Following up on our earlier post, in which we discussed strategies from Elizabeth Jaffee, M.D., for converting pancreatic cancer into a cancer that can be treated immunologically, we will now cover talks given by Jedd Wolchok,M.D.,Ph.D., an associate director of CRI’s Scientific Advisory Council (SAC) and CRI’s clinical director; Drew Pardoll, M.D., Ph.D., and Robert Vonderheide, M.D., D.Phil., both members of CRI’s SAC; and Priti Hegde, Ph.D., that also focused on how we can better set up the immune system for success against various cancers.
As Hegde showed, anti-PD-1 checkpoint inhibition works well much of the time in inflamed bladder tumors that express PD-L1 and already have immune T cells in place, as well as non-small cell lung cancers with high IFN-γ signatures that are associated with effector T cell activity. When there is a pre-existing immune response, checkpoint inhibitors are great at nudging that response in the right direction.
Unfortunately, tumors that lack inflammation and are immunologically “quiet”--often because of immunosuppression--do not respond as well.
In several types of cancers, subtypes of tumors show high levels of stromal cell markers, which have been associated with suppressed immune activity in the presence of TGF-β, an immunosuppressive cytokine. In a trial in which 28 melanoma patients were treated with anti-PD-1 therapy, patients who did not respond to treatment had higher levels of stromal markers such as TGF-β.
In bladder cancer too, different subtypes of tumors are seen that react differently to anti-PD-1 treatment. Immunosuppressive tumors are resistant to treatment, while tumors that lacked immunusuppressive cells and had low PD-L1 expression responded very well: 50% of those patients saw clinical benefit, and the response rate in the neighborhood of 30% is far above the historical rate of about 8%.
Dr. Hegde finished her talk by proposing a tumor immunity continuum-framework that can be used to guide combination therapies that address the characteristics of specific tumor subtypes. Tumors characterized by CD8+ T cells, IFN-γ signaling, and PD-L1 are usually responsive to anti-PD-1 therapy, but the stromal-defined tumors characterized by immunosuppressive myeloid-derived suppressor cells (MDSCs) and angiogenic factors are not. Combination treatments--that target both the immunosuppressive MDSCs as well as activate T cells--could be used to make these tumors more inflammatory and responsive to anti-tumor responses.
Dr. Wolchok, who oversaw the trials that led to the first and only FDA approved immunotherapy combination--the anti-PD-1 nivolumab and the anti-CTLA-4 ipilimumab--next explored how these MDSCs might be targeted to improve responses in melanoma, and stressed that complementary therapies will be crucial to achieving immunotherapy’s full potential.
In metastatic melanoma patients, Dr. Wolchok noted that many tumors possess these MDSCs as well as IDO, a checkpoint pathway that dampens immune responses. In mice, tumors genetically altered to express IDO grew more aggressively and showed MDSC accumulation. MDSCs were found to be recruited by regulatory T cells (Tregs), and when those Tregs were depleted, the recruitment of immunosuppressive cells was inhibited. Additionally, these immunosuppressive cells could be deactivated by targeting one of their receptors.
Preventing MDSC-mediated immunosuppression aids anti-tumor immunity, and in combination with anti-CTLA-4 therapy, MDSC inhibition was able to effectively treat these IDO-producing tumors. Importantly, the timing of MDSC inhibition was crucial to its ability to improve responses in combination with checkpoint immunotherapy.
Next, Dr. Vonderheide began his talk by echoing many of the same insights as Dr. Jaffee earlier in the day, including the discoveries that several different cell types can suppress T cells, including B cells. He also pointed out that a small proportion (10%) of pancreatic tumors do have T cell infiltration, and that these patients typically live longer. He then went on to highlight new clinical trials that are investigating several combination treatment strategies against pancreatic cancer, to treat the other 90% of pancreatic tumors that don’t have a pre-existing immune response.
One strategy employs the combination of GVAX, a vaccine consisting in this case of pancreatic cells genetically modified to express GM-CSF; the chemotherapy cyclophosphamide to deplete Tregs; and CRS-207, a weakened form of Listeria monocytogenes that has been engineered to express a cancer-associated protein called mesothelin. These will be used with or without the anti-PD-1 antibody nivolumab. This treatment has already produced a partial response in a patient that received nivolumab and who initially exhibited pseudoprogression, or false tumor growth that results from an infiltration of immune cells to the tumor site.
Another trial employs an anti-CD40 antibody, which helps to stimulate activation of various immune cells, with or without addition of the chemotherapy drugs gemcitabine and paclitaxel.
Dr. Vonderheide closed by enthusiastically noting that these trials are experiencing strong patient enrollment, and is hopeful that some of these strategies will be shown to trigger T cell infiltration into pancreatic tumors.
Finally, Dr. Pardoll explored several diverse topics and discussed several technologies that might allow us to improve the effectiveness of immunotherapy, based on research he and others conducted as part of the CRI-SU2C Cancer Immunology Dream Team.
He noted that tumors can develop resistance to anti-PD1 therapy in response to IFN-γ activity that acts to increase immunosuppressive PD-L1 expression on cancer cells. This PD-L1 expression is mediated through the JAK-STAT signaling pathway, and that if JAK is inhibited, the process can be short-circuited and PD-L1 expression can be prevented.
Dr. Pardoll also highlighted a study which showed that anti-CTLA-4 immunotherapy prior to surgical resection was able to convert previously immunologically “quiet” pancreatic tumors into immunogenic tumors characterized by increased PD-L1 expression and infiltration of both CD4+ and CD8+ T cells.
A new study looking at non-small cell lung cancer employed a nanotechnology and microfluidics approach to follow T cell profiles in the peripheral blood. By determining what T cells are “seeing” both before and after treatment, it was observed that while many T cell receptor clones are patient-specific, some are shared between patients. Dr. Pardoll declared that the odds of this happening by chance as “essentially zero” and suggested that there is likely an important mechanism behind this that future studies should investigate.
To conclude, Dr. Pardoll discussed adoptive T cell therapy, and explained that one of the challenges is encouraging persistence in the T cells, through combination therapies that enable the cells to survive long-term and promote durable anti-tumor immunity.
Up next in today’s report from #AACR16: CAR T cells and the personalized immunotherapy revolution.