Comparative outcomes of T-cell-depleted and non-T-cell-depleted allogeneicbone marrow transplantation for chronic myelogenous leukemia: Impact of donor lymphocyte infusion

Purpose: Donor lymphocyte infusion (DLI) can restore complete remission in patients with chronic myelogenous leukemia (CML) who have relapsed after T- cell-depleted (TCD) allogeneic bone marrow transplantation (BMT). The exist ence of salvage treatment for patients with DLI after TCD allogeneic BMT pr ompted an evaluation of overall outcome after CD6(+)-TCD allogeneic BMT for patients treated during the time when DLI has been available. Patients and Methods: We performed a retrospective analysis of outcomes of 46 patients who underwent TCD allogeneic BMT for stable-phase CML and compa red these outcomes with those of 40 patients who underwent non-TCD allogene ic BMT. All subjects were patients at one of two neighboring institutions d uring a period when DLI was available. All patients received marrow from HL A-identical sibling donors, underwent similar myeloablative regimens, and h ad similar pretreatment characteristics. Results: After BMT, the TCD group had a lower incidence of grade 2 to 4 acu te (15% v 37%, P =.026) and chronic graft-versus-host disease (GVHD) (18% v 42%, P =.024) than did the non-TCD group. The I-year treatment related mor tality rates for the TCD group and the non-TCD group were 13% and 29%, resp ectively (P =.07). The estimated 3-year probability of relapse (cytogenetic or hematologic) was higher for patients in the TCD group than for patients in the non-TCD group (62% v 24%, P =.0003). Twenty-three patients (20 in t he TCD group and three in the non-TCD group) received and were assessable f or response to DLI. After DLI, 17 of 20 patients in the TCD group and two o f three patients in the non-TCD group achieved complete remission. Donor ly mphocyte infusion induced GVHD in nine of 23 patients. Thirty (65%) of 46 p atients in the TCD group and 27 (69%) of 39 assessable patients in the non- TCD group remained alive without evidence of disease. The estimated 3-year overall survival rates were similar for the TCD group and the non-TCD group (72% v 68%, respectively; P =.38). At last follow-up, there was no differe nce in the overall prevalence of GVHD or the proportion of patients requiri ng immunosuppressive agents between groups. Conclusion: These results suggest that the combination of T-cell depletion and past-BMT DLI is a viable treatment option for patients undergoing allog eneic BMT for CML and should be prospectively compared with traditional for ms of GVHD prophylaxis. (C) 1999 by American Society of Clinical Oncology.