Integrating the melanoma 31-gene expression profile test with clinical and pathologic features can provide personalized precision estimates for sentinel lymph node positivity: an independent performance cohort

Abstract

Abstract Introduction Up to 88% of sentinel lymph node biopsies (SLNBs) are negative. The 31-gene expression profile (31-GEP) test can help identify patients with a low risk of SLN metastasis who can safely forego SLNB. The 31-GEP classifies patients as low (Class 1 A), intermediate (Class 1B/2A), or high risk (Class 2B) for recurrence, metastasis, and SLN positivity. The integrated 31-GEP (i31-GEP) combines the 31-GEP risk score with clinicopathologic features using a neural network algorithm to personalize SLN risk prediction. Methods Patients from a single surgical center with 31-GEP results were included (n = 156). An i31-GEP risk prediction 10% risk. Using the i31-GEP to guide SLNB decisions could have significantly reduced the number of unnecessary SLNBs by 19.2% (30/156, p 10% had a similar SLN positivity rate (33.3%) as patients with T3-T4 tumors (31.3%). Conclusion The i31-GEP identified patients with < 5% risk of SLN positivity who could safely forego SLNB. Combining the 31-GEP with clinicopathologic features for a precise risk estimate can help guide risk-aligned patient care decisions for SLNB to reduce the number of unnecessary SLNBs and increase the SLNB positivity yield if the procedure is performed.

Keywords

• Cutaneous melanoma
• Gene expression profiling
• 31-GEP
• Sentinel lymph node biopsy
• Prognosis