TCD alloSCT is associated with an increased relapse rate compared to non-TCD alloSCT. We and others found an association between the presence of mixed chimerism (MC) after transplantation (Tx) and an increased relapse risk. Since DLI for hematological relapse of acute leukemia cures only 20–30% of patients (pts), pre-emptive DLI might restore complete chimerism (CC) and decrease the relapse rate. Previously, we found that 50% of all relapses occurred within 6–7 months following Tx. We have demonstrated in pts with relapsed CML after TCD alloSCT that low dose DLI given 6–9 months after Tx was effective without inducing severe GVHD. Therefore, we evaluated whether low dose DLI administered 6–9 months after Tx for acute leukemia in pts with MC not suffering from GVHD could decrease the relapse rate without induction of severe GVHD. After conditioning with TBI and cyclophosfamide, 31 pts (AML 21, ALL 8, high risk MDS 2) were transplanted in CR with stem cells from an HLA-identical sibling donor (21), an HLA-matched unrelated donor (8), or a 1 or 2 antigen mismatched family donor (2). The graft was T cell depleted with 20 mg Campath ‘in the bag’. No additional GVHD prophylaxis was given. Bone marrow morphology, immunophenotyping, and chimerism analysis was performed at 3 and 6 months after Tx and at 1.5, 3, 6, 9, and 12 months after DLI. Six months after Tx 7 of 31 pts had relapsed, and 5 pts had died due to infections. In 1 of 19 remaining pts no chimerism analysis was performed, 16 pts were MC with a median of 2% (range 1–35) pt cells, and 2 pts were CC but later reverted to MC. Five of 16 MC pts suffered from GVHD prohibiting DLI and 1/16 MC pts did not receive DLI due to logistical problems, resulting in 12 eligible MC pts. A first DLI at a dose 3×10E6 CD3+ T cells/kg body weight (BW) was given at a median of 8.3 (range 6–22) months after Tx. The median follow-up after the first DLI was 10.5 (range 3–36) months. In 9 pts the percentage patient cells decreased after DLI (p<0.05, Wilcoxon rank test); in 6/9 pts donor chimerism increased to >99% showing the effectiveness of DLI to influence chimerism status. In 1 pt MC increased from 1 to 4% patient cells and in 2 pts MC remained stable. GVHD grade I developed in 1 pt and grade II in 3 pts after the first DLI. When no CC was obtained and no GVHD was present additional DLI was administered in escalating doses of 1×10E7 (1x), 3×10E7 (3x), and 1×10E8 T cells/kg BW (1x). No GVHD developed after subsequent escalated doses of DLI. Two MC pts who did not receive DLI due to GVHD at 6 months after Tx, relapsed within 9 months after Tx as did the patient in whom no chimerism was determined at 6 months. None of the 12 MC pts treated with DLI relapsed. In conclusion, our results indicate that low dose DLI projected to be administered 6–9 months after Tx increases the degree of donor chimerism in the majority of pts and may reduce the likelihood of relapse without severe GVHD. Since 7 of the 10 relapses occurred within the first 6 months after Tx, we aim to administer low dose DLI at 3–4 months after Tx to study its effect on chimerism and relapse risk in our next group of pts. Since the short interval between DLI and Tx will increase the likelihood of developing GVHD using unmodified DLI, also CD4 purified T cells will be administered.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author