| pubmed-article:2194590 | pubmed:abstractText | We prospectively evaluated 169 patients with a number of screening studies performed between 71 to 121 days after allogeneic marrow transplantation to detect the development of chronic graft-versus-host disease (GVHD). Group 1 patients (n = 78) were asymptomatic and had normal physical examinations at the time of screening and, with a minimum of 8 years follow-up, have not developed chronic GVHD. Group 2 patients (n = 38) had signs and symptoms of chronic GVHD at time of testing. Group 3 patients (n = 53) were similar to those in group 1 in having no clinically evident GVHD at the time of testing, but later developed clinical chronic GVHD. Using time to an event analysis, we compared patients in groups 1 and 3 to determine which of 17 clinical and laboratory factors evaluated at screening accurately predicted the development of subsequent chronic GVHD. Multivariate analyses showed several factors to have independent predictive value. In the first model, results of oral biopsies were excluded since these were done only in one half of the patients. Predictive factors in this analysis included: (1) histologic findings of GVHD on skin biopsy, relative risk 3.23 (95% confidence interval 1.75 to 5.94), P = .0002; and (2) history of grade II through IV acute GVHD, relative risk 3.12 (95% confidence interval 1.72 to 5.64), P = .0002. When oral biopsy results were included in the second model, independent risk factors included: (1) histologic findings of GVHD on skin biopsy, relative risk 5.96 (95% confidence interval 1.95 to 18.19), P = .0017; and (2) low numbers of immunoglobulin A (IgA)-bearing plasma cells detected by direct immunofluorescence in salivary gland areas on oral biopsy, relative risk 11.53 (95% confidence interval 2.51 to 52.03), P = .0017. Our study demonstrates the value of day 100 screening studies for predicting subsequent development of clinical chronic GVHD. | lld:pubmed |
