Converting Mixed Chimerism to Full Donor Chimerism in Recipients of Campath-Containing Reduced Intensity Transplants Reduces the Relapse Risk and Results in Significantly Improved Survival Compared to Those with Persistent Full Donor Chimerism.

Donor leucocyte infusions (DLI) are frequently required following reduced intensity conditioned (RIC) allografts to convert mixed donor chimerism (MC) to full donor chimerism (FDC). The rationale is to prevent tolerance developing and therefore to maximise the graft versus tumour (GvT) responses. However, the impact that the chimeric state has on disease relapse and transplant outcome remains controversial.

To address this we analysed the impact of the complete (global) chimerism pattern in 125 recipients of RIC transplants. The transplants were performed for a broad range of malignant haematological diseases. Conditioning regimens consisted of fludarabine, melphalan, campath (65), fludarabine, busulphan, campath (13), BEAM, campath +/− fludarabine (38) and other (9). The donors were HLA matched siblings (62, 50%), HLA mismatched siblings (6, 5%), matched unrelated (29, 23%) and mismatched unrelated (28, 22%). The median patient age was 52 and donor age was 42. The median follow up was 823 days (range 99–2674).

Four patterns of chimerism were seen: A. always 100% donor chimerism (68, 54%), B. persisting MC post transplant including cases refractory to DLI (27, 22%) C. MC post transplant with subsequently development of FDC either spontaneously or post DLI (22, 18%), D. lost DC and had autologous reconstitution (8, 6%). A number of patients in both groups B and C had FDC early post transplant.

In group A 18 (26%) patients received DLI for relapse or residual disease. A complete response was achieved in 6 (35%). In group B 10 (37%) patients had DLI: 4 for MC, 1 for residual disease, 2 for relapse and 3 for both MC and relapse; only partial responses were seen (20%). In group C DLI was given in 14 (64%) patients: 4 for MC, 5 for residual disease, 3 for relapse and 2 for both MC and relapse. A complete response was achieved in 12 (86%).

The risk of relapse at 2 years was significantly associated with the pattern of chimerism (p=0.012) and was greatest for group D (75%). Patients in group C (DLI responders) had a relapse rate of 24% compared to 61% in group B (DLI non-responders). In group A it was 37%. This resulted in a significant survival advantage for patients in group C as compared to all other groups (p=0.009). Predicted overall survival at 2 years was 95% in group C, but 54% in group A, 57% in group B, and 58% in group D. We conclude that patients with MC who later achieve FDC have a lower incidence of disease relapse than those with persistent MC, supporting the use of DLI in this group. However the observation that the patients receiving DLI to achieve FDC have a superior outcome to those patients with persistent FDC (group A) suggests that this group may have benefited from a GvT effect against minimal residual disease. The use of pre-emptive DLI in this group (despite FDC) may reduce the risk of relapse and improve transplant outcome.