in the initially low-grade tumours. diagnostic yield by obtaining samples from areas with the highest 18F-FDG uptake. However, Prognosis 18F-FDG PET also has limitations: Targeting may EANM 18F-FDG PET results are a valuable prognos- be di+cult in lesions with hypo- or isometa- tic factor in gliomas, showing that 18F-FDG bolic glucose metabolism, such as low-grade uptake correlates well with survival [5]. It has tumours. Problems can also arise in de*ning even been claimed that 18F-FDG PET might the borderline between tumoral and cortical be better than pathological grading for de- 18F-FDG uptake when the hypermetabolic le- termination of prognosis in gliomas. Although sion is located near or within the grey matter. the prognosis depends on the histological Amino acid tracers have likewise been tested tumour grade and the correlation between and reported to be helpful for determination the tumour grade and the degree of amino of the optimal biopsy site (Fig. 2). acid tracer uptake in the late uptake phase seems poor (see above), a recent comparison of 18F-FDG PET and 11C-MET PET in gliomas has shown that 11C-MET PET results are a signi*cant prognostic factor independent of tumour grading [6]. Meanwhile there is increasing evidence that amino acid uptake is an important prognostic indicator for patients with low-grade, high-grade and also pretreated recurrent tumours, and that high uptake is statistically associated with a poor survival time.

155 18F-FDG PET scans alone have been found not to be related to survival. Thus, 18F-FDG PET seems of little use in assessing response immediately after (radiation) therapy. However, looking at longer time intervals after therapy, serial evaluation of metabolic activity with PET has provided more accurate prognostic information than a single scan [8]. Data on monitoring therapy with amino acid tracers are still limited, but promising. Both 11C-MET PET, used to monitor the e&ects of radiation therapy, systemic chemotherapy, or brachy- 125 18

Courtesy of Department of Nuclear Medicine, Municipal Hospital Karlsruhe Inc, Germany therapy with I-seeds, and F-FET PET, used Figure 2: Amino acid PET (in this example to monitor therapeutic e&ects of locoregional with 18F-FET) also reliably depicts low-grade approaches such as intralesional radioimmu- tumours in the majority of cases based on notherapy or convection-enhanced delivery increased amino acid transport. Thus, it helps of paclitaxel, have been shown to address to de*ne the metabolically most active area therapeutic e+cacy during and after these for biopsy planning even in cases where modalities. standard MRI shows only minor changes Tumour recurrence versus therapy- Several studies have reported that PET-guided induced reactive changes stereotactic biopsy provides accurate histo- During long-term follow-up, the earliest possi- logical diagnoses in most individuals and al- ble reliable di&erentiation of therapy-induced lows a reduction in the number of trajectories reactive changes (e.g. radiation necrosis) in lesions located in high-risk/functional areas from tumour recurrence is necessary in order [2,7]. to determine the most appropriate further treatment. For years, 18F-FDG PET was con- Monitoring of therapy sidered the gold standard for this di&erential Immediately after therapy, the critical question diagnosis. However, more recent studies have is whether treatment has been e&ective or questioned the e+cacy and usefulness of 18F- not. Whereas high 18F-FDG uptake preceding FDG PET for this purpose and have reported therapy is a strong indicator of more aggres- low speci*cities especially in WHO grade II or sive tumours and shorter survival, results of III tumours [9]. Increased glucose uptake in immediate (within 2 weeks) post-radiation in%ammatory tissue is considered the main

156 ewe tmu rcrec ad therapy- and recurrence tumour between reliably distinguishing for better) even (or like tracers acid Amino approaches.locoregional radiation beam therapy, radiosurgery or other high-dose after especially occur might and ovninl ramn. pr fo these from Apart treatment. conventional after changes post-therapeutic reactive and e&ects and di&erentiation of recurrent tumour ders, biopsy planning, evaluation of treatment perior properties in delineation of tumour bor- su- have may acids amino radiolabelled that increasingevidence rapidly tumours, thereis malignant transformation of initially low-grade of estimation or grading tumour for choice *rst of tracer the considered still currently is While follow-up. post-therapeutic and monitoring therapy planning, biopsy prognosis, of estimation grading, and tion delinea- tumour tumours, brain of diagnosis inf indispensable At present, the routinely used PET tracers add Conclusion andfuture perspectives currence by many studies. as the agent of choice for the diagnosis of re- Generally, amino acids have been postulated approaches. both of information prognostic synergistic and independent the combining of use combined the suggested have authors Some [3]. tients pa- glioma pretreated multimodally in even benigninduced true changes,holds this and source of false-positive false-positive of source C to hapt further enhance diagnostic accuracy by accuracy diagnostic enhance further 11 er 5 – Clinical applications of PET/CT inneurology applicationser 5–Clinical ofPET/CT C-MET and C-MET ormation for the di&erentialforthe ormation 18 18 F-FET are also well suited well also are F-FET F-FDG and amino acids amino and F-FDG 18 F-FDG PE T *ndings, T 18 F-FDG 157 marker ofangiogenesis. marker ing of alpha(v)beta(3) integrin expression as a with radiolabelled choline derivates and imag- biosynthesis membrane of imaging logues, eration of brain tumour using thymidine ana- with misonidazole derivatives, imaging prolif- hypoxia tumour imaging patients, glioma in aging gene expression following gene therapy approaches are under evaluation, such as im- many Currently e&ects. their follow to tools ment strategies may also call for new imaging otn apiain, e mlclr treat- molecular new applications, routine

EANM References Chapter 5.5

References 1. Delbeke D, Meyerowitz C, et al. Optimal cuto& levels of F-18 %uorodeoxyglucose uptake in the di&erentiation of low-grade from high-grade brain tumors with PET. Radio- logy 1995;195: 47-52.

2. Pauleit D, Floeth F, et al. O-(2-[18F]%uoroethyl)-L-tyrosine PET combined with MRI improves the diagnostic assessment of cerebral gliomas. Brain 2005; 128:678-87.

3. Popperl G, Gotz C, et al. Value of O-(2-[18F]%uoroethyl)- L-tyrosine PET for the diagnosis of recurrent glioma. Eur J Nucl Med Mol Imaging 2004; 31:1464-70.

4. Popperl G, Kreth FW, et al. Analysis of 18F-FET PET for Grading of Recurrent Gliomas: Is Evaluation of Uptake Ki- netics Superior to Standard Methods? J Nucl Med 2006;47: 393-403.

5. Patronas NJ, Di Chiro G, et al. Work in progress: [18F] %uorodeoxyglucose and positron emission tomography in the evaluation of radiation necrosis of the brain. Radiology 1982;144: 885-9.

6. Kim S, Chung JK, et al. 11C-methionine PET as a prognostic marker in patients with glioma: comparison with 18F-FDG PET. Eur J Nucl Med Mol Imaging 2005;32: 52-9.

7. Pirotte B, Goldman S, et al. Combined use of 18F- %uorodeoxyglucose and 11C-methionine in 45 positron emission tomography-guided stereotactic brain biopsies. J Neurosurg 2004;101: 476-83.

8. Schifter T, Ho&man JM, et al. Serial FDG-PET studies in the prediction of survival in patients with primary brain tumors. J Comput Assist Tomogr 1993;17: 509-61.

9. Ricci PE, Karis JP, Heiserman JE, et al. Di&erentiating recurrent tumor from radiation necrosis: time for re- evaluation of positron emission tomography? AJNR Am J Neuroradiol 1998;19:407-13.

158 to anti-epileptic drugs [1]. In these patients, these In [1]. drugs anti-epileptic to tory seizures, i.e. seizures that do not respond refrac- develop patients epileptic of 20–30% Almost 0.5–1%. approximately of prevalence a is Epilepsy making [6,7]. making the epileptic foci, impacting surgical decision- PET can provide additional information about and 5] [4, 85–90% around is region brain tic of by high-resolution MRI). The overall sensitivity tients without any brain abnormality detected (i.e.pa- patients MRI-negative in abnormality leptic foci and (2) to *nd a small area of brain (1) to lateralise and, if possible, localise the epi- uation of patients with epilepsy intractable is The role of epilepsy surgery. for candidates are who patients epileptic of assessment the for centres specialised most in used currently are that techniques two interictal and SPECT Ictal epileptogenic zone is the treatment of choice. the of removal surgical epilepsy, partial tory refrac- with patients these of number small factors for di+cult-to-control seizures [3]. In a of epilepsy and is one of the major prognostic epilepsy[2]isthemostcommontype partial Introduction Chapter 5 – Clinical applications of PET/CT in neurol Franck Petyt andGregory Semah 5.6 Epilepsy 18 F-FDG PET F-FDG in PET the detection of the epilep- 18 F-FDG PET in the presurgical eval- very common disease, with a with disease, common very 18 F-FDG PET are the PET F-FDG 159 patien Localisation oftheepileptogenic focus in temporal lobe(Fig. 1)[8,9]. mus and even occasionally to the contralateral parietal and frontal cortex as well as the thala- ipsilateral the to zone epileptogenic sumed pre- the beyond extend sometimes regions hypometabolic interictal and large very ten area of hypometabolism in TLE patients is of- The (TLE). epilepsy lobe temporal refractory interictal of sensitivity The lobe. temporal the within located is zone epileptic frequent most The ts withMRIabnormality 18 F-FDG PET is high in patients with patients in high is PET F-FDG ogy EANM Courtesy of Department of Nuclear Medicine, France Lille University Hospital,

Figure 1: Interictal FDG PET scan in a patient with TLE, showing hypometabolism involving the left temporal lobe. The hypometabolism is more pronounced in the temporopolar region but involves the mesial part of the temporal lobe and extends to the lateral part (axial image in the hippocampal plane)

This may represent the epileptic network in- PET can also be very helpful in the presurgi- volved in seizure propagation [10]. 18F-FDG cal evaluation of extra-temporal lobe epilepsy PET may show occasional lateral or medial (ETLE). Frontal lobe epilepsy (FLE) is the most temporal lobe hypermetabolism, related frequent type of ETLE and may be associated to subclinical seizures or frequent interictal with an MRI-detected cortical lesion. PET may spiking. It seems that brain glucose hypome- show hypometabolism extending beyond tabolism shown by PET is a result of several the lesion or very restricted areas of hypome- e&ects and interactions, including underly- tabolism. FLE is often associated with subtle ing pathology, neuronal loss, seizure evolution structural changes, such as cortical dysplasia and interictal changes [9]. or heterotopias, which may not be apparent

160 Chapter 5 – Clinical applications of PET/CT in neurology: 5.6 Epilepsy

on MRI but may show hypometabolism on 18F-FDG PET and surgical outcome 18F-FDG PET. In FLE, the sensitivity of 18F-FDG 18F-FDG PET has been reported to be useful in PET in localising the epileptogenic zone is predicting the outcome of epilepsy surgery. lower than in TLE [11]. Ipsilateral PET hypometabolism appears to have a high predictive value for a favourable Localisation of the epileptogenic focus in surgical outcome, even in cases of normal patients with normal MRI MRI or patients with non-localised ictal scalp 18F-FDG PET plays a very important role in EEG [12]. Patients with more severe temporal the evaluation of refractory patients, particu- lobe hypometabolism appear to have better larly when MRI or other imaging modalities post-surgical seizure control [13]. The tempo- are negative. In these patients 18F-FDG PET is ral pole seems to be a signi*cant predictor of able to locate a small area of hypometabolism the postoperative outcome [14]. However, it EANM that is associated with the epileptogenic zone remains di+cult to predict who will not be- and re%ects the presence of subtle structural come seizure-free after surgery [15]. changes, such as subtle cortical malformations which may not be apparent on MRI. 18F-FDG PET scan can help the clinician to localise these areas of cortical abnormality not detected on MRI. 18F-FDG PET has been found to be most useful in cases of negative MRI or when ictal EEG is discordant with MRI or vid- eotaped seizure semiology; in these cases it improved the positive and negative predictive values of MRI and video EEG monitoring [7].

161 References Chapter 5.6

References 1. Kwan P, Brodie MJ. Early identi*cation of refractory epi- 9. Semah F, Baulac M, Hasboun D, Frouin V, Mangin JF, Pa- lepsy. N Engl J Med 2000;342:314-9. pageorgiou S, et al. Is interictal temporal hypometabolism related to mesial temporal sclerosis? A positron emission 2. Proposal for revised classi*cation of epilepsies and tomography/magnetic resonance imaging confrontation. epileptic syndromes. Commission on Classi*cation and Epilepsia 1995;36:447-56. Terminology of the International League Against Epilepsy. Epilepsia 1989;30:389-99. 10. Chassoux F, Semah F, Bouilleret V, Landre E, Devaux B, Turak B, et al. Metabolic changes and electro-clinical pat- 3. Semah F, Picot MC, Adam C, Broglin D, Arzimanoglou A, terns in mesio-temporal lobe epilepsy: a correlative study. Bazin B, et al. Is the underlying cause of epilepsy a major pro- Brain 2004;127 (Pt 1):164-74. gnostic factor for recurrence? Neurology 1998;51:1256-62. 11. Lee JJ, Lee SK, Lee SY, Park KI, Kim DW, Lee DS, et al. Fron- 4. Ryvlin P, Bouvard S, Le Bars D, De Lamerie G, Gregoire tal lobe epilepsy: clinical characteristics, surgical outcomes MC, Kahane P, et al. Clinical utility of %umazenil-PET versus and diagnostic modalities. Seizure 2008;17:514-23. [18F]%uorodeoxyglucose-PET and MRI in refractory partial epilepsy. A prospective study in 100 patients. Brain 1998;121 12. Willmann O, Wennberg R, May T, Woermann FG, Pohl- ( Pt 11):2067-81. mann-Eden B. The contribution of 18F-FDG PET in preoperative epilepsy surgery evaluation for patients with tempo- 5. Gaillard WD, Bhatia S, Bookheimer SY, Fazilat S, Sato ral lobe epilepsy: a meta-analysis. Seizure 2007;16:509-20. S, Theodore WH. FDG-PET and volumetric MRI in the evaluation of patients with partial epilepsy. Neurology 13. Theodore WH, Sato S, Kufta C, Balish MB, Brom*eld 1995;45:123-6. EB, Leiderman DB. Temporal lobectomy for uncontrolled seizures: the role of positron emission tomography. Ann 6. Ollenberger GP, Byrne AJ, Berlangieri SU, Rowe CC, Path- Neurol 1992;32:789-94. maraj K, Reutens DC, et al. Assessment of the role of FDG PET in the diagnosis and management of children with refrac- 14. Dupont S, Semah F, Clemenceau S, Adam C, Baulac M, tory epilepsy. Eur J Nucl Med Mol Imaging 2005;32:1311-6. Samson Y. Accurate prediction of postoperative outcome in mesial temporal lobe epilepsy: a study using positron 7. Uijl SG, Leijten FS, Arends JB, Parra J, van Hu&elen AC, emission tomography with 18%uorodeoxyglucose. Arch Moons KG. The added value of [18F]-%uoro-D-deoxyglucose Neurol 2000;57:1331-6. positron emission tomography in screening for temporal lobe epilepsy surgery. Epilepsia 2007;48:2121-9. 15. Uijl SG, Leijten FS, Arends JB, Parra J, van Hu&elen AC, Moons KG. Prognosis after temporal lobe epilepsy 8. Theodore WH, Newmark ME, Sato S, Brooks R, Patronas surgery: the value of combining predictors. Epilepsia N, De La Paz R, et al. [18F]%uorodeoxyglucose positron emis- 2008;49:1317-23. sion tomography in refractory complex partial seizures. Ann Neurol 1983;14:429-37.

162 EANM T URL: www.laberdruck.at E-mail: [email protected] Tel: +43-(0)6272-7135-426,Fax +43-(0)6272-7135-499 Linzer Straße358a/1/7,1140 ·Mag. Evelynekreativ Sacher Layout andDesign: I for ofthisinformation.No responsibilityistaken thecorrectness Content: URL: www.eanm.org E-mail: [email protected] Tel: +43-(0)1-2128030,Fax: +43-(0)1-21280309 Hollandstrasse 14,1020 Vienna, Austria E Publisher: ISSN: 2079-3138 ISBN: 978-3-902785-02-2 Imprint M Laber-Druck Ges.m.b.H. Printing: URL: www.kreativ-sacher.at E-mail: [email protected] Tel: +43-(0)1/4165227,Fax: +43-(0)1/4168526 nf uropean Association ofNuclearMedicine ichael-R ormation asperdate ofpreparation:ormation August 2011 echnologist Committee ottmayr-Straße Austria 46,5110Oberndorf, Vienna, Austria 163

EANM

Principles and Practice of PET/CT Part 2 A Technologist’s Guide