Practical Issues Surrounding Immunotherapy for HNSCC
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: Let’s talk about a couple of very practical issues. The first is a patient who has had prior cetuximab, and now you’re thinking about immunotherapy in the form of an anti-PD-1. Is there a reason to think about that patient any differently, Josh?
Joshua M. Bauml, MD: Yes. There were subgroup analyses that were presented at ASCO this year looking within CheckMate-141 at those patients who had previous treatment with cetuximab. It should be noted that prior treatment with cetuximab did not negate the ability to give cetuximab as an investigator’s choice—but that was a very small subgroup of patients on the trial. And for those patients who had been previously treated with cetuximab, the incremental benefit to the nivolumab was much diminished. That’s quite notable, because really what you then ended up comparing is nivolumab to patients who largely received docetaxel and methotrexate.
One of the things that’s interesting about CheckMate-141 is, the most common drug that was given on investigator’s choice was methotrexate. And so, when you’re comparing nivolumab—which we’re establishing as a highly active drug—with mostly methotrexate and you’re not seeing a survival benefit, or you’re seeing a much diminished one, that’s really confusing. I don’t know exactly how to interpret that information.
Barbara A. Burtness, MD: You’ve said a number of dismissive things about methotrexate over the course of the morning here, and I just want to point out that the hopes of gefitinib, zalutumumab, and a number of other agents are founded on the fact that methotrexate does have some activity. There are some responders, and there are some people who don’t progress immediately.
Ezra Cohen, MD: Not an inactive drug.
Joshua M. Bauml, MD: Not an inactive drug.
Jared Weiss, MD: When you’re getting into deep salvage, patients do tolerate it pretty well, other than mucositis. So, for that patient who has been repeatedly chemosensitive; who has received their PD-1; who has received platinum, taxane, or 5-FU; and who maybe doesn’t have trials available, it’s an active option. But it’s maybe not quite the paradigm of chemotherapy that we reach to first or consider. I think it’s controversial for many whether or not this should be a standard that another drug, like afatinib or nivolumab, really should be compared to.
Ezra Cohen, MD: Getting back to nivolumab and pembrolizumab, again, they’re very practical. You now have a choice of either one. Jared, what do you choose when you’ve got a patient who’s a good candidate for an anti–PD-1 immunotherapy? Which one do you go to?
Jared Weiss, MD: We’ve spoken about many controversial topics today and had a nice spectrum of opinions here. For me, this is the question for which I have the least passion of any question you’ve asked any of us today. So, in my mind, in the existing data, the toxicity, safety, and efficacy of these agents appear far more similar than different. If you worked for BMS or Merck, you might have reason to torture slight differences in cross-trial comparisons. You certainly could talk about the way the trials were done. With those decisions, there are differences. There’s no good reason to see these as dramatically different agents, in my opinion.
So, what are you left with? What you’re left with is that nivolumab is an every-2-week drug, or at least it currently is, and pembrolizumab is an every-3-week drug. In my practice, my median travel distance is quite long. It’s 3 or 4 hours for local reasons, and when that’s the case, for me it creates a strong driver to give an every-3-week drug. And for me, that’s the main differentiator. But if one of you said, “Hey, I really like nivolumab. I’ve had more experience with it. I’ve been on trials,” I wouldn’t criticize that in any way.
Ezra Cohen, MD: Any differences of opinion here?
Barbara A. Burtness, MD: No.
Ezra Cohen, MD: Very similar? More similarities than differences between the 2 drugs?
Joshua M. Bauml, MD: Yes.
Ezra Cohen, MD: Well, what about the choices between an EGFR-targeted agent, chemotherapy, and immunotherapy? I’ll go around. Josh, what are the factors in there?
Joshua M. Bauml, MD: If I have a patient who has previously failed platinum therapy, then I will preferentially use immunotherapy prior to a drug like cetuximab or other cytotoxic chemotherapy, based upon the toxicity and the activity profile.
Ezra Cohen, MD: Barbara?
Barbara A. Burtness, MD: So, if I have a patient who is highly symptomatic, with bulky disease or a mass near the airway, something like that, and we’re in frontline, even if we’re within 6 months of radiation, the response rates to chemotherapy are a little bit higher. And I think those are patients who need a response. The other group, where I’m a little reluctant about using a PD-1 inhibitor fresh out of the gate, are the patients who are PD-L1–nonexpressing, less than 1% on tumor and tumor-associated immune cells. Because that’s a group where—albeit, you can see responses—the response rate is low. The overwhelming majority of them progress right away. So, I still feel that there are people with bleeding risks or bulky mass or a lot of symptoms who are going to do well with chemotherapy and cetuximab for some time and who are probably going to do well with an I-O agent down the road.
But I agree with what Josh said earlier: that the sooner we can get to a multiplexed predictive model that might include HPV status, tumor bulk, PD-L1 expression, tumor-infiltrating lymphocytes, and mutational load—there are a lot of interferon response gene signatures, there are a lot of candidates here—I think it would be great to have that.
Ezra Cohen, MD: Jared, what factors into your decision?
Jared Weiss, MD: Like the rest of the panel, my preferred treatment option is a clinical trial. In the absence of that, if we go Neil Love style and say, “If we can’t have that, then I use these, for the most part, as indicated”—which is to say that in a platinum-sensitive first-line patient, a platinum-containing first-line agent regimen remains my standard of care. But one thing I would say about that is that I do have very strong feelings about the EXTREME regimen as written, that the toxicity is a little bit extreme. While I know others may have differences of opinions, my personal opinion is that the toxicity in that regimen is excessive for a population whose primary goal is palliation. And so, we didn’t talk much about alternative induction regimens, some of those weekly schedules. There are a series now, more than a half dozen phase II trials, looking at the strategy of weekly carboplatin with a weekly taxane and a targeted agent, usually cetuximab, that show response rates in the 60% to 93% range and show very acceptable toxicity profiles, consistently superior to what was seen in the EXTREME regimen of cisplatin/5-FU and cetuximab.
I will confess, on the record, that I have wholesale moved away from the standard-of-care regimen of cisplatin/5-FU/cetuximab, that I just can’t stomach doing that to patients. I will routinely use one of these weekly schedules for the first-line patients when I choose chemotherapy. I happen to like carboplatin/nab-paclitaxel and cetuximab, but there are multiple variants people use that tend to have the theme of platinum taxane plus or minus EGFR-targeted agent that I think are a little more humane.
So, in my practice, I’m sticking to the label in the sense that if they’re platinum-sensitive, I’m using some chemotherapy combination first, but I’m diverging quite a bit from that label in using an alternative cytotoxic regimen when I do. And then, I’m having a lower threshold, perhaps, than before the PD-L1 agents to declare failure on that regimen, whereas before I might have watched and seen a pace of growth and considered other things. I have a lower threshold now, with a more effective second-line agent to move on to—a PD-1. I really want to guarantee the patient a shot at the durable control that these agents offer, and so I have a lower threshold to call progression than I might have in the past.