Packing a Punch against PD-L1

Targeting the programmed cell death protein 1 (PD-1) or its ligand, PD-L1, and the cytotoxic T lymphocyte–associated protein 4 (CTLA-4) checkpoints reverses mechanisms used by tumor cells to evade the immune system. Antibodies directed against these inhibitory immune checkpoints constitute first-line therapy in cancers such as cutaneous melanoma, non–small cell lung cancer, and head and neck carcinomas due to their ability to produce durable tumor regression (1). Based on their complementary roles in regulating immunity, the combination of anti–PD-1 and anti–CTLA-4 antibodies has been explored in clinical trials. Despite improving response rates over monotherapy, combining antibodies produces more immune-related adverse effects (irAE). Similarly, toxicity and tolerability have been major concerns for combining chemotherapy or targeted therapy with immune-checkpoint inhibitor antibodies. Furthermore, antibodies may have limited tissue distribution and, hence, require higher dosing. In an effort to maximize clinical outcomes and minimize drug toxicities, intermittent dosing and sequential scheduling of targeted therapies and immune therapies are currently being tested (2).

In this issue of Cancer Discovery, Koblish and colleagues characterize a small-molecule PD-L1 inhibitor, INCB086550 (3). In vitro, INCB086550 blocks the interaction between PD-1 and its ligand, PD-L1, partially through dimerization and internalization of PD-L1 and by binding to the remaining PD-L1 on the cell surface (Fig. 1). These mechanisms maximize PD-L1/PD-1 checkpoint inhibition, providing an effect comparable to monoclonal antibodies. Interestingly, a higher level of PD-L1 expression was associated with increased dimer formation and internalization, suggesting that INCB086550 could be a safe inhibitor that is an effective drug for patients presenting high levels of PD-L1.

In humanized and syngeneic mouse models, INCB086550 reduced tumor growth with efficacy comparable to the FDA-approved anti–PD-L1 antibody atezolizumab, with effects observed over a period of 2 to 3 weeks. The use of humanized mouse models is an important component of the study. CD34⁺ human cell-reconstituted NSG mice were utilized, permitting examination of tumor cell–immune cell interactions in a human MDA-MB-231 xenograft model either expressing or deficient in PD-L1. The production of a robust immune response with a broad repertoire of immune cell populations from the reconstituted CD34⁺ cells is a major consideration in these types of studies. Humanized mouse models are advantageous, since they can permit testing of drugs in patient-derived xenograft models with human leukocyte antigen–matched immune cells. Thus, such models are likely to be key in testing response and resistance to new immune-checkpoint inhibitors (4). As future directions in both syngeneic and humanized mice, extending the length of the treatment and testing INCB086550 in combination with anti–CTLA-4 or next-generation immune-checkpoint inhibitors such as anti-LAG3 antibodies would expand on the full potential of this inhibitor.

Effects on tumor growth were associated with INCB086550 tumor distribution and increased cytotoxic T-cell infiltration. Cell-based assays showed that INCB086550 stimulates proinflammatory cytokine production in primary human immune cells, similar to effects of the clinically approved anti–PD-L1 antibody. Additionally, bulk RNA sequencing performed on humanized mouse-derived tumors showed upregulation of a gene signature that is known to be upregulated by the anti–PD-1 antibody nivolumab (5), albeit not in all tumors. Moving forward, analysis of the activation status of intratumoral T cells could help to further elucidate the mechanism of action of INCB086550. Additionally, single-cell RNA sequencing could reveal the remodeling occurring in the tumor immune microenvironment (TIME) following the administration of INCB086550. The composition of tumor-infiltrating immune cells is dynamic and evolves during immune-checkpoint inhibitor treatment, with differential effects between responders and nonresponders (6). Specifically, increases in the number of CD8⁺ T cells, natural killer cells, B cells, and tertiary lymphoid structures, as well as a decrease in macrophages, were observed in the tumors of responders to anti–PD-1 antibodies while on treatment (5, 7). A direct comparison of the TIME composition, as well as changes in immune cell subpopulations following administration of INCB086550 versus anti–PD-L1 antibodies, would further highlight the similarities and differences between a small-molecule and antibody approach to inhibit PD-L1.

Preliminary toxicity experiments suggest that INCB086550 is well tolerated across different mouse models and species. The low toxicity and fewer irAEs of this inhibitor, in combination with its tumor penetration and high efficacy in the presence of elevated levels of PD-L1, are key features that may give it a therapeutic advantage in the clinic. The effects of INCB086550 are being evaluated in solid tumors (NCT04629339). Ex vivo blood analysis from patients enrolled in the phase I clinical trial suggests that oral administration of INCB086550 induces dose-related pharmacodynamic immune activation comparable with that reported for anti–PD-L1 and anti–PD-1 antibodies. Furthermore, tumor size reduction as indicated by CT scan has been shown in one patient following a 24-week dosing period. Durable efficacy and tolerability in larger numbers of patients compared with anti–PD-1/PD-L1 antibodies will be the ultimate determinant of the success of INCB086550. Considering the high incidence of resistance following immune-checkpoint inhibitor therapy, it also will be important to address the efficacy of INCB086550 combined with other checkpoint inhibitors such as anti–CTLA-4 or anti-LAG3 antibodies.

This study on INCB086550 underscores the strong interest around the discovery of small-molecule inhibitors to target immune checkpoints, especially given their ability to inhibit tumor growth with favorable biosafety in comparison with monoclonal antibodies (8). Several synthetic small molecules from Bristol Myers Squibb (e.g., BMS1166 and BMS202) function via dimerization of PD-L1 and preclusion of PD-1 binding (9). CA-170 from Curis, Inc. is a small-molecule PD-L1 and VISTA inhibitor that exhibits promising tumor suppression effects associated with the activation of tumor-infiltrating T cells (10). However, most small-molecule inhibitors against immune checkpoints are still in the early stages of development and testing in preclinical models.

In summary, although anti–PD-1/PD-L1 monoclonal antibodies have achieved remarkable success in patients with cancer, there is a lot of interest around the discovery of small-molecule inhibitors to target immune checkpoints. Likely advantages are the short half-life, greater tumor penetration, increased oral bioavailability, and lower irAEs. The introduction of small-molecule immune-checkpoint inhibitors introduces new options for monotherapy targeting, combinational approaches, and sequencing regimens.