Factors associated with response to talimogene laherparepvec in the treatment of advanced melanoma

Abstract

Talimogene laherparepvec (T-VEC) is currently the only United States Food and Drug Administration-approved intralesional therapy for advanced melanoma. Recent studies have assessed the integration of T-VEC with systemic immunotherapy, though the response remains variable. Therefore, we sought to identify factors associated with the response to T-VEC by conducting a retrospective, single-center analysis involving melanoma patients treated with T-VEC. In the present study, we recorded demographic and clinicopathological data, details of T-VEC treatments, prior and concurrent treatments, and clinical outcomes. The primary endpoint was the in-field overall response rate (ORR: complete + partial). Secondary endpoints included complete in-field response, defined as complete resolution of disease or a negative biopsy; disease failure-free survival (DFFS), defined from the initiation of treatment as the time to progression in patients who did not experience a disease-free interval and time to recurrence in those who did; and overall survival (OS). We used two-sample t-tests for continuous variables and Fisher’s exact test for categorical variables. DFFS and OS were further analyzed using the Kaplan–Meier method and log-rank tests for selected variables. Among the 18 patients who met the inclusion criteria, an in-field response was observed in 14 (78%) patients. Low disease burden (<5 lesions or a total diameter <5 cm) was associated with a higher in-field ORR compared to high-burden disease (100% vs. 43%, P = 0.01). There was a trend toward decreased in-field ORR for patients with >2 prior lines of systemic therapy (P = 0.18). With a median follow-up of 386 days, DFFS was associated with BRAF wild-type melanoma (P = 0.04), an in-field ORR (P = 0.007), and a measurable bystander response (P = 0.01). Two or more prior lines of immunotherapy were associated with poorer survival (P = 0.006) and worse DFFS (P = 0.02). In conclusion, despite a low sample size, we identified covariates associated with in-field ORR, DFFS, and OS, warranting further study.