Pre-Clinical Evaluation of B7-H3-Specific Chimeric Antigen Receptor T-Cells for the Treatment of Acute Myeloid Leukemia

Background: The development of safe and effective chimeric antigen receptor T-cell (CAR-T) therapy for acute myeloid leukemia (AML) remains an elusive goal. Whereas CD19-directed CAR-T therapies have shown great promise for the treatment of B-cell malignancies, the identification of AML-associated surface antigens that can be safely and effectively targeted by CAR T-cells has remained a challenge. Because most AML-associated surface antigens are also expressed on normal myeloid progenitors, the potential for unacceptable hematopoietic toxicity has been a major limitation. The identification of cell surface antigens that are absent on all normal myeloid progenitors and yet expressed on all subtypes of AML is not likely. On the other hand, it seems plausible that some antigens not detected on early myeloid lineage cells may be preferentially overexpressed in certain AML subtypes. We have identified B7-H3 as one such candidate. B7-H3 (B7-homolog 3, or CD276) is a coreceptor belonging to the B7 family of immune checkpoint molecules. B7-H3 protein expression in normal tissues is largely restricted to certain antigen-presenting cells. In multiple human cancers, however, B7-H3 protein is significantly overexpressed. This includes a substantial subset of AML, and B7-H3 expression appears to be higher in AML with a monocytic immunophenotype. Furthermore, B7-H3 on tumor cells, and on myeloid-derived suppressor cells in the tumor microenvironment, is likely to play an immunosuppressive role, and may drive immune escape in multiple cancer types. This suggests that targeting B7-H3 could also enhance anti-tumor adaptive immune responses. We therefore hypothesized that B7-H3-specific CAR-Ts (B7-H3.CARs) could be effective in eliminating B7-H3-expressing AML cells and would not cause unacceptable hematopoietic toxicity.

Methods and Results: We obtained bone marrow aspirates from patients with monocytic/myelomonocytic AML (n=10) and demonstrated surface expression of B7-H3 on a median of 62.5% (range 27.8 to 98.5) of primary AML blasts. We also showed that B7-H3 is highly expressed on monocytic/myelomonocytic AML cell lines (THP1, U937, OCI-AML2, OCI-AML3), and that B7-H3 expression compares favorably to that of other previously identified candidate antigens for AML-directed CAR-T therapy. Next, we generated B7-H3.CARs via retroviral transduction of CD3/CD28-activated T-cells, followed by expansion in vitro with interleukin- (IL) 7 and IL-15. When B7-H3.CARs (n=3-5 donors) were cocultured with B7-H3-positive AML cell lines (listed above) and with primary AML blasts (n=10 patients), B7-H3.CARs proliferated, released high amounts of IL-2 and interferon-γ, and demonstrated efficient B7-H3-specific cytotoxicity. Autologous B7-H3.CARs also demonstrated significant cytotoxicity against primary AML blasts (n=4). Additionally, B7-H3.CARs controlled tumor cell proliferation and prolonged survival in xenograft mouse models of disseminated AML using OCI-AML2 (p=0.0025, n=5 mice per group) and THP1 (p<0.0001, n=10 mice per group).

Next, we showed that B7-H3 is not significantly expressed on hematopoietic stem cells or myeloid progenitor cell populations in normal human bone marrow samples (n=2). We also evaluated the effects of B7-H3.CARs (n=4 donors) on normal hematopoietic stem cells via in vitro colony formation assays using umbilical cord-blood (n=4 donors) derived CD34+ cells, and showed that B7-H3.CARs did not significantly inhibit the formation of myeloid progenitor colonies. We then showed in a humanized mouse model (using fetal liver-derived hematopoietic stem cells) that B7-H3.CARs did not lead to significant reductions in the populations of circulating CD45/CD14-positive or CD45/CD33-positive cells.

Conclusions: B7-H3 is expressed on a significant proportion of AML blasts from patients with monocytic AML. Adoptive transfer of B7-H3.CARs could be an effective treatment option for patients with B7-H3-positive AML, since i) we have previously demonstrated limited expression of B7-H3 in normal tissues, and ii) the present results show that B7-H3.CARs are unlikely to cause significant hematopoietic toxicity. Given variable expression of B7-H3 in AML, however, it may be necessary to develop a dual-targeting approach, combining B7-H3 with a second target AML-associated surface antigen.

View largeDownload slide

View largeDownload slide

Close modal