Clinical Significance of Severity and Subcategory of Chronic GVHD Evaluated by NIH Consensus Criteria.

Chronic GVHD (cGVHD) represents a major cause of morbidity and mortality after allogeneic hematopoietic stem cell transplantation (allo-SCT). Recently, NIH consensus criteria were proposed to standardize the diagnosis and global assessment of cGVHD with a new severity scoring system based on organ-specific manifestations taking functional impact into account. However, clinical significance of this grading system is not fully established although several studies have shown its impact on post-GVHD survival.

We retrospectively evaluated the incidence and outcome of cGVHD defined by NIH consensus criteria in 294 patients with hematologic disorders who consecutively underwent allo-SCT between January 2000 and December 2008 at our department. After excluding patients who received two allogeneic transplants, rejected graft, died or relapsed before day 100 after allo-SCT, and received donor lymphocyte infusion before day 100, a total of 213 patients with a median age of 46 (range 17-69) years, 120 males and 93 females, were evaluated in the study; 113 had myeloid neoplasms, 93 had lymphoid neoplasms, and the remaining 7 had non-malignant diseases. Non-malignant disease and hematologic malignancy in fist complete remission, first chronic phase or without prior chemotherapy were classified as standard risk disease (n=107). All other disease statuses were classified as high risk disease (n=106). A total of 129 patients received myeloablative conditioning and 84 received reduced-intensity regimens. Stem cell sources included peripheral blood (n=44) or bone marrow (n=62) from related donors, and bone marrow (n=91) or cord blood (n=16) from unrelated donors. Diagnosis and global scoring of cGVHD were performed according to NIH consensus criteria. cGVHD-specific survival (cGSS) was defined as the length from the day of cGVHD diagnosis to the day of death, relapse, or last follow-up. Competing risk regression analyses were used to evaluate risk factors associated with development of cGVHD and time from diagnosis of cGVHD to discontinuation of systemic immunosuppressive therapy (IST). Multivariate Cox regression analyses were used to evaluate factors associated with cGSS.

After a median follow-up of 1133 days (range 101-3396 days), 96 patients developed diagnostic or distinctive manifestations of cGVHD according to NIH consensus criteria with a median time to diagnosis of 205 days (range 64-911 days); 77 (80%) patients were subcategorized into classic cGVHD and 19 (20%) into overlap syndrome. In multivariate analysis, the prior occurrence of grade 2-4 acute GVHD (hazard ratio [HR]=1.50, 95% confidence interval [CI]=1.01-2.24, p=0.047) was significantly associated with the development of cGVHD. Four-year probabilities of cGSS among the patients with classic cGVHD and overlap syndrome were 88% and 63%, respectively, while those among patients graded to have mild (n=20; 21%), moderate (n=53; 55%) and severe cGVHD (n=23; 24%) were 100%, 86%, 62%, respectively. Patients with overlap syndrome had marginally lower cGSS as compared to those with classic cGVHD (HR=2.77, 95%CI=0.96-7.98, p=0.06), while patients with severe cGVHD had significantly lower cGSS as compared to those with mild or moderate cGVHD (HR=3.11, 95%CI=1.09-8.87, p=0.034) in multivariate analysis. Among 82 patients who received systemic IST as initial treatment of cGVHD, the probability of withdrawal of systemic IST at 2 years was 36 %. Multivariate analysis revealed that the use of corticosteroid at cGVHD diagnosis (HR=0.40, 95%CI=0.19-0.84, p=0.015) and high risk disease status at transplant (HR=0.39, 95%CI=0.20-0.77, p=0.007) were significantly associated with prolonged systemic IST.

Our results suggest that the global severity scoring and subcategory of cGVHD evaluated by NIH consensus criteria have prognostic value for predicting cGSS among Japanese patients with cGVHD, implying the applicability of these criteria for the study encompassing participants from various ethnic backgrounds. Larger and prospective studies are warranted to more precisely validate clinical significance of NIH consensus criteria for cGVHD.

No relevant conflicts of interest to declare.