Abstract 3295: Vitiligo and other clinical melanocyte-related adverse events following tebentafusp (IMCgp100) exposure in patients with uveal melanoma

Abstract

Abstract Background Tebentafusp, a TCR-anti-CD3 bispecific fusion protein designed to target a peptide from melanocyte lineage antigen gp100, has shown monotherapy activity in advanced cutaneous and uveal melanoma (UM) [1]. gp100 is expressed in normal melanocytes in skin and hair follicles. We previously reported a possible association between rash and overall survival (OS) in a Ph 1/2 trial of HLA-A*0201 patients with UM [1]. Anti-PD1 therapy for melanoma can induce vitiligo in 10-25% of patients, with one study showing a median time to onset of 126 days (range 52 - 453) and a potential association with clinical benefit [2]. We now investigate the association of melanocyte-related adverse events (MRAEs) and OS following tebentafusp treatment of patients with UM. Methods To test the differential potency of tebentafusp, HLA-A2+ epidermal skin derived melanocytes (5 healthy donors) and gp100+ and gp100- melanoma cell lines were incubated with PBMCs and increasing tebentafusp concentrations, and tested in IFNγ (measure of T cell activation) and GrB (measure of T cell mediated killing) ELISPOT assays. The clinical trial database (NCT02570308; Ph1 n=19, Ph2 n=23) was also queried for AEs suggestive of melanocyte origin, including vitiligo, leukotrichia, and skin hyperpigmentation. Simon-Makuch estimates were derived to visualize the association between MRAEs and OS while adjusting for immortal time bias, and Mantel-Byar method was used to estimate odds of death associated with experiencing MRAEs. Results In vitro, tebentafusp redirected and activated T cells against gp100+ melanoma and skin-derived melanocytes but not gp100- melanoma; in all cases, potency for melanocytes was less than melanoma, potentially reflecting differential peptide presentation. 24 of 42 patients (57%) developed one or more MRAEs including vitiligo/skin hypopigmentation (28%), leukotrichia (33%), and hyperpigmentation (21%), with median time to onset of 67 days (range 24-221). The estimated probability of experiencing a MRAE within 6-months of continuous treatment is 70%. Among the 24 patients with MRAEs, 22 (92%) and 20 (83%) experienced rash and pruritus, respectively. Rash preceded MRAEs in all patients with both events, while pruritus preceded MRAEs in 77% of patients. In a post-hoc analysis, experiencing a MRAE was associated with a 72% lower odds of death compared to those who did not experience these AEs. Conclusion Over half of tebentafusp-treated patients with advanced UM experienced at least one MRAE, supporting the hypothesis that tebentafusp can redirect T cells against gp100+ cells in patients. This rate of MRAEs is higher than historically reported for anti-PD1 based therapies. MRAEs appear to be associated with prior rash, and even accounting for the time bias of onset, may be associated with OS. 1. Sato T. J Clin Oncol 36, 2018, suppl, 9521 2. Hua C. JAMA Dermatol 2016;152:45-51 Citation Format: Marlana Orloff, Joseph J. Sacco, Paul Nathan, Chris Holland, Chris Cohen, Jane Harper, Shaad E. Abdullah, Takami Sato, Richard D. Carvajal. Vitiligo and other clinical melanocyte-related adverse events following tebentafusp (IMCgp100) exposure in patients with uveal melanoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3295.