Incidence, Risk Factors and Outcomes of Sclerotic Chronic Graft-Versus-Host Disease (GVHD) Phenotype After Allogeneic Hematopoietic Cell Transplantation

Abstract

Abstract 322 Background:A sclerotic GVHD phenotype represents one of the most severe late complications of allogeneic hematopoietic cell transplantation (HCT), sharing some clinical similarities with systemic sclerosis. The incidence and risk factors of sclerotic chronic GVHD and its association with major transplant outcomes are not well established. Patients and methods: We analyzed 986 consecutive patients who received systemic therapy for chronic GVHD after a first allogeneic HCT performed between 2000 – 2008 in Seattle for malignant and nonmalignant diseases. Chronic GVHD was diagnosed by the NIH consensus criteria. Sclerotic GVHD was defined when manifestations of cutaneous sclerosis, fasciitis or joint contracture were first documented in the medical record. Multivariate Cox regression analyses accounted for patient and donor age per decade, donor type, HLA matching, ABO matching, diagnosis and disease risk, stem cell source, patient gender, intensity of conditioning regimen, use of total body irradiation (TBI) or rabbit antithymocyte globulin in the conditioning regimen, and GVHD prophylaxis. Risk factors considered at the onset of chronic GVHD were prior acute GVHD, prior severe (stage 3 and 4) skin acute GVHD, eosinophilia, thrombocytopenia, progressive onset, extent of skin involvement, and bronchiolitis obliterans. Overall mortality (OM), nonrelapse mortality (NRM; malignant disease only) and recurrent malignancy (malignant disease only) were compared between patients with and without sclerotic features using time-dependent Cox models. Results: Of the 986 patients, median age was 47 (0–78) years, 776 (79%) were Caucasian, 950 (96%) had malignant diseases, 407 (41%) had HLA-matched related donors, 356 (36%) had HLA-matched unrelated donors, and 223 (23%) had HLA-mismatched related or unrelated donors. The median time from HCT to chronic GVHD was 5.3 (2.5–46) months; 73 patients (7%) presented with sclerotic features at initial diagnosis and 142 patients (14%) developed sclerosis after the onset of chronic GVHD. The cumulative incidence of sclerotic GVHD was 20% (95%CI, 18–23) at 36 months after onset of chronic GVHD, and median onset of sclerosis was 7.9 months after onset of chronic GVHD (Figure). In multivariate models (Table), factors associated with an increased risk of sclerotic GVHD were mobilized blood cell graft, female recipient, TBI >450cGy, and prior severe skin acute GVHD. Factors associated with a decreased risk of sclerotic GVHD were HLA-mismatched donors and ABO major mismatch. Risks of OM, NRM and recurrent malignancy in patients with sclerotic features did not differ statistically from patients without sclerotic features (hazard ratio (HR) 0.97, 95% CI 0.7–1.3, p=0.83; HR 0.94, 95% CI 0.6–1.4, p=0.78; HR 0.73, 95% CI 0.5–1.2, p=0.17, respectively). Conclusion: Sclerotic GVHD is an underestimated phenotype of chronic GVHD. Female predominance in sclerotic chronic GVHD is similar to that in systemic sclerosis. Risks of major transplant outcomes do not differ statistically between patients with and without sclerotic chronic GVHD phenotype. Mechanisms that account for the decreased risk of sclerosis associated with HLA and ABO mismatching and for the increased risk associated with prior severe skin acute GVHD warrant future investigation. [Display omitted] TableMultivariate analyses for risk factors of sclerotic GVHD (n=905)HR (95% CI)PCovariates defined at transplantStem cells Bone marrow1.0 PBSC2.51 (1.4–4.6).003 Cord blood–nsHLA and donor HLA-matched related1.0 HLA-matched unrelated–ns HLA-mismatched0.54 (0.3–0.9).02Patient age–nsDonor age–nsPatient gender Male1.0 Female1.61 (1.2–2.2).005Donor gender pair–ns Female donor to male recipient–nsABO mismatch None1.0 Minor–ns Major0.61 (0.4–0.9).02Disease diagnosis–nsDisease risk–nsConditioning intensity–nsTotal body irradiation None1.0 ≤ 450 cGY–ns > 450 cGY1.84 (1.3–2.6).001ATG in conditioning–nsGVHD prophylaxis–nsCovariates at onset of chronic GVHDPrior grade II-IV acute GVHD–nsPrior stage 3 to 4 skin acute GVHD1.88 (1.3–2.6).0002Eosinophilia–nsPlatelets < 100 ×109/L–nsOn steroids or progressive onset–nsSkin involvement–nsExtent of skin involvement–nsBronchiolitis obliterans–nsns, not statistically significant in univariate analysis and not included in multivariate model Disclosures:No relevant conflicts of interest to declare.