Pembro + CRT in HNSCC fails to deliver statistically significant event-free survival improvement

Share :
Published: 11 Sep 2022
Views: 20
Prof Jean-Pascal Machiels - Cliniques Universitaires Sain-Luc, Brussels, Belgium

Dr Jean-Pascal Machiels speaks to ecancer about the primary results of the phase 3 KEYNOTE-412 study.

The study explores pembrolizumab with chemoradiation therapy vs placebo plus CRT for locally advanced head and neck squamous cell carcinoma.

The results of the study showed an overall baseline characteristic that was well balanced between arms. 

At data cutoff for the final analysis (May 31, 2022), median time from randomisation to data cutoff was 47.7 mo. There was a favourable trend toward improved EFS with adding pembro vs placebo to CRT, but the difference did not achieve statistical significance.

Patients with locally advanced head and neck cancer, particularly patients unresected, are treated with chemoradiation with high dose cisplatin. The prognosis of the patients, despite this heavy treatment remains quite poor because we have only 50% of the patients alive at five years. So we know that in head and neck cancer in the recurrent setting pembrolizumab and nivolumab improve survival so it was logical to try to improve these patients in the curative setting in the locally advanced disease to try to add to chemoradiation pembrolizumab. 

We took patients with locally advanced squamous cell carcinoma of the head and neck, as I mentioned unresected, and these patients were randomised between chemoradiation plus pembrolizumab versus chemoradiation plus placebo. We included 804 patients. Pembrolizumab was started one week before chemoradiation, it was given every three weeks and we gave 17 cycles, so around one year of treatment. The primary endpoint was event free survival, so this is an endpoint that includes death, progression per RECIST and also pathologically proved local regional relapse because RECIST are not able to capture all the progression or treatment failures in this disease.

Indeed, the primary endpoint was event free survival in the intention to treat population and we did not meet the primary endpoint with a p-value of 0.04 and a hazard ratio of 0.83. But we had a trend in favour of chemotherapy plus pembrolizumab. If we look at the two-year event free survival rate it was 56% in the control and 63% in pembrolizumab. The median event free survival in the control was 45 months and it was not reached in the pembrolizumab arm. But statistically it’s a negative study because the p-value was 0.04 and the efficacy boundary was 0.02 for this trial.

If we look at the survival in the intention to treat population it was similar between the two groups. In the subgroup of patients the two-year event free survival was 62% in the control and 71% in the pembrolizumab arm. When we look at the overall survival of these patients the hazard ratio was also in favour of overall survival with a three-year overall survival rate of 73% in the control and 79% in pembrolizumab, but this is an exploratory analysis.

When we look at the safety we did not detect any new signals. We had the toxicity of chemoradiation and also the autoimmune toxicity in some of the patients but nothing new. So, in conclusion, we observed a trend in favour of pembrolizumab with chemoradiation; according to the statistics of the study it’s not significant. Our data suggests that if we select the patients for a PD-L1 expression according to the CPS score and with the methodology used in the trial we may enrich the population and have a better outcome but this needs to be investigated in further trials.

So this data won’t impact practice even for the CPS subgroup?

I don’t think that we should change our standard of care based on this study but it’s given a positive message and we should continue to do this investigation of immunotherapy in locally advanced head and neck cancer.

Why is HNSCC so hard to treat?

That’s a very good question. Indeed, that’s not an easy disease for several reasons. It’s a difficult area – we have the speech, the breathing function that you need to keep. So you have not only to cure the patient but the patient should have also a good speech function, able to swallow after the treatment, able to breathe. So this makes things a little complex. We have basically two kinds of treatment – we have surgery followed by radiation or chemoradiation based on the prognosis, or we have primary chemoradiation. We choose a treatment indeed based on the functional result that we can have on the patient – the patient is able to breathe, to swallow and to eat. So it’s indeed difficult because you need a lot of expertise to treat these patients and this expertise of different areas – you need good surgeons, you need good medical oncologists, you need good radiation therapy, you need also very good support in terms of psychology but also in terms of speech function, in terms of physiotherapy. So it’s a really multidisciplinary disease. Also for the majority it’s a poor population, meaning that they are drinkers, they are smokers and so the compliance also – you need to work with the patient to improve that.

Will the roll out of the HPV vaccine affect some head and neck cancer rates?

For head and neck cancer the only subsidiary that is concerned with HPV is oropharynx cancer. Now we’ve started to vaccinate the boys and the girls so we should expect maybe a decrease in the incidence of this disease but not before twenty or thirty years. So today the prevalence of HPV positive oropharyngeal cancer is still increasing.