Direct Integration of Anti-CD19 Chimeric-Antigen-Receptor (CAR) T Cells Early after Post-Transplantation Cyclophosphamide (PTCy) in Murine MHC-Haploidentical Hematopoietic Cell Transplantation (HCT) Exerts Anti-Tumor Effects without Causing Graft-Versus-Host Disease (GVHD)

Abstract

Anti-CD19 CAR T cells can induce remissions in many patients. Yet, relapses can occur and so some patients may proceed to HCT either prior to or after relapse to try to achieve long-term cure. Even so, relapse limits the success of HCT. We recently showed that the suppressive mechanisms induced immediately after PTCy prevent new donor T cells from causing GVHD. Thus, we hypothesized that donor anti-CD19 CAR T cells given early after PTCy would not cause GVHD but may retain an anti-tumor effect; the premise is that the CAR T cells would induce an initial deep remission, while the polyclonal graft-versus-tumor (GVT) T-cell response would ensure long-term cure. To test this approach, we modified our T-cell-replete murine MHC-haploidentical HCT model (B6C3F1→B6D2F1) by giving 1 × 10⁶ E2a-PBx leukemia cells 7 days prior to lethal irradiation and HCT [Figure 1A]. Importantly, E2a-PBx, as a pre-B-cell ALL cell line on a B6 background, is syngeneic with host and donor; thus, in this model only anti-CD19 CAR T cells can treat the leukemia (there is no allo response against it). T-cell-enriched B6C3F1 donor splenocytes were stimulated with anti-CD3/CD28 beads, transduced with an anti-CD19 CAR vector with a CD28 costimulatory domain, and infused at 1 × 10⁶ transduced T cells per mouse on day +5 (24 hours after PTCy). Mice receiving leukemia, HCT, and PTCy all died of leukemia when receiving non-transduced in vitro expanded T cells, but cleared leukemia when receiving CAR T cells [Figure 2A]; this effect occurred without clinical or histopathologic GVHD or toxicity. Leukemia relapse did occur in some CAR-treated mice and was more frequent in mice treated with PTCy (∼25% of mice). Relapse events after PTCy were associated with CD19+ extramedullary tumors but partial loss of CD19 in hematopoietic organs. Measurable CAR T cells were found but had increased expression of co-inhibitory molecules, suggesting that CAR T-cell exhaustion and leukemia antigen loss may be two mechanisms of escape. Attempts to assess the impact of having both CAR and allogeneic T cells able to treat the leukemia (to mirror the intended clinical use) in a C3H→B6D2F1 HCT model were limited by permanent clearance of leukemia by the allogeneic response alone. We also explored the impact of PTCy on GVT immunity by giving CAR T cells on day 0 prior to PTCy [Figure 1B]; leukemia was cleared by mice despite PTCy and did not relapse [Figure 2B]. CAR T cells persisted after PTCy and actually increased in numbers in the bone marrow from days +3 to +7. There was no increase of GVHD in CAR-treated mice clinically or histopathologically. Our data suggest that CAR T cells can be safely and effectively administered either prior to or after PTCy, permitting an anti-tumor response without GVHD. Furthermore, these studies show that tumor-specific T cells survive PTCy, consistent with our recent findings that alloreactive T cells persist after PTCy.