Vaccine and immunotherapy strategies for cancer and viruses causing cancer

Division Of Clinical Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

The strategies above involve 5 steps that together comprise a push-pull approach, to optimize antigen structure, improve quantity & quality of the response, & remove regulatory barriers. Regulation of tumor immunity by NKT cells: NKT cells are true T cells restricted by a non-classical class I MHC molecule, CD1d, which presents glycolipid antigens. We discovered a novel immunoregulatory pathway in which NKT cells suppress tumor immunosurveillance, using IL-13 to induce myeloid cells to make TGF-b that suppresses immunity. Translating, we completed a phase I trial of anti-TGF-b as a new checkpoint blocker with clinical benefit in melanoma patients. We discovered synergy between TGF-b blockade & 2 types of cancer vaccines in mice. We found that blockade of TGF-b1 & 2 (without 3) is sufficient to enhance immunosurveillance & vaccine efficacy, further amplified by PD-1 blockade. Type I (invariant TCR) NKT cells promote tumor immunity, whereas we found that type II (diverse TCR) NKT cells suppress tumor immunity. These subsets cross-regulate each other, defining a new immunoregulatory axis. We are investigating the relationship between this axis & other regulatory cells & molecules. We found both type II NKT cells & Treg cells can suppress tumor immunity concurrently, but type I inhibit type II NKT cells, leaving Tregs as the dominant suppressor unless type I NKT cells are blocked or absent. Thus, the balance between 2 regulatory cells is determined by a 3rd cell that regulates the regulators. Moreover, we found distinct regulatory cells dominate in the same tumor in the lung & skin. The effector cells also cross between tissues in one direction only. Thus, tissue context determines cancer immunity even for the same tumor, implying that immunotherapy for primary tumors & metastases in different tissues may need to be different, not widely recognized by oncologists. We succeeded in making sulfatide-loaded CD1d tetramers that detect a subset of type II NKT cells in the liver & lung and have characterized these cells, showing differences in markers, transcription factors & RNA expression. We are also trying to characterize the subset of type II NKT cells that respond to sulfatide by examining the cells that respond in vitro in different tissues. We are also performing a structure-function study of sulfatide analogs to activate or inhibit type II NKT cell activities. One sulfatide analog stimulates type II NKT cells when not processed, but type I NKT cells when processed in endosomes by DCs, and under that condition can prevent instead of promote cancers. We identified a new class of agonist for type I NKT cells, b-ManCer, that inhibits tumors by a mechanism distinct from that of a-GalCer, requiring TNF-a & NO synthase instead of interferon (IFN)-g. This agonist also synergizes with a-GalCer, is much less anergy-inducing than a-GalCer, & stimulates human NKT cells also. Studies using structure-specific antibodies showed that b-ManCer could assume a structure resembling a-GalCer. All of these studies are aimed to remove the roadblocks and/or improve the balance along the type I-II NKT axis to allow cancer vaccines to successfully induce tumor regression. Epitope enhancement in cancer vaccine strategies & translation to clinical trials: We first devised and carried out epitope enhancement (sequence modification to improve MHC binding) on an HLA-A2-binding epitope we discovered in a novel prostate & breast cancer antigen, TARP. The enhanced epitope induces human T cells that kill human tumor cells. We translated this to a phase I clinical trial of 2 peptides in stage D0 prostate cancer. 74% of vaccinees had a decreased PSA slope & tumor growth rate at 1 year (p = 0.0004). A randomized placebo-controlled phase II study is ongoing. All 6 vaccinated patients in the safety lead-in had flat or reduced PSA slope. We studied preclinically and published a novel adenovirus-based HER-2 vaccine expressing the extracellular (EC) and transmembrane (TM) domains of rat neu (ErbB2), which prevents tumor growth in the neu-transgenic mice, and cures large established TUBO mammary tumors (2 cm) & established lung metastases. The therapeutic effect in mice is purely antibody mediated, through inhibiting ErbB-2 function, unlike trastuzumab, which is FcR dependent, so may work in trastuzumab failures. We have translated to humans with an autologous DC vaccine transduced with an Ad5f35 vector expressing the ECand TM domains of human HER2 and have now completed a phase I/II trial of the human version of this vaccine. At vaccine doses of 10-40 million DCs, 7/16 (44%) evaluable patients showed clinical benefit (CR, PR, or SD) and complete safety. A new study combining this with checkpoint inhibitors is being planned. We have also compared combinations of checkpoint inhibitors and anti-TGF-b in enhancing cancer vaccine efficacy in mice, and found preliminary data for synergy among anti-TGFb, anti-LAG3, anti-TIGIT & anti-PDL1. Without the vaccine, the others don't work without anti-TGFb, but when used without anti-TGFb, the other checkpoint inhibitors require the vaccine for efficacy. Thus, the vaccine is essential but the other inhibitors of negative regulation amplify the efficacy in mice. We invented a new modified a-GalCer (PLS-a-GalCer) that is more potent in treating mouse tumors, and are investigating the mechanism. We also carried out a CRADA-collaborative study in mice of an intratumoral therapy that we found induces a T cell response necessary for regression and induces memory & resistance to rechallenge. This has also been translated to the clinic by our collaborators. Recently, we have begun examining the effects of the balance between cholesterol and omega-3 fatty acids in regulating tumor growth in mice. Cytokines as vaccine adjuvants for induction of high avidity T cells. Our earlier work showed that high avidity T cells were more effective at clearing viral infections & cancers. We found ways to induce them with cytokines & TLR ligands. The quality of response proved more important than the quantity. We recently found, using a novel adjuvant, CAF09, that we could lower the vaccine dose sufficiently to induce higher avidity CD4 T cells to better clear virus infection. We also found that IL-1b induces Th17 helper cells that do not help Tc1 CD8 T cells that protect against vaccinia virus. Rather, they skew the CD8 response to Tc17 cells that make IL-17 & do not protect. TGF-b blockade can prevent this problem. We also examined combinations of cytokines & TLR ligands as vaccine adjuvants and found greatest efficacy of IL-15 + TLR3 and TLR9 agonists, but also some efficacy of IL-12 + GM-CSF. Mucosal immunity, microbiome: We are also examining the effect of the gut microbiome on cancer as well as HIV. We have an ongoing project in which we are examining the effect of NKT cells on the gut microbiome and the downstream effect on cancer. We have bred Ja18 knockout (KO) mice that lack type I NKT cells only, and CD1d KO mice that lack both type I and type II NKT cells, along with wild type B6 mice that have both types. We have collected their gut microbiome samples at ages 14, 23, 30, 60 and 90 days, and done 16S RNA sequencing in collaboration with Giorgio Trinchieri and the CCR Microbiome Core. We are analyzing the changes in microbiome with age. Then, we have cohoused some of these pairwise from day 30 to day 90 and examined the ability of microbes to be transferred. This could provide an indirect mechanism by which the presence or absence of different types of NKT cells can affect cancer growth through altering the microbiome.