Received: 21 March 2022 | Revised: 12 May 2022 | Accepted: 28 May 2022 DOI: 10.1002/cam4.4924 B R I E F C O M M U N I C AT I O N FDG-­PET/CT-­guided rebiopsy may find clinically unsuspicious transformation of follicular lymphoma Aino Rajamäki1,2 | Hanne Kuitunen3 | Marc Sorigue4 | Salla-­Maarit Kokkonen5 | Outi Kuittinen2,6,7 | Kaisa Sunela8 1 Department of Oncology, Hospital Nova of Central Finland, Jyväskylä, Aim: The purpose of the study was to evaluate the clinical impact of Finland fluorodeoxyglucose-­positron emission tomography/computed tomography 2 Institute of Clinical Medicine, Faculty (FDG-­PET/CT) followed by a new biopsy from the site with maximum standard- of Health Medicine, University of Eastern Finland, Kuopio, Finland ized uptake value (SUVmax) in case of high maximal SUV values, in detecting 3 Department of Oncology, Oulu clinically unsuspected histologic transformations (HT) of follicular lymphoma University Hospital, Oulu, Finland (FL). 4 Department of Hematology, ICO-­ Methods: This retrospective study included all the patients who had undergone Hospital Germans Trias i Pujol, FDG-­PET/CT during primary diagnosis or relapse of FL between 2010 and 2020 Functional cytomics-­IJC, Badalona, Spain at Oulu University Hospital. 5 Department of Nuclear Medicine, Results: The diagnosis changed from an indolent disease to a transformed lym- Oulu University Hospital, Oulu, phoma in >10% (7/63) of the patients who underwent diagnostic FDG-­PET/CT. Finland 6 The HT risk associated with high SUVmax (>10) was 24% (7 of 29 performed bi- Medical Research Centre and Cancer and Translational Research Unit, opsies). Four out of these seven patients with verified HT had no previous clinical University of Oulu and Oulu University suspicion of transformation. Hospital, Oulu, Finland 7 Conclusion: Our results suggest that a rebiopsy based on a high SUVmax in di- Department of Oncology, Kuopio University Hospital, Kuopio, Finland agnostic FDG-­PET/CT is valuable in detecting clinically unsuspected HT of FL. 8 Department of Oncology, Tays Cancer KEYWORDS Center, Tampere University Hospital, Tampere, Finland FDG-­PET/CT, follicular lymphoma, histologic transformation, indolent lymphoma, SUVmax Correspondence Aino Rajamäki, Department of Oncology, Hospital Nova of Central Finland, Keskussairaalantie 19, 40620 Jyväskylä, Finland. Email: aino.rajamaki@hotmail.com Funding information Finnish Blood Disease Research Foundation; Northern-­Savonia Cancer Society This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. Cancer Medicine. 2023;12:407–411.  wileyonlinelibrary.com/journal/cam4   | 407 | 20457634, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4924 by Readcube (Labtiva Inc.), Wiley Online Library on [24/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 408    RAJAMÄKI et al. 1 | I N T RO DU CT ION Whole-­body PET and CT images were loaded on three-­dimensional workstations for visual evaluation and Follicular lymphoma (FL) may undergo histologic trans- data analysis (GE healthcare AW workstation or Hermes formation (HT) during the course of the disease. The Hybrid Viewer). PET/CT images were reviewed by expe- 10-­year cumulative incidence of biopsy-­proven transfor- rienced radiologists. Areas of highest abnormal FDG up- mation is approximately 8%.1,2 HT often occurs locally in take in lymphoma lesions were identified and the highest a limited number of disease sites and an indolent histol- SUV (SUVmax) within a given 3D region of interest was ogy from one site does not exclude the presence of a high-­ reported. grade disease elsewhere.3 An SUVmax of >10 was mainly considered the indica- The outcome of patients with FL has improved greatly tion for a new biopsy and rebiopsy was performed at an over the last years and the 10-­year overall survival is al- accessible site with the highest SUVmax. Because of the most 80%.4 However, HT is associated with increased real-­life nature of the study, three biopsies were taken also lymphoma-­related mortality.1 As FL may be treated with with a lower SUVmax and six patients were not re-­biopsied low-­intensity regimens that are insufficient for aggressive although their SUVmax exceeded 10 (n = 6) (Figure S1). A disease, it may be speculated that differentiating cases receiver operating characteristic (ROC) curve was used to with HT from those with purely indolent disease would determine the optimum cutoff for discriminating between have a positive impact on the survival. Among FL patients HT and no HT. Statistical analyses were performed using with early progression, the proportion of those patients IBM SPSS Statistics (version 27; IBM Corp.). with an undiscovered primary transformation also needs to be determined. The aim of this study was to retrospectively evaluate the 3 | RESULTS efficacy of FDG-­PET/CT followed by a new biopsy from the SUVmax area in detecting clinically unsuspected HTs. Patient characteristics are presented in Table 1. Patients were diagnosed between 1988 and 2020. The median fol- low-­up duration after primary diagnosis was 38 months 2 | M ET H ODS (range: 1–­382 months). Of 63 FL patients with a diagnos- tic FDG-­PET/CT done, a SUVmax >10 was noticed in 35 This retrospective study was performed in line with the patients, and rebiopsy was taken in 29 of those. HT was principles of the Declaration of Helsinki and it was re- found in seven out of 29 rebiopsies (24.1%), out of which viewed and approved by the authorities of the Northern six presented with diffuse large B-­cell lymphomas and one Ostrobothnia Hospital District. All patients with a diagno- with double-­/triple-­hit lymphoma histology. The median sis of Grade 1 to Grade 3a FL, who had at least one visit to SUVmax was higher in patients who demonstrated HT the oncology department in Oulu University Hospital be- compared to those without HT but SUVmax >10 and a tween 2010 and 2020 were screened. Patients who had un- new biopsy (27.1 (range: 10.5–­34.9) vs. 13.6 (range: 10.2–­ dergone diagnostic FDG-­PET/CT at diagnosis or relapse 24.4), respectively) (Figure 1). The optimum SUVmax were included in the data. Principally, FDG-­PET/CT was cutoff value obtained with ROC analysis was 26.5, with a performed for all new patients with a diagnosis of FL or sensitivity for HT of 0.86 and specificity of 1.0. its relapse, but the practice varied between clinicians and The median time from diagnosis to transformation was became more common during the last 5 years. Altogether, 25 months (range: 0–­183 months). Two HTs (28.6%) were 397 patients were identified. Among these, 63 had under- discovered at diagnosis, four (57.1%) at first relapse, and gone diagnostic FDG-­PET/CT and were deemed eligible one (14.3%) at second relapse. Over half of the new bi- for inclusion (Figure S1). opsies with HT (4/7) were obtained from the abdominal Patients fasted for 6 h prior to injection of FDG. A lymph nodes. Other sites (one each) included the cervi- dose of 3.43 MBq/kg of 18F-­FDG was injected intrave- cal, supraclavicular, and axillary lymph nodes. Of the nously 60 min before PET imaging. PET/CT examination non-­HT biopsies, 24.0% were obtained from the abdomi- was performed using a combined PET/CT device (GE nal, 16.0% from the cervical, and 12.0% from the axillary Discovery 690). The examination started with a low-­dose lymph nodes. The remaining biopsies were obtained from non-­enhanced CT scan from the level of the skull base the thoracic, inguinal, and supraclavicular lymph nodes, to the upper thigh regions. СT acquisition was followed bone, kidney, and miscellaneous sites. by the acquisition of PET emission images of 3 min per Among seven patients with proven HT, in four patients bed position using а matrix of 192 × 192. PET images were there was no previous clinical suspicion of transformation reconstructed using iterative algorithms and attenuation (one PET image is shown as Figure S2). HT was clinically correction was routinely applied. suspected in two patients, but the first biopsy showed | 20457634, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4924 by Readcube (Labtiva Inc.), Wiley Online Library on [24/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License RAJAMÄKI et al.    409 T A B L E 1 Patient characteristics obtained at the time of the TABLE 1 (Continued) index FDG-­PET/CT performed (at the time of the possible HT) HT (n = 7) No HT HT (n = 7) No HT Variable n (%) (n = 56) n (%) Variable n (%) (n = 56) n (%) Treatment following FDG-­PET/CT Age, years, median (range) 68 (59–­88) 60 (29–­87) Anthracycline containing 6 (85.7) 16 (28.6) Sex, male 3 (42.9) 26 (46.4) Bendamustine 0 17 (30.4) Primary diagnosis Rituximab monotherapy 0 3 (5.4) FL Grade 1 or 2 4 (57.1) 47 (83.9) Radiotherapy only 0 5 (8.9) FL Grade 3a 3 (42.9) 9 (16.1) Watchful waiting 0 13 (23.2) Stage Othera 1 (14.3) 2 (3.6) I–­II 1 (16.7) 18 (32.1) Abbreviations: B-­symptoms, systemic symptoms (unexplained weight III–­IV 5 (83.3) 38 (67.9) loss, fever, night sweats); ECOG, Eastern Cooperative Oncology Group Missing 1 0 performance status; FDG-­PET/CT, fluorodeoxyglucose-­positron emission tomography/computed tomography; FLIPI, Follicular Lymphoma FLIPI International Prognostic Index; LDH, lactate dehydrogenase. 0–­1 0 14 (25.9) a Other treatments included R-­DHAP (rituximab, dexamethasone, 2 0 18 (33.3) cytarabine, and cisplatin) in the group of HT patients, and O-­MINE (ofatumumab, mesna, ifosfamide, mitoxantrone, etoposide), and 3–­5 5 (100.0) 22 (40.7) lenalidomide (1 each) in the group of patients with no HT. Missing 2 2 Bone marrow involvement Yes 1 (14.3) 15 (29.4) Missing 0 5 Extranodal involvement Yes 3 (50.0) 15 (28.3) Missing 1 3 LDH level Normal 2 (28.6) 23 (46.0) Elevated 5 (71.4) 27 (54.0) Missing 0 6 ECOG 0 2 (28.6) 24 (55.8) 1 4 (57.1) 17 (39.5) 2 1 (14.3) 1 (2.3) 3 0 1 (2.3) Missing 0 13 Hemoglobin level, g/L F I G U R E 1 A comparison of individual SUVmax values <120 2 (28.6) 6 (11.1) between patients with histological transformation (HT) versus no ≥120 5 (71.4) 48 (88.9) HT, with the sensitivity and specificity for HT of the threshold 26.5. The graph includes patients with SUVmax <10 and a new biopsy Missing 0 2 (n = 29). B-­symptoms Yes 3 (42.9) 15 (28.3) Missing 0 3 indolent histology and the diagnosis of HT was confirmed The point when FDG-­PET/CT was made on rebiopsy after FDG-­PET/CT. In one patient, HT was Diagnosis 2 (28.6) 33 (60.0) clinically suspected and the biopsy was obtained directly based on FDG-­PET/CT. First relapse 4 (57.1) 16 (29.1) The median follow-­up after FDG-­PET/CT was 19 Second relapse or later 1 (14.3) 6 (10.9) months (range: 0–­97 months). During surveillance three Missing 0 1 other patients experienced HT. One HT was diagnosed within 6 months of the FDG-­PET/CT and two were diag- (Continues) nosed >1 year after the investigation. Median SUVmax of | 20457634, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4924 by Readcube (Labtiva Inc.), Wiley Online Library on [24/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 410    RAJAMÄKI et al. these patients in previous FDG-­PET/CT was 9.0 (range: tabsence of response to frontline therapy, and Grade 3a 8.9–­9.8). There were six patients who were not re-­biopsied FL. The use of rituximab reduced the risk of HT.1,2,7,8 although their SUVmax exceeded 10. In one patient, the Similar findings were also seen in our material (Table 1). initial biopsy was taken from the SUVmax area and the Moreover, features associated with clinical suspicion of biopsy was not repeated. In two patients, lymphoma HT include rapid growth of tumor masses, the rapid de- was treated as a transformed disease without a histolog- velopment of symptoms, and/or clinical deterioration ical verification. One patient was taking part in a clinical or change in the tempo of a previously indolent disease study, and in two patients the reason was not found in course.3 Importantly, four out of seven patients had no medical records retrospectively. During the follow-­up, no previous clinical suspicion of transformation in our study. HTs were histologically confirmed among these patients. The present study has some limitations. The study pop- Fifty out of 63 patients received treatment after FDG-­ ulation was small with only seven observed HTs, leaving PET/CT (Table 1). Among these, 76.0% received immu- room for a possibility of bias. The median follow-­up was nochemotherapy, 6.0% received rituximab monotherapy, short so the potential survival advantage of an FDG-­PET/ and one patient (2.0%) received maintenance rituximab. CT cannot be analyzed. New biopsies were mainly ob- Nine patients showed disease progression during fol- tained only from the patients whose SUVmax exceeded low-­up after FDG-­PET/CT. The 5-­year progression-­free 10. Hence, we cannot exclude the possibility of some HTs survival (PFS) was 37.9% and the median PFS was 52 with lower SUVmax values. This protocol was based on months. There were no lymphoma-­related deaths, while the existing literature suggesting that a high SUVmax two deaths were due to other reasons. (>10) on FDG-­PET/CT predicts aggressive lymphoma and therefore, could also predict transformation.9,10 Indeed, the optimum cutoff value obtained with ROC analysis 4 | DI S C USSION was surprisingly high, 26.5. However, considering the lim- ited number of patients included in this study, we spec- In this retrospective study of 63 patients with FL, a rebi- ulate that a portion of HTs might not be detected with opsy based on a high SUVmax in FDG-­PET/CT was valua- these high SUVmax cutoffs. Among the 29 patients with ble in detecting HT. As HT is coupled with poor prognosis, a SUVmax >10 and a new biopsy, 24.1% were diagnosed revealing patients with HT at an early stage is essential. with HT, suggesting that in cases where SUVmax exceeds Very few studies have evaluated the role of FDG-­PET/ 10, a rebiopsy from the area with the highest SUVmax CT in detecting clinically unsuspected transformations in might be beneficial. patients with FL. Noy et al,5 and Karam et al6 discovered The encouraging preliminary results of the benefits of a higher SUVmax in transformed tumors than in non-­ diagnostic FDG-­PET/CT followed by a new biopsy from transformed tumors. However, a recent study by Mir et al7 the SUVmax area in detecting HT call for further research including 549 patients reported that neither the intensity with larger populations and longer follow-­up periods. This of the FDG uptake (SUVmax) nor the variation in the up- would enable the analysis of the potential survival advan- take level (SUVrange) of de novo FL patients predicted tage offered by diagnostic FDG-­PET/CT and rebiopsy. subsequent HT. These findings suggest that there might be only a minor benefit in performing a rebiopsy to exclude AUTHOR CONTRIBUTIONS HT based on SUVmax alone before initiating the therapy. Aino Rajamäki, Outi Kuittinen, and Hanne Kuitunen However, in that study, patients with biopsy-­confirmed designed the research. Aino Rajamäki collected the data. HT based on PET findings at baseline were not included. Aino Rajamäki, Salla-­Maarit Kokkonen, Outi Kuittinen, The median time between baseline PET scan and biopsy-­ and Kaisa Sunela analyzed the results and wrote the origi- confirmed HT was 11.1 months. Most probably, transfor- nal draft. Aino Rajamäki and Marc Sorigue made the fig- mations were not present at baseline. All patients were ures. Hanne Kuitunen and Marc Sorigue reviewed and treated with immunochemotherapy, which might have edited the manuscript. Outi Kuittinen and Kaisa Sunela treated a proportion of undetected transformations. supervised the study. All the authors accepted the final HT is often suspected based on clinical presentation. version of the manuscript. Factors associated with an increased risk of transforma- tion include age older than 60 years, elevated lactate de- FUNDING INFORMATION hydrogenase, high Follicular Lymphoma International This work was funded by the Finnish Blood Disease Prognostic Index scores, B-­symptoms, an advanced Research Foundation and Northern-­Savonia Cancer disease, more than one extranodal site, poor Eastern Society (A.R.). Other authors disclose no special funding Cooperative Oncology Group performance status (>1), for this research. | 20457634, 2023, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4924 by Readcube (Labtiva Inc.), Wiley Online Library on [24/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License RAJAMÄKI et al.    411 5. Noy A, Schöder H, Gönen M, et al. The majority of transformed CONFLICT OF INTEREST lymphomas have high standardized uptake values (SUVs) on The authors have no relevant financial or non-­financial positron emission tomography (PET) scanning similar to diffuse interests to disclose. large B-­cell lymphoma (DLBCL). Ann Oncol. 2009;20(3):508-­ 512. doi:10.1093/annonc/mdn657 DATA AVAILABILITY STATEMENT 6. Karam M, Novak L, Cyriac J, Ali A, Nazeer T, Nugent F. Role The datasets generated during the current study are of fluorine-­18 fluoro-­deoxyglucose positron emission tomogra- available from the corresponding author for reasonable phy scan in the evaluation and follow-­up of patients with low-­ grade lymphomas. Cancer. 2006;107(1):175-­183. doi:10.1002/ requests. cncr.21967 7. Mir F, Barrington SF, Brown H, et al. Baseline SUVmax did not ETHICAL APPROVAL AND PATIENT predict histological transformation in follicular lymphoma in CONSENT the phase 3 GALLIUM study. Blood. 2020;135(15):1214-­1218. No consent from the Ethical Board or patients was re- doi:10.1182/blood.2019001091 quired for the study, as Finnish and EU legislation allows 8. Wagner-­Johnston ND, Link BK, Byrtek M, et al. Outcomes the use of register-­based data for retrospective studies of transformed follicular lymphoma in the modern era: a re- port from the National LymphoCare Study (NLCS). Blood. without the informed consent of the patients concerned. 2015;126(7):851-­857. doi:10.1182/blood-­2015-­01-­621375 Permission to use patient-­related data was applied from 9. Ngeow JYY, Quek RHH, Ng DCE, et al. High SUV uptake on the hospital authorities responsible for that data. FDG-­PET/CT predicts for an aggressive B-­cell lymphoma in a prospective study of primary FDG-­PET/CT staging in lymphoma. ORCID Ann Oncol. 2009;20(9):1543-­1547. doi:10.1093/annonc/mdp030 Aino Rajamäki https://orcid.org/0000-0003-1105-3549 10. Schöder H, Noy A, Gönen M, et al. Intensity of 18fluoro- Outi Kuittinen https://orcid.org/0000-0001-9969-0053 deoxyglucose uptake in positron emission tomography dis- tinguishes between indolent and aggressive non-­Hodgkin's lymphoma. J Clin Oncol. 2005;23(21):4643-­4651. doi:10.1200/ REFERENCES JCO.2005.12.072 1. Federico M, Caballero Barrigón MD, Marcheselli L, et al. Rituximab and the risk of transformation of follicular lym- phoma: a retrospective pooled analysis. Lancet Haematol. SUPPORTING INFORMATION 2018;5(8):e359-­e367. doi:10.1016/S2352-­3026(18)30090-­5 Additional supporting information may be found in the 2. Alonso-­Álvarez S, Magnano L, Alcoceba M, et al. Risk of, online version of the article at the publisher’s website. and survival following, histological transformation in follic- ular lymphoma in the rituximab era. A retrospective multi- centre study by the Spanish GELTAMO group. Br J Haematol. 2017;178(5):699-­708. doi:10.1111/bjh.14831 3. Mikhaeel NG. Hotter is not more aggressive: baseline How to cite this article: Rajamäki A, SUVmax in FL. Blood. 2020;135(15):1191-­1192. doi:10.1182/ Kuitunen H, Sorigue M, Kokkonen S-M, blood.2020004986 Kuittinen O, Sunela K. FDG-­PET/CT-­guided 4. Bachy E, Seymour JF, Feugier P, et al. Sustained progression-­ rebiopsy may find clinically unsuspicious free survival benefit of rituximab maintenance in patients with follicular lymphoma: long-­term results of the PRIMA transformation of follicular lymphoma. Cancer study. J Clin Oncol. 2019;37(31):2815-­2824. doi:10.1200/JCO.​ Med. 2023;12:407-411. doi: 10.1002/cam4.4924 19.01073