AbstractAbstract 3889During the last decade, the role of Interim 2-[18F]fluoro-2-deoxy-D-glucose emission positron tomography (FDG PET-CT) has increased for Hodgkin's lymphoma (HL) evaluation and it is strongly recommended for both staging and post treatment evaluation. Early treatment response assessment is a challenge for risk-adapted therapy and interim PET-CT has emerged as a powerful prognostic tool to predict treatment outcome, a negative exam predicting a favorable disease-free survival while a positive study a worse prognosis. However, FDG PET response criteria are a matter of ongoing debate. So, the aim of our study was to confirm the prognostic value of interim FDG PET-CT performed after 4 courses of chemotherapy (ABVD) but also to compare the respective performances of the different published criteria: International Harmonization Project (IHP), Gallamini criteria, 5-point scale (Gallamini A et al. Haematologica 2006; 91: 475–481; Juweid ME et al. J Clin Oncol 25:571-578, 2007; Barrington SF et al. Eur J Nucl Med Mol Imaging 2009;36(Suppl. 2):S252; Meignan M et al. Leukemia & Lymphoma, August 2009; 50(8): 1257–1260; Gallamini A, et al. Leuk Lymphoma. 2009 Nov;50(11):1761-4.) Design and methods:From 2002 to 2006, newly diagnosed patients with HL who underwent interim PET-CT after 4 courses of ABVD were eligible for the purpose of the present study. Treatment strategy after 4 courses of ABVD was then adapted according to prognostic factors at diagnosis and interim PET-CT result. PET images were interpreted prospectively visually by at least two nuclear medicine physicians with expertise in lymphoma imaging with the criteria used before 2005: negative result was defined as no residual uptake above local background. All other findings were considered as positive. Using these evaluation criteria, interim PET positive patients who were not in CR on CT underwent autologous stem cell transplantation. Interim PET positive patients who reached CR on CT and interim PET negative patients received additional courses of ABVD (patients with advanced HD or bulky disease at diagnosis) or radiotherapy alone (stage I or II without bulky disease). Retrospectively and for the purpose of the present study, interim PET results were interpreted using the more recent criteria (Gallamini criteria, IHP criteria and 5-point scale method) Results:90 patients were included. Median age at diagnosis was 34.4 (16-71). 49% of the patients presented with B-symptoms and 50% had stage III-IV disease. According to initial criteria, 31 patients had a positive interim PET result. Among these patients, 28 underwent ASCT. Fifty-nine patients had a negative interim PET-CT. Six of 31 patients with positive interim PET-CT and 7 of 59 patients with negative interim result presented treatment failure. Thus, the negative predictive value (NPV) and positive predictive value (PPV) for predicting 2 years progression free survival (PFS) was 95% and 16%, respectively. Retrospectively, interim PET-CT was analyzed according to different recent published criteria using mediastinum and liver background as threshold for positivity. With these criteria, NPV remained very high (from 95% to 96%). However, the PPV increased from 25% to 55% according to the threshold used. Taken together, the statistical analysis revealed that Interim PET-CT was significantly correlated with PFS only for two methods: the Gallamini criteria and the 5-point scale method. Conclusion:The first part of our study using the criteria used before 2005 shows the high NPV of interim PET-CT for predicting treatment outcome in HL. However, better prognostic values are obtained by using higher threshold for positivity such as liver background even after 4 cycles of ABVD. Indeed, our analysis underscores the better results reached by the Gallamini criteria and 5-point scale methods. The better method to predict PFS was obtained by the 5-point scale (scores 1, 2, 3 and 4 were classified as negative while only score 5 was classified as positive) showing a PPV and NPV of 55% and 96%, respectively. Disclosures:No relevant conflicts of interest to declare.
Read full abstract