Impact of CTLA4 Genotype and Other Immune Response Gene Polymorphisms on Outcomes after Umbilical Cord Blood Transplantation - a Eurocord, CBC-Ctiwb-EBMT, Netcord and FMRP-USP Study

Introduction: Donor and recipient polymorphisms of immune defense and inflammatory cytokine genes have been associated with outcomes after bone marrow or peripheral blood stem cell transplantation (SCT), but have not been described in the setting of umbilical cord blood transplantation (CBT).

Material and Methods: We evaluated the impact of genetic polymorphisms of recipients and CBU units on survival and other CBT outcomes using the following candidate genes related to immune response: NACHT-leucine-rich repeat with a pyrin domain at N-terminus (NLRP1-rs5862, NLRP2-rs043684, NLRP3-rs10754558), Toll-interleukin-1 receptor domain containing adaptor protein (TIRAP/Mal-rs8177374), interleukin-10 (IL10-rs1800872), V-rel reticuloendotheliosis viral oncogene homolog (REL-rs13031237), tumor necrosis factor receptor superfamily - member 1B (TNFRSF1B-rs1061622) and associated protein 4 of cytotoxic T lymphocyte (CTLA4-rs3087243). Seven Netcord banks provided pre-transplant DNA samples from 851 CBU units and 173 recipients. All patients underwent CBT at EBMT centers. Malignant (n=696) and non-malignant (n=155) diseases were analyzed separately.

Results: Among the 851 recipients, 57% were male and 61% adults. Degree of HLA matching between CBU unit and recipient (HLA-A, -B at antigen level and -DRB1 at allele level) was 6/6 in 12%, 5/6 in 40%, and ≤4/6 in 48% of transplants. Myeloablative conditioning was used in 77% of cases and anti-thymocyte globulin or monoclonal antibodies in 82% of cases. Median infused total nucleated cells (TNC) and CD34+ cells were 3.7x10⁷/kg and 1.6x10⁵/kg, respectively. In the cohort with malignant diseases, multivariable analysis adjusted for patient-, donor- and disease-related variables showed that recipients of CBU with GG CTLA4 genotype had a poorer neutrophil engraftment (HR 1.24; 95% CI 1.09-1.40; p=0.03), increased NRM (HR 1.52; 95% CI 1.35-1.72%; p<0.01) and inferior disease-free survival (DFS) (HR 1.41; 95% CI 1.06-1.88; p=0.02), whereas the AA CTLA4 genotype was associated with lower relapse rate (HR 0.64; 95% CI 0.57-0.72; p=0.02). Recipients of GG TNFRSF1B CBU had lower platelet recovery (HR 1.94; 95% CI 1.71-2.19; p=0.02). With the aim to confirm these results in a more homogenous cohort of recipients, we analyzed recipients with acute leukemias, receiving myeloablative conditioning and with available high resolution typing of HLA-A,-B,-C and-DRB1 (n=305). In this subset analysis, recipients of GG CTLA4 CBU had increased NRM (HR 1.72; 95% CI 1.10-2.70%; p=0.02), but no impact on DFS and relapse was observed. In the cohort of patients with non-malignant diseases (n=155), there was no statistical association between the selected candidate genes and outcomes. Similarly, all recipient gene polymorphisms tested (n=173) were not associated with CBT outcomes.

Conclusion: In this retrospective analysis, CBU gene polymorphisms CTLA4 and TNFSF1B impacted CBT outcomes. Importantly, CTLA4 GG genotype was associated with inferior survival and the AA genotype with lower relapse rate in patients with malignant disorders. The possible biological explanation of the impact of CTLA4 is that GG genotype is associated with decreased production of soluble CTLA4 (whose primarily functions is to inhibit T-cell activation) which contributes to decrease suppression of T-cell alloimune responses post CBT If these findings are confirmed in further studies, CTLA4 gene polymorphisms can be used among other factors to select CBU with the aim to improve survival after CBT.