Pembro/Chemo Combo Fails to Improve
Event-Free Survival in Head and Neck Cancer

Neil Osterweil

September 15, 2022

PARIS — Adding pembrolizumab (Keytruda) to chemoradiotherapy did not significantly improve event-free survival, compared with CRT plus placebo as first-line therapy for patients with locally advanced head and neck squamous cell cancers (HNSCC), reported investigators of the KEYNOTE-412 trial.

Among 804 patients with newly diagnosed, pathologically proven, unresected locally advanced head and neck squamous cell carcinomas who were followed for a median of 47.7 months, the event-free survival (EFS) rate with the pembrolizumab/CRT combination followed by maintenance pembrolizumab was 63.2%, compared with 56.2% for CRT plus placebo. This translated into a nonsignificant hazard ratio of 0.83, said Jean-Pascal Machiels, MD, PhD, at the European Society for Medical Oncology Congress 2022.

Despite the trial failing to meet its primary endpoint, Machiels expressed optimism about the results.

"Pembrolizumab with chemoradiation was associated with a favorable trend toward improved event-free survival versus placebo plus chemoradiation in patients with locally advanced head and neck cancer," he said.

He noted that the 2-year EFS rate was 63% with pembrolizumab, compared with 56% with placebo.

The data also support the hypothesis that programmed death–ligand 1 (PD-L1) expression as measured by a combined positive score (CPS) could be a predictive biomarker for identifying those patients most likely to respond to the immune checkpoint inhibitor, he added.

KEYNOTE-412 Details

The rationale for combining the checkpoint inhibitors pembrolizumab with chemoradiotherapy comes from the KEYNOTE-048 trial results of which showed a survival improvement for the use of pembrolizumab plus a platinum-containing regimen as a first-line therapy for recurrent or metastatic HNSCC, as well as pembrolizumab monotherapy for patients with PD-L1 CPS of 1 or greater.

In the current study, Machiels and colleagues studied whether adding pembrolizumab to CRT could benefit patients with treatment-naive unresected, locally advanced HNSCC.

Eligible patients included those with stage T3 or T4, N0-N3 or any N2a-3 (T1-T4) cancers of the larynx, hypopharynx, or oral cavity, and either p16-negative oropharynx cancers or T4 or N3 p16-positive oropharynx cancer. Patients were required to be eligible for high-dose cisplatin-based CRT.

A total of 804 patients were randomized, 402 in each arm, to receive either pembrolizumab 200 mg intravenously every 3 weeks for 3 cycles plus CRT followed by maintenance pembrolizumab for 14 cycles, or to placebo plus CRT followed by placebo maintenance.

As noted before, there was no significant difference between the study arms for the primary endpoint of EFS. The 24-month EFS rate was 63.2% for the pembrolizumab group, compared with 56.2% for controls. The respective 6-month EFS rates were 57.4% versus 52.1%.­

In a post hoc analysis, both EFS and overall survival were numerically with pembrolizumab among patients with PD-L1 CPS of 20 or greater. The respective 2- and 3-year EFS rates were 71.2% versus 62.6%, and 66.7% versus 57.2%.

The 24-months overall survival rates were 83.3% with pembrolizumab and 79.9% with placebo, and 36-month rates were 79.1% and 73%, respectively.

Neither EFS rates nor OS rates among patients in this subgroup differed significantly; however, there were no new safety signals with the combination, Machiels said. The incidence of grade 3 or greater adverse events was 92.2% in the pembrolizumab arm versus 88.4% in the placebo arm. Four patients in the pembrolizumab arm and six in the control arm died from treatment-related causes.

Benefit Still to Be Proven

In a media briefing held prior to his presentation, Machiels was asked how he could justify his conclusions about a benefit for adding pembrolizumab given that there was no difference between the treatment groups for the primary endpoint.

He said that when the investigators designed the trial 7 years ago, the CPS score for PD-L1 expression had not yet been developed, and that if it had been they might have designed the trial to explore the effect of the pembrolizumab chemoradiation combination according to CPS subgroups.

He also pointed to the numerically superior 2-year EFS and overall rates.

In the presidential symposium, James Larkin, MD, PhD, an invited discussant from the Royal Marsden Hospital, London, said that chemotherapy and anti–PD-1 therapies are known to offer benefit in advanced cancers despite the trial's failure.

"There is a signal, particularly as we've seen in the high PD-L1 group," he said, noting that the signal was consistent with that seen in the JAVELIN 100 study, which was also a negative trial. He cautioned against relying too heavily on the comparison, however, as JAVELIN 100 was conducted with avelumab, a PD-L1 inhibitor, whereas pembrolizumab is a PD-1 inhibitor.

"Could there be an issue here with treatment schedule? An example and a comparison might be the PACIFIC study in non–small cell lung cancer, which is a positive trial, where actually the checkpoint inhibit with durvalumab was given immediately after the chemoradiotherapy, leading to benefit, rather than being concurrent," he said.

Larkin also questioned whether, as co-discussant Sherene Loi, MD, PhD, from the Peter MacCallum Cancer Center, Melbourne, suggested radiotherapy to lymph nodes might alter the immune response to checkpoint inhibitors.

"Clearly radiotherapy is the central component of treatment in this setting, so it would be quite difficult to scale too much on that, but the question is: 'Could it be modified?' For example, just to irradiate the primary tumor and involved lymph nodes and potentially spare noninvolved lymph nodes," he said.

The KEYNOTE-412 study was funded by Merck Sharp & Dohme. Machiels reported uncompensated consulting to the company. Larkin reported consulting for and receiving honoraria from Merck and others. Loi reported uncompensated advisory board activity for Merck and others.

This article originally appeared on, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.