High-Throughput Genetic Testing for Thrombotic Microangiopathies and C3 Glomerulopathies

Abstract

The thrombotic microangiopathies (TMAs) and C3 glomerulopathies (C3Gs) include a spectrum of rare diseases such as atypical hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, C3GN, and dense deposit disease, which share phenotypic similarities and underlying genetic commonalities. Variants in several genes contribute to the pathogenesis of these diseases, and identification of these variants may inform the diagnosis and treatment of affected patients. We have developed and validated a comprehensive genetic panel that screens all exons of all genes implicated in TMA and C3G. The closely integrated pipeline implemented includes targeted genomic enrichment, massively parallel sequencing, bioinformatic analysis, and a multidisciplinary conference to analyze identified variants in the context of each patient's specific phenotype. Herein, we present our 1-year experience with this panel, during which time we studied 193 patients. We identified 17 novel and 74 rare variants, which we classified as pathogenic (11), likely pathogenic (12), and of uncertain significance (68). Compared with controls, patients with C3G had a higher frequency of rare and novel variants in C3 convertase (C3 and CFB) and complement regulator (CFH, CFI, CFHR5, and CD46) genes (P<0.05). In contrast, patients with TMA had an increase in rare and novel variants only in complement regulator genes (P<0.01), a distinction consistent with differing sites of complement dysregulation in these two diseases. In summary, we were able to provide a positive genetic diagnosis in 43% and 41% of patients carrying the clinical diagnosis of C3G and TMA, respectively.

Keywords: C3 Glomerulopathy; Genetic testing; Thrombotic Microangiopathy.

Figure 1.

Variant interpretation is often challenging and generally requires multidisciplinary knowledge and the integration of information from multiple sources. Genetic data from all patients studied in our TGE&NGS pipeline are reviewed in a multidisciplinary care conference (Renal Group Meeting) during which time genetic data are discussed in light of all phenotypic (clinical) data available to generate a consensus Final Report (see Supplemental Figure 1). Variants are labeled as: pathogenic, likely pathogenic, variant of uncertain significance (VUS), likely benign, or benign based on predetermined metrics.

Figure 2.

Ninety-two subjects were analyzed using both TGE&NGS and Sanger sequencing as a validation step for the GRP. Nearly all exonic variants have very high quality (red box), reflecting an effective enrichment strategy. The single exception was a small portion of CFH (green box), where we observed both false-positive and false-negative calls. For this reason, exons 20–22 of CFH are always Sanger-sequenced (see Supplemental Figure 1). Some low QD variants in introns of CFI and C3 (blue box) were ignored because they did not impact exons and splice sites. TP, true positive; FP, false positive; FN, false negative; ND, no Sanger sequencing data; UTR, untranslated region.

Figure 3.

VUS/likely pathogenic/pathogenic variants distributed unevenly across genes and gene groups in TMA and C3G patients. (A) Variant load in CFH, CFI, and C3 was high for both patients with TMA and C3G. (B) Variants accumulated in C3 convertase genes (C3 and CFB) in patients with C3G and in AP regulator genes (CFH, CFI, CD46, and CFHR5) in patients with TMA. (C) Patients with C3G and TMA were more likely to carry single rare/novel variants than control samples retrieved from 1000 Genomes Project (1000Gs). (D) More patients with C3G and TMA carried VUS/likely pathogenic/pathogenic variants than did patients with untargeted diseases (see Table 2). (Fisher’s exact test was used to compare intergroup differences).