The latest updates on oral azacitidine in AML

Oral azacitidine (oral-AZA), a hypomethylating agent (HMA), was approved by the European Commission and the U.S. Food and Drug Administration for the treatment of adult patients with acute myeloid leukemia (AML) in first remission after intensive induction chemotherapy.

During the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting and the European Hematology Association (EHA) 2023 Hybrid Congress, Amaya^1,2 presented a poster on a phase I trial of an all-oral azacitidine plus venetoclax regimen in patients with relapsed/refractory (R/R) AML. The phase III QUAZAR AML-001 trial (NCT01757535), which has been covered extensively by the AML Hub, showed that oral-AZA was associated with an improved overall survival (OS) and relapse-free survival (RFS) versus placebo in older patients with AML in remission. During EHA 2023, Wei³ and Roboz⁴ presented subgroup analyses from this trial. Wei et al.⁵ and Gaudy et al.⁶ also recently published further subgroup analyses from this trial in Hematologica and Clinical Pharmacology & Therapeutics, respectively.

Phase I trial of azacitidine and venetoclax^1,2

This was an open-label, phase I study in patients with R/R AML. This study had two cohorts: patients received either 200 mg (Cohort 1, n = 3) or 300 mg (Cohort 2, n = 5) of oral-AZA on Days 1–14, plus 400 mg/day of venetoclax for 28 days following an initial dose escalation of 100 mg on Day 1, 200 mg on Day 2, and 400 mg on Days 3–28.

Key findings

• Dose-limiting toxicities (DLTs) were not identified in either cohort.
• In both cohorts, 100% of patients experienced drug-related adverse events (AEs) and hematologic Grade ≥3 AEs.
• Non-hematologic Grade ≥3 AEs occurred in 33% and 20% of patients in Cohort 1 and 2, respectively.
• The most common Grade 3–4 AEs in Cohort 1 were:
  • Neutropenia (100%);
  • Anemia (66%);
  • Thrombocytopenia (33%).
• The most common Grade 3–4 AEs in Cohort 2 were:
  • Neutropenia (80%);
  • Anemia (80%);
  • Thrombocytopenia (60%);
  • Febrile neutropenia (20%).

One patient in Cohort 1 and one in Cohort 2 achieved complete remission (CR). In addition, one patient in Cohort 2 achieved partial remission, while two patients in Cohort 1 achieved stable disease.

QUAZAR AML-001

Longitudinal monitoring of mutational profile³

This subgroup analysis from the QUAZAR AML-001 trial assessed the gene mutations present at screening before maintenance. It also compared the mutational profile at relapse in both treatment arms.

Key findings

Of the 310 patients included in this analysis, 71.3% had detectable mutations:

• The most commonly observed mutations were DNMT3A (28.4%), TP53 (15.5%), IDH2 (12.3%), TET2 (11.9%), SRSF2 (11.0%), IDH1 (6.1%), and ASLX1 (5.5%).
• A persistence of clonal hematopoiesis-associated mutations (DNMT3A, ASXL1, TET2) was observed at relapse, regardless of treatment received.
• Oral-AZA was not associated with an increase in TP53 variant size at Cycle 6 or relapse.
• RAS-pathway mutations were more common in the placebo arm compared with oral-AZA arm at relapse (28.6% vs 14.5%; p = 0.03).

NPM1 measurable residual disease monitoring⁴

This subgroup analysis from the QUAZAR AML-001 trial evaluated measurable residual disease (MRD) monitoring by polymerase chain reaction (PCR) in 81 patients with NPM1 mutations.

Key findings

Following intensive chemotherapy, NPM1 mutant (NPM1^mut) was detected in 38.2% of patients by PCR positivity versus 1.2% by next-generation sequencing (NGS). This suggested NGS was not sensitive enough to detect NPM1 MRD in the patients in remission. Overall, PCR and multiparameter flow cytometry (MFC) detection of NPM1^mut MRD at screening were concordant (p < 0.0001) for all cycles. However,

• 30% of cases were discordant.
• 16% of cases were PCR positive (PCR+) and MFC negative (MFC−).
• 14% of cases were PCR negative (PCR−) and MFC positive (MFC+).

Patients who were NPM1 MRD positive detected by both PCR and MFC had inferior RFS versus patients who were MRD negative (Table 1). Multivariable analysis showed that MFC positivity had a minimal effect on RFS (p = 0.01), suggesting that PCR at screening had a stronger prognostic relationship with clinical outcomes than MFC.

Table 1. RFS by MRD status*

Longitudinal MRD assessment suggested that PCR and MFC were not concordant:

• 81% of patients who were PCR+ and MFC− became MFC+.
• 56% of patients who were PCR− and MFC+ became MFC−.

All patients who were PCR+ at screening were PCR+ at relapse; however, 40.9% of patients who were PCR− at screening remained PCR− at relapse. Median RFS was lower in patients with NPM1^mut PCR+ versus PCR− in both oral-AZA (15.6 vs 46.1 months) and placebo arm (5.5 vs 13.8 months). Multivariable analysis revealed oral-AZA remained associated with improved RFS (hazard ratio [HR], 0.56; p = 0.04) regardless of MRD status, while PCR+ status at screening had a stronger association with inferior RFS (HR, 1.95; p = 0.02) versus MFC+ status (HR, 1.31; p = 0.36).

Impact of initial chemotherapy on survival outcomes⁵

This post hoc analysis from the QUAZAR AML-001 trial assessed the impact of receiving consolidation chemotherapy on RFS and OS. A total of 212, 166, and 94 patients received 1 consolidation, ≥2 consolidation, and 0 consolidation cycles, respectively.

Key findings

Oral-AZA was associated with improved RFS and OS versus placebo, independent of whether the patients received consolidation therapy. In patients who did not receive consolidation therapy, oral-AZA versus placebo, was associated with:

• Improved median RFS (8.4 vs 3.9 months; HR, 0.58; p = 0.0258).
• Higher estimated 1-year RFS rate (40.8% vs 22.0%).
• Improved median OS (23.3 vs 10.9 months; HR, 0.54; p = 0.0103).
• Higher estimated 1-year OS (71.2% vs 40.5%).

In patients who received consolidation therapy, oral-AZA versus placebo, was associated with:

• Improved median RFS (10.2 vs 5.0 months; HR, 0.67; p = 0.001).
• Higher 1-year RFS rate (45.9% vs 28.6%).
• Improved median OS (24.7 vs 15.4 months HR, 0.74; p = 0.0147).
• Higher 1-year OS rate (73.2% vs 59.2%).

This median RFS benefit of oral-AZA versus placebo was observed in patients receiving 1 consolidation cycle (10.0 vs 4.7; HR, 0.72) as well as those receiving ≥2 consolidation cycles (13.0 vs 6.1; HR, 0.59). The estimated 1-year OS rate was also improved with oral-AZA versus placebo in patients receiving 1 consolidation cycle (68.8% vs 59.2%) and ≥2 consolidation cycles (80.0% vs 59.2%).

Oral-AZA numerically improved median RFS versus placebo by 1.5–8.5 months across all induction/consolidation cohorts. The safety profile of oral-AZA was similar across all consolidation groups.

Pharmacokinetics of oral azacitidine⁶

This study examined the pharmacokinetics of oral-AZA using patient data from the AZA-MDS-004 trial (NCT01519011; n = 31), CC-486-CAGEN-001 trial (NCT02223052; n = 30), and the QUAZAR AML-001 trial (n = 227).

Key findings

Multivariable analysis confirmed the association of log-transformed area under the concentration–time curve at steady state (logAUC_ss; HR, 0.441; p <0.001), log-transformed maximum plasma concentration at steady state (logC_max,ss; HR, 0.559; p = 0.002) with RFS and logAUC_ss (HR, 0.599; p = 0.010) with OS.

Logistic regression analyses revealed the probability of developing Grade ≥3 neutropenia with oral-AZA:

• LogAUC_ss (odds ratio [OR], 5.71; p <0.001).
• Cumulative area under the plasma concentration-time curve from Cycle 1 to Cycle 6 (logAUC_C1–C6; OR, 2.71; p <0.001).
• LogC_max,ss (OR, 2.38; p = 0.012).

Diarrhea was the only gastrointestinal AE associated with logAUC_ss (OR, 2.69; p = 0.01).

Of the 227 patients in the exposure-safety analysis, 56.8% of patients had no dose modification, 19.4% extended their treatment schedules due to AML relapse, and 22.9% had a dose reduction due to AEs. 

Conclusion

The phase I trial^1,2 did not identify any DLTs and showed responses to oral-AZA in both cohorts. A dose-expansion cohort is planned for an additional 10 patients treated at the maximum tolerated dose.

The subgroup analyses from the QUAZAR AML-001 trial demonstrated that the mutational spectrum at relapse was similar between oral-AZA versus placebo arms, except for RAS-pathway mutations that were more frequent in the placebo arm.³ The prognostic value of NPM1 MRD monitoring supports the use of PCR-based post-remission NPM1 MRD monitoring.⁴ Oral-AZA maintenance therapy improved both RFS and OS versus placebo, independent of consolidation therapy post-induction.⁵ Although the analysis was limited by small sample sizes and lack of statistical power, the number of consolidation or induction cycles received did not impact the efficacy of oral-AZA. The optimal dose of oral-AZA was identified as 300 mg daily for 14 days.⁶