Dual targeting of TGF-β and PD-L1 via a bifunctional anti-PD-L1/TGF-βRII agent: status of preclinical and clinical advances

Abstract

Immunosuppressive entities in the tumor microenvironment (TME) remain a major impediment to immunotherapeutic approaches for a majority of patients with cancer. While the immunosuppressive role of transforming growth factor-β (TGF-β) in the TME is well known, clinical studies to date with anti-TGF-β agents have led to limited success. The bifunctional agent bintrafusp alfa (previously designated M7824) has been developed in an attempt to address this issue. Bintrafusp alfa consists of an IgG₁ targeting programmed death ligand 1 (PD-L1) moiety fused via peptide linkers to the extracellular domain of two TGF-β receptor II molecules designed to 'trap' TGF-β in the TME. This agent is able to bring the TGF-β trap to the TME via its anti-PD-L1 component, thus simultaneously attacking both the immunosuppressive PD-L1 and TGF-β entities. A number of preclinical studies have shown bintrafusp alfa capable of (1) preventing or reverting TGF-β-induced epithelial-mesenchymal transition in human carcinoma cells; this alteration in tumor cell plasticity was shown to render human tumor cells more susceptible to immune-mediated attack as well as to several chemotherapeutic agents; (2) altering the phenotype of natural killer and T cells, thus enhancing their cytolytic ability against tumor cells; (3) mediating enhanced lysis of human tumor cells via the antibody-dependent cell-mediated cytotoxicity mechanism; (4) reducing the suppressive activity of T_reg cells; (5) mediating antitumor activity in numerous preclinical models and (6) enhancing antitumor activity in combination with radiation, chemotherapy and several other immunotherapeutic agents. A phase I clinical trial demonstrated a safety profile similar to other programmed cell death protein 1 (PD-1)/PD-L1 checkpoint inhibitors, with objective and durable clinical responses. We summarize here preclinical and emerging clinical data in the use of this bispecific and potentially multifunctional agent.

Keywords: TGF-β; checkpoint inhibition; immunotherapy.

Figure 1

Bintrafusp alfa (designated in previous publications as M7824) prevents and reverts TGF-β-induced mesenchymal features in lung cancer models in vitro and in vivo. (A) Schema of the structure of M7824, a bifunctional protein consisting of a human IgG₁ anti-PD-L1 antibody fused via linkers to two TGF-β receptor II molecules. (B) M7824 blocks and reverts TGF-β1-induced mesenchymal features in A549 cells in vitro. Blockade: A549 cells were treated with TGF-β1 (2 ng/mL)±anti-PD-L1 or M7824 (200 ng/mL) for 72 hours to prevent EMT induction. Reversion: A549 cells were pretreated with TGF-β1 (2 ng/mL) for 72 hours to induce EMT, followed by 72 hours of treatment with TGF-β1±anti-PD-L1 or M7824 (200 ng/mL) to revert established EMT. Indicated EMT markers and loading control GAPDH were visualized by immunoblot. (C) HCC4006 lung cancer cells (4×10⁶ cells) were implanted s.c. in C.B-17 SCID mice (day −6). tumor-bearing mice were treated with i.p. injections of vehicle (HBSS), anti-PD-L1 (400 µg), or M7824 (492 µg; days 0, 2, 4, 6, 8, 10, 14). tumors were harvested, fixed and paraffin embedded (day 15), and stained via IHC for mesenchymal vimentin (shown is a representative tumor from each group (brown=vimentin, haematoxylin counterstain). (D) M7824 blocks and reverts chemoresistance conferred by TGF-β1 in lung PC-9 cells in vitro. PC-9 cells were left untreated (control) or treated with TGF-β1 (2 ng/mL) for 72 hours followed by treatment with indicated doses of chemotherapy for 96 hours. For the blockade experiments, PC-9 cells were simultaneously treated with TGF-β1 (2 ng/mL) and M7824 (200 ng/mL) for the entire duration of the assay; for the reversion experiments, cells were treated with TGF-β1 (2 ng/mL) for 72 hours, followed by TGF-β1+M7824 during the chemotherapy assay to revert a previously induced EMT. Cell viability was assayed by the Cell-Titer-Glo luminescent viability assay. Figures adapted from David et al. EMT, epithelial-mesenchymal transition; GAPDH, glyceraldehyde3-phosphate dehydrogenase; HBSS, Hank’s balanced salt solution; IHC, immunohistochemistry; i.p., intraperitoneal; PD-L1, programmed death ligand 1; s.c., subcutaneous; SCID, severe combined immunodeficiency; TGF-βRII, transforming growth factor-β receptor II.

Figure 2

Bintrafusp alfa (designated in previous publications as M7824) reduces TGF-β signaling and growth of murine tumors. (A) Murine carcinoma EMT6 cells (2.5×10⁵ cells) were implanted s.c. in the mammary fat pad of female Balb/c mice, or (B) murine colon cancer MC38 cells (5×10⁵ cells) were implanted s.c. in the right flank of female C57BL/6 mice. Treatment was administered i.p. on days 9, 11 and 13 post-tumor implantation with vehicle (PBS), a modified version of M7824 lacking a functional anti-PD-L1 moiety (MUT, 492 µg), or M7824 (492 µg) (EMT6, n=17 mice per group; MC38, n=10 mice per group). Primary tumor growth curves (left panels) show mean±SD; tumor weights on indicated days for individual mice (right panel) with mean±SD and compared by one-way ANOVA with Tukey’s multiple comparisons test. Tables below show the number and per cent of cured mice per group. (C) M7824 reduces plasma levels of TGF-β1 in EMT6-bearing mice. EMT6 tumors were implanted as in (A); mice were treated i.p. with PBS, MUT, or M7824 (492 µg; days 10, 12, 14). Platelet-poor plasma was collected 24 hours after last treatment. Total TGF-β1 levels in the plasma were normalized to the mean of the PBS control group. (D) M7824 reduces phosphorylated SMAD2 in the TME of EMT6-bearing mice. EMT6 tumors were implanted as in (A); mice were treated i.p. with vehicle, MUT, or M7824 (492 µg; days 17, 19, 21). Six hours after the last treatment, tumors were sectioned and snap-frozen and subsequently analyzed for total and phosphorylated SMAD2 by capillary Western blot. Phosphorylated SMAD2 relative to total SMAD2 levels were then normalized to the control group. Data in (C, D) combine two independent experiments (n=3–6 mice per experiment); graphs show mean±SD and analysis by one-way ANOVA with Tukey’s multiple comparisons test. (E) Combination of M7824 and Ad-TWIST vaccine improves survival in EMT6-bearing mice. EMT6 tumors were implanted as in (A); mice were treated i.p. with either PBS, M7824 (492 µg; days 10, 12, 14), Ad-TWIST (1×10¹⁰ virus particles, days 16, 23, 30), or M7824+Ad-TWIST. Survival curves show per cent survival and table below shows median OS in days. Data are representative of 2–3 independent experiments, n=10 mice. Statistical significance: *P<0.05, **P<0.005, ***P<0.001. Figures adapted from Knudson et al. ANOVA, analysis of variance; EMT, epithelial-mesenchymal transition; i.p., intraperitoneal; OS, overall survival; MUT, M7824 lacking a functional anti-PD-L1 moiety; PBS, phosphate buffered saline; PD-L1, programmed death ligand 1; p-SMAD2, phosphorylated SMAD2; s.c., subcutaneous; TGF-β, transforming growth factor-β.

Figure 3

Bintrafusp alfa (designated in previous publications as M7824) in combination with radiation and chemotherapy. (A–C) B cell-deficient mice (μMt⁻) were inoculated i.m. with 0.5×10⁶ MC38 cells (day −8) and treated with isotype control (133 µg i.v.; day 2), radiation (3.6 g/day; days 0–3), M7824 (55 µg i.v.; day 2), or radiation+M7824 (n=10 mice per group). (A) tumor growth curves. (B) tumor weight for individual mice on day 14. (C) ELISpot of IFN-γ–producing, p15E-responsive CD8⁺ T cells: CD8⁺ T cells were isolated from spleens (day 14; n=5) followed by culture with irradiated antigen-presenting cells derived from naïve solenocytes pulsed with KPSWFTTL (p15E) peptide or irrelevant peptide (OVA). (D–F) C57BL/6 mice were inoculated i.m. in the right thigh with 0.5×10⁶ MC38 cells (primary tumor) and s.c. in the left flank with 1×10⁶ MC38 cells (secondary tumor). Seven days following tumor inoculation, mice were treated with isotype control (400 µg; days 0, 2, 4), radiation (5 g/day; days 0–3), M7824 (164 µg; day 0) or M7824+radiation. (D) Radiation was only applied to the primary tumor site; tumor measurements were taken at the primary (E) and secondary (F) tumor sites. (G–I) µMt^- mice were inoculated s.c. with 1×10⁶ MC38 cells (day −7). Mice were treated with isotype (400 µg; days 3, 6, 9, 12, 17), M7824 (164 µg; days 3, 6, 9, 12, 17), oxaliplatin/5-fluorouracil (Ox/5-FU; 5 mg/kg i.p. and 60 mg/kg i.v.; day 0), or M7824 +Ox/5-FU (n=10). tumor volume (G) and tumor weights on day 18 (H). (I) ELISpot of IFN-γ–producing, p15E-responsive CD8⁺ T cells (day 18; n=5) as described in (C). Statistical significance: **P<0.01, ***P<0.001, ***P<0.0001. (B, H) tumor weights are presented for individual mice with mean indicated by lines and compared by unpaired t-test. (C, I) IFN-γ ELISpot data show means±SEM of three replicates analyzed by two-way ANOVA. From Lan et al. Reprinted with permission from the American Association for the Advancement of Science (AAAS). ANOVA, analysis of variance; IFN-γ, interferon-γ; i.m., intramuscular; i.p., intraperitoneal; i.v., intravenous; OVA, ovalbumin; s.c., subcutaneous.

Figure 4

Phase I trial of bintrafusp alfa shows antitumor efficacy in advanced solid tumors. (A) Spider plot showing changes in sum of the longest tumor diameter from baseline according to Response Evaluation Criteria in Solid tumors (V.1.1) for each patient who received bintrafusp alfa. Threshold for PR indicated by the dotted line at −30%. Threshold for PD indicated by the dotted line at 20%. (B) A metastatic cervical cancer patient treated with bintrafusp alfa shows durable complete CR. (C) Carcinoembryonic antigen (CEA) curve for patient with confirmed, ongoing, durable CR. CR, complete response; NE, not evaluable; PD, progressive disease; PR, partial response; SD, stable disease. Figures adapted from Strauss et al.