Validation of a Clinicopathologic and Gene Expression Model for Predicting Sentinel Node Metastasis in Melanoma: A Multicenter Danish Cohort Study

Abstract

Background: Sentinel lymph node biopsy (SLNB) is crucial for staging and managing melanoma, but selecting patients for SLNB is challenging, with around 80% of procedures yielding negative results. The clinicopathological and gene expression profile model (CP-GEP) was developed to identify low-risk melanoma patients who may forgo SLNB. CP-GEP combines Breslow thickness, patient age, and a gene expression analysis to classify patients as high- or low-risk for nodal metastasis. This study aimed to validate the performance of CP-GEP in a multicenter Danish cohort. Method: Primary melanoma tissue from 536 T1-T3 patients who had undergone SLNB was retrospectively analyzed using CP-GEP. Results were compared with SLNB status and the Melanoma Institute Australia nomogram (MIA). Results: T1, T2, and T3 melanomas comprised 32.8%, 46.8%, and 20.3% of cases, respectively. The SLNB positivity rate was 18.1%. Overall, 40.9% was classified as CP-GEP low-risk (NPV 91.3%). Among T1 and T2 subgroups, 72.7% and 35.5% were low-risk, with NPVs of 94.5% and 87.6%, respectively. For 507 patients with MIA scores, CP-GEP identified 42.4% as low-risk (NPV 91.2%) versus 8.1% by MIA (NPV 95.1%). Conclusion: CP-GEP is a promising tool for supporting deselection of SLNB in melanoma patients, with a potential reduction rate of over 40%.