Response and survival of patients with chronic graft-versus-host disease treated by extracorporeal photochemotherapy: a retrospective study according to classical and National Institutes of Health classifications

BACKGROUND: Chronic graft-versus-host disease (cGVHD) classification has recently been improved by the National Institutes of Health (NIH); patients stratification with those new criteria has implications for patients' prognosis. STUDY DESIGN AND METHODS: To assess whether the NIH consensus classification (NCC) better predicts survival and response to extracorporeal photochemotherapy (ECP), and to identify variables associated with response and survival, we retrospectively analyzed 102 patients with cGVHD reclassified according to NCC treated with ECP (1997-2010) at our center. Cox regression was used in univariate and multivariate models. RESULTS: Of the 102 patients, 64 (62.7%) had classic cGVHD, 24 (23.5%) had overlap cGVHD, and seven (6.9%) patients each had late and persistent acute GVHD. The cumulative ECP-specific follow-up was 2333.3 person-years. Response was complete in 16 (15.7%), partial in 38 (37.3%), minimal in 28 (27.5%), and absent in 20 (19.6%). Of the 22 deaths, 15 (68.2%) occurred among patients with minimal or no response (p = 0.031). The only variables associated with response were nonmyeloablative transplant (hazard ratio, 3.5; 95% confidence interval [CI], 1.36-9.08; p = 0.009), donor lymphocyte infusion (hazard ratio, 2.58; 95% CI, 1.2-5.56; p = 0.015), and lung involvement (hazard ratio, 0.34; 95% CI, 0.12-0.94; p = 0.038). CONCLUSION: ECP is a safe and effective treatment for cGVHD and response to ECP is the only variable that influences survival. We found no correlation between response and NCC clinical subtype, number, or degree of organ involvement, except for lung, or the variables mentioned above. Prospective studies are needed to identify subsets of patients with higher probability of response.