346: Homeostatic and Inflammatory Processes Contribute to Elevated Plasma Levels of B Cell Activating Factor (BAFF) in Chronic Graft Versus Host Disease (CGVHD)

BAFF, a non-redundant cytokine produced by myeloid cells, plays a critical role in the normal homeostatic maintenance, activation and function of B cells. Elevated circulating levels of BAFF, however, have been observed in systemic autoimmune disorders and, in murine models, have been linked to a failure to delete auto-reactive B cells. We similarly observed elevated plasma BAFF levels in 77 patients in an ongoing NCI CGVHD natural history protocol, with a median of 2845 pg/ml (range 92 to 17058), as compared to 556 pg/ml (range 75 to 1834) in 18 normal donors. Furthermore, in a subset of 22 patients in which severity of CGVHD could be assessed by the presence of marked erythema or sclerosis on 10 to 90% of their body surface areas (BSA), the BAFF level correlated with the percentage of affected BSA (Spearman r = +.63). We then explored the factors that might contribute to elevated BAFF levels. In recipients recovering from either autologous or allogeneic transplant (without GVHD) we observed the highest BAFF levels at day 0 (median of 10534 and 12240 pg/ml respectively), when B cells were severely depleted. As B cell populations recovered to normal levels post transplant, plasma BAFF concentrations declined (Spearman r = -.80 and r = -.60, respectively), consistent with homeostatic cytokine-consumption dynamics. Despite comparably high levels of BAFF (median of 11342 pg/ml) at transplant day 0 in 16 patients who later developed CGHVD, BAFF levels in the cross-sectional, natural history patient population were only moderately correlated with the degree of post transplant B cell recovery (r = -.48). Since inflammatory triggers can induce elevated BAFF production by dendritic cells, we assessed plasma levels of cytokines indicative of an inflammatory process. In 34 patients, the plasma levels of IP-10 and sTNFRII correlated positively with BAFF levels (r = +.627 and r = +.642, respectively), consistent with active inflammatory processes in those CGVHD patients with elevated BAFF levels. In a multi-step regression model, the levels of circulating B cells, plasma IP-10 and sTNFRII combined to strongly predict BAFF levels (R = .834). These findings suggest that both homeostatic recovery of B cell populations consuming BAFF and inflammatory cytokine cascades initiated by donor-anti-host reactivity combine to regulate BAFF levels post transplant. BAFF, a non-redundant cytokine produced by myeloid cells, plays a critical role in the normal homeostatic maintenance, activation and function of B cells. Elevated circulating levels of BAFF, however, have been observed in systemic autoimmune disorders and, in murine models, have been linked to a failure to delete auto-reactive B cells. We similarly observed elevated plasma BAFF levels in 77 patients in an ongoing NCI CGVHD natural history protocol, with a median of 2845 pg/ml (range 92 to 17058), as compared to 556 pg/ml (range 75 to 1834) in 18 normal donors. Furthermore, in a subset of 22 patients in which severity of CGVHD could be assessed by the presence of marked erythema or sclerosis on 10 to 90% of their body surface areas (BSA), the BAFF level correlated with the percentage of affected BSA (Spearman r = +.63). We then explored the factors that might contribute to elevated BAFF levels. In recipients recovering from either autologous or allogeneic transplant (without GVHD) we observed the highest BAFF levels at day 0 (median of 10534 and 12240 pg/ml respectively), when B cells were severely depleted. As B cell populations recovered to normal levels post transplant, plasma BAFF concentrations declined (Spearman r = -.80 and r = -.60, respectively), consistent with homeostatic cytokine-consumption dynamics. Despite comparably high levels of BAFF (median of 11342 pg/ml) at transplant day 0 in 16 patients who later developed CGHVD, BAFF levels in the cross-sectional, natural history patient population were only moderately correlated with the degree of post transplant B cell recovery (r = -.48). Since inflammatory triggers can induce elevated BAFF production by dendritic cells, we assessed plasma levels of cytokines indicative of an inflammatory process. In 34 patients, the plasma levels of IP-10 and sTNFRII correlated positively with BAFF levels (r = +.627 and r = +.642, respectively), consistent with active inflammatory processes in those CGVHD patients with elevated BAFF levels. In a multi-step regression model, the levels of circulating B cells, plasma IP-10 and sTNFRII combined to strongly predict BAFF levels (R = .834). These findings suggest that both homeostatic recovery of B cell populations consuming BAFF and inflammatory cytokine cascades initiated by donor-anti-host reactivity combine to regulate BAFF levels post transplant.