The Pattern of T-Cell Donor Chimerism Is a Key Predictor of Outcome Following Alemtuzumab Based Reduced Intensity Conditioning Hematopoietic Stem Cell Transplantation for MDS and AML.

In animal models, a mixed lymphohematopoietic chimerism can be achieved after both myeloablative and non-myeloablative conditioning, establishing bidirectional tolerance for recipient and host antigens, with reduced graft rejection and GvHD. However to date, there has only been limited success in inducing stable MDC in human studies. Furthermore, conflicting reports exist regarding the association between mixed donor chimerism(MDC,5–95% donor chimerism), graft failure and relapse. We report a prospective study on the kinetics of T-cell engraftment in 110 patients with AML(n=50) and MDS(n=60) who received a RIC allogeneic HSCT regimen of fludarabine, busulphan and alemtuzumab with cyclosporine for post-HSCT immunosuppression. Median age was 53 years(range:19–72), and median follow-up:690 days(range:168–1470). Serial analysis of myeloid(CD15) and lymphoid(CD3) chimerism was performed and declining donor CD3 chimerism beyond day 100 was treated with pre-emptive DLI(pDLI).

The achievement of full donor chimerism(FDC,>95% donor chimerism) in either lineage at day 30,60,90 was correlated with overall outcomes, and the attainment of CD3 FDC at day 90 was most predictive of outcomes. Based on this we identified 3 distinct chimerism patterns: 1)Patients achieving CD3 FDC by day 90(n=46). 2)Patients with stable CD3 MDC(n=22) and not requiring pDLI; median CD3 donor chimerism at day 90 was 79%. 17/22 had persisting stable MDC at a median of 675 days post-HSCT. 3)Patients with declining CD3 donor chimerism(n=42). 28/42 patients received pDLI. 14/42 had regressive CD3 donor chimerism(RDC) or loss of donor chimerism by Day 90, and did not receive pDLI due to rapid decline in chimerism and relapse.

All groups were matched for age, cell dose/source, prior chemotherapy, disease type/stage; patients with stable MDC and declining MDC had a higher proportion of sibling donors(p=0.03). Patients with stable MDC had a lower incidence of GvHD compared with patients achieving FDC[grade II-IV GvHD:4(18%) vs 22(48%),extensive cGvHD:2(9%) vs 15(33%)]. 10/28 patients(36%) developed GvHD following pDLI. None of the 14 patients with RDC developed GvHD.

A day 90 landmark analysis was performed to assess the impact of CD3 chimerism on 2-year outcomes. Patients achieving FDC by day 90 had an inferior DFS and OS compared with stable MDC(43% vs 76%,p=0.02 and 44% vs 87%,p=0.01) or pDLI(43% vs 77%, p=0.01 and 44% vs 80,p=0.01). Patients with FDC had a higher TRM compared to stable MDC(37% vs 0%,p=0.02), but there was no difference in TRM between the FDC and pDLI groups(37% vs 16%,p=0.07). There was no difference in relapse between groups(p=0.21). On multivariate analysis, CD3 chimerism pattern was the only independent predictor of TRM(p=0.03). Disease stage at time of transplantation (p<0.01) was the only independent variable for relapse. Both CD3 chimerism pattern(p<0.01 and p<0.001) and disease status at transplantation(p=0.01 and p<0.01) had a significant effect on DFS and OS. Following alemtuzumab-based RIC, the pattern of T-cell engraftment allows identification of patient groups with distinct outcomes. Patients with high levels of stable CD3 MDC can achieve durable DFS with low incidence of GvHD, while attainment of FDC within 100 days was associated with poorer OS due to increased TRM. Further studies need to be directed towards establishing underlying factors which dictate T-cell engraftment, expansion and homing post-HSCT.