Dr. Burtness: Our group at Yale undertook a retrospective analysis to look at a regimen of weekly carboplatin, paclitaxel and cetuximab, which has been studied both in ECOG and MD Anderson, and is I think pretty commonly used in practice but has not been included as a control or an experimental arm in any of the large randomized trials, which have changed practice in head and neck cancer in recent years.
One of the reasons that this regimen is appealing is it's more tolerable to people with renal injury, frailty, or advanced age. And we were interested in seeing what we could learn about its activity and tolerability in the patients we had treated at Yale, including those who were elderly or frail.
Study design and patient selection
So it was a retrospective observational study. We used chart review and we had a total of 80 patients. Twenty-two of them had received this three-drug weekly regimen as an induction to try to shrink cancer that was either borderline amenable to surgery or to bridge them until they could lie flat for radiation. And 58 had received it for metastatic or recurrent disease.
The reasons that patients were offered this regimen instead of a more conventional cisplatin approach were performance status in 13 of the 80 patients, hearing loss in 11, either renal injury or multiple renal risk factors in 6, and advanced age in 5.
What we found was that in patients who received this as their first-line regimen for metastatic or recurrent disease, our mean overall survival was 15 months. We had a response rate of 22% and the intriguing thing was that in the elderly/frail subset, we also had pretty substantial overall survival at 10.9 months.
The induction cohort, this was only 22 patients, but 86% of them reached their end point at an objective response rate of 64%. Here, obviously we use modified RECIST criteria because patients couldn't have the confirmatory scan because they were going on to surgery or radiation.
In this group, the median overall survival was 29.8 months. Induction success was deemed to be 82%, response rate was 73%, and among elderly/frail, who got this regimen for induction, overall survival of 30.6 months.
About half the patients had grade 3 or higher toxicity, 67% had dose interruptions due to toxicities, but bear in mind that with this weekly regimen, it's fairly straightforward to just wait until that toxicity resolves and just delay the start at the next cycle. But there were 11 patients or 14% who stopped the treatment altogether due to toxicities.
Implications for clinical practice
This study was conducted in the pre-immune checkpoint inhibitor treated patients, and as the first-line regimen, at least for PD-L1-expressing patients, this wouldn't be a first-choice regimen any more. Nonetheless, I think demonstrating high response rates and reasonable survivals in such a complicated patient population, does indicate that this weekly paclitaxel, carboplatin, cetuximab backbone can be considered in situations where immune checkpoint inhibitors are not appropriate.
So when we thought about this regimen, for example, in comparison to the EXTREME regimen, the response rate overall was 22% in the metastatic recurrent population. So that is a little bit lower than the 36% that we expect with the EXTREME regimen or with the pembrolizumab, platinum, 5-FU combination.
On the other hand, this was a group of patients who would not have been eligible, by and large, for those trials because of comorbidities or frailty. And for the poor performance status, elderly/frail subset, the response rate of 15% and OS of nearly 11 months actually compared very favorably to the limited data that we have in that group of patients.