Neoadjuvant and Adjuvant Pembrolizumab with Standard-of-Care Significantly Improves Event-Free Survival in Head and Neck Cancer

Event-free survival (EFS) was 27% higher for patients with locally advanced head and neck squamous cell cancer (HNSCC) who received standard-of-care with neoadjuvant and adjuvant pembrolizumab compared to the control group who did not receive the immune checkpoint inhibitor, according to a new study published today in The New England Journal of Medicine. ⁽¹⁾

“As the first positive trial in two decades for this patient population, this result is very exciting,” said radiation oncologist and early drug development specialist Nancy Lee, MD, Service Chief of Head & Neck Radiation Oncology and Proton Therapy at MSK. “The study findings demonstrate that immune checkpoint inhibition before and after curative resection provides a significant, clinically meaningful benefit to patients.”

Dr. Lee, a leading international expert in head and neck cancer radiation oncology, was involved in the study design and served as Radiation Co-Chair of the international, phase 3 KEYNOTE-689 trial. The groundbreaking results were presented in a plenary session at the American Association for Cancer Research Annual Meeting 2025 in Chicago.

On June 12, 2025 the U.S. Food and Drug Administration approved pembrolizumab for adults with resectable locally advanced HNSCC whose tumors express PD-L1 ≥1, as a single neoadjuvant treatment, continued as adjuvant treatment in combination with radiotherapy with or without cisplatin after surgery, and then as a single agent. The approval is the first for HNSCC in six years and the first perioperative approval for this indication.

Head and Neck Squamous Cell Carcinoma

Locally advanced HNSCC is a challenging disease. For patients without adverse pathologic features, such as positive margins or extranodal extension, adjuvant radiotherapy is the standard of care.  ^{(2) (3)}  For patients with high-risk disease, though, the standard of care is radiotherapy with concomitant cisplatin after surgery, as established by two pivotal phase 3 studies about twenty years ago. However, about one-third of patients experience relapse within 12 months and less than half survive to five years. ^{(4)  (5) (6) (7) (8)}

The anti-PD-1 monoclonal antibody pembrolizumab is an essential first-line standard-of-care for recurrent and metastatic HNSCC. ^{(2) (3) (8)} Two phase 2 studies showed that adding perioperative pembrolizumab to traditional neoadjuvant and adjuvant regimens was associated with lower relapse rates and better disease-free survival compared to historical controls. ^{(9) (10) (11)}

Study Design

The international, open-label, phase 3 KEYNOTE-689 trial (NCT03765918) was conducted at 192 sites in North America, South America, and Europe. The study randomized 714 adult patients with stages 3 or 4a resectable HNSCC to standard of care with pembrolizumab with or without two cycles of neoadjuvant and 15 cycles of adjuvant pembrolizumab. There were 363 patients in the pembrolizumab group and 351 in the control group. The standard of care was surgery and adjuvant radiotherapy with or without concomitant cisplatin.  ⁽¹⁾

Patients were stratified by primary tumor site, tumor stage, and PD-L1 status. Patients underwent surgery within six weeks of randomization to the pembrolizumab group and within four weeks of randomization to the control group.  ⁽¹⁾

Positive margins or extranodal extension indicated a high risk of recurrence post-surgery. Planned postoperative treatment based on pathology-guided risk assessment included radiotherapy alone for low-risk disease or concomitant chemoradiotherapy plus cisplatin for high-risk disease. Postoperative (chemo)radiotherapy began after recovery and within six weeks after surgery. Patients in the pembrolizumab group began pembrolizumab treatment starting on day one of radiotherapy, whereas patients in the control group had no additional treatment.  ⁽¹⁾

The trial’s primary endpoint was EFS in participants whose tumors had a PD-L1 combined positive score (CPS) of 10 or greater, one or greater, and in all participants. EFS was defined as the time from randomization to radiographic disease progression, including during the neoadjuvant setting, local or distant progression or recurrence by imaging or biopsy, or death due to any cause.  ⁽¹⁾

Secondary endpoints included the proportion of patients with a major pathologic response and overall survival (OS) in CPS subgroups and for all participants, and safety and tolerability.  ⁽¹⁾

Study Results

The primary endpoint was met in all three study populations at the first interim analysis as follows:  ⁽¹⁾

• IIn the CPS ≥ 10 population, the EFS at 36 months was 60% in the pembrolizumab group versus 46% in the control group, with a hazard ratio (HR) for progression, recurrence, or death of 0.66 (p = 0.00434).
• In the CPS ≥ 1 population, the EFS rates were 58% and 45%, respectively (HR = 0.70, p = 0.0028).
• Across all participants, regardless of PD-L1 CPS, the EFS rates were 58% and 46%, respectively (HR = 0.73, p = 0.00822). 

Overall survival results were not mature at the time of data analysis. Estimated OS rates at three years were as follows:  ⁽¹⁾

• In the CPS ≥ 10 population, the estimated OS rate was 68% in the pembrolizumab group versus 59% in the control group (HR = 0.72).
• In the CPS ≥ 1 population, the estimated OS rates were 69% and 60%, respectively (HR = 0.72).
• Across all participants, the estimated OS rates were 68% and 61%, respectively (HR = 0.76).

Notably, about 12% fewer patients in the pembrolizumab group had high-risk pathologic features and received cisplatin post-surgery compared to the control group, suggesting that neoadjuvant pembrolizumab contributed to treatment de-escalation for some patients.  ⁽¹⁾

Grade 3 or higher treatment-related adverse events (TRAEs) occurred in 45% of patients in the pembrolizumab group and 43% in the control group, including grade 5 events in 1.1% and 0.3%, respectively. Most TRAEs were consistent between groups and in line with known toxicity profiles. Participants in the pembrolizumab group experienced more potentially immune-mediated adverse events of any grade (43% vs. 10%) and grade 3 or higher TRAEs (19% vs. 11%),  ⁽¹⁾ consistent with reports from similar trials in other cancer types.  ^{(12) (13) (14)}

Improving Outcomes for Patients with Head and Neck Cancer

MSK’s Head and Neck team of experts includes radiation oncologists, surgeons, and medical oncologists dedicated to improving outcomes for patients with head and neck cancer through novel treatment regimens and de-escalation strategies.

Dr. Lee is leading an MSK-exclusive randomized phase 3 trial (NCT06563479) comparing personalized de-escalated chemotherapy and radiation therapy in patients with good risk human papillomavirus (HPV)-positive throat cancer with standard of care radiation in patients with unselected HPV-positive throat cancer. 

Good-risk disease is determined by the absence of tumor hypoxia on FMISO PET scan results, a clinical biomarker of distant metastasis risk used by Dr. Lee and colleagues since well before 2016 to assess whether patients with head and neck cancer may safely receive de-escalated radiation therapy.  ⁽¹⁵⁾ The trial continues to recruit patients at seven MSK locations in the New York City area. Read more.

The study was sponsored by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc. Access disclosures for Dr. Lee. Learn more about MSK head and neck cancer clinical trials.