Deep learning based on standard H&E images of primary melanoma tumors identifies patients at risk for visceral recurrence and death

Prathamesh M. Kulkarni, Eric J. Robinson, Jaya Sarin Pradhan, Robyn D. Gartrell-Corrado, Bethany R. Rohr, Megan H. Trager, Larisa J. Geskin, Harriet M. Kluger, Pok Fai Wong, Balazs Acs, Emanuelle M. Rizk, Chen Yang, Manas Mondal, Michael R. Moore, Iman Osman, Robert Phelps, Basil A. Horst, Zhe S. Chen, Tammie Ferringer, David L. RimmJing Wang, Yvonne M. Saenger

Research output: Contribution to journalArticlepeer-review

39 Scopus citations

Abstract

Purpose: 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. Experimental Design: 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). Results: Area under the curve (AUC) in the YSM patients was 0.905 (P < 0.0001). AUC in the GHS patients was 0.880 (P < 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 < 0.0001). Conclusions: 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.

Original languageEnglish (US)
Pages (from-to)1126-1134
Number of pages9
JournalClinical Cancer Research
Volume26
Issue number5
DOIs
StatePublished - Mar 1 2020
Externally publishedYes

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Fingerprint

Dive into the research topics of 'Deep learning based on standard H&E images of primary melanoma tumors identifies patients at risk for visceral recurrence and death'. Together they form a unique fingerprint.

Cite this