番茄社区

ORLANDO -- Myeloablative conditioning (MAC) led to a long-term survival benefit versus reduced-intensity conditioning (RIC) for patients with acute myeloid leukemia (AML) treated with hematopoietic cell transplantation, updated results from a randomized trial showed.

After more than 4 years' follow-up, patients randomized to RIC unexpectedly had a 50% increase in the survival hazard versus the MAC group. Relapse was four times more likely with RIC, and the patients were more than twice as likely to relapse or die as compared with the MAC group.

Mortality after AML relapse was similar between the two groups, as reported here at the .

"Overall survival remained significantly superior in patients with AML who received myeloablative conditioning," said Bart Scott, MD, of the Fred Hutchinson Cancer Research Center in Seattle. "The proportional risk of transplant-related mortality [without relapse] continued to increase over time with myeloablative conditioning. Most relapses occurred during the first year in both groups."

"For patients with AML, myeloablative conditioning remains the superior choice."

Confirming earlier findings

The findings confirmed and extended those of the from the Bone and Marrow Transplant Clinical Trials Network (BMT CTN) , which upended the investigators' expectation of survival benefit with RIC.

The optimal intensity for conditioning regimens prior to allogeneic transplant has remained a moving target. Investigators thought that a reduced-intensity regimen might lead to better overall survival (OS) by reducing treatment-related mortality. BMT CTN 0901 tested the hypothesis in 270 transplant-eligible adults ≤65 with AML (N=218) or myelodysplastic syndrome (MDS, N=54).

Almost all patients in the MAC arm received fludarabine and busulfan (64%) or busulfan and cyclophosphamide (30%). Patients randomized to RIC received reduced-intensity fludarabine/busulfan (80%) or fludarabine/melphalan. Tacrolimus/methotrexate was the most common prophylaxis for graft versus host disease (GVHD) in both groups (82%).

The trial had a primary endpoint of 18-month OS, which investigators expected to be higher with RIC. Instead, the primary analysis showed almost a 10-point absolute difference in favor of MAC (77.4% vs 67.7%, P=0.07). The difference grew to almost 14% among patients with AML (76.8% vs 63.0%, P=0.027), whereas OS was similar between RIC (85.2%) and Mac (81.5%) in the MDS subgroup.

MAC led to a 21.5% difference in relapse-free survival (RFS) at 18 months (68.8% vs 47.3%, P<0.01). MAC was associated with a lower risk of relapse and disease progression in patients with MDS or AML. Grade 3/4 acute graft versus host disease (GVHD) occurred twice as often with MAC (13.6% vs 6.8%, P=0.066), and chronic GVHD occurred significantly more often with MAC (64% vs 47.6%, P=0.019).

Unanswered questions

The trial's primary results raised several questions to be addressed during continued follow-up in the trial, said Scott. Would the survival difference change over time? Would the higher rates of GVHD with MAC lead to higher treatment-related mortality? Would conditioning intensity affect outcomes after posttransplant relapse? Would the higher relapse rate with RIC for patients with MDS affect long-term survival?

Longer follow-up did answer the key question: Patients lived longer if they received MAC. The 4-year OS was 65% with MAC and 49% with RIC, resulting in a hazard ratio of 1.54 for the comparison of RIC versus MAC (95% CI 1.07-2.20). RFS also favored the MAC arm (58% vs 34%, HR 2.06, P<0.001) and patients randomized to RIC were more than four times as likely to relapse (HR 4.06, P<0.001).

An analysis by disease subtype showed that RFS differed significantly only for patients with AML. A trend toward inferior RFS in patients with MDS was evident for patients treated with RIC and might have achieved statistical significance if not for the small numbers, said Scott.

More patients in the MAC arm died without relapse (P<0.001), but patients whose disease relapsed after transplantation had similar 3-year post-relapse survival of about 25%.

"There was a steady rise in treatment-related mortality with the MAC regimen and remained proportionally higher than with the RIC regimen; however, this higher treatment-related mortality was not able to overcome the much higher relapse rate with the RIC regimen, so that overall there was still a survival benefit seen in the MAC group," said Scott.

Previously published results showed that patients who had by next-generation sequencing (NGS) prior to the start of conditioning did not benefit from more- or less-intense conditioning. Scott pointed out that relatively few patients were NGS negative before conditioning therapy.

In response to a question from the audience, Scott said previous European randomized trials of conditioning regimens had nonrelapse mortality as the primary endpoint and were not statistically powered to show a difference in OS.

Scott, answering another question, said older patients "paradoxically" benefited more from MAC. He emphasized that investigators selected "the best of the best" candidates for the trial, perhaps minimizing the impact of comorbidities. Additionally, older patients tend to have higher-risk disease, which might have influenced the magnitude of the benefit observed with MAC.

With regard to other factors that might have influenced the results, Scott said graft failure occurred infrequently in both treatment groups, and donor-specific characteristics did not have a significant impact.

The trial has had a practice changing-impact on the management of AML and MDS, said John F. DiPersio, MD, PhD, of Washington University in St. Louis.

"For anyone under 65 who is in good health, most everyone tries to think about using a myeloablative regimen first instead of reduced-intensity conditioning, for AML and for MDS," DiPersio told 番茄社区. "For patients over 65, since that study didn't include anyone over 65, it's completely unknown. Obviously the toxicities of myeloablative regimens are going to be higher in older patients, and those patients do poorly, in general."

"The few studies that have been published in older patients suggest that myeloablative and reduced-intensity conditioning regimens are relatively equivalent. To conclude that older patients would benefit from myeloablative conditioning, a study would have to be done in that age group."

Comment