Want to take quizzes and track your credits?
In patients with cancer, is the detection of pathogenic germline genetic variation improved by incorporation of automated deep learning technology compared with standard methods?
In this cross-sectional analysis of 2 retrospectively collected convenience cohorts of patients with prostate cancer and melanoma, more pathogenic variants in 118 cancer-predisposition genes were found using deep learning technology compared with a standard genetic analysis method (198 vs 182 variants identified in 1072 patients with prostate cancer; 93 vs 74 variants identified in 1295 patients with melanoma).
The number of cancer-predisposing pathogenic variants identified in patients with prostate cancer and melanoma depends partially on the automated approach used to analyze sequence data, but further research is needed to understand possible implications for clinical management and patient outcomes.
Less than 10% of patients with cancer have detectable pathogenic germline alterations, which may be partially due to incomplete pathogenic variant detection.
To evaluate if deep learning approaches identify more germline pathogenic variants in patients with cancer.
Design, Setting, and Participants
A cross-sectional study of a standard germline detection method and a deep learning method in 2 convenience cohorts with prostate cancer and melanoma enrolled in the US and Europe between 2010 and 2017. The final date of clinical data collection was December 2017.
Germline variant detection using standard or deep learning methods.
Main Outcomes and Measures
The primary outcomes included pathogenic variant detection performance in 118 cancer-predisposition genes estimated as sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The secondary outcomes were pathogenic variant detection performance in 59 genes deemed actionable by the American College of Medical Genetics and Genomics (ACMG) and 5197 clinically relevant mendelian genes. True sensitivity and true specificity could not be calculated due to lack of a criterion reference standard, but were estimated as the proportion of true-positive variants and true-negative variants, respectively, identified by each method in a reference variant set that consisted of all variants judged to be valid from either approach.
The prostate cancer cohort included 1072 men (mean [SD] age at diagnosis, 63.7 [7.9] years; 857 [79.9%] with European ancestry) and the melanoma cohort included 1295 patients (mean [SD] age at diagnosis, 59.8 [15.6] years; 488 [37.7%] women; 1060 [81.9%] with European ancestry). The deep learning method identified more patients with pathogenic variants in cancer-predisposition genes than the standard method (prostate cancer: 198 vs 182; melanoma: 93 vs 74); sensitivity (prostate cancer: 94.7% vs 87.1% [difference, 7.6%; 95% CI, 2.2% to 13.1%]; melanoma: 74.4% vs 59.2% [difference, 15.2%; 95% CI, 3.7% to 26.7%]), specificity (prostate cancer: 64.0% vs 36.0% [difference, 28.0%; 95% CI, 1.4% to 54.6%]; melanoma: 63.4% vs 36.6% [difference, 26.8%; 95% CI, 17.6% to 35.9%]), PPV (prostate cancer: 95.7% vs 91.9% [difference, 3.8%; 95% CI, –1.0% to 8.4%]; melanoma: 54.4% vs 35.4% [difference, 19.0%; 95% CI, 9.1% to 28.9%]), and NPV (prostate cancer: 59.3% vs 25.0% [difference, 34.3%; 95% CI, 10.9% to 57.6%]; melanoma: 80.8% vs 60.5% [difference, 20.3%; 95% CI, 10.0% to 30.7%]). For the ACMG genes, the sensitivity of the 2 methods was not significantly different in the prostate cancer cohort (94.9% vs 90.6% [difference, 4.3%; 95% CI, –2.3% to 10.9%]), but the deep learning method had a higher sensitivity in the melanoma cohort (71.6% vs 53.7% [difference, 17.9%; 95% CI, 1.82% to 34.0%]). The deep learning method had higher sensitivity in the mendelian genes (prostate cancer: 99.7% vs 95.1% [difference, 4.6%; 95% CI, 3.0% to 6.3%]; melanoma: 91.7% vs 86.2% [difference, 5.5%; 95% CI, 2.2% to 8.8%]).
Conclusions and Relevance
Among a convenience sample of 2 independent cohorts of patients with prostate cancer and melanoma, germline genetic testing using deep learning, compared with the current standard genetic testing method, was associated with higher sensitivity and specificity for detection of pathogenic variants. Further research is needed to understand the relevance of these findings with regard to clinical outcomes.
Sign in to take quiz and track your certificates
JN Learning™ is the home for CME and MOC from the JAMA Network. Search by specialty or US state and earn AMA PRA Category 1 CME Credit™ from articles, audio, Clinical Challenges and more. Learn more about CME/MOC
Corresponding Author: Eliezer M. Van Allen, MD, Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, 360 Longwood Ave, LC9329, Boston, MA 02215 (firstname.lastname@example.org).
Accepted for Publication: October 6, 2020.
Author Contributions: Drs AlDubayan and Van Allen had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: AlDubayan, Conway, Al-Rubaish, Al-Sulaiman, Al-Ali, Taylor-Weiner, Van Allen.
Acquisition, analysis, or interpretation of data: AlDubayan, Conway, Camp, Witkowski, Kofman, Reardon, Han, Moore, Elmarakeby, Salari, Choudhry, Al-Sulaiman, Taylor-Weiner, Van Allen.
Drafting of the manuscript: AlDubayan, Conway, Camp, Han, Taylor-Weiner, Van Allen.
Critical revision of the manuscript for important intellectual content: AlDubayan, Conway, Witkowski, Kofman, Reardon, Moore, Elmarakeby, Salari, Choudhry, Al-Rubaish, Al-Sulaiman, Al-Ali, Taylor-Weiner, Van Allen.
Statistical analysis: AlDubayan, Conway, Camp, Kofman, Reardon, Han, Elmarakeby, Salari, Choudhry, Taylor-Weiner, Van Allen.
Obtained funding: AlDubayan, Van Allen.
Administrative, technical, or material support: Moore, Al-Rubaish, Al-Sulaiman, Al-Ali, Van Allen.
Supervision: AlDubayan, Taylor-Weiner, Van Allen.
Conflict of Interest Disclosures: Dr Moore reported receiving personal fees from Immunity Health. Dr Van Allen reported serving on advisory boards or as a consultant to Tango Therapeutics, Genome Medical, Invitae, Illumina, Manifold Bio, Monte Rosa Therapeutics, and Enara Bio; receiving personal fees from Invitae, Tango Therapeutics, Genome Medical, Ervaxx, Roche/Genentech, and Janssen; receiving research support from Novartis and Bristol-Myers Squibb; having equity in Tango Therapeutics, Genome Medical, Syapse, Enara Bio, Manifold Bio, and Microsoft; receiving travel reimbursement from Roche and Genentech; and filing institutional patents (for ERCC2 variants and chemotherapy response, chromatin variants and immunotherapy response, and methods for clinical interpretation). No other disclosures were reported.
Funding/Support: This work was supported by Conquer Cancer Foundation Career Development Award 13167 from the American Society of Clinical Oncology (awarded to Dr AlDubayan), Young Investigator Award 18YOUN02 from the Prostate Cancer Foundation (awarded to Dr AlDubayan), the Challenge Award from the PCF-V Foundation (awarded to Dr Van Allen), the Emerging Leader Award from the Mark Foundation (awarded to Dr Van Allen), grant R01CA222574 from the National Institutes of Health (awarded to Dr Van Allen), and grant 12-MED2224-46 (for science and technology) from King Abdulaziz City (awarded to Drs Al-Rubaish, Al-Sulaiman, and Al-Ali).
Role of the Funder/Sponsor: The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank all the individuals who participated in this study. We also thank Eric Banks, PhD (data sciences platform, Broad Institute of Massachusetts Institute of Technology and Harvard University; no compensation was received), for his valuable insight into the underlying model of the Genome Analysis Toolkit and for his comments on the results of this study. We also thank Jeff Kohlwes, MD, MPH (general internal medicine, University of California, San Francisco; no compensation was received), Aaron Neinstein, MD (endocrinology and clinical informatics, University of California, San Francisco; no compensation was received), and Tara Vijayan, MD (infectious disease, University of California, Los Angeles; no compensation was received), for their feedback on the content in the manuscript.
Additional Information: The results are based, in part, on data generated by the Cancer Genome Atlas managed by the National Cancer Institute and the National Human Genome Research Institute. Information about the Cancer Genome Atlas can be found at http://cancergenome.nih.gov. The raw sequence data can be obtained through dbGaP (https://www.ncbi.nlm.nih.gov/gap) or as described in the original articles (details appear in the Methods section). All software tools used in this study are publicly available.
You currently have no searches saved.