Diagnostic high-throughput sequencing of 2396 patients with bleeding, thrombotic, and platelet disorders

Abstract

A targeted high-throughput sequencing (HTS) panel test for clinical diagnostics requires careful consideration of the inclusion of appropriate diagnostic-grade genes, the ability to detect multiple types of genomic variation with high levels of analytic sensitivity and reproducibility, and variant interpretation by a multidisciplinary team (MDT) in the context of the clinical phenotype. We have sequenced 2396 index patients using the ThromboGenomics HTS panel test of diagnostic-grade genes known to harbor variants associated with rare bleeding, thrombotic, or platelet disorders (BTPDs). The molecular diagnostic rate was determined by the clinical phenotype, with an overall rate of 49.2% for all thrombotic, coagulation, platelet count, and function disorder patients and a rate of 3.2% for patients with unexplained bleeding disorders characterized by normal hemostasis test results. The MDT classified 745 unique variants, including copy number variants (CNVs) and intronic variants, as pathogenic, likely pathogenic, or variants of uncertain significance. Half of these variants (50.9%) are novel and 41 unique variants were identified in 7 genes recently found to be implicated in BTPDs. Inspection of canonical hemostasis pathways identified 29 patients with evidence of oligogenic inheritance. A molecular diagnosis has been reported for 894 index patients providing evidence that introducing an HTS genetic test is a valuable addition to laboratory diagnostics in patients with a high likelihood of having an inherited BTPD.

Graphical abstract

Figure 1.

Classification of patients using clinical and laboratory phenotypes. (A) Classification of 2396 index patients from the ThromboGenomics, VIBB, and PANE cohorts into 1 of 5 disease classes: thrombotic, platelet count, platelet function, coagulation, and unexplained bleeding. (B) Representative HPO codes for patients characterized in each of the 5 disease classes.

Figure 2.

Diagnostic yield and proportion of novel variants by disease class. (A) Diagnostic yield of reported variants for 2396 index patients for each of the 5 disease classes: thrombotic, platelet count, platelet function, coagulation, and unexplained bleeding. For patients with >1 reported variant, the most pathogenic variant was used in this analysis (n = number of index patients). (B) Proportion of reported variants that were novel or known for patients in each disease class (n = number of variants).

Figure 3.

Gene ranking according to the number of reports per disease class. For each disease class, genes were ranked according to number of times they were reported (A-D). Dashed lines represents, from top to bottom, the 25th, 50th, and 75th quantiles. Recently discovered genes and changes of mode of inheritance are in blue.

Figure 4.

Oligogenic variants in patients with thrombotic (red) and coagulation (blue) disorders. From outside to inside: track 1, amino acid numbering for thrombotic (red) and coagulation (blue) genes (lighter shade denotes untranslated regions of the 3′ of the F2 gene and the 5′ of the PROS1 gene; track 2, amino acid conservation scores; track 3, variant frequency in gnomAD (minor allele frequency normalized scale to 1/106); track 4, disease-causing (red) and questionable disease-causing (orange) Human Gene Mutation Database variants; track 5 and arcs: reported variants in the 11 thrombotic (red) and 13 coagulation (blue) patients with the arcs representing oligogenic findings.