Approximately 30-50% of people undergoing allogeneic hematopoietic cell transplantation (HCT) will develop chronic graft-versus-host disease (GVHD), a syndrome characterized by multi-system inflammation and fibrosis that clinically resembles autoimmune diseases. Chronic GVHD is the leading cause of non-relapse mortality in two-year disease free survivors and a major cause of morbidity. In 2005, the NIH convened a consensus conference designed to improve clinical research methods in chronic GVHD. Recommendations from the conference have been published, and are called the ""NIH consensus criteria."" During our previous funding period, we assembled a multi-center, longitudinal, observational study of 554 patients who were assessed every six months to test the NIH consensus criteria. We have published 6 papers (12 more submitted or in preparation) evaluating the consensus recommendations. Our results support the recommended definitions for chronic GVHD diagnosis and severity scoring. Specifically, we confirmed the poor prognosis associated with concurrent acute and chronic GVHD (overlap syndrome) and showed a higher non-relapse mortality and lower overall survival associated with moderate-severe chronic GVHD. However, our analyses did not support the recommended NIH measures to assess therapeutic response. In particular, calculated overall NIH response at 6 months correlated poorly with patient and provider-reported responses, and also did not predict subsequent non-relapse mortality and overall survival. Thus, if a new, promising treatment for chronic GVHD were available, the field does not have a validated measure that can serve as a primary endpoint in a clinical trial. To address this critical gap in the field, we propose developing two new tools, the chronic GVHD activity index (CGVHD-AI) and the chronic GVHD severity score (CGVHD-SS). The CGVHD-AI is intended for use as an overall measure of disease activity, so that change in the score from enrollment to follow-up reflects treatment effects. The CGVHD- SS is intended to serve as an intermediate endpoint that predicts eventual treatment success, even if that takes months to years to reach. Eleven centers will enroll 368 patients with chronic GVHD who are just starting initial or secondary therapy. Comprehensive assessments at enrollment and 6 months will collect the candidate patient-reported, provider-reported, and laboratory tests that will be considered for inclusion in the new measures. At 18 months after enrollment, patients will again be assessed with all measures. Using this information, we will develop and validate the CGVHD-AI and CGVHD-SS, thus allowing robust clinical trials in chronic GVHD.