Aim: Outcome of unrelated donor stem cell transplantation (SCT) is influenced by matching for HLA class I and II. We investigated the effects of HLA typing on acute and chronic graft-versus-host disease (aGvHD and cGvHD) and overall survival (OS) in adult patients/donors typed for HLA-A, B, Cw, DRB1, DPB1 at high resolution (HR) and low resolution (LR).

Patients: 100 patients (pts.)(median age 32 y, 17 – 54) were treated with allogenic SCT for diseases: 55 Ph-positive chronic myeloid leukaemia in first chronic phase (CML CP1), 21 CML in advanced phase (CML AP) and 24 acute leukaemias (AL). Pre-transplant conditioning consisted in 98 pts of total body irradiation (TBI) and cyclophosphamide (91) or etoposide (7) and in 2 pts of cyclophosphamide and busulphan. In all pts. GvHD prophylaxis was provided by cyclosporin A and short-term methotrexate with in vivo T-cell depletion (TCD) by CAMPATH 1H (Anti CD52).

Results: With LR typing 68 pts./donors were fully matched but 32 had class I (31) or class II (1) mismatches (MisM). HR typing revealed 9 pts with additional class I (7) or II (7) MisM, thus 59 pts./donors were fully matched. AGvHD grade III–IV occured in 11 pts (11%) whilst extensive cGvHD was seen in 12 pts (14% out of 83). There was no statistical difference in the occurrence of aGvHD or cGvHD between fully matched and mismatched pts./donors at LR or HR. The probability of transplant-related mortality (TRM) of the whole cohort was 15% at day 100 and 23% at day 180 and engraftment failure 1%. The 3-year survival probabilities were significantly different (p=0.004) for pts. with CML CP1 at 63% (good risk group) when compared with CML AP and AL at 23% (poor risk group). In multivariate analysis patients age, diagnosis group and negative CMV status of patient and donor were the only significant predictors for survival. In the entire group the full compatibility at LR or HR for class I or class II or for a single HLA antigen or the KIR ligand mismatch in the GvHD direction did not have an impact on survival. In contrast to the CML CP1 group (n=55) we found that in the poor risk group (n=45) a MisM at HR had a negative impact on survival (p=0.03) and a trend towards worse survival was found for patients with HLA B HR mismatch. Patients with more than one HR MisM had a worse survival when the combination of HLA B and HLA D MisM was present (p=0.008).

Conclusion: In 100 pts. treated with unrelated SCT and in vivo TCD with CAMPATH 1H the HR typing did have an impact on survival only in the poor risk group but not in pts with CML CP1. Thus, T-cell depletion with in vivo CAMPATH 1H permits the use of mismatched unrelated donors in patients with CML CP. The overall rates for aGvHD and cGvHD were low compared to published data in non TCD SCT. Further investigations on larger cohorts are needed to elucidate the impact of HR typing in unrelated SCT and the significance of HR MisM at single HLA loci in transplants T-cell depleted with in vivo CAMPATH 1H.

Author notes

Corresponding author