Efficacy and Biological Correlates of Response in a Phase II Study of Venetoclax Monotherapy in Patients with Acute Myelogenous Leukemia

Abstract

We present a phase II, single-arm study evaluating 800 mg daily venetoclax, a highly selective, oral small-molecule B-cell leukemia/lymphoma-2 (BCL2) inhibitor in patients with high-risk relapsed/refractory acute myelogenous leukemia (AML) or unfit for intensive chemotherapy. Responses were evaluated following revised International Working Group (IWG) criteria. The overall response rate was 19%; an additional 19% of patients demonstrated antileukemic activity not meeting IWG criteria (partial bone marrow response and incomplete hematologic recovery). Twelve (38%) patients had isocitrate dehydrogenase 1/2 mutations, of whom 4 (33%) achieved complete response or complete response with incomplete blood count recovery. Six (19%) patients had BCL2-sensitive protein index at screening, which correlated with time on study. BH3 profiling was consistent with on-target BCL2 inhibition and identified potential resistance mechanisms. Common adverse events included nausea, diarrhea and vomiting (all grades), and febrile neutropenia and hypokalemia (grade 3/4). Venetoclax demonstrated activity and acceptable tolerability in patients with AML and adverse features.

Significance: Venetoclax monotherapy demonstrated clinical activity in patients with AML (relapsed/refractory or unfit for intensive chemotherapy) with a tolerable safety profile in this phase II study. Predictive markers of response consistent with BCL2 dependence were identified. Clinical and preclinical findings provide a compelling rationale to evaluate venetoclax combined with other agents in AML. Cancer Discov; 6(10); 1106-17. ©2016 AACRSee related commentary by Pullarkat and Newman, p. 1082This article is highlighted in the In This Issue feature, p. 1069.

Figure 1

Biologic activity of venetoclax. A and B, the median leukemia-free survival was 2.3 months (range 1.0–2.7) and the median overall survival was 4.7 months (range 2.3–6.0), in relapsed/refractory AML patients or those unfit for intensive chemotherapy treated with venetoclax monotherapy. C, 25 (78%) patients had bone marrow blast counts evaluable at the first assessment (week 4). The best percent change in bone marrow counts at the first assessment (week 4) is shown. Six patients came off study before the first assessment and 1 patient had unevaluable bone marrow (aplasia). Patients who achieved an objective response by the IWG criteria (complete response/complete response with incomplete blood count recovery) are indicated with blue, and patients who had anti-leukemic activity that did not meet IWG criteria are indicated with hashed bars. Patients with a BCL-2 family protein sensitive index at screening and/or IDH1/2 mutations are annotated with asterisks and brackets, respectively. D, 6 patients with a BCL-2 family sensitive index (green) achieved longer durations on venetoclax therapy than the 16 patients with a BCL-2 family resistant index (red) (P = 0.0381 by Wilcoxon signed rank test). The median number of days on venetoclax was 96 days for patients with a BCL-2 sensitive index vs 31 days for patients with a BCL-2 resistant index.

Figure 2

Venetoclax acts through on-target BCL-2 inhibition. Twelve samples were analyzed (>50% viability on thaw; >5% AML blasts). A, mitochondrial BCL-2 dependence correlated with release caused by BAD-HRK. Background BCL-X_L dependence was removed by subtraction of HRK for a specific measure of mitochondrial BCL-2 dependence. B, mitochondrial response to venetoclax was weakly related to days on venetoclax therapy. C and D, AML blast dependence on the anti-apoptotic protein BCL-X_L (using the HRK peptide) or the anti-apoptotic protein MCL-1 (using the MS1 peptide) negatively correlated with days on venetoclax. E, an index combining AML blast dependence on the anti-apoptotic protein BCL-X_L and the anti-apoptotic protein MCL-1 negatively correlated with days on venetoclax therapy. F, ROC analysis of dependence on MCL-1 or BCL-X_L, by arithmetically adding response to HRK peptide to that of MS1 peptide, was a perfect binary predictor of staying on venetoclax therapy more than 30 days (area under the curve of the ROC = 1.0). Correlations (r) and P-values based on one-tailed Spearman rank-order correlation tests.

Figure 3

Mechanism of action of venetoclax. Venetoclax acts as a specific inhibitor of BCL-2 and upon binding, releases pro-apoptotic proteins to induce apoptosis. BIM, BCL-2–like 11; BAX, BCL-2–associated X protein; BAK, BCL-2 antagonist/killer 1.