In a prospective, multicenter study, the 31-GEP identified patients at increased risk of tumor recurrence and added significant prognostic value to AJCC staging

Abstract

Listen

Translate article

Cutaneous melanoma (CM) guidelines base management decisions on a patient’s American Joint Committee on Cancer (AJCC) tumor stage. However, limitations in staging accuracy suggest additional tools could improve risk-aligned patient management decisions. The 31-gene expression profile (GEP) test identifies patients with CM with low (Class 1A), intermediate (Class 1B/2A), or high (Class 2B) risk for sentinel lymph node (SLN) positivity, recurrence, metastasis, and death. In this multicenter, prospective study, we validated the 31-GEP for risk of recurrence and demonstrated the added value of 31-GEP to AJCC staging. Patients were included in the prospective CONNECTION study if they were tested with the 31-GEP from 2018 to the present (n=876). Survival was estimated using Kaplan-Meier analysis and 31-GEP stratification tested with the log-rank test. Cox regression was performed to identify predictors of recurrence. ANOVA was used to compare Cox models for the most accurate recurrence prediction. Patients with a Class 1A result had significantly higher 3-year recurrence-free survival than those with a Class 1B/2A or Class 2B result (98.5% vs. 80.8% vs. 63.0%, p<0.001). The 31-GEP had higher sensitivity (75.0% vs. 37.5%) and negative predictive value (98.6% vs. 96.9%) than AJCC staging. Multivariable analysis demonstrated that Class 1B/2A (hazard ratio, HR=3.12, p=0.037) and Class 2B (HR=5.91, p=0.002) were significant predictors of recurrence in addition to stage IB (HR=4.02, p=0.02), IIA (HR=6.06, p=0.006), IIC (HR=17.14, p<0.001), and III (HR=8.84, p<0.001). Stage IIB (HR=4.63, p=0.057) was not significant. Comparing a staging-only model to a 31-GEP+staging model showed that adding 31-GEP to staging significantly improved prediction accuracy over staging alone (ANOVA: c2=10.58, p=0.005). In this prospective study, the 31-GEP stratified risk of recurrence, was a significant predictor of recurrence, and added significant predictive value to AJCC staging. The 31-GEP identifies patients at high risk of recurrence who should be managed more intensely, and adding 31-GEP to staging allows better risk-aligned care decisions, which can lead to improved patient outcomes.