Introduction: Primary graft failure (PGF) after hematopoietic cell transplant is a life-threatening complication due to the absence of early diagnostic tests, precluding expedited re-transplantation. Two primary mechanisms contribute to PGF: i) impaired donor stem cells and/or host stroma and ii) immune-mediated rejection by host lymphocytes. Thus, we hypothesized that patients with PGF and/or their donors would have higher frequency of germline deleterious variants in cancer predisposition genes and that engraftment-associated cytokines would be higher in recipients preceding the diagnosis of confirmed PGF than matched controls.

Methods: Under IRB approved protocols, a case:control study (1:3 where possible) was performed to identify likely germline pathogenic (P) or likely pathogenic (LP) variants in cancer predisposition genes (CPGs) in either the donor or host from patients enrolled on BMTCTN1202 using augmented whole exome sequencing. Controls were matched on: disease indication, preparative regimen and intensity, donor source, and recipient age, and were alive without relapse nor acute GVHD on day 100. Quality filtered variants were culled by a 0.005 gnomAD allele frequency threshold, except those with high frequency, for which 0.05 was employed. CPGs were curated for pathogenicity per established protocols. Weekly plasma (from infusion to 4 weeks) was assayed for relevant analytes using Mesoscale per manufacturers guidelines. Presence of CPGs were compared between cases and controls using an odds ratio with 95% confidence interval (CI). A mixed model with an interaction was used to compare cases and controls across time (0, 7, 14, 21, 28 days post transplant). An alpha of 0.05 was used.

Results: Among the 34 patients with graft failure, between4 and 71 years old, and 79 controls (21 matched on 3, 3 partially match, 9 unable to be matched), we observed 21% cases and 9% of controls have P/LP likely germline CPGs variants in the donor or recipient, yielding an odds ratio of 2.7 (95% CI 0.9,7.7). There was a high rate of CPGs in genes associated with Noonan syndrome (e.g., LZTR1, NRAS, and PTPN11) in pediatric cases. Of the cytokines/chemokines evaluated, FLT3L on days 14, 21, and 28 (p<0.0001), and IL-15 on days 7 (p=0.0005), 14, 21, and 28 (p<0.0001) were significantly higher in those with graft failure compared to those without graft failure. P-selectin on days 14 (p=0.0151), 21 (p<0.0001), and 28 (0.0002) was significantly lower for those without graft failure compared to those with graft failure.

Conclusions:

Recipients who developed graft failure had more P/LP CPG variants in the case or donor, although the difference was not statistically significant in this small cohort. Several cytokines linked to engraftment were statistically significantly different between cohorts and were biologically rationale. Elevated FLT3L levels have previously inversely correlated with hematopoietic stem cell number and has been linked to certain CPGs as well. IL-15 is a T cell consumptive cytokine that would remain high in those without T cell recovery. P-selectin is upregulated with immune cell activation; thus, low P-selectin could reflect the lack of immune recovery and recognition of organ injury after HCT. Our data suggest that integrating pre-HCT genetic markers with post-HCT analyte profiles may identify those at highest risk for graft failure, permitting pre-emptive interventions, decreasing the morbidity and mortality from primary graft failure.

This content is only available as a PDF.
2025
Sign in via your Institution