Development of a gene-expression panel predictive of local recurrence, metastasis, and overall survival in intermediate- to high-risk cutaneous squamous cell carcinoma: A cohort study.

Abstract

Cutaneous squamous cell carcinoma (cSCC) is one of the most common malignancies and carries risks of metastasis and death. Current staging systems have a poor positive predictive value in identifying cSCCs at risk of metastasis or patient death. Gene expression profiling (GEP) in high-risk cSCC has demonstrated promise in predicting poor outcomes. To develop and optimize a novel GEP panel using a patient cohort with stage-matched, outcome-differentiated cSCC based upon 183 functionally relevant genes identified through previous multi-omic work using the NanoString platform on 186 tumors in 183 patients. Fourteen genes were used to generate a risk score for metastasis, resulting in an area under the curve (AUC) score of 82.0% (95% CI: 75.9% to 88.2%), overall accuracy of 75.3%, sensitivity of 65.1%, and specificity of 80.5%, respectively. Three genes were found to be predictive of local recurrence, resulting in an AUC score of 74.7% (95% CI: 63.8% to 85.6%), overall accuracy of 69.4%, sensitivity of 73.9%, and specificity of 68.7%, respectively. Eight genes were associated with overall survival 1.28 (95% CI: 1.13 to 1.46, P < 0.001) after adjusting for age, immunosuppressant use, and metastasis status. Our GEP panel demonstrates the potential to predict outcomes in at-risk cSCC.