Preliminary Results of FDG-PET Scanning after GDP Chemotherapy Prior to Autologous Stem Cell Transplant (ASCT) for Relapsed/Refractory (RR) Lymphoma

Introduction: Gemcitabine, dexamethasone and cisplatin (GDP) has become a standard salvage chemotherapy (SC) regimen for relapsed/refractory (RR) lymphoma prior to autologous stem cell transplantation (ASCT) (Crump JCO 2014). Response to SC is now evaluated using 18-Fluoro-deoxyglucose positron-emission-tomography (FDG-PET) scans based on the recent Lugano classification (Cheson JCO 2014). We evaluated the response of patients (pts) with Hodgkin lymphoma (HL) and aggressive non-Hodgkin lymphoma (NHL) to GDP by PET-CT scans and attempted to determine whether PET was predictive of outcome. Methods: We performed a retrospective chart review of consecutive HL and DLBCL (diffuse large B-cell lymphoma) pts who underwent ASCT following GDP SC at our centre between January 2014 and July 2016. All pts previously received anthracycline-based chemotherapy (typically ABVD for HL or R-CHOP for NHL) for primary treatment. Pts underwent FDG-PET scans after 3 cycles of SC and scans were retrospectively reported with the Deauville Criteria Scale with Deauville scores of 1-3 considered negative, whereas scores of 4 or 5 represented a positive result. Response to GDP was documented with CT scans using 1999 Working group Criteria (JCO 1999). Pts proceeded to ASCT if they had a PR by CT imaging and no signs of PD on PET-CT. Post-PET radiation or additional chemotherapy could be given at the discretion of the treating physician. Progression free survival (PFS) and overall survival (OS) were calculated by the Kaplan-Meier method from the date of progression or date of ASCT. Statistical analysis to determine the significance of PET result on outcome was performed using the long rank test. Results: 45 pts with DLBCL and 29 pts with HL were identified: median age was 45 (range: 23-69) and 29 (range: 19-58) years respectively. Baseline patient characteristics are listed in Table 1. Following GDP SC, PET Deauville scores of 1-3 were reported in 55% of HL and 29% of DLBCL pts. Overall response rate to GDP by CT was 72% for HL pts and 64% for DLBCL pts (Table 2). For HL, 100% (16/16) of pts with a PET(-) scan and 91% (10/11) with a PET(+) result proceeded to ASCT. 85% (11/13) of DLBCL pts who had a PET(-) scan and 48% (15/31) with a PET(+) scan underwent ASCT. Additional therapy post PET scan included involved field radiation, additional chemotherapy (mini-BEAM) or novel therapy (brentuximab and bendamustine for HL patients). Median follow-up time was 14.2 months (HL) and 13.4 months (DLBCL) from relapse/progression after initial chemotherapy; and 9.6 months (HL) and 10.9 months (DLBCL) from ASCT. There was a statistically significant difference between PET(+) and PET(-) DLBCL patients from the time of initial disease progression for PFS (26% versus 69%, p=0.011) that did not reach statistical significance for OS (p=0.072). However, there was no significant difference for PFS and OS in DLCBL from ASCT when stratified by PET result (p= 0.154 and p=0.723 respectively). In HL pts, no statistically significant differences for PFS and OS from progression (p=0.480 and p=0.387 respectively) or from ASCT (p=0.579 and p=0.450 respectively) were found when stratified by PET result (see Tables 3 and 4). Conclusion: Deauville 4-5 PET scores appear to be predictive of PFS for RR-DLBCL pts. The lack of a significant difference between PET results for OS in DLBCL is likely due to sample size and the need for longer follow-up. For RR-HL pts, PFS and OS rates were both high but did not seem to noticeably differ by PET result. Additional therapy employed for Deauville 4-5 patients peri-ASCT could influence outcome. Overall, PET response assessment post GDP appears to stratify outcome in patients with DLBCL and PET adapted therapy in HL patients may improve outcome in Deauville 4-5 patients. Review of pending cases and subsequent follow-up is ongoing. Disclosures No relevant conflicts of interest to declare.