Roxanne Nelson

December 14, 2012

ATLANTA, Georgia — Children with blood cancers who receive a double umbilical cord blood (UCB) transplant do not gain a survival advantage, a new study has found.

The single-unit approach, therefore, remains the gold standard, refuting the notion that "more is better" in this particular clinical setting.

In patients who underwent transplantation, 1-year overall survival was similar in those who received a single UCB unit and those who received a double UCB unit (71% vs 66%; P = .12).

Dr. John Wagner

"There was no difference in the overall survival rate, even though the hypothesis was that we would see a better result with a double UCB," said lead author John E. Wagner, MD, director of the pediatric blood and marrow transplant program at the University of Minnesota in Minneapolis. He presented the data here at the American Society of Hematology 54th Annual Meeting.

"Single UCB should be considered the standard graft for children; double UCB graft is appropriate for those without a suitable single UCB unit," said Dr. Wagner.

Pilot data from adults with acute leukemia suggested that the infusion of 2 partially human leukocyte antigen (HLA)-matched UCB units is not only safe, but that it is associated with survival rates comparable to those seen in children.

"The hypothesis was that an increased cell dose in recipients of double UCB transplant would be associated with improved survival," said Dr. Wagner.

To assess this, the researchers conducted a randomized trial of the 2 strategies in children (adults are not as likely to have an adequate single unit).

All outcomes were similar in the 2 groups, except that there was a higher incidence of platelet recovery after single UCB transplant and a higher incidence of grade 3 to 4 acute graft-vs-host disease (GVHD) after double UCB transplant, explained Dr. Wagner.

No Difference in Outcomes

The multicenter randomized phase 3 trial enrolled 224 children with hematologic malignancies from December 2006 to February 2012. All patients were required to have 2 UCB units available. The first unit was the best available HLA-matched unit to the patient with at least 2.5 × 107 total nucleated cells/kg of body weight; the second unit was the next best HLA-matched unit with at least 1.5 × 107 total nucleated cells/kg of body weight.

In addition, each unit had to be at least a 4/6 HLA match to the patient and a 3/6 match between units (at intermediate resolution for HLA-A, HLA-B, and allele-level for HLA-DRB1).

The children were randomly assigned to be transplanted with either a conventional single UCB unit (n = 113) or a double UCB unit (n = 111), and randomization stratified by transplant center and patient age.

In the single-unit group, 2.7% crossed over to a double UCB transplant; in the double-unit group, 1.0% crossed over to a single UCB transplant.

Acute lymphoblastic leukemia was the predominant diagnosis in both groups, accounting for about half of all patients. Demographic characteristics, such as disease risk, sex, cytomegalovirus serology, performance score, and HLA match, were balanced between the 2 groups.

All patients received a uniform conditioning regimen (fludarabine 75 mg/m², total body irradiation 1320 cGy, cyclophosphamide 120 mg/kg) and GVHD prophylaxis (cyclosporine and mycophenolate mofetil).

The primary end point was 1-year overall survival in the 2 groups; secondary end points included comparisons of engraftment, acute and chronic GVHD, nonrelapse mortality, relapse, and disease-free survival.

For surviving patients in the 2 groups, median follow-up was 25 months; 92% of them were in remission and 60% were in second or subsequent remission.

At 1 year, disease-free survival was similar in the single- and double-unit groups (68% vs 64%; P = .20), as was the relapse rate (12% vs 14%; P = .37) and the nonrelapse-related mortality rate (20% vs 22%; =.45).

At day 100, the incidence of acute grade 2 to 4 GVHD was identical in the 2 groups (57%; P = .94). At 1 year, the incidence of any chronic GVHD was similar in the single- and double-unit groups (32% vs 30%; = .64).

At day 100, the overall incidence of acute grade 3 to 4 GVHD was higher in those who had undergone a double UCB transplant than in those who had undergone a single UCB transplant (23% vs 14%; P = .03).

Although "rates of chronic graft vs host disease were almost identical, there is a subpopulation with a higher rate of graft vs host disease in the double-transplant recipients," said Dr. Wagner.

Mismatches at 1 HLA locus were similar in the single- and double-unit groups (45% vs 42%), as were mismatches at 2 HLA loci (40% vs 45%). The median precryopreserved total nucleated cell dose was 4.8 × 107/kg in the single-unit group and 8.9 × 107/kg in the double-unit group.

For some years it was believed that double transplantation was associated with better outcomes, said Vanderson Rocha, MD, PhD, professor of hematology at Oxford University, United Kingdom.

"I think it is very important to expand the use of cord blood in adults," said Dr. Rocha, who moderated a press briefing where the findings were presented. "This is a very important question in adults. Unfortunately, we cannot do a randomized study in adults — it is quite complicated. Most important is that we should not use a cell dose below 2.5 × 107 total nucleated cells/kg."

The study was sponsored by the National Heart, Lung, and Blood Institute and the North Central Institute. The authors have disclosed no relevant financial relationships.

American Society of Hematology (ASH) 54th Annual Meeting: Abstract 359. Presented December 10, 2012