Investigating the Effects of Clonal Hematopoiesis in Healthy Marrow and Blood Donors on the Outcomes of Recipients Following Hematopoietic Stem Cell Transplantation

Background: Using error-corrected sequencing with a detection limit of 0.0001, Young et al. (2016) found that 95% of healthy individuals harbor somatic mutations in blood cells in genes associated with leukemia. In addition to increasing the risk of AML (Jaiswal et al. 2014), these clones have recently been associated with cardiac dysfunction (Fuster et al. 2017; Jaiswal et al. 2017). This prompted us to consider the impact of clonal hematopoiesis from healthy donors on outcomes in recipients of allogeneic hematopoietic stem cell transplantation (HSCT). Indeed, donor-derived hematological complications have been anecdotally documented where rare donor clones harboring pathogenic mutations expanded in the recipients following HSCT. As a result, hematopoietic progenitors harboring mutations conferring self-renewal or growth advantages that facilitate preferential engraftment could be unknowingly transferred from donor to recipient (Gibson et al. 2017). Several retrospective studies of secondary hematopoietic malignancies post-HSCT have shown that donor clones harboring mutations in JAK2 and DNMT3A with low variant frequency (VAF) have undergone clonal expansion in the recipients (Yasuda et al. 2014). To date, there has not been a systematic and quantitative study to assess how donor clonal hematopoiesis engrafts and potentially affects the clinical outcome of recipients. In this study, we aim to quantify the spectrum of donor clones that engraft the recipient by longitudinally tracking clonal dynamics in the recipients at three time points post-HSCT. We also aim to compare individual patient outcomes with clonal engraftment. Methods: Error-corrected sequencing (ECS) on 80 genes frequently mutated in AML was performed on 133 isolated peripheral blood leukocytes from 27 allogeneic HSCT recipients at Washington University and the Siteman Cancer Center. Five samples were obtained for each patient which include the matched donor sample from the CIBMTR (n=25) along with pre-transplant, day 30 post-HSCT (D30), day 100 post-HSCT (D100) and one year post-HSCT (D360) for all patients. Independent biological replicates were sequenced for each sample and only the somatic events found in both replicates will be considered true positives using published and validated computational pipelines and thresholds. Results and Conclusions: All D30 and D100 samples harbor clonal mutations (mean 5.67 variants in D30 samples; 5.52 variants in D100 samples) with VAFs ranging from 0.0005 - 0.11. A total of 126 variants are only observed in D30 samples while 122 variants are only observed in D100. Twelve patients have 27 filtered mutations observed in both time points with general increase in VAFs in the later time point. Interestingly, we found SRCAP to be most recurrently mutated in both D30 and D100 samples (48.1% and 33.3%, respectively). This gene has recently been implicated in therapy-related clonal hematopoiesis following cytotoxic treatment (Wong et al. 2018). SRCAP is not frequently observed in clonal hematopoiesis in individuals without preceding hematological disorders or cytotoxic treatments. Therefore, these observed SRCAP variants could either: 1) be present in donors in very low VAFs or 2) arise after transplantation. Analysis of donor sequencing data would provide insight into this. Overall, we find that ECS is a specific and sensitive method for quantitatively characterizing the dynamics of clonal engraftment and proliferation in allogeneic HSCT recipients. In addition, we have found that mutated SRCAP appears to promote clonal engraftment and expansion after conditioning therapy, similar to recent results post-chemotherapy. Future work: Computational analysis of donor and recipient pre-transplant samples along with D360 samples is underway and nearly complete. Correlating ECS results with clinical and demographic (e.g. age and gender of donor and recipient) data is also underway. Disclosures No relevant conflicts of interest to declare.