AML At First Relapse: A Real Life Picture

As clinical trials for relapsed Acute Myeloid Leukaemia (AML) patients do not accurately reflect the daily clinical reality, data regarding the outcome of these patients is scarce. We thus conducted a retrospective analysis to quantify the prospects of salvage treatment of primary refractory or first-relapse AML patients and to assess the contribution of allograft and intensive treatment regimens with respect to major risk groups in a real-life setting.

We performed a retrospective analysis of 163 patients diagnosed from 2005-2012, in 5 haematological centres in the north of France (Lille, Amiens, Roubaix, Valenciennes and Lens). We considered every patient in that time frame who was treated following an intensive pathway. Statistical analysis as performed using Kaplan-Meier survival analysis and logrank test in the SPSS software

The mean age at diagnosis was 45 (range 16-70 years) and the median age at relapse was 48 (ranging 17-70 years). The median time from diagnosis to relapse was 8 months. 20.6% of patients were considered primary refractory (relapse within 60 days from diagnosis). The median overall survival was 28 months (95% CI was 17-38 months). There was no statistically significant survival difference between primary refractory patients and first relapsed patients. Unsurprisingly, survival was significantly (p<0.05) higher in the transplanted patients (48 months) than in the non-transplanted group (19 months) and in those who achieved CR (48 months) than in those who didn’t (14 months). Risk stratification was established using the European Leukaemia Network (ELN) classification and the repartition was as followed: favourable (14.3%), intermediate I (18.3%), intermediate II (41.8%) and unfavourable (23.5%). 65% of patients achieved a complete remission (CR). This percentage varied by risk group as follows: favorable (84%), intermediate I (67%), intermediate II (59%) and adverse (39%). Median overall survivals were not statistically significantly different in between the groups but there was a trend suggesting than intermediate I performed poorly (median OS: 19months versus 48 months for favourable, 35 for intermediate II and 18 months for unfavourable patients). When considering treatment regimens, there were no significant difference between the regimens used (Mitoxantrone-Cytarabine-Quinine (MAQ) (n=26), Amsacrine-Cytarabine (n=62), Fludarabine-Cytarabine-Mitoxantrone (FLAN)(n=24), Idarubicine or Daunorubicine Cytarabine (n=21), Clofarabine-Cytarabine (n=4), Mylotarg based (n=9) and other (HDAC, Cytarabine as a single agent)(n=16). The median OS for these groups were respectively 16, 44, 24, not reached, 20, 21 and 11 months. Further analysis is hindered by the size of the groups.

This data suggests that for patients with favourable disease delaying transplant to first relapse and treating them with intensive salvage regimen is a valuable option. High risk patients still perform poorly. The relative low representation of these patients (23.6%) is probably due to the fact that these patients are transplanted upfront and often not treated using intensive regimens at relapse. Discussion remains for the intermediate patients. In our study, although numbers are small, they do not seem to behave similarly. The outcome of intermediate I patients resembles more unfavourable patients: this should be considered when discussing both upfront transplantation and management at first relapse.

This data suggests that delaying transplant for low risk patients is feasible and associated with a good outcome. Salvage intensive chemotherapy and transplant is an effective approach for these patients. Intermediate I patients behave like unfavourable patients and should thus be considered for upfront transplantation and experimental treatments.

Preudhomme:CELGENE: Research Funding. Quesnel:CELGENE: Research Funding. Berthon:CELGENE: Research Funding.