Graft-versus-host disease (GVHD) is a cause of considerable morbidity and mortality after allogeneic hematopoietic cell transplantation (HCT) following nonmyeloablative conditioning. Genetic polymorphisms in various genes, associated with the immune system have been implicated in the development of GVHD. The PTPN22 gene encodes a lymphoid tyrosine phosphatase (LYP), which is involved in suppression of T-cell receptor signalling. The 1858 C→T polymorphism in PTPN22 entails an amino acid substitution that disrupts the suppressive function of LYP, rendering T-cells hyperresponsive. The 1858 T allele has been implicated in conferring increased susceptibility to various autoimmune diseases. As in autoimmunity, T-cell activation plays a key role in the development of GVHD. To asses the influence of the protein tyrosine phospatase N22 (PTPN22) 1858 C→T polymorphism on development of GVHD after allogeneic HCT following nonmyeloablative conditioning, 100 consecutive patient-donor pairs receiving allogeneic HCT with related (n=66) or unrelated (n=34) donors for hematological malignancies (HD=13, MM=14, CLL=12, NHL=17, MDS=18, AML=24, CML=2), between March 2000 and December 2005 at Rigshospitalet, Denmark, were genotyped. With a mean follow-up of 534 days (range 38–2324 days) the overall survival (OS), progression free survival (PFS), treatment related mortality (TRM) and relapse related mortality (RRM) were 59%, 50%, 25% and 17%. The C/T and T/T genotypes were present in 16% and 1% of the recipients, respectively, and in 17% and 1% of the donors. The overall cumulative incidence of grade 2–4 acute GVHD, grade 3–4 acute GVHD and extensive chronic GVHD was 67%, 24%, and 49%, with no significant difference between patients carrying the C/C or C/T and T/T genotype or donors carrying the C/C or C/T and T/T genotype. To assess a possible gene-dosage effect, the number of T-alleles in each recipient-donor pair was cumulated, and the cumulative incidence of grade 3–4 acute GVHD increased from 20% in recipient-donor pairs carrying no or one T-allele to 50% in recipient-donor pairs carrying two or more T alleles (p=0.04), while there was no significant difference in grade 2–4 acute GVHD (66% versus 80%; p=0.47) and extensive chronic GVHD (50% versus 42%; p=0,8) between groups. In the competing risk regression analysis, the recipient-donor pair genotype with 2 or more T-alleles was an independent risk factor (hazard ratio 3.0; 95% CI 1.2–7.5; p=0.022) for development of grade 3–4 acute GVHD, even after adjusting for baseline variables known to affect GVHD rates (donor type and sex-mismatch, patient and donor age, CD34+ cell dose, single HLA locus mismatch). Overall survival, PFS, TRM or RRM were not significantly different for recipient-donor pairs for any given combination of alleles. Furthermore, patients from recipient-donor pairs carrying two or more T-alleles were hospitalized for significantly more days (p=0.01) due to GVHD (median=15 days; range 0–63 d), than patients from recipient-donor pairs with no or one T-allele (median=0 days; range 0–104 d). Collectively, our data suggest, that the PTPN22 1858 C→T polymorphism, when present in both recipient and donor, is a risk factor for development of grade 3–4 acute GVHD after nonmyeloablative conditioning allogeneic HCT.