ISHL10 Abstract T002

Overall survival (OS) impact of BEACOPP versus ABVD in advanced Hodgkin Lymphoma: a pooled analysis of 4 randomized trials

Introduction. In 4 individual randomized trials comparing ABVD and BEACOPP in advanced HL, BEACOPP significantly improved disease control but not OS. We explored the potential OS benefit of BEACOPP in a pooled analysis of these 4 trials. Methods. The primary objective was to assess the impact of BEACOPP on OS by one-stage individual patient data meta-analysis stratified for trial. Secondary endpoints were PFS, early toxicities, second malignancies and use of hematopoietic stem cell transplantation (HSCT). Results. Four trials representing a total of 1227 patients (622 ABVD and 605 BEACOPP) were included (IIL Viviani 2011, LYSA H34 Mounier 2014, EORTC 20012 Carde 2016, HD2000 Merli 2016). Patients characteristics: median age: 32 yrs, male 64%, nodular sclerosis 78%, stage II-III-IV: 13-32-55%, B symptoms: 77%, Bulky disease: 38%, IPS>3: 66%. With a median fup of 4.9 yrs, the 5-yrs OS is 86.59% for ABVD (95%CI=83.0-89.3) vs 90.8% for BEACOPP (95%CI=88.0-93.0). The data indicated that the HR is not constant over time (P<0.05) preventing the use of a simple Cox model. Per analysis plan, two time periods (< and ≥ 18m) were defined from the observed survival and hazard curves. In the first 18 months (N=1227, 54 events), no difference was detected (HR:0.75, 95%CI=0.41-1.23). After 18 months (N=1083, 68 events), the risk of death appeared higher in patients in ABVD arm than in BEACOPP arm (HRABVD vs BEACOPP=2.52, 95%CI=1.48-4.28, p<0.001). Further adjustment for IPS showed that both treatment (p<0.001) and IPS group (p=0.0326) effects are significant. The 5-yr PFS was 72.0% for ABVD (95%CI=67.3-76.1) vs 82.9% for BEACOPP (95%CI=78.7-86.4) (p<0.001). Grade 3-4 toxicities (ABVD vs BEACOPP): anemia: 4.3% vs 24.2%; leukopenia: 23.2% vs 76.8%; thrombocytopenia: 1.7% vs 26.4%; infection: 7.9% vs 34.2%. After ABVD, 17 second malignancies (0 AML) (2.7%) were reported and 21(7 AML)(3.6%) after BEACOPP. After ABVD, 86 patients (13.8%) received HSCT vs 39 (6.4%) after BEACOPP. Conclusions. This pooled analysis of 4 studies with increased power, indicated a non-constant treatment effect over time. A statistically significant OS benefit of BEACOPP was observed from 18 months after randomization. Frontline use of BEACOPP increased PFS, early hematological toxicity, secondary leukemia incidence but halved the need of HSCT. Further follow-up is needed to explore the impact of BEACOPP on fertility, second malignancies and long term survival.