Opportunities for precision medicine in hereditary disease inferred from large-scale, real-world diagnostic genetic testing results
Laboratory Genetics and Genomics
-
Primary Categories:
- Genomic Medicine
-
Secondary Categories:
- Genomic Medicine
Introduction:
A molecular diagnosis can be associated with medical actionability, informing clinical management of genetic disease, for example, by pointing to clinical guidelines for surveillance or to precision therapies. However, the rate of genetic test results that are medically actionable remains unknown in real-world clinical testing. Understanding the degree to which gene testing can inform medical care is important because it has broad relevance for clinical care, professional practice guidelines, patient advocacy, health economic research, and insurance reimbursement.
Methods:
Results from nearly 2.9 million probands referred for clinical gene panel testing or exome sequencing over the course of a 9+ year period were analyzed to determine actionability rates. Positive molecular diagnoses in this cohort were matched to expert-curated lists of genetic disorders with clinical management guidelines (i.e., medically actionable), which included ACMG secondary findings, ClinGen actionability for disorders and associated genes, the rx-genes database, and an expansive study that compiled lists of actionable disorders in the context of newborn screening. Results were stratified by clinical area, test type, ICD-10 code, and race, ethnicity, or ancestry (REA) group. Additionally, the frequency of positive results was calculated for family members of probands diagnosed with actionable conditions who participated in cascade testing.
Results:
Probands tested using one or more of 524 panel tests had a diagnostic rate of 12.0%, with 91.6% of those results being linked to medical actionability. Thus, the overall actionable yield (i.e., the product of the diagnostic yield and the actionability rate) was 11.0%, with just 1.0% of patients receiving a molecular diagnosis currently not identified as medically actionable. For patients undergoing exome sequencing, 19.0% received a molecular diagnosis, and of those, 56.0% were medically actionable. Thus, the actionable yield for exome testing was 10.6%, while 8.4% of exome patients received a molecular diagnosis without current measures of medical actionability. Diagnostic and actionability rates varied across specific panel tests and ICD-10 codes, but the overall rate of medical actionability was high (87.7%-98.1%) across 8 of 9 clinical areas, with ophthalmology being the only area with a lower rate (44.8%). There was little overlap across the four expert-curated lists, leading to substantial variation in the proportion of results considered actionable if the lists were considered individually. Across the lists, actionability rates also varied across clinical areas, e.g., the ACMG list had greater actionability in cardiology and hereditary cancer, reflecting the purpose for which it was developed. Among patients with a molecular diagnosis, the rates of actionable results did not differ among REA groups when data were stratified by gene list and clinical area. Diagnostic and actionability rates were comparable between patients for whom ICD-10 codes were provided and patients without ICD-10 codes, suggesting that clinicians utilize genetic testing appropriately. Finally, the rate of positive results in family members undergoing cascade testing for actionable disorders was 27.5%.
Conclusion:
In this study of nearly 2.9 million probands referred for clinical genetic testing for a broad range of clinical conditions, most molecular diagnoses (>90%) appeared to be associated with medical actionability, invoking opportunities for clinical care such as precision therapies, recommended surveillance, and other interventions. Our study suggests that in the setting of real-world clinical genetic testing, the gap between molecular diagnosis and medical actionability is surprisingly narrow. Our data highlight specific areas of medicine where research and professional practice guidelines are most needed to close the existing actionability gaps. Finally, these results underscore the potential high clinical utility of family cascade testing, which remains underutilized in clinical genetic testing.
A molecular diagnosis can be associated with medical actionability, informing clinical management of genetic disease, for example, by pointing to clinical guidelines for surveillance or to precision therapies. However, the rate of genetic test results that are medically actionable remains unknown in real-world clinical testing. Understanding the degree to which gene testing can inform medical care is important because it has broad relevance for clinical care, professional practice guidelines, patient advocacy, health economic research, and insurance reimbursement.
Methods:
Results from nearly 2.9 million probands referred for clinical gene panel testing or exome sequencing over the course of a 9+ year period were analyzed to determine actionability rates. Positive molecular diagnoses in this cohort were matched to expert-curated lists of genetic disorders with clinical management guidelines (i.e., medically actionable), which included ACMG secondary findings, ClinGen actionability for disorders and associated genes, the rx-genes database, and an expansive study that compiled lists of actionable disorders in the context of newborn screening. Results were stratified by clinical area, test type, ICD-10 code, and race, ethnicity, or ancestry (REA) group. Additionally, the frequency of positive results was calculated for family members of probands diagnosed with actionable conditions who participated in cascade testing.
Results:
Probands tested using one or more of 524 panel tests had a diagnostic rate of 12.0%, with 91.6% of those results being linked to medical actionability. Thus, the overall actionable yield (i.e., the product of the diagnostic yield and the actionability rate) was 11.0%, with just 1.0% of patients receiving a molecular diagnosis currently not identified as medically actionable. For patients undergoing exome sequencing, 19.0% received a molecular diagnosis, and of those, 56.0% were medically actionable. Thus, the actionable yield for exome testing was 10.6%, while 8.4% of exome patients received a molecular diagnosis without current measures of medical actionability. Diagnostic and actionability rates varied across specific panel tests and ICD-10 codes, but the overall rate of medical actionability was high (87.7%-98.1%) across 8 of 9 clinical areas, with ophthalmology being the only area with a lower rate (44.8%). There was little overlap across the four expert-curated lists, leading to substantial variation in the proportion of results considered actionable if the lists were considered individually. Across the lists, actionability rates also varied across clinical areas, e.g., the ACMG list had greater actionability in cardiology and hereditary cancer, reflecting the purpose for which it was developed. Among patients with a molecular diagnosis, the rates of actionable results did not differ among REA groups when data were stratified by gene list and clinical area. Diagnostic and actionability rates were comparable between patients for whom ICD-10 codes were provided and patients without ICD-10 codes, suggesting that clinicians utilize genetic testing appropriately. Finally, the rate of positive results in family members undergoing cascade testing for actionable disorders was 27.5%.
Conclusion:
In this study of nearly 2.9 million probands referred for clinical genetic testing for a broad range of clinical conditions, most molecular diagnoses (>90%) appeared to be associated with medical actionability, invoking opportunities for clinical care such as precision therapies, recommended surveillance, and other interventions. Our study suggests that in the setting of real-world clinical genetic testing, the gap between molecular diagnosis and medical actionability is surprisingly narrow. Our data highlight specific areas of medicine where research and professional practice guidelines are most needed to close the existing actionability gaps. Finally, these results underscore the potential high clinical utility of family cascade testing, which remains underutilized in clinical genetic testing.