HLA Matching Is More Critical Than CMV Matching in Predicting Non-Relapse-Mortality of Patients Undergoing HSCT From An Unrelated Donor: A Retrospective Monocenter Study.

for patients in search of an unrelated donor for haematopoietic stem cell transplantation (HSCT), a 10/10 allele-matched donor represents the best option; for a CMV- patient, a CMV- donor is generally preferred to a CMV+ donor due to low risk of CMV infection when both donor and patients are CMV-, however it is not known at the present how the CMV serological status affects transplant outcome compared to patient-donor HLA matching, and this is particularly true when more than one donor is available. The aim of the present study is to analyze the effect of CMV matching with respect to HLA matching on HSCT outcome, in patients who are CMV-.

CMV-patients who underwent a HSCT from an unrelated donor between 1^st Jan 2000 and 31^st Mar 2009 at our Institution were included in the analysis. Patients receiving HSC from cord blood were excluded. Evaluation of overall survival (OS), progression-free survival (PFS), non-relapse mortality (NRM) and acute graft-versus-host disease (GvHD) were performed with Kaplan-Meier method or cumulative incidence analysis; comparisons among groups were performed using log-rank test. Data about most relevant patients' and donors' factors (patients' age, gender matching, AB0 incompatibility, diagnosis, disease status at HSCT, CMV matching, HLA matching, HSCT conditioning, use of ATG, TBI, source of HSCs) were collected and included in a multivariate Cox regression analysis: the relative impact of CMV and HLA matching on HSCT outcomes were expressed as hazard ratio (HR) together with the 95% confidence interval (CI), after adjustment for the above mentioned factors.

sixty-six patients were eligible for the analysis, median age was 35 (16-64), diagnoses were acute leukemia (n=33: n=23 CR1 and n=10 CR2 or more), MDS (n=8), multiple myeloma (n=8), non Hodgkin's lymphoma (n=6), CML (n=5: n=2 chronic phase, n=3 acceleterated or acute phase), myeloproliferative syndrome other than CML (n=3), aplastic anemia (n=1), CLL (n=1), Hodgkin's lymphoma (n=1). A high-resolution 10/10 HLA matching was present in 55 (83%) HSCTs while one mismatch (9/10-matched donor) was present in 11 (17%) HSCTs: n=8 at HLA-A, n= 3 at HLA-C. All recipients were CMV-. Donors were CMV- in 40 of 55 (73%) 10/10-matched pairs and in 6 of 11 (54%) 9/10-matched pairs. Median follow-up was 225 days (range: 82-2014). In univariate analysis, no significant survival differences were detected among patients with HLA matched and mismatched donors (1y-OS= 42% and 20% respectively, p=0.23) or with CMV- and CMV+ donors (1y-OS= 34% and 48% respectively, p=0.26); no significant differences in occurrence of grade 2-4, grade 3-4 acute GvHD or PFS were observed as well. However, a significantly increased probability of NRM was seen when a HLA mismatched donor was present (1y-NRM= 80% vs 33%, p=0.03); this was not the case of CMV serological status (p=0.31). Multivariate analysis for NRM showed a HR = 5.22 (95% CI: 1.47-18.56) with p= 0.01 for HLA mismatched vs. matched and HR = 0.72 (95% CI: 0.23-2.24) with p= 0.57 for CMV mismatched vs. matched.

these results confirm that HLA matching is critical in determining the incidence of NRM and should continue to be preferred to CMV matching in CMV seronegative patients undergoing HSCT from an unrelated donor. However, a larger, multicenter analysis could better clarify this issue and help to define the relative importance of HLA matching vs. CMV serology in this setting of patients.

No relevant conflicts of interest to declare.