Panelists:Ezra Cohen, MD, FRCPSC, FASCO, UC San Diego; Joshua M. Bauml, MD, University of Pennsylvania; Jared Weiss, MD, UNC Lineberger Comprehensive Cancer Center; Barbara A. Burtness, MD, Yale University School of Medicine
Ezra Cohen, MD: There is, of course, another way that we can de-escalate in a sense, and that is with new surgical techniques. Jared, I know this is something that you’re familiar with. What are some of the surgical options, or let’s say perioperative options, in patients with locally advanced disease?
Jared Weiss, MD: Just as we do with radiation and chemoradiation approaches, I would divide the use of surgery into definitive surgery and salvage surgery. So, definitive surgery when surgery is the primary modality used to cure patients. I think that the most recent major advance is probably with transoral approaches. So, these are just what they sound like: Approaches through the mouth to the oral cavity, the oropharynx—perhaps for some high hypopharynx cancers—that eliminate a lot of the functional and cosmetic morbidity of some of the jaw-splitting surgeries of years ago with dramatically less blood loss, on par with perhaps a blood draw and without a cosmetic defect other than what’s done in the lymph node dissection.
There’s a lot of work that needs to be done in an era of improved radiation with IMRT, proton therapy, and improved systemic agents, to say which patients are best served with a transoral surgery for oropharynx cancer versus one of these modified radiation or improved radiation approaches with the newer techniques, lower doses, and all of these wonderful things that we’re talking about today. But certainly, I think it sometimes falls out of the conversation, and there are dramatic advances and trials ongoing addressing how to integrate these newer surgeries with some of our newer agents. So, there was an abstract at ASCO looking at induction chemotherapy prior to transoral surgeries, followed by a risk-adapted approach based on the pathologic stage taken out. And there are further studies planned on that paradigm of adjusting as you go.
The other way we use surgery is as salvage. So, patients get chemoradiotherapy, and then there’s either persistent or recurrent disease. If it’s local or locoregional, we know from historical data that there’s a real rate of salvage cure, if not as high as we’d like. It’s very situation dependent, but I think when there’s true, clear recurrent disease, most of us in the very generic situation are quoting cure rates around 20%, which we’re then adjusting based on the exact situation. They’re real, but much lower than we’d like them to be. One of the common clinical dilemmas in our tumor boards is the patient who has had chemoradiotherapy or radiotherapy and then has apparent neck nodes residually. So, we look at these by clinical exam, physical exam, CT and MRI imaging, and PET imaging. And we’ve had a series of studies over the years that have suggested prospectively, or retrospectively, that the negative predictive value of PET imaging seems quite good.
We had another one of these recently at ASCO that again replicated the finding of the excellent negative predictive value of PET. But I also found it interesting that it compared the false positive rates and true positive rates of suspicion of nodes, based on a comprehensive assessment in physical exam, CT, MRI, and PET versus PET alone. And when you looked at the comprehensive examination, the false positives were more than 3:1 and the true positives—and you’re all smiling and laughing a little bit, but the PET, which was the better one they’re advocating for, actually didn’t do so well either. There was still, if I remember right, roughly a double, 2:1, on false positives compared to true positives.
So, I think this comes back to what we should do in the real world: Not every suspicion of residual neck disease should be jumped on for surgical dissection, that sometimes the approach is very careful observation and watchful waiting. And that doesn’t mean doing nothing. It means bringing the patient back for a physical exam and then bringing them back for another imaging exam very often can prove fruitful. I think I can certainly say for my practice that almost every time that I’ve watched a patient carefully, I can’t recall having regretted doing it, and in many cases, we were quite grateful to have spared that patient the surgery.
Barbara A. Burtness, MD: And just to put that in perspective, the patients with N2b and N3 disease always used to go for a neck dissection, right?
Jared Weiss, MD: Yes.
Barbara A. Burtness, MD: A few false positives are still way better than those 85% of neck dissections we looked at that had no tumor in them. Can I go back to what you were saying in the beginning, how do we do the best decision making for taking a person to chemoradiation or transoral surgery? I think comparative studies will ultimately be needed. But I wanted to highlight what the criteria were in the recently completed ECOG trial, which was E3311. This was a trial for HPV-positive patients, and they had to meet certain eligibility criteria to look as if they’d be somebody who’d be likely to get negative margins from the transoral resection. And then, we were hoping not to have a lot of people need to go on to chemoradiation, because of that whole concern about functional outcome with trimodality therapy.
So, it was with a requirement for only using transoral surgery for T1 or T2, well lateralized cancers—the surgeon declaring that, up front, they thought the patient had at least a 70% chance of getting negative margins—and excluding matted nodes: not finally parsing from the CAT scan, “Oh, we think we see a few threads of cancer coming out of the node,” but truly matted nodes. We were able to keep the proportion of patients who needed to go to chemoradiation under 30%. I think simple clinical decision making based on T stage, excluding matted nodes, and making sure the tumor is lateralized does do a pretty good job at least of picking out patients who can get transoral surgery and radiation without too high a chance of needing platinum.
Ezra Cohen, MD: So, clearly, we’re seeing surgery become less morbid in some patients with transoral approaches. I think the negative predictive value of PET scans after definitive therapy now is really becoming something that we’ve accepted as very high, and we can be reassured by a negative PET scan.