Abstract LB510: Low baseline macrophage to T cell ratio is predictive of overall survival with tebentafusp in a phase 3 trial in metastatic uveal melanoma

Abstract

Abstract Introduction: Tebentafusp, an investigational TCR bispecific (gp100xCD3) fusion protein (ImmTAC) that targets gp100 and activates T cells, had superior overall survival (OS) versus investigator’s choice (HR 0.51) in a randomized Phase 3 trial in previously untreated metastatic uveal melanoma [1]. We previously reported that CD163+ M2 macrophages can inhibit ImmTAC-mediated T cell killing of tumor cells in vitro and demonstrated that a low CD163:CD3 ratio in tumor biopsies was associated with better OS in a Phase 2 trial of tebentafusp in previously treated metastatic uveal melanoma (NCT02570308) [2]. Here we sought to confirm this finding using biopsies collected from the randomized Phase 3 trial (NCT03070392). Methods: Previously untreated HLA-A*02:01+ mUM patients were randomly assigned 2:1 to either the tebentafusp or investigator’s choice arm. Tumor samples were obtained prior to dosing from both primary (eye) and metastatic (predominantly liver) tumors. Expression of CD3, CD8 and CD163 was determined by immunohistochemistry (IHC), and digital image analysis enumerated immune cells per mm2 (up to N=178 in the tebentafusp arm, N=81 in the control arm). A low quartile cut-off (low) vs ≥ lower quartile (higher) was used for biomarker correlations with OS. Results: Individually, the numbers of tumor infiltrating CD3+ or CD8+ T cells and CD163+ macrophages at baseline were not associated with better OS in either the tebentafusp or investigator’s choice arm. Consistent with data reported from the previously treated tebentafusp trial, patients randomized to tebentafusp with low baseline CD163:CD3 ratio had improved OS compared to patients with higher ratio (HR=0.6, p=0.03; 1-year OS low=80% vs higher=65%). This association with OS was not as strong in the investigator’s choice arm (HR=0.7, p=0.3; 1-year OS low=52% vs higher=48%) in which the majority of patients were treated with pembrolizumab. Furthermore, tebentafusp showed superior OS compared to investigator’s choice in patients with low (HR=0.4, p=0.01) or higher (HR=0.5, p=0.001) CD163:CD3 ratio at baseline. Discussion: In patients with previously untreated metastatic uveal melanoma, tebentafusp was superior to investigator’s choice regardless of the macrophage to T cell ratio in tumor biopsies. Low baseline macrophage to T cell ratio was more strongly associated with OS on the tebentafusp arm than investigator’s choice arm. These results demonstrate the importance of the tumor micro-environment in the anti-tumor response on tebentafusp in metastatic uveal melanoma. Reference: 1.Nathan P., et al. NEJM 2021; 2.Hassel J., et al. AACR #1673 2021 Citation Format: Sophie Piperno-Neumann, Sarah Stanhope, Camille Britton-Rivet, Simone Bischetti, Shaad Abdullah, Laura Collins, Koustubh Ranade, Emma Leach, Piotr Rutkowski. Low baseline macrophage to T cell ratio is predictive of overall survival with tebentafusp in a phase 3 trial in metastatic uveal melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr LB510.