Although chemotherapy is the treatment of choice for many types of cancer, it is rarely curative in most solid tumors. Immune therapy represents a potentially attractive approach to increase the efficacy of chemotherapy by targeting cancer cells that escape chemotherapy. However, it has been unclear to date whether any chemotherapy drugs are more suitable than others for such combinations, and empirical use has produced mixed results. For example, although higher objective response and disease control rates, along with elevated frequencies of cytolytic tumor antigen-specific T cells, were seen in patients with metastatic colorectal carcinoma receiving polychemotherapy with gemcitabine plus oxaliplatin, fluorouracil, and folinic acid (FOLFOX-4) followed by granulocyte-macrophage colony-stimulating factor (GM-CSF) and low-dose interleukin-2 (IL-2) , addition of IL-2 and interferon-alpha2b did not increase the efficacy of cisplatin, vindesine and dacarbazine in melanoma patients . Thus, understanding the mechanisms underpinning positive chemo-immunotherapy interactions is a critical task for the development of effective cancer therapy.
Previous reports have suggested that the exposure of tumor cells to chemotherapeutic drugs can sensitize them to immune effector cells [3–6]. Theoretically, to achieve synergy with immune therapy and increased tumor killing, chemotherapy should sensitize to immune killing tumor cells that are destined to survive chemotherapy. Depending on their mechanism of action, the efficacy of chemotherapy drugs may be influenced markedly by the time of exposure (phase-specific or time-dependent drugs) or by the dose that can be administered (phase-nonspecific or dose-dependent drugs). The efficacy of phase-specific anticancer drugs is time-dependent, as only a fraction of tumor cells are in appropriate cell cycle phase for chemotherapy-mediated killing at any given time. Thus, a fraction of tumor cells remains alive following administration of each chemotherapy dose and can eventually repopulate the tumor following completion of chemotherapy [7–10]. We hypothesized that because of this property, time-dependent chemotherapy drugs are more likely to benefit from combination with immune therapy.
Interleukin 18 (IL-18) is a pleiotropic cytokine, originally described as interferon (IFN)-γ inducing factor, that can mediate immunostimulatory effects on immune cells of the adaptive and innate immune system . Its multiple immunologic activities include the induction of IFN-γ, TNF-α, IL-1α, and GM-CSF production; augmentation of natural killer (NK) cell cytotoxicity; and promotion of Th1 differentiation of naive T cells. These features render IL-18 an interesting candidate for tumor immunotherapy. As a single agent, IL-18 was shown to elicit anti-tumor reactivity when administered at high doses in mice with established tumors . The immunostimulatory activity of IL-18 in vivo has been demonstrated in non-human primates  and humans . In phase I clinical evaluation, recombinant human (rh)IL-18 was safely administered as monotherapy to 28 patients with solid tumors, with minimal dose-limiting toxicities and two partial tumor responses . Toxicity has generally been mild to moderate even with repeat administration and a maximum tolerated dose has not been reached to date . IL-18 enhanced activation of peripheral blood CD8+ T cells, NK cells and monocytes and induced a transient increase in the frequency and expression level of Fas ligand (FasL) in peripheral blood CD8+ T cells and NK cells . The relatively minor toxicity of rhIL-18, compared with other immunostimulatory cytokines that have undergone clinical development, is remarkable and renders IL-18 a well suited drug for combinatorial approaches with chemotherapy.
In the current study, we characterized the immune effects on tumor cells of four common anticancer chemotherapy drugs utilized in ovarian cancer and other solid tumors, two phase-specific (time-dependent) drugs, paclitaxel and topotecan, and two phase-nonspecific (dose-dependent) drugs, gemcitabine and carboplatin. Both paclitaxel and topotecan exert their actions on dividing cells, acting as phase-specific chemotherapeutic drugs. Paclitaxel inhibits the dissolution of microtubules, enhances tubulin polymerization and produces a block in the metaphase of mitosis, leading to growth inhibition and cell apoptosis . Topotecan, a topoisomerase I inhibitor, stabilizes the covalent complex of enzyme and strand-cleaved DNA, which is an intermediate of the catalytic mechanism, thereby inducing breaks in the protein-associated DNA single-strands, resulting in cell death . Carboplatin is a classic cycle phase non-specific drug . The main mechanism of action of gemcitabine is inhibition of DNA synthesis. The killing effects of gemcitabine are however not confined to the S-phase of the cell cycle and the drug is equally effective against confluent cells and cells in log-phase growth . Incorporation of gemcitabine into RNA is another action, which is time- and concentration-dependent and leads to inhibition of RNA synthesis. In human tumor cell lines displaying different degrees of resistance to gemcitabine, sensitivity to this drug was related to differences in RNA incorporation . Moreover, several metabolites of gemcitabine can inhibit various enzymes, leading to self-potentation of gemcitabine action . Thus, the overall mechanism of action of gemcitabine is phase non-specific.
Because the effect of immune therapy becomes clinically relevant only if immune mechanisms target the tumor fraction surviving chemotherapy, we focused on the fate of tumor cells escaping direct killing by chemotherapy. We hypothesized that these cells are sensitized to immune therapy, which enables a cooperation between immunotherapy and time-dependent (phase-specific) drugs. Thus, we hypothesized that among chemotherapy drugs, time-dependent (phase-specific) drugs are more likely to benefit from IL-18 therapy combination. We investigated these interactions in a mouse model of ovarian cancer . IL-18 alone had a modest antitumor effect, while it exhibited positive interaction with select chemotherapeutic drugs, improving their therapeutic effect in vivo. Chemotherapeutic agents upregulated immune molecules in tumor cells and sensitized them to immune-mediated killing. Importantly, this effect was translated to a significant increase in total killing fraction and better outcome only for time-dependent drugs. Interestingly, combination of IL-18 with dose-dependent drugs did not increase their efficacy in vivo. In this study however we sought to mainly explore the ability of those drugs to render tumor cells more susceptible to immunotherapy with IL-18, and specifically focused on effector mechanisms, and not the drugs' overall effects on the immune system. Moreover, our findings indicate for the first time that the difference of chemotherapeutic drugs in their ability to interact with immunotherapy might be attributed to their mechanism of action directly on the tumor cell. These data suggest that tumor immunotherapy with IL-18 may potentiate the antitumor effect selectively of time-dependent chemotherapeutics used in various types of cancer.