Impact of macrophage infiltration of skin lesions on survival after allogeneic stem cell transplantation: a clue to refractory graft-versus-host disease.

We retrospectively reviewed 104 biopsy specimens of previously untreated skin acute graft-versus-host disease (GVHD) within 100 days after allogeneic stem cell transplantation, and analyzed the relationship between types of infiltrating cells and clinical outcomes. Counting the total number of CD8(+) T cells, CD163(+) macrophages, and CD1a(+) dendritic cells in 4 fields under original magnification x200, the infiltration of more than 200 cells of CD163(+) macrophages (many macrophages [MM]) was the only significant predictor for refractory GHVD (odds ratio, 3.79; 95% confidence interval, 1.22-11.8; P = .02). In 46 patients given steroid treatments, MM was the only significant predictor for refractory acute GVHD (odds ratio, 5.05; 95% confidence interval, 1.19-21.3; P = .03). Overall survival of patients with MM was significantly lower than that of those with an infiltration of less than 200 cells of CD163(+) macrophages. Macrophage infiltration of skin lesions could be a significant predictive factor for refractory GVHD and a poor prognosis.

Impact of the basal metabolic ratio in predicting early deaths after allogeneic stem cell transplantation.

Early deaths after allogeneic stem cell transplantation (allo-SCT) are of major concern. On the assumption that both decreased and increased basal metabolism might relate to early deaths, we analyzed the risk factors for overall survival to days 30 (OS30) and 60 (OS60). The Harris-Benedict equation was used to calculate basal metabolism. Comparing a patient's basal metabolism (PBM) calculated from pretransplant body weight with the standard basal metabolism (SBM) calculated from standard body weight (body mass index (BMI) = 22), we defined the basal metabolic ratio (BMR) as a parameter (BMR = PBM/SBM). We retrospectively analyzed 360 adult patients transplanted between 1997 and 2006 at a single center in Japan. A multivariate analysis of OS30 showed risk factors to be: BMR < or = 0.95 (low BMR; LBR) (P = 0.01), BMR > 1.05 (high BMR; HBR) (P = 0.005) and non-complete remission (non-CR) (P 5 0.001), whereas a multivariate analysis of OS60 showed those risk factors to be: LBR (P = 0.02), HBR (P = 0.04), non-CR (P = 0.002), and performance status < or = 1 (P = 0.01). OS30 and OS60 were found to be favorable in 0.95 < BMR < or = 1.05 (average BMR; ABR) (96.8 and 90.3% for ABR, 87.1 and 76.2% for LBR, and 87.8 and 81.1% for HBR). In conclusion, BMR could prove to be a predictor of early death after allo-SCT.