Engineering Immune Response With In-Situ Vaccines in iNHL

Joshua Brody, MD

In-situ vaccines could be a novel way to harness immune response and extend the reach of immunotherapy, explained Joshua Brody, MD, citing a study in Nature Medicine, which showed antitumor T-cell responses in patients with advanced-stage indolent non-Hodgkin lymphoma.

In the study, investigators injected Flt3 ligand, radiotherapy, and a TLR3 agonist into lymphoma cells, drawing antigen-loaded and activated dendritic cells to the tumor.

“There’s no question that the immune system has the ability to get rid of our patients’ tumors and provide long remissions. We shouldn't shy away from [imagining] cures either,” said Brody. “All we need to do is optimize how we teach the immune system to recognize lymphoma cells and how to avoid some of these resistance mechanisms.”

In an interview with OncLive, Brody, assistant professor of medicine, hematology and medical oncology, and director of the Lymphoma Immunotherapy program at the Icahn School of Medicine at Mount Sinai, discussed research with in-situ vaccination in non-Hodgkin lymphoma and strategies being developed to overcome acquired resistance.

OncLive: Could you discuss the potential utility of in-situ vaccines?

Brody: We presented data from a couple different projects we're working on in the lab and in early phase clinical trials. The first part is studying in-situ vaccination where we’re trying to develop a vaccine [that can be delivered] to a patient's tumor. When we say, “cancer vaccines,” we don't mean preventive vaccines for measles, mumps, or rubella. All cancer vaccines are therapeutic vaccines or vaccines that we use to teach a patient’s immune system to [kill] cancer.

Many types of cancer vaccines exist. The in-situ approach tries to take advantage of the fact that we already have the things we want the immune system to recognize right at the tumor site. The tumor antigens are collected there. If we can just mobilize [those antigens] in an immunogenic way, we can make the immune system recognize them and we can induce this immune response at 1 tumor site. Then, those antitumor T cells can travel throughout the body to [eliminate every] tumor. It's really not that different from other vaccines, such as the flu vaccine, which you receive in your shoulder. If someone coughs on you the next month, you don't put your shoulder in front of their mouth because you are systemically protected against [the virus]. It’s the same idea.

Could you elaborate on the preclinical research that suggests the potential for in-situ vaccines?

Our work stems from preclinical work that was dedicated to making patients’ immune systems recognize these tumor antigens. The approach pivots on this critical immune cell called the dendritic cell. We inject a dendritic cell differentiation factor called Flt3 ligand into the tumor. We all have a Flt3 ligand. Just like we say erythropoietin is for red blood cells, Flt3 ligand is for dendritic cells; it's the primary growth and differentiation factor of dendritic cell precursors. We inject Flt3 ligand into the patient's tumor, which attracts dendritic cells to the tumor site. Then we load those dendritic cells with tumor antigens. You can do that in many ways. Very simply, we did it with a very low dose of radiotherapy. Then, we activate those dendritic cells with toll-like receptor (TLR) agonists which, when injected into the site, activate the dendritic cells. That [process] lets us develop the vaccine right at the tumor site.

We published our findings about 1 year ago in Nature Medicine. Some patients had partial and complete remissions in their systemic tumor burden, some with very bulky tumors throughout the body. By treating 1 site with an in-situ vaccine, we could [eradicate] their lymphoma. Some of those remissions have been going on for years, so it has been very gratifying for us and, of course, for those patients.

How is the vaccine tolerated?

It’s been satisfying to see that, as we’d expect with preventive vaccines, [therapeutic] vaccines are safe. Some people are developing transient fevers that last maybe half a day or 1 day. If patients take a couple of Tylenols and drink some fluids or Gatorade, even in patients who had high fevers, [fevers] were pretty transient. Most of the fevers took place after 1 of the injections rather than throughout the [time they were administered].

To date, no one has developed any sort of autoimmune-like adverse event (AE), which is a big concern with other immunotherapies, mostly with checkpoint inhibitors where we can see autoimmune-like colitis, dermatitis, hepatitis, and pneumonitis. We haven't seen anything like that with the vaccine approach, and we haven't seen any of the other inflammatory-like immune AEs that we see with CAR T-cell therapy, such as cytokine release syndrome or neurotoxicity. Overall, [the vaccine] has been pretty safe.

What is being done to address acquired resistance?

Despite these findings, we [know] patients’ tumors can become resistant to vaccines or any type of immunotherapy through antigen escape. With CAR T-cell therapy, it can mean CD19 loss from the tumor. With bispecific antibodies against CD20, it can be CD20 loss. With every other type of immunotherapy where T cells recognize tumors through their major histocompatibility (MHC) class 1 expression, it can be MHC class 1 loss; this has been well described for many types of cancer after receiving a PD-1 inhibitor. Class 1 loss is probably one of the most common types of resistance to checkpoint inhibitors.

We started to observe that in some of these patients who had been treated with the vaccine. We talked about some data we developed in the lab trying to address antigen escape. There are many ways to solve the problem of antigen escape in CAR T cells. After giving an anti-CD19 CAR T cell, we can try to treat with an anti-CD22 CAR T cell in case [the patient] lost the CD19 antigen. We have come at it from a very different approach. If a tumor cell has lost every possible antigen that we could target, such as CD19, CD22, or MHC class 1, we still believe there is something you could target [even without] the antigen loss, specifically, the geography [of the tumor]. It is geographically localized within a tumor mass that has the antigens. We were very lucky to discover bystander killing. T cells kill the target cell that they are geared to via antigen recognition; however, there’s also an ability to kill just the cells next door. Bystander killing has been known about for some time, but a very brilliant student in my lab discovered that bystander killing is exclusively dependent on 1 molecule called Fas.

Tumor cells that express Fas are sensitive to bystander killing and tumor cells that don't express Fas are completely resistant to bystander killing. We know a lot about how Fas signals, and we can modulate Fas signaling; we can increase it. Preclinically, we could increase Fas signaling in tumor cells using a couple of FDA-approved small molecule inhibitors like BCL-2 inhibitors, HDAC inhibitors, and even inhibitors of the family of proteins called IAP. By inhibiting some of these Fas signaling regulators, we can increase bystander killing, and even kill tumors that had a mixture of antigen-negative cells. We believe geographically targeting antigen loss tumor cells with increased bystander killing may be a way to avoid antigen loss; it's very practical and very simple to do. We're very excited to see some of those positive results in the lab.

Even more excitingly, patients who were treated with CAR T cells in the ZUMA-1 trial and expressed high levels of Fas, were more likely to respond to therapy if they had high levels of Fas in their tumors. These patients were more likely to survive 1 year after receiving the CD19-directed CAR T-cell therapy. Already, we’re seeing that Fas expression in the tumor is very important in the real world. If we can modulate Fas signaling, we believe we can prevent tumor relapses from some of these immunotherapies.

How do you see immunotherapy evolving in the coming years?

In lymphoma, we're very lucky to already have so many good therapies, including chemotherapy, antibody therapy, small molecule inhibitors, and BCL-2, BTK, and PI3K inhibitors. What we really lacked was active immunotherapy, drugs like PD-1 inhibitors, which are incredibly effective in melanoma, lung cancer, and even other types of lymphoma like Hodgkin lymphoma. In non-Hodgkin lymphoma, there was a real lack of those types of adaptive immunotherapies. Probably in the next year or 2, we'll have some more of these potent immunotherapies, the most obvious example being bispecific antibodies, which have shown very promising results in diffuse large B-cell lymphoma and some low-grade lymphomas like follicular lymphoma. We’re seeing very high remission rates and very durable remissions in some of these patients. With the bispecific antibodies, we’re still struggling with antigen loss. It’s nice to have developed some [agents that have the potential] to address antigen loss before it becomes a mechanism of tumor escape.

What is the key takeaway regarding immune therapy in lymphoma?

Different schools of thought used to exist. Either the immune system can get rid of our patients’ tumors or the immune system cannot. Now, there’s only 1 school of thought. With CAR T-cell therapy, we’re seeing patients with persisting, durable remissions. It’s almost certain that some of these patients are cured.

Hammerich L, Marron TU, Upadhyay R, et al. Systemic clinical tumor regressions and potentiation of PD1 blockade with in situ vaccination. Nature Medicine. 2019;25(5):814-824. doi: 10.1038/s41591-019-0410-x