AlDubayan SH, Conway JR, Camp SY, et al. - This study was carried out to assess whether deep learning technology distinguish more germline pathogenic variants in patients with cancer vs standard methods. Between 2010 and 2017, a cross-sectional study of a standard germline detection method and a deep learning method in two convenience cohorts with prostate cancer and melanoma was conducted in the US and Europe. December 2017 was the final date of clinical data collection. One-thousand seventy-two men (mean [SD] age at diagnosis, 63.7 [7.9] years; 857 [79.9%] with European ancestry) were included in the prostate cohort and 1,295 patients (mean [SD] age at diagnosis, 59.8 [15.6] years; 488 [37.7%] women; 1060 [81.9%] with European ancestry) were included in the melanoma cohort. In comparison with the current standard genetic testing method, germline genetic testing using deep learning was correlated with higher sensitivity and specificity for detection of pathogenic variants in this convenience sample of patients with prostate cancer and melanoma. Future study is required to understand the relevance of these findings with regard to clinical outcomes.