The Future of Cancer Immunotherapy Lies in Combinational Approaches

Whole-cell vaccines, antigen-specific vaccines, and neoantigen vaccines could represent the next wave of immunotherapeutic advances in gastrointestinal (GI) malignancies, explained Arsen Osipov, MD, who added that finding the right adjunctive therapies to combine with these agents will be key to eliciting deep and durable antitumor immune responses.

“We often get pigeonholed into thinking about a particular agent or a particular approach, but it’s really going to come down to not one drug, but a combination, particularly for tumors that are responsive or not responsive to immunotherapy,” said Osipov.

In an interview with OncLive® during an Institutional Perspectives in Cancer webinar on gastrointestinal malignancies, Osipov, an assistant professor of medicine at Cedars-Sinai, discussed the potential for ongoing research with immunotherapeutic approaches in gastrointestinal malignancies, including CAR T-cell therapy, bispecific T-cell engagers (BiTEs), and vaccines.

OncLive®: Where could cancer vaccines make the greatest impact in oncology?

Osipov: There have been various iterations of vaccine therapy. We have whole-cell vaccines, antigen-specific vaccines, and the newer iteration are the neoantigen vaccines, which have a lot of potential. When you have a wholesale vaccine, like GVAX, for example, it’s a great vaccine, but its immune response is broad, and that may not be enough to elicit a very specific response towards the tumor.

Now, when you think of neoantigen vaccines, specifically, the ones that are being developed for pancreas cancer and colorectal cancer that are targeting the most common mutation, which is KRAS—and there are multiple forms of KRAS, but these vaccines have the ability to target those specific KRAS mutations. As a result of that, the vaccine, because it goes to these novel epitopes, is more specific, and then as a result, it can elicit a better immune response. Now, by itself, the vaccine probably isn’t enough; you need an adjunctive agent, and usually that adjunctive agent could be a checkpoint inhibitor.

A lot of these are done in the setting of a clinical trial, and these are relevant for tumor types that we know are fairly aggressive, pancreas cancer being one of them. Despite the fact that a lot of these patients go to surgery, the rate of recurrence still remains at 80%.

These trials are a good thing to think about for your patients, particularly those who are interested in moving away from the standard of care or adding onto the standard of care, so we can reverse that statistic, so these patients can have durable, long-term remissions.

Where do you see CAR T-cell therapy headed in solid tumors?

CAR T-cell therapy has a lot of success in hematologic malignancies. That’s where all of the approvals exist, but CAR T-cell therapy has been investigated and continues to be investigated in solid cancers. The issue is that a lot of these solid tumors have characteristics that make CAR T-cell therapy unsuccessful, and that comes from the immunosuppressive tumor microenvironment. There are genetic instability changes that lead to antigen expression changes, and there’s a lot of tumor heterogeneity. If the therapy is fit to a very specific part of the tumor antigen, it may not work.

There is this phenomenon known as on-target/off-therapy reactivity, which entails this cross reaction between the healthy tissue that expresses the antigen of interest. That leads to tissue damage and significant cytokine release. The tumor microenvironment adds on to that. We’re trying to overcome this by designing CARs that are newer generation where you can actually, within the design of the CAR, overcome the tumor microenvironment or combine it with another agent.

There are CARs that are designed to antigens that are highly expressed in GI malignancies—for example, CEA, MUC1, and HER2, which are in active investigation. The idea is to try to design CARs that can overcome the antitumor environment.

There are these newer fourth-generation CAR T-cells, known as armored CARs, and they’ve been engineered to have a constituent secretion of interleukin-12 in order to overcome the tumor microenvironment and enhance the antitumor efficacy that you need for solid tumors rather than hematologic malignancies. We’re not quite there yet, but it’s a huge area of investigation.

Do bispecifics present some of the same challenges as CAR T-cell therapy?

Bispecifics fall within the same space as CAR T-cell therapy because they are T-cell–directed therapies. You think of BiTEs as a bridge. You have a fragment of the T-cell receptor, then the other one to the antigen, you bring the T-cell close, and that activated T-cell leads to an antitumor response. The only one that is approved is blinatumomab [Blincyto], which is a CD19-targeted agent for B-cell acute lymphoblastic lymphoma.

However, we run into the same issues with BiTEs as with CAR T-cell therapy in solid malignancies. The tumor microenvironment has these immune escape mechanisms. The BiTEs by themselves will probably not be enough in solid tumors, and pretty much most of the studies that exist as of today, are looking at synergistic combinations with oncolytic viruses or checkpoint inhibitors.

How can we overcome some of these challenges? Will it require a closer look at combinatorial and sequencing strategies?

One of the most rational combinations that has led to multiple FDA approvals is anti–PD-1/anti–CTLA-4 combinations. That combination allows for T-cell priming. The CTLA-4 inhibitor has a different mechanism of action and acts predominantly in the lymph node, whereas PD-1/PD-L1 inhibitors act at the site of the tumor. Although this combination leads to improvements in certain tumor types, it comes at the cost of toxicity.

Checkpoint inhibitors and radiotherapy [could be another synergistic combination] that could lead to a nonredundant improvement in T-cell response and T-cell priming. Probably the best place where the combinatorial strategies are important aren’t tumor types where single-agent immunotherapy or dual checkpoint blockade does not work. We have to figure out how we can emulate the responses we see in melanoma or mismatch repair deficient colorectal cancer with checkpoint inhibitors.

There are 2 ways to overcome some of the obstacles we’re seeing in doing that: one is the sequencing of therapy, and two is the combination itself. Pancreas cancer is a good example of that. We know pancreas cancer has a very dense stroma. It has all these immunosuppressive elements through these myeloid cells. Therefore: Is there a way where we can attack both and then potentially add a checkpoint inhibitor to get a better effect? A study at Cedars-Sinai is giving patients chemotherapy, then a FAK inhibitor, which is a tumor microenvironment modulator. That agent has effects that we know of from our preclinical models, as well as a lot of the translational work we have done on the stroma as well as immunosuppressive cells. You want to decrease those bad immunosuppressive cells, and then give an agent that will then lead to a better immune response. That’s the premise behind combinatorial strategies. You have to attack it on every front in order to turn tumors that otherwise don’t respond to immune therapy to ones that could potentially respond.

Sequencing is also important. Why we should still think of chemotherapy as part of our treatment paradigm is because it can synchronize the tumor in preparation to for it to receive immunotherapy. It also allows us to study these patients in a more uniform fashion by exposing them to a similar agent. We think that some tumors will never respond to immunotherapy, but that’s really not the case. We just need to do more research and find out how we can change the tumors for them to become responsive to immunotherapy.

In melanoma, non–small cell lung cancer, and microsatellite instability–high colorectal cancer, immunotherapy has changed the paradigm and changed patient outcomes with the most advanced diseases. There is no reason in my mind that we cannot emulate that in other tumor types, such as pancreas cancer.

The challenges are different, but the question is: How do you overcome those other obstacles, whether it’s the tumor microenvironment or these inherent genetic changes? The only way to really do that is through good research through these combinatorial strategies. Even though we have research in each one of these areas––CAR T, BiTEs, vaccines, oncolytic viruses––at the end of the day, the solution is the rational combination of those agents, along with even chemotherapy or radiation.