Synergistic effect of CTLA-4 blockade and cancer chemotherapy in the induction of anti-tumor immunity

Abstract

Several chemotherapeutics exert immunomodulatory effects. One of these is the nucleoside analogue gemcitabine, which is widely used in patients with lung cancer, ovarian cancer, breast cancer, mesothelioma and several other types of cancer, but with limited efficacy. We hypothesized that the immunopotentiating effects of this drug are partly restrained by the inhibitory T cell molecule CTLA-4 and thus could be augmented by combining it with a blocking antibody against CTLA-4, which on its own has recently shown beneficial clinical effects in the treatment of patients with metastatic melanoma. Here we show, using two non-immunogenic murine tumor models, that treatment with gemcitabine chemotherapy in combination with CTLA-4 blockade results in the induction of a potent anti-tumor immune response. Depletion experiments demonstrated that both CD4(+) and CD8(+) T cells are required for optimal therapeutic effect. Mice treated with the combination exhibited tumor regression and long-term protective immunity. In addition, we show that the efficacy of the combination is moderated by the timing of administration of the two agents. Our results show that immune checkpoint blockade and cytotoxic chemotherapy can have a synergistic effect in the treatment of cancer. These results provide a basis to pursue combination therapies with anti-CTLA-4 and immunopotentiating chemotherapy and have important implications for future studies in cancer patients. Since both drugs are approved for use in patients our data can be immediately translated into clinical trials.

Figure 1. Combination of CTLA-4 blockade and gemcitabine chemotherapy results in synergistic anti-tumor effect.

(A) Tumor surface in mm² (mean ± SD) of AB1-HA tumors that were injected on day 0, mice (n = 87) were treated on day 9/12/15/18 with 75 µg anti-CTLA-4 and with 120 µg/g gemcitabine on day 9/12/15/18/21, or with PBS (pooled data of 5 separate experiments are shown). (B) Kaplan-Meier survival plot of the same experiment. (C) Tumor surface in mm² (mean ± SD) of AB1-HA tumors that were injected on day 0, mice (n = 65) were treated on day 9 with 200 µg anti-CTLA-4 and 6 µg/g cisplatin, or with PBS (pooled data from 3 separate experiments are shown). (D) Kaplan-Meier survival plot of the same experiment.

Figure 2. Combination of CTLA-4 blockade and gemcitabine chemotherapy results in enhanced T cell activation and proliferation and is dependent on CD4+ and CD8+ T cells.

A comparison is shown of peripheral blood T cell activation and proliferation markers on day 29 after inoculation for the different treatment groups (*p≤0.05; **p<0.01***p<0.001). ICOS⁺/CD4⁺ Th cells (A); Ki-67⁺/CD4⁺ Th cells (B); CD8⁺/ICOS⁺ CTLs(C) and CD8⁺/Ki-67⁺ CTLs (D). (E and F) Flow cytometric analysis of proliferating CD8⁺ T cells and Treg in tumor, tumor-draining lymph node and spleen on day 15. Depicted are the percentage of Ki-67⁺CD8⁺ of CD3⁺ cells and Foxp3⁺CD4⁺ of CD3⁺ cells (F). Six mice per group were tested for control and anti-CTLA-4, 12 mice per group for gemcitabine-containing regimes pooled per 2 mice because of the small tumor size in these groups. Means with SEMs are shown (n = 36). (G) Kaplan-Meier survival plot of AB1-HA tumors that were injected on day 0, mice (n = 57) were treated with anti-CTLA-4 and/or gemcitabine, or with PBS in combination with depleting antibodies against CD4 or CD8 (pooled data of 2 separate experiments are shown).

Figure 3. Combination of CTLA-4 blockade and gemcitabine chemotherapy results in the induction of protective T cell memory.

(A) Kaplan-Meier survival plot of mice that had been cured by either anti-CTLA-4 alone or combination therapy and that were subsequently rechallenged with AB1 mesothelioma cells, showing protective immunity in 80% and 92% respectively. T cell subset analysis in tumor-draining lymph nodes in these mice (*p<0.05; **p<0.01***p<0.001): CD44⁺/CD62L⁺/CD4⁺ T central memory cells (B); CD44⁺/CD62L^−/CD4⁺ T effector memory cells (C); CD44⁺/CD62L⁺/CD8⁺ T central memory cells (D); CD44⁺/CD62L^−/CD8⁺ T effector memory cells (E).

Figure 4. The efficacy of combining CTLA-4 blockade with gemcitabine critically depends on timing.

(A) Tumor area in mm² (mean ± SD) of AB1-HA tumors that were injected on day 0, mice (n = 86) were treated with different schedules of anti-CTLA4 and gemcitabine (see Figure S2), or with PBS (pooled data of 3 separate experiments are shown). (B) Kaplan-Meier survival plot of the same experiment.