Comparison of Induction Strategies and Responses for Acute Myeloid Leukemia Patients after Resistance to Hypomethylating Agents for Antecedent Myeloid Malignancy

Background

Outcomes in patients (pts) with secondary acute myeloid leukemia (sAML) (therapy related myeloid neoplasms and AML with myelodysplasia related changes (MRC) per WHO 2016 classification (Arber et al, Blood 2016)) are poor. Pts treated with hypomethylating agents (HMAs) have suboptimal responses to induction chemotherapy (IC) upon transformation to AML. Previously, it was retrospectively demonstrated that the IC with cladribine, cytarabine, filgrastim, and mitoxantrone (CLAG-M) yields significantly higher response rates (64%) than 7+3 (cytarabine and anthracycline) (29%) in pts with prior HMA exposure (Jaglal et al, Leukemia Research 2014). Following the recent approval of CPX-351 for induction in sAML subgroup, we investigated outcomes after CPX-351 to cladribine based regimens and 7+3 in pts with sAML with prior HMA exposure.

Methods

We identified pts with sAML who had prior HMA treatment for an antecedent hematologic malignancy (AHM) and later received induction chemotherapy upon AML transformation from Moffitt Cancer Center (MCC) (n=229), Memorial Sloan Kettering Cancer Center (n=11) and Roswell Park Comprehensive Cancer Center (n=2). Patients were divided into 3 cohorts based on induction regimen: (A) cladribine based (CLA+/-G+/-M) (B) standard 7+3 and (C) CPX-351. Demographics, disease-specific variables, and outcomes were collected in accordance with the institutional review board approved protocol. Responders (R) were defined as pts achieving CR or CRi as defined by the 2003 International Working Group (IWG) criteria after 1 or 2 cycles of the either induction regimen whereas non-responders (NR) were defined as responses other than CR/CRi. Pts receiving a second induction with a different regimen were considered NR. Fisher's exact test and the ANOVA test were used to determine significance for continuous and categorical variables. Kaplan-Meier analysis with log-rank test was performed to estimate overall survival (OS).

Results

Among 242 pts who received IC for AML after HMA failure for prior AHM, 114 were treated with (A) cladribine based regimen (B) 94 pts with standard 3+7 and (C) 34 pts with CPX-351 (Cohort C). Baseline characteristics for all 3 cohorts are outlined in Table 1A. Median age for cohort A, B, and C were 65 (33-82), 66 (26-81), and 69 (36-82), respectively. Males comprised of 68.4%, 63% and 52.9% of the cohorts A, B and C, respectively. No pts had favorable-risk karyotype as defined by European LeukemiaNet (ELN) 2017 criteria. Adverse risk karyotype was noted in 42.1% of cohort A, 34.6% of cohort B and 22.7% of cohort C (p=.337). The majority of pts received azacitidine as their HMA for their AHM (88.7%, 84.9% and 82.4% in cohorts A, B, C, respectively) and median number of cycles administered prior to transformation to AML were 6, 4 and 5 for cohorts A, B, and C, respectively.

Response rates in each cohort are summarized in Table 1B. The CR/CRi rate was 53% in cohort A, 32% in cohort B and 41.1% in cohort C (p=.005 between cohort A and B) (p=.329 between cohorts A and C) (p=.526 between cohorts B and C). The early death rates (<60 days of induction) were not significantly different among the 3 cohorts, at 12%, 8% and 2.9% in cohorts A, B and C respectively (p=.200).

In pts who received ≤ 4 cycles of HMAs prior to AML transformation, response rates to CPX-351 were higher (64.3%) than in pts who received >4 cycles of HMAs (25.0%) (p=.0397). Cohort A (56.5% vs. 50.0%, p=.288) and B (39.1% vs. 25.5%, p=.175) did not demonstrate such a difference (Table 1C and 1D). There was a trend towards better OS (19.9 vs. 5.5mo) with CPX-351 treated pts with ≤ 4 cycles of HMAs compared to >4 cycles (p=.092) (Figure 1).

To date, 70.0% of responding pts in cohort A have undergone an allogeneic stem cell transplant compared to 31.0% in cohort B and 28.6% in cohort C (p=.15). There was no significant difference in median OS among the 3 groups, cohort A (7.27 months), cohort B (7.63 months) and cohort C (7.07 months) (p=.887). Among responders, the mOS did not differ (12.93, 21.7, and 19.9 months for cohorts A, B, and C respectively, p=.635).

Conclusions

We demonstrate that cladribine-based induction regimens and CPX-351 yield higher CR/CRi rates compared to 7+3 in pts with sAML after HMA failure. Prolonged duration of HMA exposure may lower response potential with CPX-351 upon AML transformation. Median OS remains poor and did not differ among the 3 groups illustrating the unmet need for therapy for sAML pts after HMA failure.

View largeDownload slide

View largeDownload slide

Close modal