LB1663 A multi-gene prognostic signature associated with cutaneous squamous cell carcinoma metastasis

cSCC staging methods are reported to have sub-optimal performances in metastasis prediction. Accurate identification of patients with tumors at high risk of metastasis would have a significant impact on management. Archival FFPE tissue from primary cSCC with perilesional normal tissue from 237 immunocompetent patients (151 non-metastasizing and 86 metastasizing) were collected retrospectively from four centers. All pathology was centrally reviewed and T-staged by consensus. TempO-seq was used to probe the whole transcriptome and machine learning algorithms were applied to derive predictive signatures, with a 75%:25% split for training and testing datasets. A 20-gene prognostic model was developed and validated, with an accuracy of 86.0%, sensitivity of 85.7%, specificity of 86.1%, and positive predictive value of 78.3% in the testing set. A linear predictor was also developed, significantly correlating with metastatic risk. This has provided biological insights into the process of metastasis and potential therapeutic targets. There are biological and genomic mechanisms common to cSCC across different tissue types and the prognostic signature may provide further insights into common differentiation and stem-like pathways underpinning cSCC. Ultimately, the 20-gene prognostic signature has the potential to be incorporated into clinical workflows for cSCC to significantly improve risk stratification, identify patients with high-risk cSCC who may benefit from adjuvant treatment and reduce overtreatment for patients with low-risk cSCC.