Favorable alteration of tumor microenvironment by immunomodulatory cytokines for efficient T-cell therapy in solid tumors

Abstract

Unfavorable ratios between the number and activation status of effector and suppressor immune cells infiltrating the tumor contribute to resistance of solid tumors to T-cell based therapies. Here, we studied the capacity of FDA and EMA approved recombinant cytokines to manipulate this balance in favor of efficient anti-tumor responses in B16.OVA melanoma bearing C57BL/6 mice. Intratumoral administration of IFN-α2, IFN-γ, TNF-α, and IL-2 significantly enhanced the anti-tumor effect of ovalbumin-specific CD8+ T-cell (OT-I) therapy, whereas GM-CSF increased tumor growth in association with an increase in immunosuppressive cell populations. None of the cytokines augmented tumor trafficking of OT-I cells significantly, but injections of IFN-α2, IFN-γ and IL-2 increased intratumoral cytokine secretion and recruitment of endogenous immune cells capable of stimulating T-cells, such as natural killer and maturated CD11c+ antigen-presenting cells. Moreover, IFN-α2 and IL-2 increased the levels of activated tumor-infiltrating CD8+ T-cells concomitant with reduction in the CD8+ T-cell expression of anergy markers CTLA-4 and PD-1. In conclusion, intratumoral administration of IFN-α2, IFN-γ and IL-2 can lead to immune sensitization of the established tumor, whereas GM-CSF may contribute to tumor-associated immunosuppression. The results described here provide rationale for including local administration of immunostimulatory cytokines into T-cell therapy regimens. One appealing embodiment of this would be vectored delivery which could be advantageous over direct injection of recombinant molecules with regard to efficacy, cost, persistence and convenience.

Fig 1. IFN-α2, IFN-γ and IL-2 augment anti-tumor efficacy but do not increase tumor-accumulation of transferred cells.

Mice bearing syngeneic B16.OVA tumors were adoptively transferred with 2x10⁶ CD8a⁺ enriched, polyclonally activated OT-I lymphocytes intraperitoneally and tumors were either left non-injected or injected with PBS or recombinant cytokine in PBS (n = 10). (a) Tumor growth was monitored every 2–3 days with an electronic caliper. Due to variation in tumor sizes at the beginning of the experiment, the results are represented as relative change compared to day 0 volume, which was set at 100%. (b-c) Levels of OT-I cells in tumors were quantified on days 4 (b) and 14 (c) post-transfer by pentamer staining and flow cytometry. (d) Proportion of major histocompatibility complex (MHC) class I molecules presenting OVA-derived peptide SIINFEKL and (e) mean fluorescence intensity (MFI) of mouse MHC class I H-2kb from tumor samples was assessed by flow cytometry on day 14 post-transfer (n = 5). Data presented as mean ± SEM. *P≤ 0.05, **P≤ 0.01, ***P≤ 0.001, ****P≤ 0.0001 by repeated measures ANOVA (a) or one-way ANOVA followed by Tukey’s post-hoc test (b-e).

Fig 2. Recombinant cytokines induce intratumoral, endogenous secretion of cytokines associated with immune cell activation.

B16.OVA-bearing mice were treated intraperitoneally with 2x10⁶ CD8a⁺ enriched OT-I lymphocytes and treated intratumorally with either PBS or recombinant cytokine (in PBS) or left non-injected. Levels of (a) IFN-γ, (b) IL-2, (c) TNF-α, (d) IL-12p70, (e) GM-CSF, (f) IL-1β, (g) IL-6 and (h) IL-10 from tumor homogenates were measured with CBA Flex sets on day 14 post-transfer (n = 3–5). Horizontal lines represent median values. *P≤ 0.05 and **P≤ 0.01 byone-way ANOVA followed by Tukey’s post-hoc test.

Fig 3. Intra-tumor accumulation of antigen-presenting cells (APCs) is increased by GM-CSF and IL-2.

B16.OVA bearing mice were treated with adoptive transfer of 2x10⁶ CD8a⁺ enriched OT-I lymphocytes intraperitoneally and tumors were either injected with PBS or recombinant cytokine in PBS or left non-injected (n = 5). (a) Levels of CD11c⁺ dendritic cells and (b) proportion of dendritic cells expressing maturation marker CD86 on cell surface were analyzed on day 14 post-transfer from tumors. Data presented as mean ± SEM. *P ≤ 0.05 and **P≤ 0.01 by one-way ANOVA followed by Tukey’s post-hoc test.

Fig 4. Intratumoral myeloid cell subsets are influenced by local cytokine therapy.

Mice bearing subcutaneous B16.OVA tumors received intraperitoneal transfer of 2x10⁶ CD8a⁺ enriched OT-I lymphocytes and intratumoral injections of either PBS or recombinant cytokine in PBS (n = 5). Levels of tumor-infiltrating (a) CD11b⁺ myeloid cells, (b) NK1.1⁺ natural killer cells, (c) CD11b⁺ F4/80+ macrophages, (d) suppressive M2 macrophages (characterized by surface expression of CD206), (e) CD11b⁺ Gr-1⁺ myeloid-derived suppressor cells (MDSC) and (f) ratio of monocytic (M) to polymorphonuclear (PMN) MDSCs were assessed from tumors on day 14 post-transfer by flow cytometry. Data presented as mean ± SEM. *P ≤ 0.05, **P≤ 0.01, ***P≤ 0.001 and ****P≤ 0.0001 byone-way ANOVA followed by Tukey’s post-hoc test.

Fig 5. IFN-α2, IFN-γ and IL-2 treatment leads to changes in CD8+ TIL phenotypes.

Mice harboring subcutaneous B16.OVA tumors were treated intraperitoneally with 2x10⁶ CD8a⁺ enriched OT-I lymphocytes and injected intratumorally with either PBS or recombinant cytokine in PBS or left non-injected (n = 5). (a) Levels of tumor-infiltrating endogenous (non-OVA) CD8⁺ T-cells and (b) count of central memory (T_CM) and effector memory (T_EM) T-cells were assessed from tumors on day 14 post-transfer by flow cytometry. (c) Activation status of tumor-infiltrating CD8+ T-cells was evaluated on day 14 by expression of CD69 and IFN-γ following PMA/Ionomycin stimulation ex vivo. Data presented as mean ± SEM. *P ≤ 0.05 and **P≤ 0.01 by one-way ANOVA followed by Tukey’s post-hoc test.

Fig 6. Expression of anergy markers on CD8+ TILs are downregulated following IFN-α2, IFN-γ and IL-2 treatments.

Mice bearing subcutaneous B16.OVA tumors were injected with 2x10⁶ CD8a⁺ enriched OT-I lymphocytes into peritoneal cavity and beginning on the same day, tumors were injected with either PBS or recombinant cytokine in PBS or left non-injected (n = 5). Proportion of CD3⁺ CD8⁺ TILs expressing surface anergy markers (a) CTLA-4 and (b) PD-1 was analyzed by flow cytometry on day 14 post-transfer. Data presented as mean ± SEM. *P ≤ 0.05, **P≤ 0.01, ***P≤ 0.001, ****P≤ 0.0001 by one-way ANOVA followed by Tukey’s post-hoc test.