Metronomic chemotherapy with low-dose cyclophosphamide plus gemcitabine can induce anti-tumor T cell immunity in vivo

Abstract

Several chemotherapeutic drugs have immune-modulating effects. For example, cyclophosphamide (CP) and gemcitabine (GEM) diminish immunosuppression by regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), respectively. Here, we show that intermittent (metronomic) chemotherapy with low-dose CP plus GEM can induce anti-tumor T cell immunity in CT26 colon carcinoma-bearing mice. Although no significant growth suppression was observed by injections of CP (100 mg/kg) at 8-day intervals or those of CP (50 mg/kg) at 4-day intervals, CP injection (100 mg/kg) increased the frequency of tumor peptide-specific T lymphocytes in draining lymph nodes, which was abolished by two injections of CP (50 mg/kg) at a 4-day interval. Alternatively, injection of GEM (50 mg/kg) was superior to that of GEM (100 mg/kg) in suppressing tumor growth in vivo, despite the smaller dose. When CT26-bearing mice were treated with low-dose (50 mg/kg) CP plus (50 mg/kg) GEM at 8-day intervals, tumor growth was suppressed without impairing T cell function; the effect was mainly T cell dependent. The metronomic combination chemotherapy cured one-third of CT26-bearing mice that acquired tumor-specific T cell immunity. The combination therapy decreased Foxp3 and arginase-1 mRNA levels but increased IFN-γ mRNA expression in tumor tissues. The percentages of tumor-infiltrating CD45(+) cells, especially Gr-1(high) CD11b(+) MDSCs, were decreased. These results indicate that metronomic chemotherapy with low-dose CP plus GEM is a promising protocol to mitigate totally Treg- and MDSC-mediated immunosuppression and elicit anti-tumor T cell immunity in vivo.

Fig. 1

Anti-tumor effects induced by intermittent injections of low-dose CP. a BALB/c mice were injected s.c. with 5 × 10⁵ CT26 cells into the right flank. On day 10, mice were injected i.p. with either CP (100 mg/kg) with an 8-day interval or CP (50 mg/kg) with a 4-day interval between injections. Arrows indicate the CP injection. Tumor size (mm²) was measured twice weekly. Each group consisted of six or eight mice, and lines represent tumor growth. The mean ± SD of the results on day 34 after tumor inoculation are shown. Similar results were obtained in three experiments. NS, not significant by ANOVA with Dunnett’s post hoc test. b BALB/c mice were injected s.c. with 5 × 10⁵ CT26 cells into the right flank. On the indicated days, mice were injected i.p. with CP (50 or 100 mg/kg) and draining LNs were harvested, pooled, and stimulated in vitro with the AH1 or the control peptide. After 3 days, IFN-γ levels in the supernatant were determined by ELISA. Each group consisted of three mice. *P < 0.05 indicates statistical significance by Student’s t test

Fig. 2

Injection of low-dose GEM suppresses tumor growth in vivo. BALB/c mice were injected s.c. with 5 × 10⁵ CT26 cells into the right flank. On day 10, mice were injected i.p. with GEM (50 or 100 mg/kg). Arrows indicate the injection of GEM. Tumor size (mm²) was measured twice weekly. Each group consisted of five or six mice, and lines represent tumor growth in each mouse. The mean ± SD of the results on day 30 after tumor inoculation are also shown. Similar results were obtained in two experiments. *P < 0.05 indicates statistical significance by ANOVA with Dunnett’s post hoc test. NS, not significant

Fig. 3

No suppressive effect of the combination chemotherapy with low-dose CP plus GEM on splenic T cell function. a Naïve BALB/c mice were injected i.p. with CP (50 mg/kg) and GEM (50 mg/kg) on the indicated days. Spleen cells were harvested, counted, and analyzed by flow cytometry. Each group consisted of four mice. *P < 0.05 indicates statistical significance by Student’s t test. b Spleen cells from individual mice were cultured in anti-CD3 mAb-coated wells for 2 days, and IFN-γ levels in supernatants were determined by ELISA. Numbers represent individual mice

Fig. 4

Injections of low-dose CP and GEM with an 8-day interval suppress tumor growth. a BALB/c mice were injected s.c. with 5 × 10⁵ CT26 cells into the right flank. Ten days later, these mice were injected i.p. with CP (50 mg/kg) and/or GEM (50 mg/kg) with an 8-day interval between injections. Arrows represent the injections. Tumor size (mm²) was measured twice weekly. Each group consisted of six or seven mice, and lines represent tumor growth. b The mean ± SD of the results on day 42 after tumor inoculation are shown. *P < 0.05 indicates statistical significance by ANOVA with Dunnett’s post hoc test. Similar results were obtained in three experiments. c BALB/c nude mice were injected s.c. with 5 × 10⁵ CT26 cells into the right flank. On days 8 and 16, these mice were injected i.p. with CP (50 mg/kg) and GEM (50 mg/kg). Arrows indicate the injections of CP and GEM. Each group consisted of five mice. *P < 0.05 indicates statistical significance by Student’s t test

Fig. 5

Tumor-specific T cell immunity in CT26-cured mice after metronomic combination chemotherapy. a BALB/c mice were injected s.c. with 5 × 10⁵ CT26 cells into the right flank. On days 10, 18, and 26, these mice (n = 24) were injected i.p. with CP (50 mg/kg) and GEM (50 mg/kg). Arrows indicate the injections. Tumor size (mm²) was measured twice weekly. Lines represent tumor growth. Six mice that cured of CT26 were inoculated s.c. with 2 × 10⁵ CT26 cells 60 days after the initial tumor inoculation. Inset shows the tumor growth in naïve BALB/c mice that were inoculated s.c. with 2 × 10⁵ CT26 cells. b Spleen cells from naïve or CT26-cured mice were cultured with MMC-treated CT26 or RENCA cells for 3 days, and IFN-γ levels in supernatants were determined by ELISA. Numbers represent individual mice. *P < 0.05 indicates statistical significance by Student’s t test

Fig. 6

Effects of the combination therapy of CP and GEM on tumor-infiltrating immune cells. a BALB/c mice were injected s.c. with 5 × 10⁵ CT-26 cells into the right flank. On day 10, mice were injected i.p. with CP and/or GEM at a dose of 50 mg/kg. Two days later, tumor tissues were collected and mRNA expression was evaluated by real-time PCR. Each group consisted of five mice. *P < 0.05 and **P < 0.01 indicate statistical significance by Student’s t test. b Tumor tissue cell suspensions were stained with FITC-conjugated anti-Gr-1 mAb, PE-conjugated anti-CD11b mAb, and APC-conjugated anti-CD45 mAb, and the percentage of tumor-infiltrating immune cells was determined (upper, left). CD45⁺ cells were sub-divided (upper, right) into Gr-1^low CD11b⁺ or Gr-1^high CD11b⁺ cells. Each group consisted of four mice. *P < 0.05 and **P < 0.01 indicate statistical significance by Student’s t test