OBJECTIVES : Brentuximab vedotin (BV) is currently used to treat patients with relapsed and refractory (RR) classical Hodgkin’s lymphoma (cHL). There is currently no consensus on how best to manage patients who fail BV. Pembrolizumab is one possible solution to the high unmet need among patients previously treated with BV.

The objective of this study was to conduct unadjusted and population-adjusted indirect comparisons of pembrolizumab versus standard-of-care (SOC) in previously BV treated RRcHL patients using time-varying hazard ratios (HRs).

METHODS : A systematic literature review identified one observational study (Cheah et al., 2016) of SOC in a population that matched the majority of patients in KEYNOTE-087. All patients in KEYNOTE-087 who had previously failed treatment with BV were included in the primary analysis. Both unadjusted and population-adjusted (using outcomes regression) pairwise indirect comparisons between Cheah et al., 2016 and KEYNOTE-087 were conducted. Fractional polynomial models were used to estimate time-varying progression-free survival (PFS) HRs with 95% credible intervals (CrIs) for pembrolizumab versus SOC.

RESULTS : Both the unadjusted and population-adjusted comparisons showed evidence that the HR between the two treatments varied over time. The HRs indicated that pembrolizumab is more efficacious than SoC, although the HR was decreasing over time. Statistically meaningful differences in PFS were observed up to 17 months when using the observed data (HR_month3