Background: Distinct systemic treatments exist for metastatic uveal melanoma. Tebentafusp and combined immune checkpoint blockade (ICB) with ipilimumab plus anti-PD-1 antibodies are the most commonly used treatment options but their comparative efficacy is unclear. The aim of this study is to compare currently available systemic treatments regarding overall survival (OS) and progression-free survival (PFS) with a focus on the comparison of tebentafusp versus combined ICB.
Methods: The protocol for this study was defined a priori and registered online in the PROSPERO international prospective register of systematic reviews (CRD42022308356, date of registration: 7.2.2022). We performed a systematic literature search in Medline, Embase, and Central to identify eligible studies reporting Kaplan-Meier curves or individual-level survival data showing OS and PFS for metastatic uveal melanoma patients treated with systemic treatments. Kaplan-Meier curves were digitized using the "WebPlotDigitizer" program. Individual-level survival data were subsequently remodelled and pooled for distinct treatment groups. To compare the OS of tebentafusp versus combined ICB, we used matching-adjusted indirect comparison (MAIC), two-stage MAIC (2SMAIC), and simulated treatment comparison (STC) together with digitized individual-level survival data as population-adjusted models.
Results: Overall, 55 independent studies were included of which 2,682 patients were evaluable for OS and 2,258 for PFS. Tebentafusp showed the highest median OS (mOS) of 22.4 months (95% confidence interval (CI): 19.9-29.6) compared to combined ICB (mOS: 15.7 months (95% CI: 14.4-17.9)), anti-PD-(L)1 antibody (mOS: 10.9 months (95% CI: 9.8-13.4)), chemotherapy (mOS: 9.95 months (95% CI: 8.9-11.2)), targeted therapies (mOS: 8.86 months (95% CI: 7.5-10.8)), and anti-CTLA-4 antibody (mOS: 7.8 months (95% CI: 6.8-9.3). The median PFS (mPFS) was similar among the treatment groups ranging from 2.7 months to 3.4 months. For the comparison of tebentafusp versus combined ICB, the hazard ratio (HR) was 0.641 (95% CI: 0.449-0.915) in the unadjusted model, whereas the population-adjusted models showed a HR of 0.386 (95% CI: 0.236-0.631) using MAIC, 0.378 (95% CI: 0.234-0.612) applying 2SMAIC and 0.284 (95% CI: 0.184-0.440) using STC.
Conclusions: Tebentafusp achieved the best results compared to combined ICB and other systemic treatments, although these results have to be interpreted with caution due to the approximative methodical approach and high heterogeneity of included studies.
Keywords: Immune checkpoint blockade; Individual patient data; Population-adjusted models; Survival; Tebentafusp; Uveal melanoma.
Copyright © 2023 Elsevier Ltd. All rights reserved.