Abstract 2834: BH3 profiling predicts clinical response in a phase II clinical trial of ABT-199 (GDC-0199) in acute myeloid leukemia

Abstract

Abstract Background: ABT-199 is a potent and selective small molecule inhibitor of the anti-apoptotic protein BCL-2, currently in a Phase II clinical trial for relapsed/refractory acute myeloid leukemia (AML). Results from the trial indicate that AML patients have a heterogeneous response to ABT-199 used as a single agent. Therefore, it would be beneficial to prospectively identify patients who will be most sensitive to ABT-199 treatment. We have developed BH3 profiling, which is a rapid, functional assay that measures dependence on any of the anti-apoptotic BCL-2 family proteins for an individual sample. We have previously shown in many cancer cell lines and primary samples that BH3 profiling predicts in vitro sensitivity to ABT-199. Here, we test for correlation between clinical response to ABT-199 and BH3 profiling. Methods: Twelve pre-treatment AML bone marrow or peripheral blood samples were BH3 profiled using a panel of peptides which included BAD (measures BCL-2, BCL-XL, BCL-W dependence), MS1 (MCL-1 dependence), HRK (BCL-XL dependence) and mitochondrial ABT-199 (BCL-2 dependence). In addition, three samples obtained after 4 wks on study were also BH3 profiled to assess the in vivo mechanism of action of ABT-199. Results: BH3 profiling predicted clinical response in several ways. BCL-2 dependence predicted reduction of blast count at 4 wks, with an AUC of the ROC of 0.86. Myeloblast dependence on the anti-apoptotic protein MCL-1 or BCL-XL alone predicted low blast count reduction at 4 wks, with an AUC of the ROC of 0.80 and 0.97 respectively. Since dependence on MCL-1 or BCL-XL both predict poor clinical performance, we asked whether we could combine the information to make a superior predictor by arithmetically adding response to HRK to that to MS1. We found that this metric was a perfect binary predictor of blast reduction at 4 wks (AUC = 1.0). Additionally, we asked whether ABT-199 was working on-target, at the mitochondria in vivo. We BH3 profiled bone marrow aspirates collected while patients were on treatment and analyzed the response to the MS1, HRK and BIM peptides. We found that the AML blasts remaining following 4 wks of treatment have an increased response to the HRK, MS1 and BIM peptides compared to pre-treatment samples. This suggests that ABT-199 is working on target as it displaces pro-apoptotic proteins from BCL-2 to BCL-XL and MCL-1 and diminishes the overall anti-apoptotic reserve at the mitochondrion. Conclusions: The results of this study indicate that BH3 profiling can be used to exclude AML patients destined to respond poorly and enrich for those patients destined to respond well to single agent ABT-199. Since we have previously found that the level of mitochondrial apoptotic priming predicts clinical response to conventional chemotherapy, we speculate that ABT-199 could facilitate myeloblast killing via conventional chemotherapy in vivo. Citation Format: Leah Hogdal, Brenda Chyla, Evelyn McKeegan, Joel Leverson, Jalaja Potluri, Nancy Falotico, Justin Ricker, Rod Humerickhouse, Mack Mabry, Glenna Foight, Amy Keating, Ilene Galinsky, Richard Stone, Daniel DeAngelo, Marina Konopleva, Anthony Letai. BH3 profiling predicts clinical response in a phase II clinical trial of ABT-199 (GDC-0199) in acute myeloid leukemia. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2834. doi:10.1158/1538-7445.AM2015-2834