INTEGRATION BETWEEN METABOLIC TUMOUR VOLUME AND METABOLIC HETEROGENEITY PREDICTS OUTCOME OF DLBCL LYMPHOMA PATIENTS IN THE SAKK 38/07 STUDY COHORT

Introduction: Recent studies suggest that the metabolic tumor volume (MTV) assessed in baseline (18) F-fluorodeoxyglucose (18FDG) positron emission tomography/computed tomography (PET/CT) images may predict the prognosis of patients with DLBCL. The present study was aimed at assessing the ability of the baseline functional PET parameters considered alone or in association in predicting the efficacy of conventional immunochemotherapy treatment administered every 14 (R-CHOP14) or 21 days (R-CHOP21) in two cohorts (testing and validation sets) of DLBCL patients. Here, we report results of the testing set. Methods: The testing set included 141 DLBCL patients treated with R-CHOP14 in the prospective SAKK38/07 study with available baseline PET/CT data. The lymphoma lesions were segmented by a fixed threshold algorithm, with a 2.5 SUVmax value as cut-off to estimate the metabolic tumor volume (MTV). Maximum standardized uptake value (SUVmax), and total lesion glycolysis (TLG) were estimated. MH was also measured using the area under the curve of cumulative SUV-volume histogram (AUC-CSH). The optimal MTV cut off values to discriminate subgroups with different progression free (PFS) and overall survival (OS) have been defined by ROC curves analysis. Cell of origin was determined by immunohistochemistry (Hans algorithm). Results: After a median follow-up of 63 months, 30 progressions and 23 deaths were recorded. Among quantitative PET parameters baseline MTV was the most powerful predictor of outcome (Cox model p< 0.001 for PFS and OS). At 5 years, PFS was 83% for patients with low MTV vs. 59% for those with high MTV (log-rank test, p=0.0005). OS was 91% vs. 64%, respectively (log-rank test, p=0.0001). High MTV values predicted shorter PFS only in patients with non-germinal center B (GCB) subtype (n = 84) but not in patients with GCB subtype (n = 29) (log-rank test p = 0.006 vs. 0.7). In patients with high MTV, the presence of high MH increased the prognostic discrimination and the combined analysis of MTV and MH demonstrated that increased values of these two parameters might be used to identify patients with different risk of progression (Figure 1). Patients with both high MTV and high MH had a 5-years PFS of 42%, while those with both low MTV and low MH had a 5-year PFS of 83% (log-rank p< 0.0001). Keywords: diffuse large B-cell lymphoma (DLBCL); positron emission tomography (PET); R-CHOP.