Venetoclax Exposure-Efficacy and Exposure-Safety Relationships in Subjects with Treatment-Naïve Acute Myeloid Leukemia Who Are Ineligible for Intensive Chemotherapy

Introduction: Venetoclax (VEN), an orally bioavailable selective BCL-2 inhibitor, was granted accelerated approval by the FDA for the treatment of acute myeloid leukemia (AML) in combination with a hypomethylating agent (HMA) or low-dose cytarabine (LDAC). The objective of this analysis was to characterize the exposure-efficacy and exposure-safety relationships using data from the nonrandomized Phase 1/2 studies and the confirmatory Phase 3 studies in subjects with treatment-naïve AML who are ineligible for intensive chemotherapy.

Methods: A population pharmacokinetic (PK) model for VEN was characterized using data from 771 AML subjects across 5 Phase 1 - 3 studies including subjects with relapsed/refractory AML and subjects with treatment-naïve AML who were ineligible for intensive chemotherapy. PK parameters from this model were used to calculate the exposure metric, the area under the plasma concentration-time curve at steady-state (AUCss), for exposure-efficacy and exposure-safety analyses. Logistic regression and Cox proportional-hazards models were used to characterize the exposure-response relationships for overall survival, event-free survival, and conversion to post-baseline transfusion independence for red blood cells or platelets as well as key safety endpoints of treatment-emergent Grade ≥ 3 neutropenia, thrombocytopenia, and infections. Additionally, best overall response of complete remission (CR), CR + CR with incomplete blood count recovery (CRi), and CR + CR with partial hematological recovery (CRh) were also characterized in the exposure-response analyses. Separate analyses were performed for VEN + HMA and VEN + LDAC, including subjects taking placebo (PBO) + HMA or PBO + LDAC, respectively, to inform the intercepts. A total of 575 subjects with treatment-naïve AML were included in the exposure-response analyses for VEN + HMA and 279 subjects with treatment-naïve AML were included in the exposure-response analyses for VEN + LDAC.

Results:

VEN + HMA: For all efficacy endpoints studied, there was a clear trend for higher efficacy in subjects treated with VEN + HMA as compared to PBO in combination with azacitidine (AZA). When PBO patients were included, significant exposure-response relationships between VEN exposure and overall survival and event-free survival were identified, and a maximal effect (E_max) model best described the statistically significant positive relationships between VEN exposure and best overall response of CR + CRi and CR + CRh. However, there were no apparent relationships between VEN exposure and any efficacy endpoint without the inclusion of PBO subjects, suggesting that the beneficial effect of VEN is already maximized in the dose range studied (400 to 1200 mg daily [QD]). No exposure-response relationship was identified between VEN exposure and treatment-emergent Grade ≥ 3 thrombocytopenia or infections. An E_max model best described the statistically significant relationship between VEN exposure and treatment-emergent Grade ≥ 3 neutropenia, consistent with the clinical finding that subjects in the VEN + HMA treatment arms had higher rates of reported neutropenia than those in the PBO + AZA arm.

VEN + LDAC: For all efficacy endpoints studied, there was a clear trend for higher efficacy in subjects treated with VEN + LDAC as compared to PBO + LDAC. With PBO subjects included, an E_max model best described the statistically significant relationships between VEN exposure and best overall response of CR, CR + CRi, and CR + CRh, with probability of response plateauing at exposures corresponding to 600 mg QD of VEN. There were no apparent relationships between VEN exposure and any efficacy endpoint when the PBO subjects were excluded from analysis, indicating that the beneficial effect of VEN is already maximized at 600 mg QD. Although reported rates of treatment-emergent Grade ≥ 3 neutropenia were higher in the VEN + LDAC treatment arms as compared to PBO + LDAC, no significant exposure-response relationships were found between VEN exposure and treatment-emergent Grade ≥ 3 neutropenia, thrombocytopenia, or infections.

Conclusion: The VEN exposure-efficacy and exposure-safety analyses confirm the benefit of adding VEN to either HMA or LDAC and support the dose regimens of VEN 400 mg QD in combination with an HMA and VEN 600 mg QD in combination with LDAC in treatment-naïve AML patients who are ineligible for intensive chemotherapy.