Antitumor efficacy of combined CTLA4/PD-1 blockade without intestinal inflammation is achieved by elimination of FcγR interactions

Abstract

Background: Programmed cell death protein 1 (PD-1) and CTLA4 combination blockade enhances clinical efficacy in melanoma compared with targeting either checkpoint alone; however, clinical response improvement is coupled with increased risk of developing immune-related adverse events (irAE). Delineating the mechanisms of checkpoint blockade-mediated irAE has been hampered by the lack of animal models that replicate these clinical events.

Methods: We have developed a mouse model of checkpoint blockade-mediated enterocolitis via prolonged administration of an Fc-competent anti-CTLA4 antibody.

Results: Sustained treatment with Fc-effector, but not Fc-mutant or Fc-null, anti-CTLA4 antagonist for 7 weeks resulted in enterocolitis. Moreover, combining Fc-null or Fc-mutant CTLA4 antagonists with PD-1 blockade results in potent antitumor combination efficacy indicating that Fc-effector function is not required for combination benefit.

Conclusion: These data suggest that using CTLA4 antagonists with no Fc-effector function can mitigate gut inflammation associated with anti-CTLA4 antibody therapy yet retain potent antitumor activity in combination with PD-1 blockade.

Keywords: CTLA-4 antigen; immunotherapy; inflammation.

Figure 1

Anti-CTLA4-mediated intestinal inflammation is Fc-dependent and more severe in combination with PD-1 blockade. Balb/c mice were treated twice a week with antibodies as indicated for 50 days. (A) Mean body weights over the time. (B) Intestinal permeability assessed by fluorescence measurement of Fluorescein isothiocyanate (FITC)-dextran in the serum at day 50. a.u. Arbitrary units. (C) Intestinal inflammation score and (D) representative photomicrographs of H&E stained histological section of the small intestine at day 50 (n=8 mice/ group). Scale bars represent 100 µm. (E) RT-qPCR gene expression in small intestine at day 50 post treatment. Results are shown as mean log2 fold change relative to isotype control treated mice. Results are representative of two independent experiments (n=8 mice per group). ***P<0.001, ****p<0.0001 (one way analysis of variance test). Error bar ±SEM.

Figure 2

T cells and macrophages are key drivers of anti-PD1/anti-CTLA4-mediated intestinal inflammation. (A) Log fold change gene expression in colon comparing isotype control to anti-PD1/anti-CTLA4 (Fc-effector) treated mice at day 35 post treatment. (B–D) Balb/c mice were treated twice a week with antibodies as indicated for 34 days. (B) Representative photomicrographs of H&E stained histological section of the colon and (C) colon inflammation score. Arrows indicate immune cell infiltration and loss of goblet cells, indicative of inflammation. Scale bars represent 100 µm. (D) Fold change gene expression in colons comparing naïve mice to day 34 post-treatment groups. Log2 mean fold change values are represented. (E) Small intestine inflammation score. (F) Log2 fold change gene expression in small intestine comparing naïve mice to day 34 post-treatment groups. Log2 mean fold change values are represented. Results are representative of 2–3 independent experiments (n=8–16 mice per group). ns: not significant *p<0.05, **p<0.01. ***p<0.001, ****p<0.0001 (unpaired t-test for panel A and one-way analysis of variance test for panel C) and E). Error bar ±SEM.

Figure 3

Fc-effector anti-CTLA4 antibody does not deplete colon lamina Tregs. (A) Mean fluorescence intensity (MFI) of intracellular CTLA4 in Foxp3⁺ Tregs from indicated organs and peripheral blood mononuclear cell (PBMC) from CT26 tumor-bearing mice when tumors were ~100 mm³. (B) Proportion of colon lamina propria and CT26 tumor-infiltrating Foxp3⁺ Treg 24 hours after treatment as indicated. Results are representative of 2–3 independent experiments (n=6–12 mice per group) ns: not significant, *p<0.05, ***p<0.001, ****p<0.0001 (one way analysis of variance test).

Figure 4

Fc-effector anti-CTLA4 impaired Treg-mediated suppression of colitis. Splenic CD45Rb^high naïve T cells and CD25 +Tregs were transferred into CB17-SCID (Severe combined immunodeficiency) recipient mice and treated with Fc-effector anti-CTLA4 antibody or Fc-null CTLA4 VHH antagonist as indicated. (A) Mouse weight over the time of the experiment. (B) Photomicrographs of H&E stained histological sections of the colon (left panel) and pathology score (right panel) at day 47 (n=14–18 mice per group). Scale bars represent 50 µm. (C) Fold change gene expression in whole colon comparing to day 47 post naïve T cell transfer (n=6 mice per group) treated groups. Data are representative of two independent experiments. ns: not significant ****p<0.0001 (One way analysis of variance test). Error bar ±SEM.

Figure 5

CTLA4 antagonists have a potent antitumoral efficacy only in combination with anti-PD-1. (A) CT26 tumor-bearing mice received the indicated antibody (at 20 mg/kg) or VHH (30 mg/kg) q4d×5 when tumors reached an average size of 100 mm³ (range 78–125 mm³). (B) Tumor volumes at day 16. The combined results from three independent experiments (n=11–21 mice per group) are shown. (C) Gene expression in whole tumor comparing isotype control treated mice to day 8 post-treatment groups Mean log2 fold change values re-represented. Results are representative of two independent experiments (n=5 mice per group). *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001 (unpaired t-test). Error bar ±SEM.