The Impact of the Bone Marrow Microenvironment on T Cell Redirection Therapeutics

Redirecting T cells to specifically target and kill malignant cells has been validated as an effective anti-cancer strategy in the clinic with the approval of CD19xCD3 BiTE (Blincyto) for acute lymphoblastic leukemia. However, the immunosuppressive nature of the tumor microenvironment potentially poses a significant hurdle to T cell therapies. For instance, the bone marrow (BM) niche is appreciated to be a site of immune privilege at steady state to allow for normal hematopoiesis and immune cell generation. Additionally, in hematological malignancies, the BM niche is protective to leukemic stem cells, a phenomenon that has minimized the efficacy of several anti-cancer drugs including chemotherapy, targeted small molecule inhibitors and antibody based therapies.

In this study, we investigated the impact of the BM microenvironment on T cell redirection. Using antibodies made with the Genmab DuoBody^® technology targeting specific tumor antigens (CD123 and BCMA) and CD3, we observed that co-culture of acute myeloid leukemia (AML; KG-1, MOLM-13 and OCI-AML5) or multiple myeloma (MM; H929 and RPMI8226) cell lines with bone marrow stromal cell lines (HS-5 and HS-27a) significantly protected leukemic cells from DuoBody^®-T cell mediated lysis in vitro. Specifically, co-culture of bone marrow stromal cells in killing assays led to a 20-50% decrease in the maximum observed cytotoxicity and a 3-10 fold weaker EC₅₀, reflecting an impact on both the efficacy and potency of bispecific DuoBody® antibodies. Similar results were also observed with primary stromal cells obtained from healthy donors. Furthermore, presence of stromal cells in a humanized xenograft AML model attenuated tumor growth inhibition (TGI) observed with DuoBody^® treatment (78% TGI with MOLM-13 vs 15% TGI with MOLM-13+HS-5). Impaired TGI correlated with reduced T cell activation (7 fold decrease in CD25 upregulation) and production of granzyme B (8 fold reduction), providing one potential mechanism to explain loss of activity of the DuoBody^® antibody. In vitro trans-well redirection assays revealed that cell-cell contact with stromal cells was crucial for reduced T cell activation and target cell survival relative to controls. Additionally, leukemic cells not killed by T cells were observed to preferentially cluster around stromal cells. We propose that target cells can evade T cell death by a stromal cell dependent mechanism involving activation of multiple pro-survival and anti-apoptotic pathways in leukemic cells in addition to suppressed activation of T cells. We are currently studying pathways mediating the cross-talk between cancer, immune and stromal cells. A better understanding of the mechanisms underlying the protective and immunosuppressive nature of the BM microenvironment will be instrumental to the design of more effective CD3 redirected therapeutics or novel combinatorial regimens for robust anti-cancer responses.