Good morning. My name is Barbara Burtness, I’m a medical oncologist at Yale Cancer Centre, Yale University School of Medicine, and I focus on head and neck cancer. I’d like to review a few of the abstracts that were shown at ASCO 2025 that were relevant to head and neck cancer and I’ll start with NIVOPOSTOP. This was actually a plenary session presented by Jean Bourhis and this was a study that looked at the addition of ten doses of nivolumab to postoperative radiation and concurrent high-dose cisplatin for resected stage 3 or 4b head and neck cancer that had high risk features. So those patients who after resection would meet the normal criteria for postoperative platinum and radiation. This was defined as either microscopically involved or close margins, so the definition of close for GORTEC, which ran this study, is a millimetre or less; evidence of extra-nodal extension; those patients who had four or more involved nodes; and also the finding of multiple foci of perineural invasion.

Like the KEYNOTE-689 trial that had been reported at AACR, this was a study that was predominantly conducted in HPV-negative patients, resectable patients with high-risk features often are HPV negative. Only 5% of the patients had p16 positive oropharynx cancer. They also had a slightly higher proportion of tumours that did not express PD-L1, 13% and 10% respectively in the two arms of the study, although I will note that they did not use the 22C3 assay that is more widely used in the United States.

What they found in this trial was that 3-year disease free survival was significantly improved by the addition of nivolumab, from 52.5% to 63%. The hazard ratio for that was 0.76 and the p-value was statistically significant at 0.034. Interestingly, the difference here was most strongly driven by a reduction in local regional recurrence which fits the fact that many of these patients had closer involved margins which is a risk factor for local regional recurrence. The data were not mature yet for testing of overall survival differences and we expect that information to be forthcoming.

So this trial makes clear that if you are seeing a patient who did not receive neoadjuvant immunotherapy before resection and they have an HPV negative cancer with high-risk features and you’re planning to give them postoperative chemoradiation, that nivolumab should be added for ten doses. A few things to comment on, maybe for a US audience, the use of high-dose cisplatin during chemoradiation in the post-op setting might be a little bit less common. I think the Japanese postoperative trial that demonstrated the equivalence of 40mg/m2/week to high-dose cisplatin has been widely accepted in this country. So investigators will have to wrestle themselves with whether or not they’re comfortable moving their practice back to high-dose cisplatin or if they’re comfortable giving weekly cisplatin, together with the postoperative nivolumab.

The other thing to say is that if you think back to the KEYNOTE-689 trial that was presented at AACR and recently published in The New England Journal that recently led to FDA approval for perioperative pembrolizumab in resectable head and neck cancer, you may remember from that trial that patients with stage 3 cancer actually derived substantial benefit from the addition of pembrolizumab. There were some patients who were able to avoid receiving platinum with their postoperative therapy when they had neoadjuvant pembrolizumab. The patient population might have been a little bit more real world in the neoadjuvant trial in that you didn’t have to have somebody who had resection and then was fit for beginning high-dose platinum and radiation within the right accrual window to be included in the survival curves on KEYNOTE-689.

So for those patients who meet a multidisciplinary team before surgery, if they’re stage 3/4 PD-L1 expressing, again that was predominantly HPV-negative disease, for those patients the new FDA standard of pembrolizumab for two doses before resection then risk-based postoperative therapy and more pembrolizumab is probably preferable. But for those patients who either have unexpectedly advanced disease at resection or who just miss the opportunity for neoadjuvant immunotherapy, nivolumab post-op establishes a new standard of care.

The next trial I’d like to turn to is KEYNOTE-412. This is a large randomised trial of chemoradiation, again with high-dose cisplatin, with or without the addition of pembrolizumab. Pembrolizumab was started a week before chemoradiation, continued during chemoradiation and given for 14 cycles of maintenance. This trial had previously been reported as a negative trial. The event-free survival hazard ratio, which was the primary endpoint of the study, was 0.83 and that was an analysis that was done in all comers, it was not done in a PD-L1 driven fashion. Now, with an additional two years of follow-up, there were actually a substantial number of additional events, 40 new events. This again tended to indicate that there was a benefit, albeit not a planned statistically-powered analysis for this trial. But it looked as if there was a benefit for the addition of pembrolizumab.

So for all comers the median overall survival for chemoradiation was 49.8 months, which I think is quite typical of the sort of population that was enrolled here, and the addition of pembrolizumab raised this to 71.8 months with a hazard ratio of 0.79. Of course there was no p-value generated because this wasn’t the primary analysis but the confidence intervals for this hazard ratio went from 0.65 to 0.96. It kind of makes you think that if this trial had had its primary analysis on an event-driven endpoint rather than a time-driven endpoint, we might have seen the results differently. 5-year event free survival went from 47% to 55%, although the median overall survival had not been reached in either group. The hazard ratio for OS was 0.86 and if you looked at the patients who were PD-L1 expressing, using a cut-point of CPS 1.0, the median event free survival for the pembrolizumab arm was 70.9 months compared to 48.3 months for the placebo arm. Again here the median overall survival was not reached but the hazard ratio was 0.84.

So these data fail to change the standard of care but I think they do show an impact of the addition of pembrolizumab in the definitive chemoradiation setting for locally advanced head and neck cancer. I think that they appropriately should drive further studies that will look at the use of PD-1 inhibition in those patients whose tumours express PD-L1 which, as we’ve seen time and again, is a moderately powerful predictor of who will be benefitting from pembrolizumab or, at least let’s say, those PD-L1 negative tumours are less likely to benefit. So this is something that hopefully will be moving forward in a number of trial designs. We do already have the JADE trial looking at dostarlimab after chemoradiation which is currently accruing. The idea that we maybe have that giving pembrolizumab concurrently with chemoradiation might be deleterious is not borne out by the findings of either KEYNOTE-689 or NIVOPOSTOP. So, again, not a practice-changing trial but I think a strong signal that we ought to be re-exploring the use of pembrolizumab for chemoradiation patients.

Another area of genuine interest in head and neck cancer at the 2025 ASCO meeting was EGFR inhibition. Of course we’ve been using cetuximab for several decades in head and neck cancer but there has always been a recognition that the activity of cetuximab is limited and people have been interested in how one could find combinations or novel approaches that might address that sort of resistance. So there has been development of two EGFR-directed bispecific antibodies: petosemtamab, also known as MCLA-158, co-targets both EGFR and a leucine-rich repeat-containing G-protein coupled receptor known as LGR5 that has been associated with a cancer stem cell phenotype and that is known to be expressed in head and neck cancer.

So the trial that was presented this year is a combination of petosemtamab with pembrolizumab as first-line treatment for PD-L1 positive recurrent metastatic head and neck cancer, it was presented by Carla van Herpen. In this trial the primary endpoint was objective response rate, 45 patients were treated, they looked fairly standard for a trial of this type, a median of eight cycles of treatment was administered. The objective response rate was 60% in the 43 patients who were evaluable, so 26 out of the original 45 patients.  It was also notable that they saw five complete responses, the median duration of response was 11 months and, in contrast to the other compound from Bicara, here there seems to be activity in HPV-associated disease with an objective response rate among the eight evaluable patients with p16-positive oropharyngeal cancer of 50%.

So this is something that’s moving forward in a phase III trial, both in first line and second line. I think that the data will be very closely watched. We are in bad need of second-line options for patients with head and neck cancer and in that regard I’d like to also turn to a phase II trial that was reported by my colleague, Aarti Bhatia, looking at the combination of cetuximab with afatinib. So here several observations came into play: one, that EGFR can be translated to the nucleus where antibodies such as cetuximab and indeed even the bispecifics probably will have less activity. And as well that in resistance upregulation of other HER family members such as HER2 and HER3 can be seen. This is maybe particularly true in HPV-associated disease.

So here 50 patients with previously treated head and neck cancer, all had previously received platinum chemotherapy and the majority had also received pembrolizumab, 45% of them had p16-positive cancers. So for this combination the objective response rate was 23% which is not necessarily higher than cetuximab had been shown to have in the post-IO setting, for example, in the monalizumab trial. But what was notable was that two of these responses were complete responses and two responses were seen in HPV-positive disease. Median progression free survival was 3.8 months.

So both these trials highlight that EGFR remains an important target and people are very enthusiastic that petosemtemab may be of interest, either coming out of its phase III first-line trial or the second-line trial.

So thank you very much for your attention and this was a very exciting year for head and neck cancer.