The Impact of Transplant in AML in 2nd CR: A Prospective Study of 741 in the MRC AML 10 and 12 Trials.

MRC AML10 and 12 Trials recruited 5425 patients of whom 4557(84%) entered CR and were treated with chemotherapy only. 741 relapsed and achieved a second CR. These patients were recommended for salvage by stem cell transplantation either sibling or MUD allograft or autograft. In a non-randomised comparison, Mantel-Byar and delayed entry proportional hazards regression were used to investigate outcome and which factors were predictive for survival. Of the 741 patients in CR2 480 received a SCT (116 sib Allo: 192 Autograft and 154 MUD 18 other/unknown) at a median of 107 days from second CR. The overall survival at 5 years was significantly better for those who received SCT (39% vs 22% p=0.00001). Of the 116 patients who received a sibling transplant 54% are alive at 5 years; of the 154 patients who received a MUD graft 40% are alive and of 192 who received an autograft 33% are alive and each transplant type was significantly better than chemotherapy alone (p=0.0001, p=0.002, p=0.002). Much of the benefit derives from reduced relapse risk (5yr RR: allograft 32%; autograft 59%; MUD 38%: no transplant 71% (0.0001). Only in MUD transplant was there a greater mortality (allograft 27%; autograft 27%, MUD 42%, no transplant 25% p=0.003). There was, however, no evidence of heterogeneity in the beneficial effects of transplant when analyses were stratified by age, duration of first remission, time taken to achieve second CR, or cytogenetics. Proportional hazards regression shows that duration of 1st CR, cytogenetics and age are important prognostic factors; adjustment for these factors to some extent changes the results of the analyses. After adjustment, there is still strong evidence of the benefits of allograft (p<0.0001), but the evidence for autograft (p=0.09) and MUD (p=0.02) becomes unclear. As a non-randomised comparison, these results cannot provide absolute evidence of benefit. After adjustment for other prognostic factors, the position for autograft and MUD transplants remains unclear especially as not all selection factors can be accounted for. However, the allograft results appear robust to adjustment and imply significant survival benefit for patients receiving allograft in 2nd CR.