Does intensifying post-remission chemotherapy improve the outcomes of older patients with AML?

Whilst approximately 50–70% of older, fit patients with acute myeloid leukemia (AML) are able to achieve complete remission (CR), preventing subsequent relapse is crucial.¹ Whilst the post-remission standard of care for younger patients is repeated courses of intermediate- or high-dose cytarabine (I/HDAC), many older patients have limited treatment options due to higher susceptibility to comorbidities and lower treatment tolerance. In this population of patients, no optimal consolidation therapy has been identified and finding effective consolidation approaches remains a challenge.

To better understand the impact of different post-remission therapies on older patient’s outcomes, Benjamin Bouchacourt and colleagues conducted a retrospective analysis of patients aged 60 or older treated at Institut Paoli-Calmettes, Marseille, FR.¹

Read more about real-world experience with non-intensive regimens in older patients with AML here.

Study design

• Older patients (≥ 60 years; n = 441) in first CR treated between 1995–2014 with at least one cycle of consolidation therapy were included in the analysis
• All patients were given one or two cycles of anthracycline-cytarabine-based induction chemotherapy to achieve CR. Regimens included
  • idarubicin (8 mg/m²) intravenous (IV) injection on Days 1–5 and cytarabine (100 mg/m²) continuous infusion on Days 1–7
  • idarubicin (12 mg/m²) IV injection on Days 1–3 and cytarabine (100 mg/m²) continuous infusion on Days 1­–7
  • daunorubicin (45–60 mg/m²) on Days 1–3 and cytarabine (100 mg/m²) continuous infusion on Days 1–7
• Patients were divided into subgroups by post-remission therapy:
  • Allogeneic hematopoietic stem cell transplant (allo-HSCT): 14% (n = 61)
  • Autologous hematopoietic stem cell transplant (auto-HSCT): 11% (n = 51)
  • Intensive chemotherapy with I/HDAC: 16% (n = 70)
  • Standard dose chemotherapy (1+5 regimen) and no HSCT: 43% (n = 190)
  • Low dose chemotherapy (LDC): 6% (n = 26)
  • Other (immunotherapy, targeted therapy, etc.): 10% (n = 43)

Patient characteristics

• Median age, years (range): 67 (60–84)
• Median follow-up from CR: 77.7 months (95% CI, 73.7–95.3)
• Most patients had de novo AML (87%)
• Most (67%) had intermediate-risk cytogenetics, though 21% had unfavorable cytogenetics
• Most patients had a performance status (PS) of 0–1 (64%)
• Patients who underwent allo-HSCT were significantly younger (p < 0.0001) and had more recently received a second induction therapy (p < 0.0001)
• Maintenance therapy was given to 135 patients

Results

Total cohort

• Median leukemia-free survival (LFS), months: 11.5 (range, 9.7–12.7)
• Median overall survival (OS), months: 19.5 (range, 16.1–22.2)
  • No difference in OS by year of transplant
• Univariate analysis found consolidation with allo-HSCT had a favorable impact on OS (hazard ratio [HR] 0.54 [0.38–0.77]; p < 0.001) whilst age, PS ≥ 2, and unfavorable cytogenetics had a negative impact on OS:
  • Age ≥ 65 years: HR 1.38 (1.10–1.74), p = 0.006
  • PS ≥ 2: HR 1.63 (1.18–2.24), p = 0.003
  • Unfavorable cytogenetics: HR 1.85 (1.44–2.38), p < 0.0001
• Multivariate analysis: allo-HSCT, cytogenetic profile, and PS had a significant impact on OS and LFS:
  • OS: allo-HSCT (HR 0.57 [0.38–0.84]; p = 0.005) had a positive impact on OS whilst unfavorable cytogenetics (HR 2.03 [1.52–2.71], p < 0.001) and PS ≥ 2 (HR 1.56 [1.11–2.18]; p = 0.009) were significantly associated with a poor OS. However, age over 65 years was no longer significant
  • LFS: allo-HSCT was associated with a longer LFS (HR 0.43 [0.29–0.63]; p < 0.001) whilst unfavorable cytogenetics and PS ≥ 2 were associated with a shorter LFS (HR 2.12 [1.58–2.83]; p < 0.001 and HR 1.43 [1.02–1.99]; p = 0.04, respectively)

Analysis of outcomes in patients who were not transplanted

[Given as I/HDAC vs 1+5]

• Seventy patients received I/HDAC whilst 190 received 1+5
  • Significantly more patients in the I/HDAC group were ≤ 65 years and had favorable cytogenetics, and more patients had additional maintenance therapy in the 1+5 group
  • Average number of cycles of therapy: 1.8 (± 9) vs 3.5 (± 2.1)
• At a median follow-up of 81 months (range, 71.9–93.4), 207 patients had relapsed (55 vs 152) and 215 had died
  • No statistically significant difference in LFS and OS between groups:
    • Median LFS, months (range): 8.8 (7.0–11.8) vs10.6 (8.2–13.1)
      • HR 0.99 (0.74–1.34), p = 0.96
    • Median OS, months (range): 18.2 (12.6–26.1) vs 18.7 (14.7–22.2)
      • HR 0.89 (0.66–1.21), p = 0.47
    • In subgroup analysis, patients with favorable cytogenetics had an OS benefit with I/HDAC treatment (I/HDAC vs 1+5: 27.1 vs 16.0 months, p = 0.04) whilst patients with unfavorable cytogenetics had an OS benefit with the 1+5 regimen (1+5 vs I/HDAC: 12.9 vs 8.6 months, p = 0.04). There was no difference by age group
    • Toxicity:
      • Patients treated with I/HDAC spent more time in hospital (32 vs 8 days, p < 0.0001)
      • Patients treated with I/HDAC received more units of blood (I/HDAC vs 1+5: 9 vs 4, p < 0.0001) and platelets (I/HDAC vs 1+5: 3 vs 0, p < 0.0001)
      • Ten patients developed a secondary cancer, although the 5-year incidence was not significant between treatment groups (I/HDAC vs 1+5: 11.7% vs 9.0%, p = 0.92)
        • Two of these secondary cancers were non-Hodgkin lymphomas
        • One was essential thrombocythemia

Analysis of outcomes in patients who were transplanted

[Given as I/HDAC vs 1+5]

• Patients who underwent allo-HSCT (n = 61):
  • Received ≥ 1 cycle of I/HDAC prior to transplant (n = 32)
  • Received ≥ 1 cycle of 1+5 prior to transplant (n =18)
    • Patients tended to be older
  • Eleven patients received alternative treatment and were excluded from the analysis
  • Median OS and LFS are shown in Table 1. Neither were significantly different between groups
• Patients who underwent auto-HSCT (n = 51):
  • Received ≥ 1 cycle of I/HDAC prior to transplant (n = 19)
  • Received ≥ 1 cycle of 1+5 plus etoposide and cyclophosphamide prior to transplant (n = 32)
  • Patients receiving 1+5 tended to be older
  • Median OS and LFS are shown in Table 1. There was a non-significant trend to improved outcome with I/HDAC
• Toxicity of I/HDAC and 1+5 for transplant:
  • Patients who received I/HDAC prior to transplant
    • spent more time in hospital
    • received more red blood cells and platelet transfusions
  • Post-transplant outcomes, in relation to 5-year non-relapse mortality (NRM), were not different between the two treatment groups after allo-HSCT or auto-HSCT (Table 1)

Table 1. Analysis of outcomes in patients who were transplanted¹

Allo-HSCT, allogeneic hematopoietic stem cell transplantation; Auto-HSCT, autologous hematopoietic stem cell transplantation; I/HDAC, intermediate-/high-dose cytarabine; LFS, leukemia-free survival; NRM, non-relapse mortality; OS, overall survival
	Total	I/HDAC	1+5	p value
Allo-HSCT
OS, months	53.0	55.8	53.0	0.62
LFS, months	25.8	45.3	53.0	0.73
5-year NRM, %	—	28.7	44.4	0.78
Auto-HSCT
OS, months	29.4	33.8	17.7	0.08
LFS, months	13.3	24.9	11.7	0.20
5-year NRM, %	—	0	10.0	0.37

Conclusion

This large, single-center, retrospective study suggests that intensifying chemotherapy during first complete remission for older patients may not significantly improve patient outcome when compared to standard-dose therapy, irrespective of transplant status. Additionally, using I/HDAC or 1+5 as bridging therapies to transplant in an elderly population was not shown to provide a benefit to OS or LFS.