Predictors of hematologic malignancy relapse in patients with advanced chronic graft-versus-host disease.

Malignancy relapse remains a major barrier to treatment success in patients after allogeneic hematopoietic stem cell transplantation (allo-HSCT). Chronic graft-versus-host disease (cGVHD) markedly reduces hematologic malignancy relapse risk, but relapses still occur in these patients. Patients (n = 275) with moderate or severe cGVHD were enrolled on the National Cancer Institute (NCI) prospective cross-sectional natural history study (NCT00092235). Subjects were median 36 months after allo-HSCT and were followed subsequently for malignancy relapse and survival. Seventeen patients experienced relapse. In a multivariable model including time-dependent influences on relapse, risk factors associated with increased risk of relapse included shorter time from transplant to cGVHD evaluation (HR 0.279, 95% CI 0.078-0.995) and lower number of prior lines of systemic immunosuppressive therapy for cGVHD (HR 0.260, 95% CI 0.094-0.719). In a model excluding time-dependent influences on relapse risk, lower number of prior lines of systemic immunosuppressive therapy for cGVHD (HR 0.288, 95% CI 0.103-0.804), lower C4 complement level (HR 0.346, 95% CI 0.129-0.923), and higher body mass index (HR 3.222, 95% CI 1.156-8.974), were all associated with increased relapse risk. Parameters indicating cGVHD severity and activity are associated with risk of malignancy relapse. Classical predictors of relapse after allo-HSCT do not seem to be prognostic.

Impact of the 2014 NIH chronic graft-versus-host disease scoring criteria modifications assessed in a large cohort of severely affected patients.

In 2005, the National Institutes of Health (NIH) chronic graft-versus-host disease (cGVHD) consensus project provided diagnosis and staging criteria, based mostly on clinical experience and expert opinion. These criteria were revised in 2014, aiming to provide enhanced specificity and clarity. However, the impact of 2014 changes to the original NIH cGVHD severity scoring criteria has not been reported. In this study, 284 patients, prospectively enrolled on the National Cancer Institute's cross-sectional cGVHD natural history study, were scored using the 2005 NIH cGVHD criteria and then rescored according to the 2014 modifications. In comparing the two criteria, 2014 cGVHD global severity scoring resulted in a tendency toward being categorized as milder scores (75 vs. 72% of severe score per 2014, p = 0.0009), with a statistically significant shift in NIH liver and lung scores toward milder categories (p < 0.0001). 2005 and 2014 NIH global severity scores showed a significant association with reduced grip strength (p < 0.0001), reduced joint range of motion (p = 0.0003), and the subspecialist evaluation score (p < 0.0001). Poor survival prediction of the severe NIH lung score is also retained in the new criteria (p = 0.0012). These findings support the use of 2014 cGVHD scoring criteria in continuous efforts to develop better classification systems.