Performance of a prognostic 31-gene expression profile test in patients with node-negative cutaneous melanoma.

Abstract

e22071 Background: The National Comprehensive Cancer Network guidelines recommend considering sentinel lymph node biopsy (SLNB) for cutaneous melanoma (CM) patients with a 5-10% risk of SLN positivity and offering it to those with > 10% risk. However, SLNB limitations include identification of only 1/3 of patients who end up with distant metastasis and a regional metastasis false negative rate ranging from 5-21% indicating a need to detect the risk of metastasis in patients with a negative SLNB. The 31 gene expression profile (31GEP) test uses the molecular biology of the primary tumor to predict 5-year recurrence-free (RFS), distant metastasis-free (DMFS), melanoma-specific (MSS), and overall survival (OS). Therefore, the objective of this study is to determine whether the 31GEP can stratify risk for patients with SLN-negative results to help guide CM management. Methods: 607 primary CM tumors were collected from patients with SLN-negative status who were enrolled in multi-center IRB-approved studies. Tumors were staged according to the AJCC 8th edition and analyzed by 31GEP testing to differentiate low-risk (Class 1A), intermediate-risk (Class 1B/2A), and high-risk (Class 2B) tumor biology. Clinical data were recorded and Kaplan-Meier (KM) and Cox regression analyses were performed to assess the relationship between 31GEP class and patient outcomes including RFS, DMFS, MSS, and OS. Results: KM analysis for 5yr RFS, DMFS, MSS, and OS stratified by 31GEP class are listed in Table. Multivariate Cox regression analysis demonstrated that the 31GEP Class 2B status was an independent predictor of RFS, DMFS, MSS, and OS with hazard ratios of 4.4 (95% CI, 2.7-7.2; p < 0.001), 4.4 (95% CI, 2.3-8.5; p < 0.001), 15.6 (95% CI, 3.4-71.2; p < 0.001), and 5.4 (95% CI, 2.6-10.9; p < 0.001), respectively. Conclusions: The results indicate that the 31GEP can stratify risk for a subset of patients with SLNB-negative results with the highest 5-year survival rates being associated with a Class 1A result. Moreover, GEP Class 2B is a significant independent predictor of metastatic risk in patients who are node negative. [Table: see text]