乌鸦传媒

ORLANDO -- Stem cells derived from unrelated umbilical cord blood (UCB) or haploidentical bone marrow led to similar progression-free survival (PFS) in patients with acute leukemia or lymphoma, according to a study reported here.

Patients had a 2-year PFS of 35% with UCB transplantation and 41% with haploidentical bone marrow, a difference that did not reach statistical significance. A multivariate analysis adjusted for stem-cell source and transplant center also showed no significant difference in 2-year PFS. Investigators had hypothesized a 15% improvement in PFS with haploidentical bone marrow.

Treatment-related mortality (TRM) and overall survival at 2 years favored the patients who received haploidentical bone marrow, Claudio G. Brunstein, MD, PhD, of the University of Minnesota in Minneapolis, said at the .

"No significant difference was observed in the 2-year progression-free survival in the double umbilical cord blood and haploidentical bone marrow arms, suggesting both donor types extend access to transplantation," he said in conclusion. "Although the trial did not record the expected 15% difference in 2-year progression-free survival between treatment arms in adults with hematologic malignancy, lower non-relapse mortality and higher overall survival favor haploidentical bone marrow transplantation."

The trial might have ended in a draw, but Juliet Barker, MD, of Memorial Sloan Kettering Cancer Center in New York City, saw the results as a no-win for patients.

"As a field ... I think that any transplant center that would accept a (progression-free) survival of, say, less than 70% in this type of patient population, then you really have a problem" she said during the discussion that followed the presentation. "Regardless of whether you're saying -- just for the sake of argument -- that haplo is the answer, my interpretation is that you shouldn't be doing either of these things, because the progression-free survival is just not good enough. We should be aiming for double what this study shows."

The results were in line with previously reported data from similar trials, said Brunstein, acknowledging that "of course, there is room for improvement. As for the design and for this patient population, these are the results."

The shortage of perfectly matched donors for stem-cell transplantation fueled research into a variety of alternatives that had potential to expand the donor pool. A decade ago, investigators in the Blood and Marrow Transplant Clinical Trials Network (BMT CTN) conducted that established double UCB and partially matched (haploidentical) bone marrow as suitable alternatives for allogeneic transplantation for patients with leukemia or lymphoma and no suitable matched donors.

Brunstein reported findings from a follow-up phase III trial that compared the two nonmatched strategies head to head in adult patients with acute leukemia or lymphoma. BMT CTN investigators designed a trial with statistical power to detect a 15% improvement in 2-year PFS with haploidentical bone marrow versus a null hypothesis of no difference between the two strategies.

Patients with suitable donors for double UCB or haploidentical bone marrow transplantation received nonmyeloablative conditioning therapy and were randomized to one of the two strategies. The primary endpoint was 2-year PFS. Key secondary endpoints included TRM, relapse/progression rate, and OS. Data analysis comprised 368 patients, 343 of whom underwent transplantation, including 326 who received the assigned treatment.

In the subgroup of patients with acute leukemia, 75-85% were in first remission at the time of transplantation, 40-45% had intermediate cytogenetic risk, and a third had poor-risk cytogenetics. Among the patients with lymphoma, about 35% had complete responses and 50-55% had partial responses.

Analysis of the primary endpoint showed an absolute improvement in 2-year PFS of 6.1% with haploidentical bone marrow (P=0.41). A multivariate analysis adjusted for donor type yielded a hazard ratio (HR) of 1.30 for double UCB versus haploidentical bone marrow (P=0.060). An analysis adjusted for donor and transplant center produced an HR of 1.27 for the cord-blood group, which also did not reach statistical significance (P=0.162).

Using acute leukemia in first remission as a reference, investigators found only one other significant predictor of PFS: lymphoma in partial remission at the time of transplant (P=0.032).

The 2-year incidence of relapse did not differ between the two groups (47-48%, P=0.97). TRM at 2 years favored patients who received haploidentical bone marrow (11% vs 18%, P=0.04), as did 2-year OS (57% vs 46%, P=0.04). Neutrophil recovery (ANC >500) also favored the half-matched marrow group (99% vs 95%, P=0.05), whereas platelet recovery to >20,000 (84% vs 78%, P=0.15), acute graft versus host disease (GVHD, 28% vs 35%, P=0.14), and chronic GVHD (26% vs 22%, P=0.36) did not differ between treatment groups.

Comment