AUTHORS
Winnie Ida Nguy, Joanne Seoyoung Baek2, Elliot Goepfert-Waterman3, Feng Shen, MD4, Brandon William Simone, PhD5, Saar Gill, MD, PhD6, Anand Bhagwat, MD7
AFFILIATIONS
1 Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, United States
2 University of Pennsylvania, Philadelphia, Pennsylvania, United States
3 University of Pennsylvania, Philadelphia, Pennsylvania, United States
4 University of Pennsylvania, Philadelphia, Pennsylvania, United States
5 University of Pennsylvania, Philadelphia, Pennsylvania, United States
6 University of Pennsylvania, Philadelphia, Pennsylvania, United States, University of Pennsylvania, Philadelphia, Pennsylvania, United States
7 Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, United States, University of Pennsylvania, Philadelphia, Pennsylvania, United States
ABSTRACT
BACKGROUND:
Despite the success of chimeric antigen receptor (CAR)-T cell therapy for hematologic malignancies, relapse remains common in B-cell acute lymphoblastic leukemia (B-ALL) and responses in acute myeloid leukemia (AML) are limited. Venetoclax (VEN), a selective BCL-2 inhibitor, could be an attractive partner for T cell–based immunotherapies but is generally considered incompatible because resting lymphocytes are sensitive to VEN. Previous approaches to circumvent this perceived hurdle have relied on transgenic overexpression of anti-apoptotic proteins in the BCL-2 family. However, since some lymphoid malignancies are associated with unregulated BCL-2 overexpression, this approach raises safety concerns. Here, we sought to determine whether physiologic changes accompanying antigen-driven T-cell activation can open a therapeutic window to enable combination of VEN with T cell immunotherapy.
METHODS AND RESULTS:
To test the hypothesis that activation of T cells by leukemia would render them resistant to apoptosis and to VEN, we activated T cells with various stimuli, including CD3/CD28 beads and PMA/ionomycin to characterize apoptosis priming of T cells and their sensitivity to VEN. We found that resting healthy donor T cells were sensitive to VEN (mean IC50 = 229nM, n = 3), whereas activation by any method conferred marked resistance to VEN (mean IC50 > 2000nM, n = 3). Similarly, activation of CD123 directed CAR T cells (CART123) by CD3/CD28 beads or by CD123+ leukemias conferred resistance to VEN (mean IC50 = 2207nM, n = 3) compared to resting CART123 cells (mean IC50 = 155nM, n = 3).
Mechanistically, we found that T cell activation is associated with increased expression of BCL-xL (mean fold increase in MFIbead activated = 3.3, MFItarget activated = 4.7, n = 3 each), but not of BCL-2 or MCL-1. Importantly, this effect is transient and is subject to physiologic regulation with return to baseline BCL-xL expression within a week of stimulation. Intracellular BH3 profiling confirmed that activation of T cells reduces their overall apoptotic priming and shifts their apoptosis susceptibility toward combined BCL-2/BCL-xL inhibition. The BCL-2/BCL-xL redundancy can be overcome in activated T cells with dual inhibitors such as navitoclax (NAV). We found that activated CART123 are more sensitive to NAV than their resting counterparts (mean IC50bead activated = 64 nM, mean IC50target activated = 83 nM, mean IC50resting = 126 nM, n = 3).
Consistent with this mechanism, VEN does not impair viability or expansion of activated CAR T cells or blinatumomab (blina)-stimulated healthy donor T cells. Instead, VEN enhances proliferation of T cells at concentrations up to 1000nM (27.75% percent divided vs. 20.90%, p=0.0139). VEN also enhances killing of AML and ALL when combined with CAR T cells or bispecific T cell engagers, including against VEN-resistant leukemia cell lines. Calculation of Bliss scores reveals synergy between T cell therapies and VEN (AML Bliss score 790; B-ALL Bliss score 11, 90% CI ±2.1).
To assess whether continuous VEN exposure improves long-term anti-leukemia T cell fitness, we performed chronic stimulation assays with T cells, blina, and B-ALL cells. Across three T cell donors, high-dimensional spectral flow cytometry revealed reduced frequency of exhausted T cells in VEN-treated conditions.
Finally, we demonstrate that CAR T cells targeting CD123 (CART-123) on primary patient derived AML xenografts in the NSG-SGM3 mouse strain could be safely combined with VEN. CART-123 plus VEN resulted in reduced tumor burden compared with either monotherapy alone. Crucially, addition of VEN to CART-123 had no impact on viable CART-123 cells measured in murine bone marrow aspirates 7 weeks after infusion, while it did eliminate untransduced (non-activated) T cells.
SIGNIFCANCE:
Activated T cells physiologically develop relative resistance to VEN through upregulation of BCL-XL, permitting co-administration of VEN and anti-leukemia T cell therapies without impairment of T cell viability or function. Combination treatment with VEN enhanced anti-leukemic activity in both CAR-T and blina-based models of leukemia treatment, including in VEN-resistant leukemias, and demonstrated improved anti-tumor activity against primary AML in vivo. These findings support implementation of VEN as a combinatorial strategy to augment T cell-directed immunotherapies in blood cancers in patients with relapsed and refractory disease.
DISCLOSURES
Winnie Ida Nguy: No relevant conflicts of interest to declare.