Impact of PRF 1 A91V germline carrier status in patients with aggressive b-NHL treated with rituximab

Abstract

Abstract INTRODUCTION Rituximab mediated antibody-dependent cellular cytotoxicity (ADCC) is an essential component of therapy in B-cell malignancies. The efficiency of cellular cytotoxicity depends on many factors including rituximab binding, lytic granule release and perforin/granzyme concentration, or expression of death-inducing ligands. Here we analysed whether selected variants in key effector molecules affect outcomes in B-cell non-Hodgkin lymphoma (B-NHL). METHODS PRF1 A91V (rs35947132) was genotyped in 501 patients enrolled in the RICOVER-60 trial (NCT0052936), comparing 6 versus 8 cycles of CHOP chemotherapy with or without rituximab in untreated elderly patients with aggressive B-NHL (Pfreundschuh et al., Lancet Oncol 2008). We evaluated associations with event-free, progression-free, and overall survival. The NHL-B2 (Pfreunschuh et al. Blood 2004) trial served as validation cohort for the CHOP-only arm and the interim analysis of the OPTIMAL>60 (NCT01478542) (Pfreundschuh et al., J Clin Oncol 2017) trial as validation cohort for the R-CHOP arm of the discovery RICOVER-60-cohort. RESULTS Within the RICOVER-60 trial 63 of 501 patients (13%) were carriers of the PRF1 A91V. Carriers showed significant better outcomes, with a 36-month overall survival of 81% [95% CI: 68%-95%] compared to 64% [95% CI: 57%-71%] in wildtype patients treated with CHOP alone. However, no additional benefit was observed for these carriers of PRF1 A91V by the addition of rituximab: 36-month EFS 61% [95% CI: 44%-77%] with CHOP and 63% [95% CI: 46%-81%] with R-CHOP; 36-month OS 81% [95% CI: 68%-95%] with CHOP and 73% [95% CI: 57%-89%] with R-CHOP, respectively. These results were validated in independent cohorts in aggressive B-NHL. In NHL-B2 trial the positive prognostic impact of PRF1 A91V was confirmed for CHOP-only therapy with a significantly higher 36-month OS of 87% [95% CI: 73%-100%] in PRF1 A91V carriers compared to 60% [95% CI: 54%-66%] (p=0.030) in wildtype carriers. In the interim analysis of OPTIMAL>60, where all patients received rituximab with chemotherapy, there were no significant differences between PRF1 A91V carriers and wildtype carriers in PFS (HR: 0.8 [95% CI: 0.3 - 1.9], p=0.635) or OS (HR: 0.7 [95% CI: 0.2 - 2.2], p=0.526). CONCLUSION Patients with aggressive B-NHL carrying the PRF1 A91V germline variant had a favourable outcome with CHOP only chemotherapy, but no benefit from the addition of rituximab. These results suggest PRF1 A91V as a negative predictive marker for rituximab-mediated cellular cytotoxicity in aggressive B-NHL, and it may have potential implications for other immune effector cell-based therapies.