Transcript:Benjamin Levy, MD: So let’s move on to monoclonal antibodies specifically in squamous cell patients. We think about monoclonal antibodies in lung cancer and, of course, we all turn to bevacizumab, a drug that has shown efficacy and survival advantages in the non-squamous patient population. But toxicity signals, perhaps wrongly characterized, were seen in the squamous cell population; maybe it wasn’t the squamous cells that really predicted toxicity to this drug. Nevertheless, bevacizumab is not approved in the squamous cell population. Things have changed now. We have two new monoclonal antibodies that are approved, one specifically in combination with platinum in frontline, necitumumab, in a squamous cell population. And the other is ramucirumab for which we have to remember included a fair number of squamous cell patients in the trial; that the benefit was there as well. I want to talk a little bit about these two drugs because it’s more options for our patients, both in the frontline setting and the chemo refractory setting and maybe put these in context at least for the second-line with immunotherapy. Paul, you want to talk a little bit about necitumumab and how it works and the SQUIRE data that got necitumumab approved by the FDA?

Paul K. Paik, MD: Sure. The regimen of cisplatin/gemcitabine and necitumumab is built actually on a foundation of work that’s been going on dating back to original reasons to target EGFR in non-small cell lung cancer starting in the late 90s, with the protein’s overexpression. So, necitumumab is a fully humanized monoclonal antibody against the EGF receptor. Its predecessor, cetuximab, was a chimeric antibody for which we have data from the FLEX trial. And the FLEX trial subset analyses were promising among other subgroups that showed improved efficacy. We’ll get into the H-score in a little bit for EGFR expression. But squamous lung cancer came in the subset analyses as a group that seemed to have more benefit than non-squamous patients. This was the backbone and rationale, particularly in conjunction with EGFR overexpression, for the SQUIRE study.

So, the SQUIRE study was a randomized phase III trial of cisplatin/gemcitabine and placebo versus cisplatin/gemcitabine and necitumumab. Necitumumab was continued as a maintenance therapy for patients who had at least stable disease on the regimen. And the data were presented first at ASCO a couple of years ago, published last year in Lancet Oncology, and showed that there was some modest improvement. And I think most of us would classify it as modest improvement in terms of overall survival and PFS benefit, with hazard ratios that hovered around 0.85, essentially, for both PFS and OS. For a first-line regimen, certainly something that’s not the solution and particularly in the face of the hazard ratios and benefits we saw with nivolumab in the second-line setting against docetaxel also makes it a little bit less compelling. But the data are the data, and these are data that led to an FDA approval, in part because there really is nothing competing in the first-line setting for something that’s similar and because the thought is that, though the results were modest, there’s still some small amount forward. And so that decision has been made. The response rates, important to note, were not significantly different, also essentially in the low 30s which, again, if you put the regimens side by side is still a lot less than carboplatin/Abraxane.

Benjamin Levy, MD: Now, are you using necitumumab? You clearly, as you’ve said, shown that there’s an FDA approval now. There was a modest but significantly different, overall survival advantage between the two arms with the addition of necitumumab. Are you using it on a day to day basis? Reserving it for patients? If so, which patients? Is there a role for this drug routine clinical use?

Paul K. Paik, MD: It’s hard to know because we’d like to have a good reason to give something, some specific clinical situation where the data are going to be able to guide you, we should give this, particularly as we said, in a landscape that doesn’t have a lot of head-to-head comparisons. The problem is that giving this requires giving cisplatin which not all of our patients are eligible for, for any number of different reasons. The side effect profile also is a little bit worse, of course, than cisplatin and gemcitabine, as we saw which gave some concern. In addition to that, while overall survival benefit is the most compelling, that’s a little bit hard to frame on an individual patient level, as we’ve been talking about, an individualized level while response rate is easier to conceptualize is important and to select for it in a given patient, overall survival based on a median is a little more abstract. And so that’s a little bit more challenging in terms of selecting. As a result, I do believe it will be a niche. I think for anyone in whom we’re considering cisplatin/gemcitabine, certainly the addition of necitumumab is there. Outside of that, it doesn’t have a well-defined space, unfortunately.

Benjamin Levy, MD: Right. Ed, your thoughts on the necitumumab data?

Edward S. Kim, MD, FACP: It’s interesting. I think I’ve been as happy as I have been sad with the landscape in the last year or so. We should be really excited that we have four drugs approved for squamous. And, although we have one that’s really popular that we’ll talk about later, the checkpoint inhibitor, one that’s less popular, and two that are not popular at all, we’ve gone away from where our science is and that’s the biomarker. And we’ll get into more of that later, obviously, but I’m one who likes to have options. I like when I walk into an ice cream store, I don’t want just Neapolitan flavors, I want 35 to 100 flavors and I want to be able to choose. And so I don’t ever disparage approvals or indications because they’re so inconsistent. Frankly, I would love to have cetuximab available and based on the data in FLEX, I thought it should reasonable that it would be there. But we just placed necitumumab with cisplatin/gemcitabine into our pathways. It is an acceptable regimen. As Paul has stated, there are some challenges with it using cisplatin. Some people will freely sort of substitute that out or not. I do believe that there is a role. We just haven’t really found it yet with monoclonal antibodies to EGFR. The results have been less than spectacular in a lot of the subsets. Maybe there will be a biomarker that arises that will make sense, but I think we can learn a lot from the checkpoint inhibitor that again, sort of the science is that unless the biomarker really drives decision making, it’s going to be tough to incorporate one. And I think that’s the same thing we’re going to have with EGFR antibodies. Unless you have a really definitive biomarker that separates a curve from a 60% response rate versus a 20% response rate, it’s going to be hard to implement.

Benjamin Levy, MD: Right. We’ll get to the biomarker in a second. I think my impressions of this data, I agree with both of you, it’s going to be hard. We have not used it routinely yet. I think the benefits are modest. I think some of the nice things about this study is that 10% of the patients who enrolled in the study was ECOG 2, and this is kind of representative of some of the patients I see, so it’s at least tolerable in this patient population. It did employ a maintenance approach in the experimental arm. We don’t know if the benefit seen was due to the maintenance or the induction part, similar to the bevacizumab story but certainly showed a benefit for the first time ever, I believe, in a squamous, specifically in a frontline setting with a platinum doublet showing an overall survival advantage. I think to touch upon your point, however, the lack of a molecular enrichment strategy is key here. And I agree, I think if we’re going to use these drugs, if we had a predictive biomarker it would help us out a lot. I think there have been efforts. I think at least at the time of publication of the study, they had looked at H-score. It did not turn out to be statistically significant like it had in the FLEX data with cetuximab. But more recently there has been a presentation on EGFR copy number by FISH at World Lung that may glean a little bit of insight into who is going to respond or who is going to do better with necitumumab. Paul, you want to talk about that biomarker?

Paul K. Paik, MD: Yeah. It’s not surprising. H-score is always difficult because the cutoff is a little bit arbitrary in terms of a 0-300 scale. The cutoff they had used was 200 and there was not a significant interaction between the two groups that panned out. It’s the nature of H-scores. EGFR copy number has a similar problem because it’s a continuous variable. At least in terms of the biology, it may conceivably make sense also. It’s something that should translate into protein overexpression also. And there was, as you had said, some signal that a higher EGFR copy number may predict for response. EGFR amplification by itself occurs maybe at most in about 10% of squamous lung cancer patients. So there’s a lot of overexpression, but the copy number in terms of amplification happens in a subset of patients. And so I do think, as you had mentioned, really focusing on these subgroups that make sense and then seeing it out is really going to have a lot of benefit and will help everyone, academic oncologists, community oncologists really figure out whether or not this has a specific niche for that population, but, again, it’s a small subset. It’s going to be difficult to sort of see this to its end, I think.

Benjamin Levy, MD: We’re still missing that. We’re still missing the molecular enrichment strategy. I think we’re hungry for that in this era of personalized and targeted therapies where we wed targeted drugs to actionable mutations. I think that’s what we’re looking for.

Transcript Edited for Clarity