Ovarian cancer is responsible for the fifth most cancer deaths among women. The American Cancer Society estimates that, in 2017 in the U.S., approximately 22,000 women will receive a new diagnosis of ovarian cancer and nearly 14,000 will die from this disease. For ovarian cancer patients of all stages, the 5-year survival rate is approximately 46 percent. However, for those having a distant disease, the rate drops to 28 percent. Among ovarian cancer patients, 62 percent have stage III or IV disease, and of these, approximately 60-70 percent will have disease recurrence. This is especially sobering, as there are no current efficacious therapies for women having recurring ovarian cancer.
Ovarian malignancies have been shown to induce spontaneous antitumor immunological responses. The presence of tumor-infiltrating lymphocytes (TILs) has been shown to have a strong positive correlation with survival in ovarian cancer patients, as those ovarian cancer patients having TIL in their tumors had 5-year overall survival rates of 38 percent. Conversely, for those patients with tumors lacking TILs, the corresponding rate is only 4.5 percent.
Natural killer (NK) cells are one very important component of these TILs. However, in an immunosuppressive tumor microenvironment, these cells do not display the typical features or behavior of healthy NK cells. Several methods for enhancing the activation and/or proliferation of NK cells in cancer patients are currently being investigated. One such method for stimulating the antitumor effects of these cells is the use of ALT-803, an IL-15N72D/IL-15Rα-Fc superagonist.
This investigational treatment is being evaluated in preclinical studies by a team led by Martin Felices, PhD, Assistant Professor of Medicine in the Department of Medicine, Division of Hematology, Oncology, and Transplantation, University of Minnesota, Minneapolis, and Melissa Geller, MD, MS, Associate Professor in the Department of Obstetrics, Gynecology, and Women's Health, Division of Gynecologic Oncology, University of Minnesota. Descriptions of these studies have been reported at the 2017 Society of Gynecologic Oncology annual meeting (Abstract 33) and are currently in press (Gynecologic Oncology 2017; doi:10.1016/j.ygyno.2017.02.028).
When discussing the role of NK cells, Felices explained, “As part of the innate immune system, NK cells do not rely on HLA (human leukocyte antigen)-mediated recognition. NK cells recognize activating ligands and cognate major histocompatibility complex (MHC) class I on target cells via unique NK cell receptors that either activate or inhibit NK cell function.”
About 5-10 percent of circulating lymphocytes are NK cells, for which the mature phenotype is CD56+CD3−. One subpopulation of NK cells is the CD56 brightCD16− class associated with the release of high levels of cytokines and low cytotoxicity. Another important subpopulation is the CD56dimCD16+ class, which is characterized by direct cell killing via cytotoxicity receptors or CD16-mediated antibody-dependent cell-mediated cytotoxicity (ADCC) without prior antigen priming.
NK-Cell Phenotype Changes/Proliferation
Previous studies have shown healthy donor-derived NK cells can, in in vitro tests, recognize and kill ovarian cancer cells. In contrast, “NK cells isolated from the ascites (buildup of peritoneal cavity fluid) of ovarian cancer patients are characterized by an immature phenotype and reduced function, presumably a result of the immunosuppressive tumor microenvironment,” Felices explained. “Relative to the NK cells from healthy donors, in the ovarian cancer patient NK cells, there was a higher proportion of the less mature CD56bright cells and a lower number of the mature CD56dim cells. The ascites-derived NK cells also had lower per cell surface densities of CD16, the receptor necessary for ADCC.”
To assist the function of NK cells in a cancer setting, Felices and Geller evaluated the use of ALT-803, an IL-15 super-agonist complex. “This complex contains two mutated IL-15/IL-15Rα sushi domains that enhance trans-presentation of the cytokine to NK cells. Additionally an Fc portion, that does not bind to complement or induce ADCC, was added to this complex to improve pharmacokinetics and prolong function in vivo,” Felices noted.
To test the effectiveness of ALT-803, ascites-recovered NK cells were subjected to in vitro tests that assessed the agonist's ability to recover proliferation and function against a panel of ovarian cancer cell lines. NK cellular proliferation was gauged using a dye-based assay. In this model, NK cells from either healthy donors or the ascites of ovarian cancer patients were labeled with CellTrace violet dye. The respective cells were then cultured for a period of 1 week with media alone (RPMI, 10%), media with monomeric IL-15, or media with ALT-803.
In principle, cell proliferation would be shown by dilution of the dye upon visualization by flow cytometry of the harvested NK cells. The control (media alone) samples clearly did not undergo as much proliferation and were present in smaller numbers than the treated samples. This was thought to be the combined result of increased cell death with diminished proliferation. Compared to the NK cells from healthy volunteers, the ascites-derived NK cells had more variable results to treatment with both IL-15 and ALT-803, although they clearly had proliferation with both treatments. “This indicates that ALT-803, which has superior in vivo pharmacokinetics to monomeric IL-15, could be used to drive expansion of peritoneal NK cells in ovarian cancer patients,” Felices said.
In Vitro NK-Cell/ALT-803-Ovarian Cancer Assays
To determine the ability of ALT-803 to enhance the function of NK cells against ovarian cancer, cells were taken from healthy volunteers and divided into two groups. The control group NK cells were incubated in vitro for 4 hours with a panel of five different human ovarian cancer cell lines, and the other group was incubated overnight with ALT-803 prior to the 4-hour incubation with the cancer cells. Cell function was assessed via flow cytometry as shown by CD107a expression, used to determine NK cell degranulation, on the surface of the NK cells as well as the intracellular levels of two relevant inflammatory cytokines, IFNg and TNFα. The control group showed little degranulation or cytokine production after co-incubation with the panel of ovarian cancer cell lines; however, those NK cells incubated overnight with ALT-803 prior to treatment with the cancer cells showed potent expression of CD107a and production of IFNg and TNFα. This enhanced NK cell function was shown to translate into cancer cell killing via a radioactive chromium (51Cr) release assay. Prior treatment with ALT-803 produced enhanced killing against three of the tested cell lines (A1847, SKOV3, and MA-148).
When incubated in a medium containing the fluid from the ascites from ovarian cancer patients, the function of NK cells from healthy donors was substantially diminished. Treatment of these cells with K562 myelogenous leukemia targets, used to test NK cell function, resulted in reduced degranulation and cytokine production. However, when NK cells were co-incubated with ALT-803 and the ascites fluid, substantial degranulation and IFNg and TNFα production were observed upon K562 product exposure. “This result unequivocally showed that ALT-803 can overcome the challenges presented by a soluable immunosuppressive tumor microenvioronment,” Felices explained.
In Vivo ALT-803/NK Tests
To confirm in vivo activation, human NK cells from a CD3/CD19-depleted product were injected with or without ALT-803 i.p. into immunodeficient NSG mice. After 3 days, the mice were sacrificed and NK cell activation was assessed. Those cells treated with ALT-803 showed greater activation via increased CD69, CD25, and CD56 expression.
In vivo treatment of ovarian cancer was evaluated with a xenogeneic mouse model that utilized an MA-148 cell line that expressed luciferase, which permitted real-time monitoring of the tumor via luminescence measurements. Although all NK cell treatments showed significant reduction in tumor growth, the addition of i.p. ALT-803 produced the greatest effect.
“We also evaluated subcutaneous dosing of ALT-803, and the results for this mode of delivery were similar to those for i.p. administration,” Felices explained. It should be noted the results obtained for the in vivo models were not as robust as those seen in the in vitro tests. “The in vivo results were obtained in immunocompromised mice, and thus the immunotherapy depended solely on the action of the NK cells; the ideal comparison would be in an immunocompetent humanized mouse model,” he stated.
PD-L1 Expression & Pembrolizumab
As a consequence of activation, NK cells release IFNg, an inflammatory cytokine. This poses a potential pitfall for cancer therapies. “The presence of IFNg can trigger the expression on cancer cells of PD-L1, the endogenous ligand for the immunosuppressive PD-1 receptor,” Felices noted. “IFNg induced PD-L1 expression on all of the ovarian cancer cell lines we used.”
To counteract this immunosuppressive mechanism, in vivo experiments were performed with the anti-PD-1 monoclonal antibody pembrolizumab. Coadministration of pembrolizumab with ALT-803 in their xenogeneic mouse model showed the best control of tumor growth, outperforming single line therapy (i.e., ALT-803 or pembrolizumab alone). “Theoretically, one could also utilize anti-PD-L1 antibodies to disrupt the immunosuppressive PD-1/PD-L1 interaction,” Felices explained.
When asked about the next steps for this promising therapy, Felices commented, “Most likely we will look to perform phase I/II studies to determine the maximum tolerable dose and/or dose-limiting toxicities as well as evaluate its efficacy for relapse prevention.”
Richard Simoneaux is a contributing writer.