Skip to main content

Conference Program

Subpage Hero

Loading

Loeys-Dietz Syndrome: Phenotyping by Two Genotype-First Methods

Clinical Genetics and Therapeutics
  • Primary Categories:
    • Genomic Medicine
  • Secondary Categories:
    • Genomic Medicine
Introduction:
INTRODUCTION: Large-scale genomic sequencing provides an opportunity to identify people at risk for potentially covert but life-threatening genetic conditions such as Loeys-Dietz syndrome (LDS). However, it remains unknown whether genomically ascertained individuals, who may be unaware of their genetic risk, nevertheless receive clinical diagnoses and evaluations consistent with LDS. We identified individuals heterozygous for LDS-related genetic variants in two large genomic sequencing databases and performed phenotyping through either review of electronic health record (EHR) data or recall-by-genotype and tailored clinical testing, also known as reverse phenotyping.

Methods:
METHODS: We used a paired clinical informatics and reverse phenotyping approach to identify individuals with putative deleterious variants in SMAD2, SMAD3, TGFB2, TGFB3, TGFBR1 and TGFBR2. Over 245,000 adults in the All of Us Research Program with genomic sequencing data were assessed for variants identified in an LDS specialty clinic population or >=two-star Pathogenic or Likely Pathogenic variants from ClinVar. Phenotype-wide association studies (PheWAS) and individual-level EHR data were analyzed for LDS-related phenotypes. Separately, adults with potentially deleterious variants based on ACMG variant classification guidelines were recontacted for tailored LDS clinical testing from a large genomic research sequencing database at the National Institutes of Health.

 

Results:
RESULTS: From 178 clinically observed variants and 115 >=two-star Pathogenic or Likely Pathogenic variants, we identified 24 variants in 5 genes across 37 All of Us participants, 30 of whom had EHR data available for abstracting. PheWAS demonstrated significantly increased odds of cerebral and other aneurysms (Odds Ratio (OR) 20.1 [95% CI 5.87 – 68.85] and 14.65 [5.19-41.39], respectively) and other and unspecified congenital anomalies (OR 23.75 [5.62-100.38]). On an individual level, participants with LDS-associated variants had a range of 0-19 unique LDS-affiliated diagnostic codes in their EHR, with a median of 4.5 codes and mean of 4.73 codes per person. There were multiple potential misdiagnoses of Marfan syndrome. Age at first documentation of LDS-associated phenotypes ranged across the lifespan, indicating the importance of ongoing surveillance for individuals that do not initially present with hallmark overt features of LDS. However, the phenotypic yield of this clinical informatics approach was complicated by a substantial proportion (N<20) of participants who did not have record of specialized diagnostic procedures such as an echocardiogram, computed tomography angiography, or magnetic resonance angiography. From four putative deleterious variants identified in the reverse phenotyping cohort, two out of four participants agreed to complete follow up clinical testing for features of LDS. Both participants had previously been diagnosed with stable aortic root dilation. One participant heterozygous for TGFB3 NM_0003239.5:c.638del p.(Leu213*) met molecular diagnostic criteria for LDS after reverse phenotyping. One participant heterozygous for SMAD3 NM_005902.3:c.1091A>G  p.(Tyr364Cys) did not meet diagnostic criteria but did have a positive family history for a clinical diagnosis Marfan syndrome. This participant had previously tested negative for variants in COL3A1, TGFBR1 and TGFBR2.

Conclusion:
CONCLUSIONS:  Genomic ascertainment through EHR-linked biobanks like All of Us provides an opportunity to assess genotype-phenotype associations and trends in diagnosis and screening at large scale, while reverse phenotyping enables more comprehensive evaluation.  Given the phenotypic heterogeneity and potential for unrecognized phenotypes in LDS, genomic ascertainment can provide a unifying diagnosis for individuals with presenting symptoms and the opportunity to screen at-risk individuals lacking obvious features of LDS.

Agenda

Sponsors