Population-adjusted indirect comparison (PAIC) of avelumab (AVE) and retifanlimab (RETI) in the first-line (1L) treatment of patients (pts) with metastatic Merkel cell carcinoma (mMCC).

Abstract

e21513 Background: In 2017, AVE was the first immunotherapy (IO) to receive regulatory approval for mMCC. The IO agent RETI has also since been approved. To inform healthcare decision-making and in the absence of head-to-head trials, there is a need to indirectly compare the efficacy of new vs existing treatments. We conducted an unanchored PAIC to compare the efficacy of AVE with that of RETI in 1L-treated pts with mMCC. Methods: The JAVELIN Merkel 200 part B (JAVELIN; AVE) and 1L mMCC POD1UM-201 (RETI) populations were compared via simulated treatment comparison. The JAVELIN data cutoff was Feb 2, 2022 (N=116; median follow-up [mFU], 54.3 months). The POD1UM-201 data cutoff was Mar 10, 2023 (N=91; mFU, 17.6 months). Progression-free survival (PFS), overall survival (OS), and duration of response (DOR) were compared using individual patient-level data from JAVELIN and aggregate data from POD1UM-201. The base case was adjusted for mutually reported treatment effect modifiers (TEM) and prognostic variables (PV), including age, ECOG performance status, baseline visceral metastases, and PD-L1 and MCPyV status. Given the differing timings of PFS recorded at scheduled study visits between the studies (AVE every 6 weeks and RETI every 8 weeks) and potential bias favoring the treatment with longer intervals between visits, assessment-schedule matching (ASM) adjustment was applied to PFS for AVE. Results: Before PAIC adjustment, the results consistently favored RETI over AVE; however, the adjusted analyses identified numerically longer PFS and OS with AVE vs RETI (Table). In both the unadjusted and adjusted analyses, AVE showed no significant difference in DOR compared with RETI. ASM adjustments showed minimal change for PFS analyses when compared with the unadjusted results. Conclusions: Despite the small population size of the included studies and limited data for POD1UM-201, these results suggest that after adjusting for imbalances in PVs and TEMs and assessment-time bias, no meaningful differences in OS, PFS, or DOR were observed with 1L AVE vs RETI among pts with mMCC. These results further support the ongoing role of AVE as a standard of care for pts with mMCC worldwide. Outcome AVE, n RETI, n AVE vs RETI hazard ratio (95% CI) Median survival (95% CI), months Survival probability at 12 months (95% CI) PFS  Unadjusted 105 91 1.65 (1.16-2.34) AVE: 4.07 (2.74-6.11)RETI: 11.22 (6.82-24.05) AVE: 0.31 (0.23-0.41)RETI: 0.48 (0.38-0.60)  Adjusted 105 91 0.95 (0.66-1.38) AVE 15.09 (8.25-21.94)RETI: 11.22 (6.82-24.05) AVE: 0.54 (0.45-0.65)RETI: 0.48 (0.38-0.60) OS  Unadjusted 105 91 2.22 (1.40-3.52) Not reached (NR) AVE: 0.60 (0.51-0.70)RETI: 0.79 (0.71-0.88)  Adjusted 105 91 0.93 (0.56-1.57) NR AVE: 0.81 (0.73-0.89)RETI: 0.79 (0.71-0.88) DOR  Unadjusted 42 48 1.57 (0.80-3.08) Not applicable (NA) NA  Adjusted 42 48 1.48 (0.74-2.96) NA NA