Apr 2, 2011 - ... London and Surrey, UK; 5Institut Catala` d'Oncologia, Barcelona, Spain; 6BC Cancer ..... Our data do n
Annals of Oncology
original articles Annals of Oncology 23: 59–64, 2012 doi:10.1093/annonc/mdr052 Published online 2 April 2011
2-18fluoro-deoxy-D-glucose positron emission tomography (FDG-PET) for postchemotherapy seminoma residual lesions: a retrospective validation of the SEMPET trial M. Bachner1, Y. Loriot2, M. Gross-Goupil2, P. A. Zucali3, A. Horwich4, J.-R. Germa-Lluch5, C. Kollmannsberger6, F. Stoiber7, A. Fle´chon8, K. Oechsle9, S. Gillessen10, J. Oldenburg11, G. Cohn-Cedermark12, G. Daugaard13, F. Morelli14, A. Sella15, S. Harland16, M. Kerst17, J. Gampe18, C. Dittrich1, K. Fizazi2 & M. De Santis1* 1
ACR-ITR VIEnna/CEADDP, LBI-ACR VIEnna, and KFJ-Spital, Vienna, Austria; 2Institut Gustave Roussy, Villejuif, France; 3Istituto Clinico Humanitas IRCCS, Rozzano (Milan), Italian Germ Cell Cancer Group; 4The Royal Marsden Hospital, London and Surrey, UK; 5Institut Catala` d’Oncologia, Barcelona, Spain; 6BC Cancer AgencyVancouver Cancer Center, Vancouver, Canada; 7Krankenhaus der Barmherzigen Schwestern, Linz, Austria; 8Centre Leon Berard, Lyon, France; 9Universita¨tsklinikum Hamburg-Eppendorf, Hamburg, Germany; 10Kantonsspital St. Gallen, St. Gallen, Switzerland; 11The Norwegian Radium Hospital, Oslo, Norway; 12Department of Oncology-Pathology, Karolinska Institute and University Hospital, Stockholm, Sweden; 13Department of Oncology, 5073 Rigshospitalet, Copenhagen, Denmark; 14 Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy; 15Assaf Harofeh Medical Center, Zerifin, Israel; 16University College Hospital London, London, UK; 17The Netherlands Cancer Institute, Amsterdam, The Netherlands; 18Max Planck Institute for Demographic Research, Rostock, Germany
Received 9 December 2010; revised 1 February 2011; accepted 1 February 2011
Background: 2-18fluoro-deoxy-D-glucose positron emission tomography (FDG-PET) has been recommended in international guidelines in the evaluation of postchemotherapy seminoma residuals. Our trial was designed to validate these recommendations in a larger group of patients. Patients and methods: FDG-PET studies in patients with metastatic seminoma and residual masses after platinum-containing chemotherapy were correlated with either the histology of the resected lesion(s) or the clinical outcome. Results: One hundred and seventy seven FDG-PET results were contributed. Of 127 eligible PET studies, 69% were true negative, 11% true positive, 6% false negative, and 15% false positive. We compared PET scans carried out before and after a cut-off level of 6 weeks after the end of the last chemotherapy cycle. PET sensitivity, specificity, negative predictive value (NPV), and positive predictive value were 50%, 77%, 91%, and 25%, respectively, before the cut-off and 82%, 90%, 95%, and 69% after the cut-off. PET accuracy significantly improved from 73% before to 88% after the cut-off (P = 0.032). Conclusion: Our study confirms the high specificity, sensitivity, and NPV of FDG-PET for evaluating postchemotherapy seminoma residuals. When carried out at an adequate time point, FDG-PET remains a valuable tool for clinical decision-making in this clinical setting and spares patients unnecessary therapy. Key words: chemotherapy, FDG-PET, follow-up, postchemotherapy residual lesions, seminoma, testicular cancer
introduction In patients with metastatic seminoma, postchemotherapy residual masses are present in 55%–80% [1–9]. Resection of these residuals mostly reveals necrotic tissue [4, 6, 8, 10]. In lesions >3 cm, viable tumor is expected in 11%–37% of cases [7, 11–15]. Surgery is technically demanding due to fibrosis and desmoplastic reactions and it is often incomplete and associated with increased morbidity [2, 10, 11, 16–21]. The optimal management of postchemotherapy seminoma residual masses *Correspondence to: Dr M. De Santis, ACR-ITR VIEnna/CEADDP, LBI-ACR VIEnna, and KFJ-Spital, Kundratstrasse 3, A-1100 Vienna, Austria. Tel: +43-1-60191-2308; Fax: +43-1-60191-2329; E-mail:
[email protected]
remains controversial. Some centers perform surgery for all lesions >3 cm [3, 4, 7], while others opt for observation and use salvage treatment only if the lesions fail to shrink or for clearly documented relapse [8, 17]. To date, three prospective and two retrospective series evaluating 2-18fluoro-deoxy-D-glucose positron emission tomography (FDG-PET) in this setting have been published, showing conflicting results [12, 14, 22–24]. The largest prospective trial (SEMPET trial) included 56 PET studies with eight positive scans—all of them true positive (TP) [14]. The results of this trial with a specificity of 100% and a sensitivity of 80% were the basis of recommendations in several international oncological and urological guidelines [25–29].
ª The Author 2011. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email:
[email protected] Downloaded from https://academic.oup.com/annonc/article-abstract/23/1/59/162131/2-18fluoro-deoxy-D-glucose-positron-emission by guest on 06 October 2017
original articles The objectives of using FDG-PET more frequently in this clinical setting were the following: (i) to have a more reliable tool than just computed tomography (CT)-based discrimination of residual tumor size for predicting necrosis/ fibrosis versus viable tumor, (ii) to spare patients unnecessary surgery, and (iii) to diagnose residual viable tumor early. However, with the more frequent use of PET, a higher than expected number of false positive (FP) scans was reported. This concern prompted us to design the current ‘SEMPECON’ trial in order to retrospectively validate the current PET recommendations in a larger group of patients.
patients and methods patients Consecutive patient data from 19 oncological centers in Europe, Israel, and Canada were contributed to this retrospective study. Patients with metastases of pure seminoma (testicular or extragonadal) after first-line or salvage platinum-containing chemotherapy who had completed chemotherapy after December 2001 (to ensure that recent CT and PET scan techniques were used) were included. Inclusion criteria were as follows: negative tumor markers (human chorionic gonadotropin and lactate dehydrogenase) on completion of chemotherapy, CT evidence of at least one clearly defined and measurable residual mass ‡1 cm in greatest diameter on completion of chemotherapy, visual interpretation of a postchemotherapy FDG-PET scan, and postchemotherapy management with either observation of at least 24 months or until relapse or surgery of the residual lesion(s). Patients after irradiation of residual lesions or with primary tumors containing any nonseminomatous elements, an elevated serum a-fetoprotein at any time, or earlier publication of the same FDGPET results were excluded from the study.
methods The protocol required anonymized data of consecutive eligible patients. PET results were classified by the local centers as positive, negative, or—whenever a clear discrimination was impossible—equivocal. Equivocal scans were not included in this analysis (see Figure 1). Following the original SEMPET trial protocol [13, 14], visual interpretation was chosen rather than standardized uptake value (SUV)based interpretation of PET scan results because of the lack of a clearly defined, ubiquitous accepted SUV cut-off as well as the general reservations against SUV as the sole criterion for classifying PET scans [30–32]. The postchemotherapy management of the residual lesions, i.e. surgery or
Annals of Oncology
observation, was left to the discretion of the treating centers and the patients’ preference. Clinical follow-up included tumor marker assessment and imaging techniques such as chest X-ray, CT, and/or ultrasound of the abdomen carried out according to the treating centers’ routine followup schedule. Positive PET scans were rated TP, if either viable tumor was found histologically or progression was seen clinically or on CT; all other positive PET scans were rated FP. Negative PET scans with necrosis or fibrosis in the resected specimen and/or no clinical or radiological signs of disease progression during a minimum follow-up period of 24 months were rated true negative (TN); negative PET scans with viable tumor in the resected residual lesion and/or clinical or radiological progression were rated false negative (FN). The results of FDG-PET studies were correlated with either the histology of the resected lesion (presence or absence of viable residual tumor tissue) or the outcome on clinical/radiological follow-up after at least 24 months, as late recurrences after 2 years are very rare in seminoma patients [33, 34]. Follow-up was calculated from the day of the PET study to either the date of the last follow-up visit or to the date of relapse. Formal approval of this retrospective noninterventional trial by local ethics committees was left to the local investigators; it was not required according to the regulations of the ethics committee of the city of Vienna. As the timing of PET after completion of chemotherapy is considered an important factor [32, 35–45], we grouped all PET scans according to the interval between day 1 of the last chemotherapy cycle and the date of the PET scan. According to the most common and precise guidelines, we chose a cut-off of 6 weeks after day 21 of the last chemotherapy cycle (i.e. day 63 of the last chemotherapy cycle) and compared the results of the PET scans before (inadequate timing) and after (adequate timing) that cut-off.
results Between December 2008 and May 2010, 177 consecutive PET studies from 19 centers were evaluated. Of these, data of 127 scans were eventually included into the final analysis. Fifty PET scans were excluded from the final analysis. The processing of data and reasons for exclusion are outlined in Figure 1. The 127 eligible PET scans derived from 125 patients, as two patients were included with two PET scans each after first-line and salvage dose chemotherapy, respectively. To simplify matters, we use the term ‘patients’ below. Patient characteristics are summarized in Table 1. The median follow-up for all patients was 34 months (range: 0–88 months). For patients who underwent surgery, median follow-up was 31 months (range: 2–85 months). Median Table 1. Patient characteristics, n = 127 n (%) Primary tumor
Intention of chemotherapy Dosing of chemotherapy Lesion size on CT
Figure 1. Data processing. fup, follow-up; pos, positive; neg, negative.
60 | Bachner et al.
Gonadal Mediastinal Retroperitoneal First-line Salvage Conventional High-dose
