Delayed Donor B-Cell Reconstitution After Allogeneic Stem Cell Transplantation Is Associated with Increased Risk of Relapse but Protects From Chronic Graft Versus Host Disease.

Abstract 2228

Poster Board II-205

B lymphocyte homeostasis may have a role in regulating the immune response, both stimulatory & regulatory/tolerogenic, in Allogeneic Stem Cell Transplantation (AlloSCT). The predictive relationship of donor B-cell reconstitution with relapse risk, chronic Graft versus Host Disease (cGvHD) & overall survival (OS) was investigated. We performed prospective comprehensive immune reconstitution studies in 71 patients undergoing AlloSCT (Matched Related Donor=33, Matched Unrelated Donor =34, Umbilical Cord Blood =3, Haplo-identical Related Donor =1) for haematological malignancy & marrow failure syndromes who were available for follow-up beyond Day+100. 27 patients received full intensity conditioning & 44 reduced intensity conditioning with 34 patients receiving Alemtuzumab & 8 patients receiving ATG as in vivo T-cell depletion with a median PAM score of 22 (range 8-36). The grafts were bone marrow in 23, peripheral blood stem cells in 40 and umbilical cord blood in 3 with a median CD34⁺cell dose of 4.3×10⁶/kg (range 0.8-10.5) & median duration of immune-suppression of 7.1 months (range 2, 49). Samples were analysed for lymphocyte subsets by multi-parameter flow cytometry & cell subset-specific Complete Donor Chimerism (CDC) by single tandem repeat PCR at day (D)+100, D+180, D+270 and D+365. The proportion of patients with a CD4⁺, CD8⁺, CD19⁺ & NK cell count within the normal range at D+100 & D+180 were: 10% & 27% for CD4⁺, 28% & 57% for CD8+, 24% & 57% for B-cells, 68% & 83% for NK cells, respectively. The proportion of patients with a CD4⁺, CD8⁺, CD19⁺ & NK cell CDC at D+100 & D+180 were: 80% & 78% for CD4⁺, 86% & 80% for CD8+, 89% & 91% for B-cells, respectively. 28 patients (38%) experienced cGvHD (22/28 extensive) with a median time to onset of 4 months (range 3-12) & median duration of immune suppression of 14.5 months (range 6.9-48.3). In univariate analysis, only the B-cell CDC at D+100 was associated with a higher incidence of cGvHD (mean 99.8%±0.2% vs. 90.4%±4.4 CDC, respectively; p=0.044) though a trend was seen in regard to D+180 B-cell CDC. No influence of mature/immature B-cell reconstitution at D+100 or D+180 was detected. 24 patients (34%) have relapsed at a median of 5.8 months (range 3-34). A trend to lower absolute B-cell counts at D+100 & D+180 was seen in those who relapsed with significantly lower B-cell CDC at D+100 and D+180 in those who relapsed compared to those who did not (D+100: mean 87.9%±2.4% vs. 98.3%±1.5 CDC, respectively; p=0.038; D+180: mean 80±5.6 vs. 100%±0 CDC, respectively; p=0.013). With a median follow-up of 16.7 months (range 5.1-51.8), 27 (38%) have died (disease progression n=12, infection n=4, GvHD-related n=6). In univariate analysis, no effect on OS was demonstrated by the absolute B-cell count or B-cell CDC at D+100 or D+180, though a trend was seen to lower levels in those who did not survive. In conclusion, delayed donor B-cell reconstitution was associated with a higher relapse risk. On the other hand, cGvHD was associated with higher early donor B-cell reconstitution, which suggests that residual recipient B-cells after AlloSCT may protect from cGvHD. This data provides further clinical evidence for a role of B-cells in the immune-modulation associated with the graft-versus-tumour/graft-versus-host disease paradigm of AlloSCT.