D. Effiom1, T. Aprati2, A. Alimohamed1, A. Karneris1, A. Lawless1, T. Sharova1, X. Bai1, R. Sullivan1, K. Flaherty1, G. Boland1, S. Cohen1 1Massachusetts General Hospital, Oncology, Boston, MA, USA 2Dana Farber Cancer Insititute, Oncology, Boston, MA, USA
Introduction:
Phase III clinical trials have shown adjuvant systemic therapy increases recurrence free survival (RFS) in patients with locoregional cutaneous melanoma (stage III). However, an estimated 30% of patients still recur within a year, indicating that additional treatment options are needed. Staging and surgical management have changed since trial publications, with the adoption of the 8th edition AJCC and preference for nodal observation rather than completion lymphadenectomy. These shifts differentiate the trial cohorts from current treatment paradigms. Furthermore, data from the advanced setting is often extrapolated to treatment in the adjuvant setting. To investigate this, we assessed adjuvant systemic therapy in a modern cohort of patients with stage III melanoma.
Methods:
We collected clinical and histogenomic data from patients diagnosed with stage III cutaneous melanoma between 2015 and 2022. Outcomes of interest were RFS and distant-metastasis free survival. Survival analysis was performed using the Kaplan-Meier method. Univariate and multivariate analysis was performed using cox regression analysis. Furthemore, 290 stage IV cutaneous melanoma patients were reviewed, with progression free survival (PFS) to first line therapy investigated.
Results:
215 patients with resected stage III melanoma were included; 65 and 76 had BRAF/MEK and PD1 adjuvant systemic therapy respectively, with 74 undergoing active surveillance. Adjuvant therapy with PD1 was associated with a longer RFS (HR: 0.54, p < 0.05) compared to surveillance; similarly, a longer median RFS was observed in the BRAF/MEK cohort compared to surveillance (statistically insignificant). We assessed somatic mutations associated with response to adjuvant therapies. Among patients treated with PD1, BRAF V600 mutants had a significantly shorter RFS than BRAF wild type patients (HR: 4.4 p < 0.01). Analysis of 79 stage IV patients treated with first line PD1 monotherapy showed that those with BRAF V600 mutation had a worse PFS compared to BRAF WT (HR: 2.3, p = 0.021).
Conclusion:
No direct prospective study has compared outcomes of adjuvant therapies for resected stage III melanoma in BRAF V600 mutants. Retrospective analyses consistently show improved RFS in mutants treated with targeted therapy compared to PD1, which our data confirms. Importantly an unbiased analysis of factors associated with PD1 response found that BRAF V600 mutation predicts a poor response to adjuvant single-agent PD1, irrespective of stage and surgical factors. Our stage IV results and previously published work support this. Our data suggests that in the adjuvant setting BRAF V600 mutant patients should preferentially be treated with targeted therapy rather than PD1 monotherapy.