CYP3A inhibitors and impact of these agents on outcomes in patients with acute myeloid leukemia treated with venetoclax plus azacitidine on the VIALE-A study

Background: In the Phase 3 randomized, placebo-controlled VIALE-A study, venetoclax (Ven) + azacitidine (Aza) improved overall survival (OS) and complete response (CR) rates vs Aza + placebo (Pbo; DiNardo et al. N Engl J Med. 2020. In Press). Neutropenia and related infections are common in acute myeloid leukemia (AML) and exacerbated by Ven combinations. Antimicrobial prophylaxis is used variably in patients (pts) with AML receiving low-intensity therapies. Ven dose modifications are required for concomitant use of commonly used antimicrobials that are strong or moderate cytochrome P450 3A inhibitors (CYP3Ai). A previous report’s findings showed that Ven dose reductions with CYP3Ai do not affect Ven exposure (Wei et al. Blood. 2020;135:2137). This analysis evaluated the use of prophylactic anti-infective CYP3Ai and these agents’ effects on infections and efficacy outcomes in the VIALE-A study (NCT02993523).

Methods: Pts with newly diagnosed AML, aged ≥75 y or 18-75 yand ineligible for induction therapy were enrolled. Pts were randomized 2:1 to receive Ven (daily, orally) + Aza (75 mg/m² d 1-7, subcutaneously/intravenously) or Pbo + Aza in 28-d cycles. Anti-infective prophylaxis was required for pts with absolute neutrophil count <500/µL. For concomitant use of CYP3Ai, Ven dose was reduced from the intended target dose of 400 mg to 200 mg (Cycle 1 ramp-up: d 1, 50 mg; d 2, 100 mg; d 3, 200 mg) for moderate CYP3Ai and 50 mg (Cycle 1 ramp-up: d 1, 10 mg; d 2, 20 mg; d 3, 50 mg) for strong CYP3Ai. Use of anti-infective CYP3Ai agents reported as being given for prophylaxis during Cycles 1 and 2 was assessed; there was a possibility that CYP3Ai agents were used for nonprophylactic purposes not documented as such. The effects of concomitant use of these agents with the required Ven dose reductions on infection and efficacy outcomes were evaluated.

Results: In total, 286 pts were randomized to Ven + Aza and 145 to Pbo + Aza. Within the first 2 cycles of therapy, concomitant anti-infective prophylaxis agents considered moderate CYP3Ai were received by 41/286 pts (14%) in the Ven + Aza arm and 18/145 (12%) in the Pbo + Aza arm. Concomitant anti-infective prophylaxis agents considered strong CYP3Ai were received in the first 2 cycles by 22/286 pts (8%) treated with Ven + Aza and 13/145 pts (9%) treated with Pbo + Aza. The median duration of each prophylactic CYP3Ai agent use was 12.5 d (range, 1–614) in the Ven + Aza arm and 15 d (range, 1–731) in the Pbo + Aza arm.

The rates of CR+CR with incomplete marrow recovery (CRi) as a best response during the study were similar with concomitant use of moderate (61%; CR, 24%; CRi, 37%) or strong (64%; CR, 27%; CRi, 36%) CYP3Ai with adjusted-dose Ven vs no use of CYP3Ai (67%; CR, 39%; CRi, 27%; Table 1). The relatively lower CR and higher CRi rates appear to be due to delay in recovery of peripheral blood counts in pts who received strong CYP3Ai during the first 2 cycles (Table 2). The median time to first CR+CRi was 1.2 mo (range, 0.6–9.9), 1.4 mo (range 1.0–5.5), and 1.4 mo (range, 0.9–5.4) in those receiving no, moderate, and strong CYP3Ai agents, respectively, in the Ven + Aza arm. Median OS was not statistically different regardless of use of moderate or strong CYP3Ai in both arms, as noted by overlapping CIs (Table 1). The 24-month estimated OS was 26.2% (95% CI, 8.1–48.9) for pts receiving strong CYP3Ai vs 37.9% (95% CI, 30.2–45.6) for those receiving no CYP3Ai. Rates of any-grade or Grade 3/4 infections within the first 2 cycles varied between those with or without use of moderate CYP3Ai reported as being given as prophylaxis in both treatment arms (Table 1). Rates of invasive fungal infections were 3%, 12%, and 9% with Ven + Aza and 0%, 0%, and 15% with Pbo + Aza in pts receiving no, moderate, and strong CYP3Ai agents, respectively. Rates of discontinuation from infections were similar in both arms regardless of CYP3Ai use.

Conclusions: Antimicrobial prophylaxis with moderate or strong CYP3Ai was used in neutropenic pts in the VIALE-A study, with overall similar composite remission rates with Ven dose reductions. The use of CYP3Ai did not increase rates of discontinuation from infections. Analysis of differences in baseline characteristics, rates of infections, and efficacy outcomes for pts who began anti-infective prophylaxis at the initiation of therapy is ongoing.