Neoadjuvant Immuno-oncology in the Treatment of Locally Advanced CSCC

Glenn J. Hanna, MD: I’ll put up the phase 2 study of neoadjuvant cemiplimab before surgery in locally regionally advanced squamous cell carcinoma cutaneous. Dr Emerick makes a great point. Often we’re trying to look ahead. Yes, there’s a response rate, and it can be exciting, and we can see pathologic response. But there are scenarios where we have to think about the risk of progression on the drug and what surgical option would be left if we didn’t employ surgery out the gate. What’s the timing of that? How many doses do we think about, and when do we stop and pause and give that drug time to work? Maybe you can talk to us about the data for neoadjuvant I/O [immuno-oncology] in this setting.

Gino K. In, MD, MPH: First, neoadjuvant therapy is an increasingly important approach with the advent of immune checkpoint inhibitors. Historically, when we look at using neoadjuvant therapy in something like breast cancer, 1 of the big understandings was that if a patient achieves a pathologic response, that probably equates with longer recurrence-free survival and overall survival. That’s the goal. If you think about it scientifically, the rationale for neoadjuvant therapy builds upon this concept that if you have tumor burden in vivo, the immune system can recognize and allow for greater activation of a diversity of T cells that may allow for continued immune surveillance following surgery. We think neoadjuvant approaches may be better than adjuvant approaches alone in terms of greater immune response but also longer recurrence-free survival. That’s what we’re trying to study.

I agree with Dr Emerick. It’s important to carefully select these patients. If somebody has a rapidly progressing tumor—where you’re concerned about losing that window for resectability or where the patient is so symptomatic that they need surgery up front because it will be something that’s faster-acting—those are not patients who would be good for a neoadjuvant study. But in this phase 2 study, the investigators used neoadjuvant cemiplimab every 3 weeks at the flat dose of 350 mg. It was given for 2 cycles. Essentially, they postponed surgery by 6 weeks. So we needed to identify patients we think are appropriate for this. Patients who can wait 6 weeks for surgery with curative intent were selected.

The data are interesting. The primary end point was looking at overall response rate radiographically, by RECIST. The secondary end points were looking at pathologic response. What the investigator showed was that the overall response rate was 30%, which as we mentioned is a little lower than what we saw on the other prospective studies using either cemiplimab or pembrolizumab. After surgery what we saw was that pathologically there was a very high response, a pathologic complete response of 55%. Also, another 15% of patients had what was considered a major pathologic response. If you add those together at 70%, it does appear that a lot of patients had significant benefit from this approach.

The other intriguing point is that with only 2 cycles of PD-1, no patients had any grade 3+ AE [adverse event] or surgical delays. It’s very intriguing. It’s a small study, I believe, with only 20 patients. That’s something we need to explore further. You’re getting the best of both worlds when you’re getting a little immunotherapy combined with surgery. Hopefully with longer follow-up, we’ll see that these patients also have no recurrences. Something that’s not shown here is that of the patients who had pathologic responses to date, none of those has had recurrence yet. Hopefully the immune response continues as well.

Glenn J. Hanna, MD: That’s a great segue into my question for Dr Bakst. Let’s say a patient does benefit and has a nice pathologic response, whether complete or near complete. Are these data enough, or are they starting to compel you to think about deescalating the adjuvant therapy? Is there a role to decrease some of the radiation dose, or do we need to drop the chemotherapy in some instances? That’s the bigger question. Many of us on these kinds of trials initially will plan to treat with the intention of the original tumor. Whatever we saw when we started—if we see shrinkage, we still treat with that intention of whatever the staging was. I’m curious, as a radiation specialist in this space, how you start to think about interpreting these responses and whether you’ll adjust your postoperative treatment.

Richard Bakst, MD: It’s a great question. The first experience you had with this was breast cancer in neoadjuvant therapy. Postmastectomy, there’s no cancer because they still need postmastectomy radiation. The dogma has been if you’re going to treat with curative intent using standard doses. This space is so new but very encouraging, and I hope I’m going to give a different answer if you ask me the same question in 10 years. As it stands now, the answer is to treat. If you’re going to pull the trigger, treat with standard of care.

Glenn J. Hanna, MD: I’ll give Dr Cohen the chance to comment on his interpretation of these neoadjuvant data. It does seem as though there are several other immunotherapy agents and larger trials emerging in this space. What are your thoughts in context about how we use these agents and whether you think it will translate into improved outcomes?

Ezra Cohen, MD, FRCPSC, FASCO: I agree with everything that’s been said. Hard to know if single-agent cemiplimab, presumably taken into a comparative randomized study, would improve the critical outcomes of either progression-free survival or overall survival. We haven’t seen that in other diseases, except for possibly melanoma. Of course, the 1 thing that cutaneous squamous cell carcinoma and melanoma have in common is the tumor mutational burden. Maybe we’ll see that in this. However, what I’m eager to see—and this gets to Dr Bakst’s point—is combination immunotherapies in the neoadjuvant setting, where perhaps we can get even higher pathologic complete responses. Then we really begin to think about changing the locally advanced or the local therapies for these patients. Without a doubt, these are incredibly encouraging data. These strategies should go forward.

Glenn J. Hanna, MD: I appreciate the thoughtful and informative discussion. It was quite enriching. It’s always meaningful when we get specialists from the nonmelanoma skin cancer space and several different disciplines together. It illustrates what we deal with in terms of the complexity of these cases and how it illustrates that we need multidisciplinary involvement from almost all these cases. I thank you, the audience, as well. We hope you found this OncLive®panel discussion to be engaging and helpful to your understanding of the optimal management of patients with nonmelanoma skin cancers.

This transcript has been edited for clarity.