Deep Learning Based on Standard H&E Images of Primary Melanoma Tumors Identifies Patients at Risk for Visceral Recurrence and Death

<jats:title>Abstract</jats:title> <jats:sec> <jats:title>Purpose:</jats:title> <jats:p>Biomarkers for disease-specific survival (DSS) in early-stage melanoma are needed to select patients for adjuvant immunotherapy and accelerate clinical trial design. We present a pathology-based computational method using a deep neural network architecture for DSS prediction.</jats:p> </jats:sec> <jats:sec> <jats:title>Experimental Design:</jats:title> <jats:p>The model was trained on 108 patients from four institutions and tested on 104 patients from Yale School of Medicine (YSM, New Haven, CT). A receiver operating characteristic (ROC) curve was generated on the basis of vote aggregation of individual image sequences, an optimized cutoff was selected, and the computational model was tested on a third independent population of 51 patients from Geisinger Health Systems (GHS).</jats:p> </jats:sec> <jats:sec> <jats:title>Results:</jats:title> <jats:p>Area under the curve (AUC) in the YSM patients was 0.905 (P &lt; 0.0001). AUC in the GHS patients was 0.880 (P &lt; 0.0001). Using the cutoff selected in the YSM cohort, the computational model predicted DSS in the GHS cohort based on Kaplan–Meier (KM) analysis (P &lt; 0.0001).</jats:p> </jats:sec> <jats:sec> <jats:title>Conclusions:</jats:title> <jats:p>The novel method presented is applicable to digital images, obviating the need for sample shipment and manipulation and representing a practical advance over current genetic and IHC-based methods.</jats:p> </jats:sec>