The Genetic Fingerprint of Susceptibility to Transplant Associated Thrombotic Microangiopathy

Background:  Transplant associated thrombotic microangiopathy (TMA) is frequent after HSCT, and in severe cases causes significant morbidity and mortality.  Currently there are no data addressing individual susceptibility to transplant associated TMA.  Frequent genetic variants in ADAMTS13 and complement genes have been described in other microangiopathies such as aHUS and TTP.  We hypothesized that polymorphic variants in complement genes increase susceptibility to transplant associated TMA.

Methods: 90 consecutive allogeneic HSCT patients were enrolled on a prospective TMA biomarker study and categorized as having TMA or no TMA using rigorous diagnostic criteria (Jodele et al, Blood 2014).  Genomic DNA was isolated from pre-HSCT recipient blood.  We used a candidate gene approach to identify 12 genes within the complement pathway likely to play a role in terminal complement activation, the likely effector mechanism for vascular damage in TMA.  All exons, flanking intronic and untranslated regions of ADAMTS13, CFH, CFI, CFB, MCB, THBD, C3, C5, CFD, CFHR1, CFHR3 and CFHR5 were sequenced using next generation sequencing technology. The resulting sequence reads were aligned against the reference DNA sequence and the variants were detected using NextGENe software. Observed variants were compared against dbSNP (NCBI). Novel variants were further evaluated using laboratory developed bioinformatic tools. Eight variants previously described as likely pathogenic and 34 variants of unknown pathogenic significance were identified.   Results were compared in patients with and without TMA.

Results: 77 patients had DNA available for analysis (34 with TMA and 43 without TMA); 23 of 34 (68%) subjects  with TMA had complement gene variants as compared to 4 of 43 (9%) controls without TMA (p<0.0001).   Figure shows a “heat map” of gene variants in recipients with and without TMA  illustrating the wide distribution of variants in multiple genes in the TMA cases, while few variants are seen in those without TMA.   Likely pathogenic variants previously described in other microangiopathies, were seen in ADAMTS13 (n=2), CFH, (n=2), CFB (n=1), CFHR5 (n=1) and CFI (n=3), all in recipients with TMA.  No known pathogenic variants were seen in subjects without TMA.  Two recipients with TMA had homozygous deletion of CFHR3/R1, which was not seen in the recipients without TMA.  The median number of gene variants (of known or unknown significance) seen in recipients with TMA was 1 (0-7), and 0 (0-2) in those without TMA (p<0.0001).  Two subjects had 5 variants and one 7 variants and all three had fatal TMA.

Conclusion: The incidence of complement gene variants was higher in patients with TMA as compared to patients without TMA, indicating that genetic susceptibility importantly alters risk. Additional larger studies are needed to define the mechanistic importance of variants of unknown clinical significance.