Genetic Evaluation of Short Stature: Examination of Ordering Practices and Diagnostic Yield
Clinical Genetics and Therapeutics
-
Primary Categories:
- Clinical Genetics
-
Secondary Categories:
- Clinical Genetics
Introduction:
Short stature represents a common indication for genetics evaluation. In its revised clinical practice resource, the American College of Medical Genetics and Genomics (ACMG) outlines a stepwise approach to the genetic interrogation for these patients that includes chromosome analysis by karyotype, CGH and/or SNP microarray, next generation sequencing (NGS), and methylation analyses according to patient’s phenotype. This approach is intended to prioritize performance of the most sensitive methodologies for the pathologic aberrations most commonly associated with the pattern of growth impediment seen in the patient. Here, we examine testing strategies for patients presenting with short stature at our institution and the corresponding yield of each respective testing modality to measure our adherence to the ACMG’s recommended framework.
Methods:
We identified patients who were referred for internal laboratory genetics evaluation between January 2021 and October 2024 whose referral included ICD-10 R62.50 for short stature. EMR records were reviewed to extrapolate patient demographic information, ordered testing, ICD-10 codes applied to the order, and test outcomes. Patients whose testing preceded dates of service or received testing outside our institution were excluded.
Results:
The final cohort included 77 pediatric patients: 58 females, 18 males, and 1 whose EMR sex was listed as ‘X.’ Test methodologies employed for initial analysis included methylation-specific PCR or MLPA (n=2), 50-cell chromosome analysis (n=45), Fragile X repeat expansion studies (n=6), NGS panel analysis (n=10), SHOX sequencing with deletion/duplication analysis (n=2), and SNP array (n=9). The majority of females (n=44; 75.9%) and all non-gendered patients were institutionally approved for 50-cell chromosome analysis, whereas the majority of males had orders placed for a SNP array either as a standalone (n=6; 33.3%) or in combination with Fragile X repeat analysis (n=5; 27.8%). Six (7.8%) tests were cancelled by reason of either provider or family request, or insurance denial. An additional 17 (22.1%) patients did not provide a sample for analysis. Seven of the 54 patients whose samples were analyzed received a molecular or cytogenetic diagnosis. The diagnostic yield according to test type is as follows: 8.8% (n=3) for 50-cell chromosome analysis, 33.3% (n=2) for NGS panel, 50% SHOX sequencing + del/dup analysis (n=2), and 20% for SNP array as a standalone test (n=1). 20.8% (n=16) of patients underwent more than one test, performed either reflexively or in tandem with another assay. Subsequent testing showed a 28.6% yield (n=2).
Conclusion:
With the exception of chromosomal abnormalities, the diagnostic rates within our cohort are largely concordant with figures in the published literature; however, our data is impacted by the volume of patients who were lost to follow-up before provision of a sample. Additionally, a minority of patients undergoing testing after receiving non-diagnostic results, potentially further limiting the diagnostic potential of testing. Though concurrent testing may solve for this issue, the cumulative cost of the analyses may become prohibitive to patients seeking diagnosis. Instead, improvements to bioinformatics pipelines built for NGS analysis may provide more robust justification to pursue molecular analysis as an initial evaluation. Platforms optimized for detection of copy number variants (CNVs) with tools to adjust for regions with high sequence homology that includes methods for orthogonal confirmation of low-confidence calls can provide a comprehensive analysis that simplifies provider ordering decisions while improving overall test yield. Testing at commercial laboratories that facilitate at-home collection may also improve patient follow-through to ensure higher catchment. However, more data will be needed to measure the efficacy of a revised testing paradigm.
Short stature represents a common indication for genetics evaluation. In its revised clinical practice resource, the American College of Medical Genetics and Genomics (ACMG) outlines a stepwise approach to the genetic interrogation for these patients that includes chromosome analysis by karyotype, CGH and/or SNP microarray, next generation sequencing (NGS), and methylation analyses according to patient’s phenotype. This approach is intended to prioritize performance of the most sensitive methodologies for the pathologic aberrations most commonly associated with the pattern of growth impediment seen in the patient. Here, we examine testing strategies for patients presenting with short stature at our institution and the corresponding yield of each respective testing modality to measure our adherence to the ACMG’s recommended framework.
Methods:
We identified patients who were referred for internal laboratory genetics evaluation between January 2021 and October 2024 whose referral included ICD-10 R62.50 for short stature. EMR records were reviewed to extrapolate patient demographic information, ordered testing, ICD-10 codes applied to the order, and test outcomes. Patients whose testing preceded dates of service or received testing outside our institution were excluded.
Results:
The final cohort included 77 pediatric patients: 58 females, 18 males, and 1 whose EMR sex was listed as ‘X.’ Test methodologies employed for initial analysis included methylation-specific PCR or MLPA (n=2), 50-cell chromosome analysis (n=45), Fragile X repeat expansion studies (n=6), NGS panel analysis (n=10), SHOX sequencing with deletion/duplication analysis (n=2), and SNP array (n=9). The majority of females (n=44; 75.9%) and all non-gendered patients were institutionally approved for 50-cell chromosome analysis, whereas the majority of males had orders placed for a SNP array either as a standalone (n=6; 33.3%) or in combination with Fragile X repeat analysis (n=5; 27.8%). Six (7.8%) tests were cancelled by reason of either provider or family request, or insurance denial. An additional 17 (22.1%) patients did not provide a sample for analysis. Seven of the 54 patients whose samples were analyzed received a molecular or cytogenetic diagnosis. The diagnostic yield according to test type is as follows: 8.8% (n=3) for 50-cell chromosome analysis, 33.3% (n=2) for NGS panel, 50% SHOX sequencing + del/dup analysis (n=2), and 20% for SNP array as a standalone test (n=1). 20.8% (n=16) of patients underwent more than one test, performed either reflexively or in tandem with another assay. Subsequent testing showed a 28.6% yield (n=2).
Conclusion:
With the exception of chromosomal abnormalities, the diagnostic rates within our cohort are largely concordant with figures in the published literature; however, our data is impacted by the volume of patients who were lost to follow-up before provision of a sample. Additionally, a minority of patients undergoing testing after receiving non-diagnostic results, potentially further limiting the diagnostic potential of testing. Though concurrent testing may solve for this issue, the cumulative cost of the analyses may become prohibitive to patients seeking diagnosis. Instead, improvements to bioinformatics pipelines built for NGS analysis may provide more robust justification to pursue molecular analysis as an initial evaluation. Platforms optimized for detection of copy number variants (CNVs) with tools to adjust for regions with high sequence homology that includes methods for orthogonal confirmation of low-confidence calls can provide a comprehensive analysis that simplifies provider ordering decisions while improving overall test yield. Testing at commercial laboratories that facilitate at-home collection may also improve patient follow-through to ensure higher catchment. However, more data will be needed to measure the efficacy of a revised testing paradigm.