Rapid Genomic Testing: A Retrospective Study of Institutional Utilization and Outcomes
Health Services and Implementation
-
Primary Categories:
- Health services and Implementation
-
Secondary Categories:
- Health services and Implementation
Introduction:
Rapid exome sequencing (rES) and rapid genome sequencing (rGS) have transformed the landscape of achieving timely diagnosis in critically ill patients. Recent improvements in technology, cost, turn-around time, and insurance coverage have increased the utilization of genomic testing. While the uptake of rapid genomic testing is increasing, ensuring appropriate utilization of these tests is essential. This single institution analysis of rapid genomic test utilization provides unique insights into ordering trends and future utilization management aims.
Methods:
The Stanford Genetic Testing Optimization Service (GTOS) database, which contains information about all genetic test orders reviewed by the GTOS, was queried for orders reviewed between January 1, 2020 and June 30, 2024. Relevant rES and rGS cases were identified using key words. Information about the clinical indication, test order specifics, and cost was collected. Tests that were cancelled or modified from rGS or rES prior to resulting were excluded. The test results and turn-around time were then ascertained through chart review. Test results were categorized as diagnostic or nondiagnostic based on the presence of variants with classification and inheritance patterns consistent with genetic disease.
Results:
A total of 11,363 database entries were identified in the specified date range, of which 198 were rES or rGS orders (2%). The vast majority of tests were ordered inpatient (98% - 194/198) and were ordered as rGS (75% - 149/198). Patients receiving rapid genomic testing were a median age of 58 days old (range: 1 day – 58 years). The average turn-around time from sample collection to test resulting was 13.1 days. At the time of returning results for inpatient orders, the majority (63% - 122/194) of patients were still admitted in the hospital, while 32% patients were discharged and the remaining 5% of results were delivered to the family postmortem.
The overall diagnostic yield was 22.2%. The yield of rES (18%) was not significantly lower than rGS (23%; p=0.41). Most diagnoses (54% - 31/57) were autosomal recessive conditions, followed in frequency by de novo variants associated with X-linked or autosomal dominant conditions (28% - 16/57). The number of rapid testing orders has increased year-over-year: 9 in 2020, 12 in 2021, 34 in 2022, 75 in 2023, and 68 in the first half of 2024. We have not observed the number of diagnoses from rES/rGS to proportionally increase: there were 2 diagnostic rES/rGS in 2020 (22% yield), 4 in 2021 (33%), 15 in 2022 (40%), 15 in 2023 (20%), and 11 in the first half of 2024 (16% yield).
Conclusion:
While rapid genomic testing was utilized with increasing frequency over the past years, a commensurate increase in diagnoses was not observed in our cohort. Instead, there appears to be an inverse relationship between the frequency of rapid genomic testing and its diagnostic yield. This apparent decrease in test utility is concerning from a utilization perspective. We suspect that this is a multifactorial phenomenon and may be in part due to more liberal test utilization. Additionally, our data suggest that results do not always return while the patient is in the hospital and, therefore, do not impact inpatient management. Better cohort selection, faster turn-around time, and more clinician education on the strengths and weaknesses of rES/rGS may improve its clinical impact. The findings herein highlight the need for additional studies to determine the most effective strategies for test utilization management and policy development, and to identify factors that influence diagnostic yield of rapid genomic testing.
Rapid exome sequencing (rES) and rapid genome sequencing (rGS) have transformed the landscape of achieving timely diagnosis in critically ill patients. Recent improvements in technology, cost, turn-around time, and insurance coverage have increased the utilization of genomic testing. While the uptake of rapid genomic testing is increasing, ensuring appropriate utilization of these tests is essential. This single institution analysis of rapid genomic test utilization provides unique insights into ordering trends and future utilization management aims.
Methods:
The Stanford Genetic Testing Optimization Service (GTOS) database, which contains information about all genetic test orders reviewed by the GTOS, was queried for orders reviewed between January 1, 2020 and June 30, 2024. Relevant rES and rGS cases were identified using key words. Information about the clinical indication, test order specifics, and cost was collected. Tests that were cancelled or modified from rGS or rES prior to resulting were excluded. The test results and turn-around time were then ascertained through chart review. Test results were categorized as diagnostic or nondiagnostic based on the presence of variants with classification and inheritance patterns consistent with genetic disease.
Results:
A total of 11,363 database entries were identified in the specified date range, of which 198 were rES or rGS orders (2%). The vast majority of tests were ordered inpatient (98% - 194/198) and were ordered as rGS (75% - 149/198). Patients receiving rapid genomic testing were a median age of 58 days old (range: 1 day – 58 years). The average turn-around time from sample collection to test resulting was 13.1 days. At the time of returning results for inpatient orders, the majority (63% - 122/194) of patients were still admitted in the hospital, while 32% patients were discharged and the remaining 5% of results were delivered to the family postmortem.
The overall diagnostic yield was 22.2%. The yield of rES (18%) was not significantly lower than rGS (23%; p=0.41). Most diagnoses (54% - 31/57) were autosomal recessive conditions, followed in frequency by de novo variants associated with X-linked or autosomal dominant conditions (28% - 16/57). The number of rapid testing orders has increased year-over-year: 9 in 2020, 12 in 2021, 34 in 2022, 75 in 2023, and 68 in the first half of 2024. We have not observed the number of diagnoses from rES/rGS to proportionally increase: there were 2 diagnostic rES/rGS in 2020 (22% yield), 4 in 2021 (33%), 15 in 2022 (40%), 15 in 2023 (20%), and 11 in the first half of 2024 (16% yield).
Conclusion:
While rapid genomic testing was utilized with increasing frequency over the past years, a commensurate increase in diagnoses was not observed in our cohort. Instead, there appears to be an inverse relationship between the frequency of rapid genomic testing and its diagnostic yield. This apparent decrease in test utility is concerning from a utilization perspective. We suspect that this is a multifactorial phenomenon and may be in part due to more liberal test utilization. Additionally, our data suggest that results do not always return while the patient is in the hospital and, therefore, do not impact inpatient management. Better cohort selection, faster turn-around time, and more clinician education on the strengths and weaknesses of rES/rGS may improve its clinical impact. The findings herein highlight the need for additional studies to determine the most effective strategies for test utilization management and policy development, and to identify factors that influence diagnostic yield of rapid genomic testing.