Inhibition of AKT signaling uncouples T cell differentiation from expansion for receptor-engineered adoptive immunotherapy
Christopher A. Klebanoff, … , Steven A. Feldman, Nicholas P. Restifo
Christopher A. Klebanoff, … , Steven A. Feldman, Nicholas P. Restifo
Published December 7, 2017
Citation Information: JCI Insight. 2017;2(23):e95103. https://doi.org/10.1172/jci.insight.95103.
View: Text | PDF
Research Article Immunology Oncology

Inhibition of AKT signaling uncouples T cell differentiation from expansion for receptor-engineered adoptive immunotherapy

Abstract

Adoptive immunotherapies using T cells genetically redirected with a chimeric antigen receptor (CAR) or T cell receptor (TCR) are entering mainstream clinical practice. Despite encouraging results, some patients do not respond to current therapies. In part, this phenomenon has been associated with infusion of reduced numbers of early memory T cells. Herein, we report that AKT signaling inhibition is compatible with CAR and TCR retroviral transduction of human T cells while promoting a CD62L-expressing central memory phenotype. Critically, this intervention did not compromise cell yield. Mechanistically, disruption of AKT signaling preserved MAPK activation and promoted the intranuclear localization of FOXO1, a transcriptional regulator of T cell memory. Consequently, AKT signaling inhibition synchronized the transcriptional profile for FOXO1-dependent target genes across multiple donors. Expression of an AKT-resistant FOXO1 mutant phenocopied the influence of AKT signaling inhibition, while addition of AKT signaling inhibition to T cells expressing mutant FOXO1 failed to further augment the frequency of CD62L-expressing cells. Finally, treatment of established B cell acute lymphoblastic leukemia was superior using anti-CD19 CAR–modified T cells transduced and expanded in the presence of an AKT inhibitor compared with conventionally grown T cells. Thus, inhibition of signaling along the PI3K/AKT axis represents a generalizable strategy to generate large numbers of receptor-modified T cells with an early memory phenotype and superior antitumor efficacy.

Authors

Christopher A. Klebanoff, Joseph G. Crompton, Anthony J. Leonardi, Tori N. Yamamoto, Smita S. Chandran, Robert L. Eil, Madhusudhanan Sukumar, Suman K. Vodnala, Jinhui Hu, Yun Ji, David Clever, Mary A. Black, Devikala Gurusamy, Michael J. Kruhlak, Ping Jin, David F. Stroncek, Luca Gattinoni, Steven A. Feldman, Nicholas P. Restifo

×

Figure 1

Pharmacologic inhibition of AKT signaling permits expansion of CD62L-expressing receptor-engineered human peripheral blood T cells.

Options: View larger image (or click on image) Download as PowerPoint
Pharmacologic inhibition of AKT signaling permits expansion of CD62L-exp...
(A) Schema for the anti-CD3 (50 ng ml–1) activation, retroviral transduction (RV Td), and expansion of human peripheral blood T lymphocytes (PBL) in the continuous presence of IL-2 (300 IU ml–1) and AKT inhibitor VIII (AKTi; 1 μM) or vehicle control (Veh). (B) Representative phosphoflow cytometry plots and (C) graphical summary of the time-dependent phosphorylation of kinases involved AKT/mTOR or MAPK signaling in PBL expanded in the presence or absence of AKTi immediately prior to and following stimulation with an anti-CD3 antibody. Results from 1 of 2 representative experiments are displayed. (D) Fold expansion and (E) transduction efficiency of unfractionated PBL genetically engineered with a second-generation 28z anti-CD19 chimeric antigen receptor (CAR) following ex vivo expansion over 10d in the continuous presence IL-2 and AKTi or Veh. Pooled results from 6 independent donors are shown after gating on viable, transduced CD3+CD4+ and CD3+CD8+ cells. (F) Representative FACS plot and (G) graphical summary of CD62L expression on CAR-modified PBL expanded for 10d in AKTi or Veh control. Results shown in panels D–G are based on patient-derived samples, while results in panels B and C used healthy donor (HD) T cells. Data in panels C–E and G are presented as mean ± SEM with n = 3, n = 6, n = 6, and n = 3 per condition, respectively. All statistical comparisons were performed using an unpaired 2-tailed Student’s t test. ***P < 0.001; **P < 0.01; *P < 0.05.