Clinical Outcomes of Patients with Secondary Acute Myeloid Leukemia (sAML) Treated with Hypomethylating Agent Plus Venetoclax (HMA-Ven) or Liposomal Daunorubicin Cytarabine (CPX-351)

The “7+3” regimen is recommended for treatment in patients with new diagnosis of acute myeloid leukemia (AML) who are fit for intensive chemotherapy. Patients with secondary AML (sAML) [i.e. AML evolving from antecedent hematologic disorders (AHD-AML) or after previous exposure to chemo/radiation therapy for unrelated cancer (t-AML)], have inferior outcomes with “7+3” regimen. A recent phase 3 study demonstrated superior CR rates and overall survival (OS) with upfront use of CPX-351 compared to “7+3” regimen in older patients (≥60 years) with sAML (Lancet et al JCO 2018). The combination of HMA+Ven is FDA approved for upfront treatment in newly diagnosed AML patients > 75 or those unfit for intensive chemotherapy based on CR+ CRi rates of 67% and median OS of 17.5 months. Herein, we compared the outcomes of older patients with sAML who received upfront treatment with either HMA+Ven or CPX-351 at our institution.

Our analysis includes 47 consecutive patients with previously untreated sAML treated between 2018-2020 . Patients were treated with either HMA+Ven (n=27) or CPX-351 (n=20) based on physician preference. WHO criteria were used for the diagnosis AHD-AML and review of medical records for documenting exposure to leukemogenic agent for t-AML. Complete remission (CR) was defined by presence of <5% blasts in bone marrow (BM) aspirates. CR with blood count recovery (i.e., absolute neutrophil counts >1000/µL and platelets ≥100,000/µL) were defined as CRh (hematologic recovery) and CR without blood count recovery as CRi (incomplete blood count recovery). Minimal residual disease (MRD) assessment was done on day-28 BM aspirate using multiparametric flow cytometric assay with lower limit of sensitivity of 0.01%.

Patients demographic and disease features are summarized in Table 1. Mean age (p=0.39), mean blast percentage in BM aspirate (P=0.82), high-risk cytogenetics (P=0.37) and high-risk molecular mutations (P=0.737) were similar in both treatment groups. Of the 27 cases of sAML in HMA+Ven group, 8 were t-AML arising after prior chemotherapy (Hodgkin’s Disease n=2; paraganglioma, desmoid tumor, breast cancer, NHL, multiple myeloma, ALL: one each) while 19 were AHD-AML. In CPX-351 group, 6 cases were t-AML arising after prior chemotherapy (NHL n=2, breast cancer n=2, T-Lymphoblastic Lymphoma and colon cancer one each) while 14 were secondary to AHD. The mean number of cycles were 3.3 (range 1-18) in HMA+Ven group and 1.45 (1-3) in CPX-351 group. Two-sample t test was used to compare continuous and normally distributed covariates, such as age and BM blasts, between HMA+Ven or CPX-351 arms. Pearson Chi-square or Fisher exact test was used to assess the associations between treatment and clinical outcomes. Kaplan-Meier method and log-rank test were used to assess OS or LFS. A P value of ≤ 0.05 was considered as statistically significant.

The CR rate in patients treated on HMA+Ven group was 78% (n=21; 95% CI: 58-91%) vs 50% (n=10; 95% CI: 27-73%) in CPX-351 group (P=0.047). CRi was achieved in 52% (n=14) patients in HMA+Ven group compared to 25%(n=5) patients in CPX-351 group(p=0.064). MRD negative remission was achieved in 52% (n=14) patients in HMA+Ven group and in 25% (n=5) patients in CPX-351 group (p=0.064). In HMA+Ven group, 52% patients (n=14) achieved remission after one cycle of therapy compared to 45% (n=9) patients in CPX-351 arm (p=0.642). With a median follow-up of 6.7 months for all patients, the median leukemia free-survival (LFS) for HMA+Ven vs CPX-351 treatments is 16.2 vs NA months (P = 0.098) and the median OS is 13.2 vs NA months (P = 0.395). Ten patients in each group (37% in HMA+Ven and 50% in CPX-351; p=0.47) underwent allo-HCT. At last follow up, 14 patients (52%) have died in HMA+Ven group from: relapsed AML (n=10), sepsis (n=2), congestive heart failure (n=1) and unknown (UK) in one patient, whereas in CPX-351 group, 8 patients (40%) have died from relapsed AML in (n=5), respiratory failure (n=2) and UK causes in one patient. In patients resistant to initial therapy, the median OS is 3.5 vs 6.0 months between HMA+ Ven and CPX-351 groups (P = 0.224).

Conclusion: In patients presenting with sAML, upfront treatment with HMA+Ven is feasible and associated with significantly better CR rates and a favorable trend for higher rates of negative MRD compared to CPX-351. A randomized prospective trial in patients with sAML is warranted to determine the most effective frontline regimen in this high-risk AML subgroup.