Herbst Highlights Next Steps With Immunotherapy in Lung Cancer

Roy S. Herbst, MD, PhD, discusses where future research is headed for immunotherapy in lung cancer.

Roy Herbst, MD, PhD

Immunotherapy has garnered major traction as a treatment approach for patients with lung cancer, but there are still several unanswered questions that need to be addressed in this space, according to Roy S. Herbst, MD, PhD.

Because only a small percentage of this patient population benefits from immunotherapy, ongoing work is being geared toward acquiring a stronger understanding on the different mechanisms of resistance to this approach. For example, about half of patients with lung cancer develop resistance to immunotherapy because they have “cold” tumors; as such, investigators need to identify better strategies to bring immune cells into these tumors.

OncLive: Please provide an overview of your presentation on emerging therapies in lung cancer.

In an interview with OncLive at the 2018 New York Lung Cancers Symposium, Herbst discussed where future research is headed for immunotherapy in lung cancer.Herbst: I spoke about the future of systemic therapy, and I must say, the future is really all about immunotherapy. We have seen a paradigm shift in the last 5 years or so with immunotherapy taking the patients with the most advanced stage IV disease and significantly prolonging their survival. We have a patient at Smilow Cancer Hospital who has been alive for 8 years with stage IV disease thanks to immunotherapy.

I should mention that it is really only 15% to 20% of the patients who benefit extremely well from this approach; many others do not. Therefore, what is next? How are we going to find ways to target primary resistance and acquired resistance? What are the combinations we will use?

Could you expand on the ongoing work of overcoming immune resistance to immunotherapy?

We have to combine immunotherapy with chemotherapy and other immunotherapy agents. I made the point that we have to understand the entity; we have to understand the immune microenvironment, so we can best target it. We need to personalize immunotherapy. This type of treatment is phenomenal, but just like targeted therapy or radiation therapy, it needs to be used in a personalized way. We also have to be patient—it is still early in this endeavor—and we need to continue bringing science into the clinic. It is an exciting time because we are helping patients, but we need to continue to improve.There are multiple mechanisms. Generally, patients with lung cancer develop resistance to immunotherapy because they have “cold” tumors—their tumors are not inflamed and are without T cells. For those 50% or 60% of patients, we need to figure out ways to bring immune cells into the tumor, possibly by trapping them with cytokines or using other agents to modulate the microenvironment. That would be the strategy there.

Where do we stand in terms of biomarkers in lung cancer?

Another issue is that some tumors do not use PD-L1 as a primary checkpoint; maybe there are other checkpoints involved. If we knew that, we could use some of the other agents out there, such as LAG-3 and TIM-3. On the other hand, there are even some tumors that have everything you want—the T cells and the PD-L1 expression—and they are still resistant to immunotherapy. This is because of myeloid-derived suppressor cells, inhibitory macrophages, or regulatory cells, and so, we need to look at strategies to overcome this as well. We have a lot of work to do, but you can see even from the interest here at the 13th Annual NY Lung Cancers Symposium that we can get it done.PD-L1 is a good biomarker—you have to have PD-L1 in order to block PD-1 or PD-L1 interaction. The problem is that it is heterogeneous and hard to measure; it is gamma interferon upregulated, so depending on when you measure it, you might get a different result. That has certainly been a major issue.

What does a diagnosis of lung cancer mean today versus 5 years ago?

Other biomarkers such as TMB have been helpful to some extent, although the data on that are still mixed. We are still awaiting phase III data. The immune microenvironment is another biomarker we could use—looking at the percentage and type of T cells. The genetics of the tumor are also important.A diagnosis of lung cancer is still a tough one. No one is quite willing to say that we can cure advanced disease yet, although we are coming close in some patients. There is certainly more hope than there was in the past. I remember as recently as the 1990s that if you had metastatic disease, it was uniformly fatal.

Of course, with the targeted agents, we have had some effect, but resistance remains a problem to this day. With the immunotherapies, we are seeing some long-term benefit. With this increased hope, unfortunately, we might be setting the bar a little too high. There is still so much resistance and other problems we need to work out.


PD-L1 is an important biomarker for immune response, added Herbst, who is chief of medical oncology and a professor of Medicine at Yale Cancer Center, Smilow Cancer Hospital, but there are still some issues with it that need to be worked out. Additionally, more phase III data are needed before tumor mutational burden (TMB) can be considered a standard biomarker in lung cancer.