Abstract 5479: A six-protein signature predicts response and survival in patients with advanced cutaneous melanoma treated with immunotherapy

Abstract

Abstract Introduction: Cutaneous melanoma (CM) is one of the most aggressive types of skin cancer, with a poor prognosis for advanced stages. Despite the advances in immunotherapy (IO) treatment, robust predictive biomarkers are still required. The aim of this study was to identify a prognostic and predictive protein-based signature for advanced CM. Methods: A training cohort of 142 pretreatment tumor samples from 71 patients with metastatic CM treated with IO (anti-PD-1) and an independent validation cohort from 67 patients with stage III treatment-naïve CM were analyzed by data-independent acquisition mass spectrometry (DIA-MS). Survival analysis based initially on univariate regression and subsequently on 100 runs of 20-fold cross-validation of multivariate Cox regression with Least Absolute Shrinkage and Selection Operator (LASSO) was performed to obtain a reduced list of candidate proteins associated with melanoma-specific survival. A risk score was built from the final six proteins. The prognostic ability of this 6-protein signature for melanoma-specific survival was demonstrated by a time-dependent receiver operating characteristic curve (AUROC) and validated by four independent datasets. The predictive ability of the 6-protein signature was also explored in the training dataset using response to IO as the outcome of interest. Finally, differential expression analyses were conducted on the training data to identify the top proteins associated with response to IO. Proteins that were highly correlated (Pearson R2 >0.9) with the 6-protein signature were identified and subjected to pathway enrichment analysis (PEA). Results: Proteomic analyses identified 4298 and 4577 proteins in the training and validation cohorts respectively, with 81.1% overlap between the two sets. Using LASSO multivariate Cox modeling, six proteins were identified in the training cohort, from which a risk score was calculated that dichotomized patients into high- and low-risk groups (Hazard ratio (HR) 2.7, 95% confidence interval (CI) 1.9-3.9, and, AUROC 0.86). The 6-protein signature’s prognostic performance was validated using the RNAseq dataset from the same training cohort (HR 2.7, p<0.001), a second RNASeq dataset from The Cancer Genome Atlas (HR 1.7, p<0.001), a single-cell RNASeq dataset from patients treated with IO (HR 1.2, p<0.001) and finally a proteomic dataset in an independent cohort (stage III treatment-naïve CM) (HR 2.4, p<0.001). The 6-protein signature was also associated with response to IO (HR 2.3, p=0.005). PEA showed that the highly correlated proteins were mostly related to DNA repair and DNA metabolic pathways. PEA revealed activation of immune-related pathways in patients who achieved a good response to IO. Conclusion: A 6-protein signature identified a sub-group of patients with advanced CM who are at higher risk of progression on IO and death from melanoma. Citation Format: Srikanth Manda, Adel T. Aref, Erin K. Sykes, Steven G. Williams, Jennifer M. Koh, Erin M. Humphries, Daniel Bucio-Noble, Daniela Lee-Smith, Natasha Lucas, Dylan Xavier, Alexander Menzies, Ines Da Silva, Felicity Newell, Rosemary Balleine, Peter G. Hains, Graham Mann, Phil J. Robinson, Georgina V. Long, James Wilmott, Qing Zhong, Richard A. Scolyer, Roger R. Reddel. A six-protein signature predicts response and survival in patients with advanced cutaneous melanoma treated with immunotherapy. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5479.