Nivolumab With Ipilimumab vs Reference Standard EXTREME for Metastatic Recurrent Head and Neck Cancer
PracticeUpdate: What forms the basis of this trial and what is the EXTREME reference standard?
Dr. Burtness: This is a very, very eagerly awaited first-line trial for recurrent metastatic head and neck cancer. It built on the success that had been seen with nivolumab in platinum-refractory disease with pembrolizumab in platinum-refractory disease, and with pembrolizumab in the first-line setting for metastatic recurrent head and neck cancer. It asked the question, how active was the combination of nivolumab with a CTLA-4 directed drug ipilimumab? How active was that combination relative to the reference standard of the EXTREME regimen, platinating chemotherapy, 5-FU, and cetuximab? In terms of overall survival with a biomarker driven analysis, particularly for the patients with PD-L1 expression in tumor and immune cells with a cut point for that CPS of 20. People may remember that CPS 20 population had done very well with pembrolizumab monotherapy in the KEYNOTE-048 trial.
PracticeUpdate: Could you elaborate on the patient group and the trial design?
Dr. Burtness: This was a large trial, about 475 patients per arm, and the groups were very well balanced with all the usual characteristics. Similarly to KEYNOTE-048, about 80% of the patients were PD-L1 expressing, although it's important to say that it was a different antibody that was used, this was the pharmDx 28-8 assay. About 20% of the patients had p16 positive oropharynx cancer. About 77% of them were smokers, and the majority had either local regional recurrence or local regional recurrence together with distant metastasis. The group that had distant metastasis alone was about 40% of patients. When patients progressed on this therapy, there was considerably better access to immune checkpoint blockade as second-line therapy for the EXTREME patients than there had been at the time of the KEYNOTE-048 trial, reflecting the many approvals for pembrolizumab in the first-line setting since the reporting of KEYNOTE-048 in 2019.
PracticeUpdate: What were the study outcomes and what are the trial implications?
Dr. Burtness: The trial failed to meet its primary endpoint. The overall survival for the total population, so that is without enriching for high PD-L1 expression, was 13.9 months for the patients who received dual immune checkpoint blockade and 13.5 months for the patients who received EXTREME, and so the hazard ratio there was 0.95 and no suggestion really of a significant benefit. If you looked at the CPS 20 population, there was a hint of a benefit, the overall survival went from 14.6 months to 17.6 months, and the hazard ratio for that was 0.78, did miss statistical significance based on the alpha sharing in that trial. The interesting thing here is that the control arm performed better. The EXTREME treated patients here had better median overall survival than had been the case in KEYNOTE-048. The obvious explanation for that is the fact that 46% of those patients were able to go on to get subsequent immune checkpoint blockade after they progressed on the first-line therapy.
PracticeUpdate: How are these trial results compared to the KEYNOTE-048 results?
Dr. Burtness: This really was a different population, that in some way, perhaps related to the burden of disease or performance status or in some other way that isn't readily captured in the patient demographics, this was a group of patients who had somewhat more treatment responsive disease. If we look at the progression-free survival for nivo plus ipi, it was 3.3 months, and that compared with 6.7 months for the EXTREME, which was longer than we had seen in KEYNOTE-048. Response rates were the same, but duration of response again appeared to be better in the EXTREME patients treated on this trial at 5.9 months. Even looking at the CPS greater than 20 population, PFS there was 5.4 months, considerably better than you'd expect for nivo or pembro alone, compared with 7 months in the control arm.
Again, that 7 months is longer than was seen in KEYNOTE-048. Duration of response was also better for the EXTREME regimen in this trial at 5.9 months for all comers, 7 months for the CPS 20 population. But for those patients with CPS 20, the duration of response was 32.6 months, which is clearly the longest that we have ever seen in this disease.
PracticeUpdate: What would be your key take home message from this study?
Dr. Burtness: This trial did not change the standard of care. There's no clear overall survival advantage, and even in the CPS 20 population the hazard ratio is not as striking as it was for pembrolizumab. There, I think, is the question of the different biomarker, the question of the different patient population.
Seeing the higher response rate and the longer duration of response in the CPS 20 population is intriguing and I think probably justifies more study maybe with more power and maybe with a more homogeneous biomarker selected group of patients for this dual regimen. The one other comment that I'll make is that some of the theory behind the addition of a CTLA-4 drug to a PD-1 inhibitor had been that this might be sensitizing in a PD-L1 non-expressing population, and looking at the forest plot, that group did not benefit seemingly at all from the combination therapy. I think that we are left with pembrolizumab or pembrolizumab plus chemotherapy is the standard first-line therapy for PD-L1 expressing cancers, but there is an intriguing hint here that for some subsets of patients, the combination of nivolumab with ipilimumab may have real lasting benefit.
PracticeUpdate: Were there any adverse effect signals that might increase your utilization of this in the CPS 20 population?
Dr. Burtness: The adverse effect profile here did not seem all that much different from immunotherapy monotherapy in the KEYNOTE-048 trial. There was a little bit more high-grade immune-related toxicity and a little bit more colitis, but really not. It's still a much less toxic regimen than the EXTREME regimen is, or the pembro with chemo is. I think had you seen real activity for this in the CPS less than 1 population, where pembro monotherapy is not approved and not advised, that would have been really exciting, but that did not emerge. The difference in the CPS 1 to 19 population didn't, that group also does benefit from pembro or pembro chemo. I think that the intriguing results were really in the CPS 20 and higher group. Yet they weren't necessarily as powerful as pembro chemo, so I think very intriguing, but more work needs to be done.
Click on any of these tags to subscribe to Topic Alerts. Once subscribed, you can get a single, daily email any time PracticeUpdate publishes content on the topics that interest you.
Visit your Preferences and Settings section to Manage All Topic Alerts
Additional Info
Disclosure statements are available on the authors' profiles: