John Haanen, MD, PhD, discusses why the best combination regimens and the ideal patient population for immunotherapy treatments in lung cancer are still largely unknown.
John Haanen, MD, PhD
Immunotherapy is more than just hype, says John Haanen, MD, PhD, head of the Division of Medical Oncology, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital. However, for oncologists to apply it to its full potential in lung cancer, a deeper understanding of how and for whom it works is needed.
“Immunotherapy in lung cancer is here to stay, and it will be more and more important in the coming years—as a single agent, but probably have more significance in combination,” says Haanen. “Therefore, understanding the background and the mechanism of action of immunotherapy is going to be more important if you want to utilize it. It will explain why immunotherapy can or cannot work, and it will also explain what you can expect, as far as toxicities are concerned.”
In addition to safety, several other challenges regarding immunotherapy remain in lung cancer, including how the dynamic nature of PD-1/PD-L1 works, the role of PD-L1 testing, and what effective biomarkers will emerge for immunotherapy in lung cancer.
OncLive: Is there a learning curve for oncologists, in terms of handling toxicities with immunotherapies?
The best combination regimens and the ideal patient population for those treatments are also still largely unknown. To shed light on some of these questions, OncLive spoke with Haanen, who focuses his research efforts on immunotherapy.Haanen: Anti—PD-1 drugs appear to be quite safe; toxicities are usually not that severe, and only 10% of patients will develop grade 3 or 4 toxicities. However, it is still the early days. I suspect that the longer we are able to give this type of drug, the more toxicity we will see.
In other types of tumors, there is accumulation of side effects occurring with so-called “safe drugs,” such as anti—PD-1 agents. Once we step into combination immunotherapies—for instance, the combination of CTLA-4 blockade and PD-1 blockade—the toxicity rates will suddenly increase. Then, we are confronted with toxicities that pulmonary oncologists have never seen.
What role does PD-L1 testing play in lung cancer?
It is important to realize that, if you start combining different things, more challenges will happen. You need to have, at home, a team of doctors that can help you manage these toxicities. For instance, with combination therapy, you need an endocrinologist and a gastroenterologist, because patients might develop hypophysitis or colitis. You really need those experts to deal with that.PD-L1 testing is a very complicated story. We’ve learned that PD-L1 expression can be very dynamic, so it depends at what point in time you look at PD-L1. We know that PD-L1 expression by tumor cells within the tumor microenvironment can be induced by an interaction between the immune system, the T cells, and the tumor. Therefore, when T cells are arriving at the tumor site, they are activated and they may recognize the tumor.
Because of that, they may induce PD-L1 expression. It is not something that is always there. There is oncogenic-driven PD-L1 expression, which is just due to an oncogenic event that occurs in the tumor. Mostly, we think that PD-L1 expression is induced by an ongoing immune response. Knowing that means that PD-L1 expression tells you something about the interaction between the immune system and the tumor, so it could be a very positive indication that an anti—PD-1/PD-L1 drug may work.
However, because it is dynamic and a complicated test, it means that we may see heterogeneous expression or expression may come up later after we have already looked. If we use tumor material from 1 year before, the expression may be completely different by the time we give the drug.
Without a clear biomarker, how should oncologists decide which patients should receive immunotherapeutic agents?
Those are things we have to realize when we think about PD-L1 expression. If you look at the outcomes of patients, you see a difference between those with low PD-L1 expression versus high PD-L1 expression. However, there is not a clear-cut threshold for the PD-L1 expression you need, in order to get a good response to an anti¬—PD-1/PD-L1 drug. Therefore, I don’t think PD-L1 expression is a very good biomarker to select on.If you look at the targeted agents, especially in non—small cell lung cancer, you need a mutation for them to be very effective. In immunotherapy, it is much more complex. We need several things. We know that the tumors that are highly mutated respond better, so the smokers with EGFR and ALK mutations will have a higher response rate.
We know that T cells are very important. If there are no T cells in the tumor, chances are the immunotherapy is not going to work. It is different. For the mutated cancers, where we have targeted agents, it is quite simple. The biomarkers are obvious.
What is on the horizon for immunotherapy in lung cancer?
Without good biomarkers in immunotherapy, we really have to do our best and find biomarkers, because we know that there are a certain number of patients who will never respond to immunotherapy. Moreover, we do not want to expose those patients to a toxic, and perhaps, very expensive treatment. If we are able to select patients who are not responding and had no chance of responding, we will really make a leap forward in the selection process.If we look at the future, I think that a combination of therapies is certainly something that we are moving toward. That can be combinations with a lot of things. There are many new drugs in development. There are drugs that are targeting the immune system, stimulating the immune system, or taking away the breaks, but there are also classical drugs like chemotherapy. We consider chemotherapy to be a cytotoxic agent, but I think a lot of chemotherapies also have immunologic effects. We are trying to learn what these are, and these may differ from chemotherapy to chemotherapy.
Radiotherapy also has some effect on the tumor microenvironment and targeted agents do, too. These targeted agents may work on the tumor, but they may also have an effect on the immune system itself. For instance, these work for BRAF inhibitors with BRAF-mutated tumors. This may block the proliferation of the tumors; at the same time, they have an effect on the T cells. T cells are very sensitive to BRAF inhibition and they may even stimulate their activity. Therefore, we need to know a little bit more about which combinations to make.
Adoptive cell therapy is something that we are also exploring. This is highly effective in patients with acute lymphoblastic leukemia, and there is a lot of ongoing research looking at them in solid tumors. Those can also be combined with checkpoint inhibitors. Once T cells are activated, they start expressing PD-1; if you combine it with an anti—PD-1 drug, you may add to the effect of the activation.
There are many combination possibilities, but one of the problems that we are facing is determining how to select patients for each combination. We don’t know that at the moment; we need to get as much information on patients as possible. Therefore, we need to collect material, blood, and biopsies not only prior to treatment, but also once they have failed treatment. We need to get other material, such as feces and plasma, and do as much research as possible on all of these materials in order to find out what a patient at a specific point in time needs. With that, we will hopefully get a little bit further.