Genetic Susceptibility in Sinusoidal Obstruction Syndrome/Veno-Occlusive Disease

Novel biologics have reinstated our interest in sinusoidal obstruction syndrome/veno-occlusive disease (SOS/VOD) as a severe complication of allogeneic hematopoietic cell transplantation HCT. In parallel, we have pioneered significant advances on complement dysregulation in patients with transplant-associated thrombotic microangiopathy, identifying complement-related gene variants. There is a high unmet need for personalized prophylaxis treatment in patients who are at high risk of SOS/VOD and early treatment thus we developed this multinational and multicenter effort to investigate genetic susceptibility in patients with SOS/VOD.

We enrolled consecutive adult SOS/VOD patients diagnosed with moderate, severe, or very severe SOS/VOD based on the revised EBMT criteria and TA-TMA controls diagnosed based on the International Working Group (IWG) criteria, in 1:1 ratio by 3 JACIE-accredited Units. Genomic DNA was isolated from peripheral blood samples before HCT and analyzed using Next-Generation Sequencing (NGS) for a customized panel of genes related to the complement system (MiniSeq, Illumina): complement factor H/CFH, CFH-related, CFI, CFB, CFD, C3, CD55, C5, CD46, thrombomodulin/THBD, including the ADAMTS13 gene to identify mutations in key genes associated with endothelial function and coagulation pathways. The probes were designed with DesignStudio (Illumina, San Diego, California) to cover all exons and an additional 15 bases within the intronic region (98% coverage). For library preparation, 10 ng of input DNA was used. The libraries were quantified using Qubit and sequenced on the MiniSeq System (Illumina). Bioinformatics and statistical analysis were subsequently conducted to identify statistically significant differences between the groups and correlations with clinical-laboratory data. Complement database and ClinVar were used to characterize likely pathogenic or pathogenic variants. Additionally, the Ensembl Variant Effect Predictor (VEP) tool detected rare variants with a minor allele frequency (MAF) less than 1% (0.01) in a European combined population. Four bioinformatic tools (SIFT, PolyPhen-2, PROVEAN, CADD) were used to evaluate variants’ clinical significance. Variants predicted to be deleterious from at least 2 tools (50%), were also considered pathogenic.

We studied 30 SOS/VOD and 30 TA-TMA patients. Baseline pre-transplant characteristics were similar between groups, with the exception of pre-transplant gemtuzumab or inotuzumab treatment which was recorded in 4 patients with SOS/VOD, compared to 1 with TA-TMA.

In total, 426 variants were recorded after the removal of duplicates, from which 20 pathogenic variants were identified. Regarding SOS/VOD, four variants were detected in ADAMTS13 gene (rs28647808 in 5 patients, rs28503257 in 1 patient, rs36090624 in 1 patient, rs41314453 in 2 patients), three in CFH (rs35343172 in 1 patient, rs35274867 in 1 patient, rs3753396 in 6 patients), two in C3 (rs117793540, rs149850773 in 1 patient each), and one variant in CFB (rs12614 in 10 patients). Although functional characterization is not complete for all variants, these have previously described in complement-related, thrombotic and autoimmune disorders. Interestingly, rs41314453 has been also identified as the strongest genetic predictor of ADAMTS13 activity.

In TA-TMA, patients exhibited a higher frequency of variants in complement-related genes, including CFH gene (rs148403790 in 2 patients, rs149474608 in 1 patient, rs145975787 in 1 patient, rs513699 in 3 patients, rs61822181 in 1 patient, rs460897 in 2 patients), CFI (rs7437875, rs114013791, rs41278047, rs146444258 in 1 patient each), and C3 (transcript ENST00000245907 in 1 patient), mostly involved in complement-related disorders.

Our findings reveal pathogenic germline variants in pre-transplant samples of patients that developed SOS/VOD, highlighting the role of multiple pathways involving the endothelium, coagulation, and complement. Pathogenic genetic variants identified in TA-TMA were not common with those observed in SOS/VOD, suggesting distinct mutational patterns between the SOS/VOD and TA-TMA groups. Association of these variants with phenotypical characteristics is needed, along with potential exploration of effects due to combination of variants. Validation of the results is warranted to in larger patient and control cohorts.