Mismatched Family Member Donor Transplantation for Patients with Refractory Hematologic Malignancies: Long-Term Followup of a Prospective Clinical Trial.

Patients with refractory hematologic malignancies and those who relapse after allogeneic hematopoietic stem cell transplantation (HSCT) have a dismal prognosis. Both disease recurrence and regimen-related toxicities contribute to the high mortality in this population. We designed a reduced intensity conditioning regimen using haploidentical donors in an effort to improve overall survival rates, by both decreasing disease recurrence and transplant-related mortality. From 2003 to 2005, we enrolled 25 patients with refractory hematologic malignancies. The median age was 11.9 years (range 2.8 to 26.5); 15 were male. Diagnoses included ALL (n= 9), AML (n= 10), MDS (n=1), NHL (n=3), CML (n=1), and JMML (n= 1). The majority (n=24) had persistent or refractory disease at the time of HSCT; only one was in remission (n= 1). Nine patients had no prior HSCT, 15 had 1 prior HSCT, and 1 had 3 prior HSCT. Donors were fathers (n=16), mothers (n=6), or siblings (n=3). Donor-recipient pairs matched at 3 (n=18), 4 (n=6), or 5 (n=1) of 6 HLA loci. All grafts were cytokine-mobilized peripheral blood stem cells, using GM-CSF and G-CSF. Grafts were depleted of T-lymphocytes on the CliniMACS device using the OKT3 antibody. Grafts contained a median of 14.89 X10⁶ CD34⁺/kg (range, 2.23 to 51.20) and 1.4 X10⁵ CD3⁺/kg (range, 0.1 to 4.5). The conditioning regimen consisted of fludarabine 40 mg/m²/day for 5 days, melphalan 60 mg/m²/day for 2 days, and thiotepa 10 mg/kg/day for 1 day. OKT3 was administered in an escalating and de-escalating fashion from day −9 to day +17. Rituximab was administered on day ) as EBV prophylaxis. MMF was initiated on day −2 for GVHD prophylaxis. The median time to engraftment of neutrophils was day +10 (range, 7–12); the median time to recovery of platelets to 20,000/mm³ and 50,000/mm³ was day +17 (range, 12–36) and day +17 (range, 12–76), respectively. 3 patients did not engraft:

• recovered with JMML, and

• were salvaged with grafts from their original HSCT donor.

The cumulative incidences of grade 2–4 and grade 3–4 acute GVHD were 44.0% and 8.0%, respectively; the cumulative incidence of chronic GVHD was 28.0%. No patient developed VOD. 13 patients died of disease recurrence with a median time to relapse of 105 days (range, 26–714). The cumulative incidence of relapse at 2 years post-HSCT was 50%. With only 4 patients dying of non-relapse causes, the cumulative incidence of non-relapse mortality plateaued at 20% at 11 months post-HSCT with 1 patient dying of each of the following: hemorrhage, infection, cardiomyopathy, and chronic GVHD. With a mean follow-up time of 387 days post-HSCT, the overall survival at one year and 2 years post-HSCT was 40% and 35%, respectively. The disease-free survival at one year and 2 years post-HSCT was 36% and 30%, respectively. Eight patients remain alive as outpatients with performance scores ≥90. Haploidentical HSCT with a reduced intensity conditioning regimen is safe and feasible in this high-risk patient population, resulting in prompt engraftment, acceptable GVHD rates, and promising survival rates. Measures to further prevent disease recurrence must be incorporated into future clinical trials to improve outcomes while maintaining low regimen-related toxicity rates.