The 31-gene expression profile as a guide to better risk-aligned care decisions for patients with stage I–III cutaneous melanoma: An NCI-SEER analysis.

9573 Background: Current management guidelines for cutaneous melanoma (CM) are based on AJCC staging to stratify patients into risk groups. However, this approach can result in under- and over-estimation of risk for individual patients. The 31-gene expression profile (31-GEP) test has been prospectively validated to provide a more personalized likelihood of sentinel lymph node positivity, as well as risk of recurrence, distant metastasis, and mortality, than AJCC staging alone (Class 1A=low risk, Class 1B/2A=intermediate risk, and Class 2B=high risk). Using expanded SEER registry cohorts, we assessed the ability of the 31-GEP to independently stratify patients with high and low mortality risk categories, and to evaluate whether 31-GEP testing itself was associated with improved patient outcome. Methods: SEER registry data for patients with stage I–III CM (2013–2019) were linked to patients with 31-GEP test results provided by Castle Biosciences (N=13,560) using a registry trusted independent third party. Five-year melanoma-specific survival (MSS) was estimated using Kaplan-Meier analysis; survival differences between groups were compared using log-rank test. Multivariable analysis was performed to determine significant predictors of melanoma-specific mortality (MSM). Survival differences between 31-GEP tested and untested patients were performed by matching tested and untested patients according to clinicopathological factors, diagnosis year, ethnicity, and socioeconomic status. Results: Patients with a Class 1A 31-GEP result had a significantly higher 5-year MSS than those with Class 1B/2A or Class 2B results (99.1%, 92.5%, vs. 85.9%, p<0.001). Multivariable analysis showed that a Class 2B result (HR=4.20, p<0.001), Class 1B/2A result (HR=3.21, p<0.001), a positive lymph node (HR=2.78, p<0.001), Breslow thickness (HR=1.10, p=0.001), ulceration (HR=1.41, p=0.033), age (HR=1.04, p<0.001), and mitotic rate (HR=1.05, p=0.022) were significant predictors of MSM. In staging subset analyses, patients with a Class 1A 31-GEP result had a significantly higher 5-year MSS than those with Class 2B results in stage I-IIA CM (1A=98.8%, 1B/2A=95.5% vs. 2B=93.0%, p<0.001), stage IIB–IIC CM (1A=94.2%, 1B/2A=91.2% vs. 2B=82.3%, p=0.002), and stage III CM (1A=94.8%, 1B/2A=75.0% vs. 2B=77.6%, p<0.001). Among all stages, 31-GEP-tested patients had a lower MSM (HR=0.68, p<0.001) than untested propensity score-matched patients. Conclusions: In a large, real-world cohort of clinically tested patients with stage I–III CM, the 31-GEP stratified mortality risk within all staging groups, which could better help clinicians and patients make risk-aligned treatment and clinical management decisions.