Local CTLA4 blockade effectively restrains experimental pancreatic adenocarcinoma growth in vivo

Abstract

Antibody-mediated blockade of CTLA4 has been shown to be effective in treating a select group of patients with late-stage melanoma. The precise mechanism underlying the clinical activity of CTLA4 immunotherapy is poorly understood, although recent experimental findings indicate that antibody-mediated depletion of regulatory T cells (Tregs) in the tumor microenvironment plays a key role in efficacious antitumor responses. In the current study, we used an experimental model of pancreatic adenocarcinoma to compare the antitumor efficacy of peritumoral low-dose anti-CTLA4 monoclonal antibody (mAb) administration to that of a commonly utilized systemic high-dose anti-CTLA4 regimen. We selected pancreatic adenocarcinoma as it presents a particular challenge to clinicians due to its aggressive behavior, metastatic spread and limited treatment options. Furthermore, Fc gamma receptor (FcγR)-dense myeloid cells commonly infiltrate pancreatic tumors, such that these tumor types exhibit increased susceptibility to CTLA4 antibody-targeted Treg depletion via antibody-dependent cell-mediated cytotoxicity (ADCC). Locally administered anti-CTLA4 mAb effectively reduced tumor growth at a low dose and no additional anti-tumor effects were apparent when increasing the dose or number of injections. No significant difference in overall survival was seen when comparing locally administered low-dose with standard systemic high-dose CTLA4 blockade therapy, and both delivery routes led to increased tumor-infiltrating effector T cells and reduced Treg cells. As opposed to low-dose peritumoral treatment, high-dose systemic therapy stimulated the accumulation of Tregs in secondary lymphoid organs, an effect that could potentially counteract the antitumor immunotherapeutic benefit of CTLA4 blockade. Our study confirms previous findings that local administration of low-dose anti-CTLA4 antibody generates sustained antitumor effects and provides rationale to devise ultrasound-guided intratumoral anti-CTLA4 antibody injection regimens to treat patients with pancreatic adenocarcinoma and other types of solid tumors. In support, clinical relevancy could include reduced immune-related adverse events by limiting systemic antibody spread to immune cell-dense organs.

Keywords: anti-CTLA4; checkpoint blockade; immunotherapy; local therapy; pancreatic cancer.

Figure 1. Localized low-dose anti-CTLA4 antibody therapy is efficacious. (A–C) Mice (n = 10–11 per group) were inoculated subcutaneously with 2.5 × 10⁵ Panc02 cells and treated with anti-CTLA4 blocking antibody by peritumoral injections either 3 times (day 5, day 8, and day 11) or 6 times (day 5, day 8, day 11, day 14, day 17, and day 20), as indicated. (A and B) To determine the optimum dosage and injection frequency of locally administered anti-CTLA4 monoclonal antibody (mAb), a dose-response experiment was performed by 3 (A) 30 μg, 60 μg, and 90 μg or 6 (B) 30 μg peritumoral injections of anti-CTLA4 mAb. Tumor growth was measured with caliper and calculated by 4/3π*a(radius of length)*b(radius of width)*c(radius of depth). Lines indicate individual animals. Grey bar indicates d30. (C) Kaplan–Meier survival curve from data presented in (A and B). Cumulative results from 2 independent experiments. Statistical analysis of survival was performed by log-rank test with ***P < 0.001. CR, complete responder; ns, not significant

Figure 2. Antitumor efficacy and circulating serum levels of locally delivered vs. systemically administered anti-CTLA4 blocking antibody. (A–C) Mice (n = 12 per group) were inoculated subcutaneously with 2.5 × 10⁵ Panc02 cells and treated with anti-CTLA4 blocking antibody by either peritumoral injections of 30 μg or intraperitoneal injection of 200 μg anti-CTLA4 monoclonal antibody or a phosphate-buffered saline (PBS) control day 5, day 8, and day 11. (A) Tumor growth was measured with caliper and calculated by 4/3π*a(radius of length)*b(radius of width)*c(radius of depth). Lines represent individual mice. Grey bar indicates d40. (B) Kaplan–Meier survival curve of mice in (A). Statistical analyses were performed by log-rank test with **P < 0.01, ***P < 0.001. (C) Tumor growth of naïve mice and complete responders from 2 independent experiments rechallenged with 2.5 × 10⁵ Panc02 in the contralateral flank (n = 4-8). Averaged tumor volumes (measured as in A) are shown per group. (D) Naïve mice were locally (30 μg s.c.) or systemically (200 μg, i.p. or i.v.) treated with anti-CTLA4 mAb with 3 days intervals. The levels of circulating therapeutic anti-CTLA4 antibodies were monitored during the course of treatment and long-term in the serum of treated animals via enzyme-linked immunosorbant assay (n = 1–5 per time point, data shown are the mean ± SEM). CR, complete responders; ns, not significant; i.p., intraperitoneal; i.v., intravenous; p.t., peritumoral; s.c., subcutaneous.

Figure 3. Systemic accumulation of Tregs after high-dose anti-CTLA4 mAb therapy. (A–D) Mice were inoculated subcutaneously with 2.5 × 10⁵ Panc02 cells and treated with anti-CTLA4 blocking antibody (as indicated below) or a phosphate-buffered saline (PBS) control on day 5, day 8, and day 11. (A–C) The levels of Tregs (CD4⁺FoxP3⁺) residing in secondary lymphoid organs were evaluated by immunofluorescent staining and flow cytometry on day 14 after low-dose peritumoral (p.t.) or high-dose intraperitoneal (i.p.) anti-CTLA4 mAb therapy in animals carrying Panc02 tumors. (A) Data represent cumulative results from 3 independent experiments visualized as percent Tregs in the tumor-draining lymph node (TDLN) normalized against the PBS treated control (median, whiskers: min to max; n = 16–17). Statistical analysis was performed by ANOVA and the Bonferroni multiple comparison test. (B) Average percent FoxP3⁺ Tregs of total CD4⁺ cells present in the spleens of animals treated as indicated. Data represent cumulative results from 2 independent experiments (median, whiskers: min to max, n = 12–13). Statistical analysis was performed by ANOVA and the Bonferroni multiple comparison test. (C) Animals were treated twice with 30 μg or 90 μg local peritumoral anti-CTLA4 mAb, day 5 and day 8 after Panc02-tumor inoculation. The percentage of Tregs (CD4⁺ FoxP3⁺) present in the spleen was quantified by immunofluorescent staining and flow cytometry on day11 (median, n = 6). Statistical analysis was performed by unpaired Student’s t test. (D) The levels of circulating therapeutic anti-CTLA4 antibodies were monitored in serum via enzyme-linked immunosorbant assay (n = 2–5 per time point, data shown are the mean ± SEM) *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant.

Figure 4. Anti-CTLA-4 shifts the balance between tumor-infiltrating Teff cells and Tregs. Tumor-infiltrating effector T (Teff) cells and regulatory T cells (Tregs) were evaluated on day 14 after low-dose p.t. and high-dose i.p. aCTLA-4 therapy in animals carrying Panc02 tumors. Data are visualized as (A) percent CD4⁺ Teff cells (CD4⁺FoxP3^-), (B) percent CD8⁺ Teff cells (CD8⁺FoxP3^-) and (C) percent Tregs (CD4⁺FoxP3⁺) in tumor tissue (data shown are the median; n = 6). Statistical analysis was performed by Mann–Whitney. *P < 0.05, **P < 0.01; ns, not significant.