Patient-Reported Outcomes and Mortality in Cutaneous Chronic Graft-vs-Host Disease

Abstract

Chronic graft-vs-host disease (GVHD) is associated with impaired quality of life and symptom burden. The independent association of skin involvement with patient-reported outcomes (PROs) and their utility as a clinical prognostic marker remain unknown. Identification of patients with cutaneous chronic GVHD and impaired PROs could assist in initial risk stratification and treatment selection. To compare the association of sclerotic and epidermal-type chronic GVHD with longitudinal PROs and to evaluate whether PROs can identify patients with cutaneous chronic GVHD at high risk for death. This multicenter prospective cohort study involved patients from the Chronic GVHD Consortium of 9 US medical centers, enrolled between August 2007 and April 2012, and followed up until December 2020. Participants included adults 18 years and older with a diagnosis of chronic GVHD requiring systemic immunosuppression and with skin involvement during the study period. Patient-reported symptom burden was assessed using the Lee Symptom Scale (LSS) skin subscale with higher scores indicating worse outcomes. Quality of life was measured using the Functional Assessment of Cancer Therapy-Bone Marrow Transplantation (FACT-BMT) instrument with lower scores indicating worse outcomes. Nonrelapse mortality, overall survival, and their association with PROs at diagnosis were also assessed. Among 436 patients with cutaneous chronic GVHD (median [IQR] age at transplant, 51 [41.5-56.6] years; 261 [59.9%] male), 229 patients had epidermal-type chronic GVHD (52.5%), followed by 131 with sclerotic chronic GVHD (30.0%), and 76 with combination disease (17.4%). After adjusting for confounders, patients with sclerotic chronic GVHD had mean FACT-BMT scores 6.1 points worse than those with epidermal disease (95% CI, 11.7-0.4; P = .04). Patients with combination disease had mean LSS skin subscale scores 9.0 points worse than those with epidermal disease (95% CI, 4.2-13.8; P < .001). Clinically meaningful differences were defined as at least 7 points lower for FACT-BMT and 11 points higher for LSS skin subscale. At diagnosis, clinically meaningful worsening in FACT-BMT score was associated with an adjusted odds of nonrelapse mortality increased by 9.1% (95% CI, 2.0%-16.7%; P = .01). Similarly, for clinically meaningful worsening in LSS skin subscale score, adjusted odds of nonrelapse mortality increased by 16.4% (95% CI, 5.4%-28.5%; P = .003). These associations held true after adjusting for clinical severity by the National Institutes of Health Skin Score. The results of this cohort study demonstrated that skin chronic GVHD was independently associated with long-term PRO impairment, with sclerotic and combination disease carrying the highest morbidity. The degree of impairment at skin chronic GVHD diagnosis was a prognostic marker for mortality. Therefore, PROs could be useful for risk stratification and treatment selection in clinical practice and clinical trials.