The combination of a low-dose chemotherapeutic agent, 5-fluorouracil, and an adenoviral tumor vaccine has a synergistic benefit on survival in a tumor model system

Abstract

Standard cancer therapies, particularly those involving chemotherapy, are in need of modifications that both reduce short-term and long-term side effects as well as improve the overall survival of cancer patients. Here we show that combining low-dose chemotherapy with a therapeutic vaccination using an adenovirus encoding a model tumor-associated antigen, ovalbumin (Ad5-OVA), had a synergistic impact on survival in tumor-challenged mice. Mice that received the combinatorial treatment of Ad5-OVA plus low-dose 5-fluorouracil (5-FU) had a 95% survival rate compared to 7% and 30% survival rates for Ad5-OVA alone and 5-FU alone respectively. The presence of 5-FU enhanced the levels of OVA-specific CD8(+) T lymphocytes in the spleens and draining lymph nodes of Ad5-OVA-treated mice, a phenomenon that was dependent on the mice having been tumor-challenged. Thus 5-FU may have enhanced survival of Ad5-OVA-treated mice by enhancing the tumor-specific immune response combined with eliminating tumor bulk. We also investigated the possibility that the observed therapeutic benefit may have been derived from the capacity of 5-FU to deplete MDSC populations. The findings presented here promote the concept of combining adenoviral cancer vaccines with low-dose chemotherapy.

Figure 1. Anti-tumor effect and survival in mice treated with Ad5-OVA and/or 5-FU.

C57BL/6 mice were challenged with E.G7 and then, 7 days post tumor challenge, given: no treatment (naïve); Ad5-OVA; Ad5-OVA +5-FU; or 5-FU. A–D. The tumor volumes for each mouse from one representative experiment are shown. E. Survival graph representing four pooled experiments. Total number of mice for each treatment was: n = 15 for naïve group, n = 14 for Ad5-OVA alone group, n = 21 for Ad5-OVA +5-FU group, n = 10 for Ad5-LacZ +5-FU group, and n = 15 for the 5-FU alone group. Statistical analysis (Log-rank (Mantel-Cox) test) of survival data revealed that mice survived significantly longer, compared to untreated mice, when treated with Ad5-OVA alone (p<0.001 (***⁽¹⁾)), 5-FU alone (p<0.001(***⁽²⁾)) or Ad5-LacZ +5-FU (p<0.001(***⁽³⁾), and that mice treated with Ad5-OVA +5-FU in combination survived significantly longer than untreated mice (p<0.001 (***⁽⁴⁾) and mice treated with Ad5-OVA alone (p<0.001 (***⁽⁵⁾)), 5-FU alone (p<0.001(***⁽⁶⁾)) and Ad5-LacZ +5-FU (p = 0.001 (***⁽⁷⁾).

Figure 2. OVA-specific CD8⁺ T cell responses in tumor challenged and non-tumor-challenged mice.

C57BL/6 mice were challenged with E.G7 and then, 7 days post tumor challenge, were treated as indicated. A and B. On day 9 post treatment the levels of OVA-specific CD8+ T cells, as a percentage of total CD8+CD3+ cells, were determined using a fluorescently-tagged OVA tetramer (see materials and methods for further details) in samples from the spleen (A) and the lymph node (B). The percentages were normalized against the mean values obtained for the naïve groups. Mean percentage values for the naïve group prior to normalization were 0.26% (range 0.16–0.39%) for the lymph node and 0.68% (range 0.54–0.89%) for the spleen. Data represents pooled data from four separate experiments, where the number of mice/treatment group in each experiment was n = 1, n = 2, n = 2 and n = 4 and therefore a total of n = 9 mice from each pooled treatment group were analyzed. Statistical significance was determined using an ANOVA one way analysis of variance (with Tukey post-test) (**P<0.01, ***P<0.001). C . Non-tumor-challenged mice (n = 4 per group) were vaccinated with indicated treatments and then the levels of OVA-specific CD8⁺ T cells in the peripheral blood were measured on days 7, 15 and 21. These results were shown to be reproducible. Error bars represent standard error of the mean.

Figure 3. Treg levels in spleen and draining lymph node.

C57BL/6 mice were challenged with E.G7 and then, 7 days post tumor challenge, were treated as indicated. On day 5 post treatment the levels of Tregs (Foxp3⁺), as a percentage of total CD4⁺CD3⁺ cells, was determined using direct immunofluorescence (see materials and methods for further details) on the samples from the spleen (A) and the lymph node (B). Results displayed are from one representative experiment (n = 4 mice per group). These results were reproducible. Using an ANOVA one way analysis of variance (with Tukey post-test) no statistically significant differences were observed in independent experiments nor after pooling of the data. Error bars represent standard error of the mean.

Figure 4. Levels of gMDSCs and mMDSCs in the spleen and draining lymph node.

C57BL/6 mice were challenged with E.G7 cells and then, 7 days post tumor challenge, were treated as indicated and then 5 days later, spleen and draining lymph nodes were harvested and stained using direct immunofluorescence for detection mMDSCs and gMDSCs and then analyzed using flow cytometry (see materials and methods for further details). A. Example of FloJo-generated dot-plots showing how mMDSCs and gMDSCs were delineated in samples obtained from the spleen. Total splenocytes were gated using a forward scatter (FSC) versus side scatter (SSC) dot-plot (top row). These cells were further gated to select for CD11b^+hi cells using a SSC versus CD11b dot-plot (middle row). Then these CD11b+hi cells were further defined using a Ly6G versus Ly6C dot-plot (bottom row). mMDSCs were defined as CD11b⁺Ly6G⁻Ly6C^+hi (circled (left side)) and gMDSCs were defined as CD11b⁺Ly6G⁺Ly6C^+low (circled (right side)). A similar method of analysis was used for cells obtained from the draining lymph node. B–E. On day 5 post treatment, the levels of mMDSCs (B and D) and gMDSCs (C and E), as a percentage of total cells, were determined, in both the spleen (B and C) and the draining lymph node (D and E). Mean percentage values (and range) for the naïve groups prior to normalization were as follows: mMDSCs in the lymph node = 0.24% (range: 0.02–0.84%); mMDSCs in the spleen = 0.26% (range: 0.02–0.63%); gMDSCs in the lymph node = 0.03% (range: 0.01–0.08%); gMDSCs in the spleen = 1.35% (0.87–4.1%). Results displayed are derived from pooled data from four separate experiments, where the number of mice/treatment group in each experiment was n = 1, n = 2, n = 2, n = 4 and therefore a total of n = 9 mice from each pooled treatment group were analyzed. Statistical significance was determined using an ANOVA one way analysis of variance (with Tukey post-test) (*P<0.05, **P<0.01, ***P<0.001)). Error bars represent standard error of the mean.

Figure 5. Effect of depleting MDSCs from variously vaccinated tumor-bearing mice on survival and tumor-specific T lymphocyte responses.

A. Tumor-bearing mice were treated with multiple i.p. doses of RB6-8C5 (as described in materials and methods section) and draining lymph node and spleen cells were isolated after 6 days of depletion (day 12 post-tumor challenge) and stained for the presence of gMDSCs (CD11b⁺Ly6C^+lowLy6G⁺) and mMDSCs (CD11b⁺Ly6C^+hiLy6G⁻). Data is presented as a percentage of total lymph node or spleen populations. Student T-test was used to determine statistical significance. B. Survival analysis: C57BL/6 mice were challenged with E.G7 and then, 7 days post tumor challenge, were treated as indicated. Survival curve represents pooled data from 2 independent experiments where a total of n = 8 mice/treatment group were used. Statistical analysis (Log-rank (Mantel-Cox) test) of survival data revealed that only the Ad5-OVA plus 5-FU treatment to be significantly different from all other treatments (*** P<0.001). C. Two weeks post-treatment the levels of OVA-specific CD8⁺ T cells, as a percentage of total CD8⁺CD3⁺ cells, was determined using a fluorescently-tagged OVA tetramer (see materials and methods for further details), in PBLs. Results displayed represent pooled data from 2 independent experiments and were performed in conjunction with the survival studies (see above (B.)). Statistical significance was determined using an ANOVA one way analysis of variance (with Tukey post-test) (*P<0.05). All error bars represent standard error of the mean.