Impact of Post-Remission Consolidation Modalities in the Clinical Outcomes of Acute Myeloid Leukemia Patients in First Remission - Is There a Role for Autologous Hematopoietic Stem Cell Transplant?

The optimal post-remission consolidation in Acute Myeloid Leukemia (AML) remains a topic of some debate, especially in the intermediate risk group of patients. We undertook a retrospective analysis of newly diagnosed AML patients treated in Singapore General Hospital from 1999 to 2012. Patients with M3 AML and patients who did not receive treatment of curative intent were excluded.

286 patients were identified for analysis, median age was 48 (range 15 to 79), ELN risk grouping: Favorable (FR) (n=84), Intermediate-1 (IR1) (n=120), Intermediate-2 (IR2) (n=57), and High risk (HR) (n=25). Post-remission therapy was administered as follows, 107 received high dose cytarabine (HiDAC) consolidation, 67 received autologous hematopoietic stem cell transplant (HCT), and 112 received allogeneic HSCT. Conditioning for allogeneic HCT was administered according to widely accepted protocols: bulsuphan and cyclophosphamide (Bu/Cy) for patients 50 years and younger while reduced intensity or non-myeloablative protocol was administered for older patients or the physically unfit. Patients for Autologous HCT were mobilized with VP16 and cytarabine and conditioned with Bu/Cy. HiDAC was administered at 3g/m² q12h on days 1, 3 and 5; the dose was reduced to 1 to 1.5g/m² for those above 60 years old, median number of HiDAC cycles was 4 (range 1 – 4).

After a median follow up of 110 months, 5 year overall survival (OS) between patients who received HiDAC, autologous HCT and allogeneic HCT was 43.2 %, 63% and 48.3% (p =.04), 5 year leukemia free survival (LFS) was 40.3%, 61% and 43.2% and (p = .01). The apparent difference between autologous and allogeneic HSCT (p=.119) disappeared when we excluded HR patients, in which only 1 received autologous HSCT. However, the OS difference between autologous HSCT and HiDAC remained significant (p=.03). When patients were stratified according to ELN risk, OS and LFS were similar in FR patients regardless of consolidation modalities (p = .67 and p = .18) while IR1 and 2 patients receiving HiDAC had poorer outcomes when compared with autologous and allogeneic HSCT (OS: 26.9%, 57.6%, 54.1%; LFS: 23.4%, 56.3%, 45.9% respectively, p < .0001 for OS and LFS), this remained significant when patients who did not complete at least 3 cycles of HiDAC consolidation were excluded (p = .02 for OS and LFS).

In those who underwent autologous HSCT, predictors of better survival were age < 40 years (5yr OS 74.2% vs 56.7% p = .04, 5yr LFS 70% vs 56.7% p = .09), ELN favorable risk (5yr OS 80% vs 55% p = .03, 5 yr LFS 70% vs 55% p = .08) and bone marrow (BM) blast percentage of less than 70% in conjunction with a diagnostic white blood count (WBC) of less than 30 x 10⁹ (5yr OS 75.7% vs 40% p = .007, 5yr LFS 70.9% vs 40% p = .01). On multivariate analysis, only BM blast of < 70% and diagnostic WBC of less than 30 x 10⁹ remained predictive of OS (p = .048) and LFS (p = .049).

Long term outcomes of patients who remained disease free 2 years post consolidation were excellent, with 84.5% alive and disease free 10 years after diagnosis. 4 (9.3%) patients relapsed in the HiDAC group, 6 (14.3%) in autologous HSCT and 2 (4%) in allogeneic HCT group. There is a tendency towards late AML relapse (> 4 yrs) in patients receiving HiDAC and autologous HCT (4/43 and 4/42 respectively) compared with those receiving allogeneic HSCT 1/49, although the differences were not statistically significant (p = .36). 8 patients in both HiDAC and autologous HSCT group developed a secondary malignancy while 3 in the allogeneic HCT group had a second cancer. 10yr OS in those who were alive and disease free 2 years post consolidation was 88.7% in the HiDAC group, 82.6% in the autologous group and 85.6% in the allogeneic group.

In conclusion autologous HCT remains a viable post-remission treatment modality for AML in first complete remission, with acceptable short and long term outcomes compared with allogeneic HCT and HiDAC consolidation. The benefit of autologous HCT appears to be most pronounced in the IR group and patients with low BM blast and WBC at diagnosis. We recommend that autologous HCT be considered in these groups of patients without a matched donor.

Overall Survival for intermediate risk patients.  P = < 0.001                                                                                      Leukemia Free Survival for intermediate risk patients.  P = < 0.001