Optimizing induction therapy in older patients with AML undergoing allo-HSCT

Survival outcomes in older patients with acute myeloid leukemia (AML) are historically poor, mainly due to factors such as poor tolerance to conventional chemotherapy and a higher incidence of cytogenetic risk in this population.¹ Lower-intensity venetoclax-based approaches have shown promising results in older patients with AML who are unfit for intensive chemotherapy (IC). However, the type of induction therapy used in older patients with newly diagnosed AML who are fit for allogeneic hematopoietic stem cell transplantation (allo-HSCT) is controversial and the impact on outcomes remains largely unknown.¹

Short et al.¹ recently published a retrospective study in Blood Advances evaluating the impact of the type of induction therapy in older patients with newly diagnosed AML after allo-HSCT. We are pleased to present the key highlights from this study here.

Study design

This was a retrospective study in older patients with newly diagnosed AML undergoing allo-HSCT at the University of Texas MD Anderson Cancer Center between November 2012 and July 2021. The study design is shown in Figure 1.

allo-HSCT, allogeneic hematopoietic stem cell transplantation; AML, acute myeloid leukemia; CR, complete remission; CRi, CR with incomplete recovery; IC, intensive chemotherapy; LIT, lower intensity therapy; MLFS, morphologic leukemia-free state; VEN, venetoclax.
*Data from Short, et al.¹

Responses, defined by the European LeukemiaNet consensus guidelines were as follows:

• Minimal residual disease (MRD) negativity — absence of a distinct cluster of ≥20 cells showing altered expression of ≥2 antigens
• Relapse-free survival (RFS) — time from allo-HSCT until relapse or death from any cause
• Overall survival (OS) — time from allo-HSCT until death from any cause
• Non-relapse mortality (NRM) — death occurring after allo-HSCT in the absence of relapse
• Graft-versus-host disease — graded per standard criteria
• Cumulative incidence of relapse (CIR)

Results

Baseline characteristics

A total of 127 patients were included, of which 44 patients received IC, 29 received low-intensity therapy (LIT) without venetoclax (VEN), and 54 patients received LIT with VEN. Patients in the IC cohort versus LIT without VEN or LIT with VEN were younger and more likely to have an Eastern Cooperative Oncology Group performance status of 0 at the time of AML diagnosis (p = 0.02) and de novo AML (p = 0.03). Most patients had intermediate or adverse-risk AML; however, two patients had core-binding factor AML in the IC cohort (Table 1).

Table 1. Baseline characteristics*

AML, acute myeloid leukemia; ECOG PS, Eastern Cooperative Oncology Group performance status; ELN, European LeukemiaNet; IC, intensive chemotherapy; LIT, lower-intensity therapy; MLFS, morphologic leukemia-free state; VEN, venetoclax. *Adapted from Short, et al.1 †Statistically significant values are shown in bold. ‡Core-binding factor AML (n = 2).
Characteristic, % (unless otherwise stated)	IC without VEN (n = 44)	LIT without VEN (n = 29)	LIT + VEN (n = 54)	p value†
Median age (range), years	63 (60–76)	67 (61–75)	68 (60–77)	<0.0001
ECOG PS
0	34	10	17	0.06
1	61	72	72	—
2	5	17	11	—
AML type
de novo	80	69	56	0.04
Secondary	20	31	44	—
ELN 2017 risk
Favorable‡	20	14	19	0.66
Intermediate	36	38	26	—
Adverse	43	48	56	—
Mutations
NPM1	27	24	22	0.85
FLT3-ITD	20	24	11	0.26
IDH1/IDH2	25	10	26	0.22
KRAS/NRAS	11	10	22	0.23
RUNX1	7	3	22	0.02
ASXL1	9	0	17	0.06
TP53	18	21	13	0.62

Outcomes by induction therapy

The median duration of follow-up post allo-HSCT was 59.8 months, 76.8 months, and 20.3 months in the IC, LIT without VEN, and LIT + VEN cohorts, respectively. The 2-year CIR, NRM, RFS, and OS for all three cohorts are shown in Figure 2.

CIR, cumulative incidence of relapse; IC, intensive chemotherapy; LIT, lower-intensity therapy; NRM, non-relapse mortality; OS, overall survival; RFS, relapse-free survival; VEN, venetoclax.
*Data from Short, et al.¹

Cumulative incidence of relapse and non-relapse mortality

• Patients achieving MRD negative status prior to allo-HSCT showed a higher CIR in the LIT without VEN versus the IC or LIT + VEN cohorts (45% vs 23% and 17%, respectively).
• In patients with adverse-risk AML, the 2-year CIR was higher in the LIT without VEN versus IC or LIT + VEN cohorts (57% vs 22% and 23%, respectively).
• Patients in the IC cohort showed a higher rate of neutropenic fever compared with the LIT without VEN or LIT + VEN cohorts combined (61% vs 43%, respectively, p = 0.06).
• Patients with TP53 mutations in the LIT without VEN cohort showed a particularly high 1-‍year CIR of 67% and those in the IC cohort showed a high 1-year NRM of 63%.

Relapse-free survival and overall survival

Figure 3 shows that the 2-year RFS and OS were higher in the LIT + VEN cohort compared with IC and LIT without VEN  cohorts in patients with:

• MRD negative status
• Adverse-risk AML

In addition, patients with TP53 mutations in the LIT + VEN cohort also showed improved RFS and OS compared with those in the IC and LIT without VEN cohorts (RFS, p = 0.05; OS, p = 0.02).

AML, acute myeloid leukemia; IC, intensive chemotherapy; LIT, lower-intensity therapy; MRD, minimal residual disease; OS, overall survival; RFS, relapse-free survival; VEN, venetoclax.
*Adapted from Short, et al.¹

In a univariate analysis, HSCT comorbidity index (HSCT-CI) ≥3 was associated with significantly increased risk of relapse (sub-distributional hazard ratio [SHR], 2.21; p = 0.03).

• Patients with adverse-risk AML showed a trend towards increased risk of relapse (SHR, 1.97; p = 0.07) as well as those in the LIT without VEN cohort (SHR, 2.16; p = 0.09).
• Adverse-risk AML (hazard ratio [HR], 1.70; p = 0.05), MRD positivity prior to allo-HSCT (HR, 1.79; p = 0.03), and HSCT-CI ≥3 (HR, 2.54; p = 0.001) were associated with inferior RFS.
• MRD positivity prior to allo-HSCT (HR, 1.78; p = 0.04) and HSCT-CI ≥3 (HR, 2.62; p = 0.001) were associated with inferior OS.

In a multivariate analysis, only HSCT-CI ≥3 was associated with inferior RFS (HR, 1.90; p = 0.02) and OS (HR, 2.09; p = 0.01).

Conclusion

This retrospective study showed favorable RFS and OS with LIT + VEN in older patients with newly diagnosed AML fit for allo-HSCT, providing a rationale to consider LIT + VEN in this population. The study also highlighted better survival outcomes with LIT + VEN compared with IC induction for patients with adverse-risk AML. This appears to be the largest cohort assessing LIT + VEN after allo‑HSCT and comparing IC and LIT + VEN in older allo-HSCT eligible patients. However, the findings of the study should be considered with several caveats such as confounding factors due to younger and fitter patients in the IC cohort and inclusion of patients who underwent allo-HSCT in first remission. The patients in the cohorts were treated in different time periods, and this may have impacted the outcomes due to advances in treatment for relapsed/refractory AML, better supportive care, and changes in allo-HSCT procedures.