Clinical Perspectives on New Data in Advanced NSCLC - Episode 3
Benjamin P. Levy, MD: I have 2 more questions. Does this change the calculus of how we approach patients with surgery for stage IIIa disease? I think I’ve asked you this question before. A patient who’s surgically resectable should potentially have surgery at some point along the treatment continuum if they’re a stage IIIa and the surgeon deems they are surgically resectable. Does this improved outcome we see with durvalumab change the way we think about surgery in stage IIIa disease?
Suresh S. Ramalingam, MD: What it does for me is create the possibility of improving outcomes, even for patients with surgically resectable disease, with immune checkpoint inhibition. But that is a question for a clinical trial.
Benjamin P. Levy, MD: Yes.
Suresh S. Ramalingam, MD: I think to move this paradigm into the resectable setting at this point would be premature. What we have seen already are some promising data coming out where in patients with resectable lung cancer, giving immunotherapy as preoperative treatment has some very favorable results. There are already randomized trials looking at how does this fit in with the surgically resectable population? Even though there is a little bit of overlap, the surgically unresectable stage III and the resectable stage III cases are distinctly biologically different groups of patients, and we need to treat them as such.
Solange Peters, MD, PhD: It’s interesting here because we have some differences in the way we qualify resectable stage III disease in Europe and in the United States. I think we are more liberal in the way we still consider a patient operable or resectable in Europe. We go for stage IIIa, multistation, and stage II. We sometimes go for very bulky disease even at the limit of stage T4 invading some other organs. So, our surgeons are quite strong in Europe. It means that we have this tendency, or these trends, to prefer surgery first. By definition, radiation is really a second choice.
I think this trial has changed a little bit of the game. For all these marginally resectable patients, we have a 10- or 15- or 20-minute discussion at the multidisciplinary tumor board. We can maybe cut the discussion down to how we have a good treatment to give to this patient, which is probably simple and probably more rational in terms of age, PS [performance status], and tumor extent. I think it really offers us a very promising new strategy to maybe spare some patients from very invasive, very large surgery, and that’s something that is probably a positive move for Europe.
Sanjay Popat, PhD, FRCP: I think that’s right. The question about borderline resectability has been one that we debate a lot of the time. In fact, with multizone N2 disease, there are a lot of surgeons in the United Kingdom and many parts of Europe who wouldn’t even consider resecting that patient, in my mind appropriately. I think this has entirely shifted the balance of the sands, because we now have a treatment that has been proven to improve survival compared to discussing with colleagues whether the surgery is bringing any benefit or not. In my mind, we can appropriately see less surgery in these borderline cases and implement far more radiation. But one of the areas that I wanted to mention, Ben, is how there’s a lot of chemoradiation that occurs in the sequential setting in many parts of Europe, certainly in the United Kingdom.
Benjamin P. Levy, MD: Yes. Is it still the standard for stage III disease for the most part in the United Kingdom?
Sanjay Popat, PhD, FRCP: I think there’s a lot of sequential chemoradiation, mainly because of the previous concern about the additional toxicities for marginal benefit with concurrent versus sequential chemoradiation. But now we have a dataset, which mandates concurrent chemoradiation. We need to ensure that all our patients who are suitable for concurrent chemoradiation are getting it.
Benjamin P. Levy, MD: I have one more question regarding this dataset, then we’ll dovetail into another section. The EGFR-positive subset in this trial is the one group that really didn’t benefit, and I think we have to be mindful of subset analysis and hazard ratios from forest plots. This is a question that continues to come up. Patients who are EGFR-positive didn’t seem to be a group in the trial that garnered a statistically significant benefit. We know that EGFR mutations may not do well with at least single-agent checkpoint blockade. What are people’s thoughts about using durvalumab in a stage III patient who is EGFR positive and completes concurrent chemoradiation? I don’t know if there’s a right or wrong choice, but this is why we have these perspectives here. Solange?
Solange Peters, MD, PhD: The first answer is to say that in most of these patients, we simply don’t know. Usually, EGFR is not routinely tested for in locally advanced disease or in early disease. But would you have to know about it? I think there are very few data telling you about this strategy, which is probably creating some immunogenicity. We remember there are data from IMpower150; when you combine with bevacizumab, you still observe a benefit in EGFR-mutated disease. It might be the case for this trial. Like Ram was saying before, this kind of question should not be answered by our intuition. It requires a trial if we want to decline or to prevent a patient from receiving this drug if he’s EGFR-mutated. I wouldn’t decline this treatment to any patient without a dedicated trial because the combination of chemotherapy and radiation may create an environment where some immunogenicity is gained. This is really something that I would be very careful about. Remember, this trial will be a huge debate in Europe. Almost half of the patients have an unknown PD-L1 [programmed death-ligand 1] status, about 40% of them. It means that many of them have no EGFR. This is not the setting where biomarkers are a big player. The big player is locally advanced disease.
Benjamin P. Levy, MD: In a nice way, if it’s for all patients. They see everyone benefited, at least in PFS [progression-free survival]. I agree; I’ve been more comfortable. I think we need trials. I think that it’s tough to draw conclusions from subset analysis in EGFR-mutated cases in that trial. I’ve been inclined to give it to them outside of a clinical trial.
Sanjay Popat, PhD, FRCP: I think you’re right though, you have to have a good reason not to give it to them, and the data do not really support omitting them, given the quality of the data that we have. At the end of the day, it’s a subset, an unplanned analysis of a relatively small number of patients with wide confidence intervals and a point estimate showing some benefit. There aren’t enough data to withhold that treatment from that group.
Benjamin P. Levy, MD: All very good points.
Transcript Edited for Clarity