Perioperative Pembrolizumab Enhances Survival in Locally Advanced HNSCC

“Neoadjuvant and adjuvant pembrolizumab as a [standard of care] significantly improved [event-free survival] in [patients with] resectable locally advanced head neck cancer in all 3 prespecified analysis populations,” Ravindra Uppaluri, MD, PhD, said during a presentation of the data.

The combination of neoadjuvant pembrolizumab (Keytruda) followed by surgery and adjuvant pembrolizumab with and after postoperative radiation with or without cisplatin represents a new standard of care (SOC) for the treatment of patients with resectable locally advanced head and neck squamous cell carcinoma (HNSCC), according to data from the first interim analysis from the phase 3 KEYNOTE-689 trial (NCT03765918) presented at the 2025 AACR Annual Meeting.1

At the July 25, 2024, data cutoff, the 12-, 24-, and 36-month event-free survival (EFS) rates in all patients from the pembrolizumab plus SOC (n = 363) vs SOC (n = 351) arms were 75.1% vs 62.5%, 65.0% vs 54.6%, and 57.6% vs 46.4%, respectively. The median EFS was 51.8 months (95% CI, 37.5-not reached [NR]) in the pembrolizumab/SOC arm compared with 30.4 months (95% CI, 21.8-50.1) in the SOC arm (HR, 0.73; 95% CI, 0.58-0.92; P = .0041). EFS events were observed in 37.5% vs 45.3% of patients from the pembrolizumab/SOC and SOC arms, respectively. These events specifically included death (pembrolizumab/SOC, 18.5%; SOC, 18.2%), distant progressive disease (7.2%; 14.5%), local and distant progressive disease (1.1%; 2.0%), and local progressive disease with recurrence (10.7%; 10.5%).

“Neoadjuvant and adjuvant pembrolizumab as a SOC significantly improved EFS in [patients with] resectable locally advanced head neck cancer in all 3 prespecified analysis populations,” Ravindra Uppaluri, MD, PhD, said during a presentation of the data. “The overall survival [OS] benefit did not reach statistical significance in the first interim analysis, but additional follow-up is ongoing.”

Uppaluri is the director of Head and Neck Surgical Oncology at the Brigham and Women’s Hospital and Dana-Farber Cancer Institute and associate professor of Otolaryngology at Harvard Medical School in Boston, Massachusetts.

Background and the KEYNOTE-689 Trial Design

In October 2024, Merck, the developer of pembrolizumab, announced that the study had met its primary end point of EFS after neoadjuvant pembrolizumab followed by adjuvant pembrolizumab plus radiotherapy with or without cisplatin, along with maintenance pembrolizumab, led to a significant improvement in EFS vs SOC adjuvant radiotherapy with or without cisplatin in patients with newly diagnosed resectable stage IIA or IVA, locally advanced HNSCC.2

The active-controlled, open-label study evaluated the pembrolizumab combination in treatment naive patients with newly diagnosed resectable stage IIIA or IVA, locally advanced HNSCC, including larynx/hypopharynx/oral cavity cancer (stage IIIA or IVA), oropharyngeal cancer (stage III or IVA p16–), or oropharyngeal cancer (stage III T4 N0-2 p16+).1,3 Patients were also required to have an ECOG performance status of 0 or 1 and tumor tissue for PD-L1 testing. Stratification factors included primary tumor site (oropharynx/oral cavity vs larynx vs hypopharynx), tumor stage (III vs IVA), and PD-L1 tumor proportion score (≥ 50% vs < 50%).1

Patients were randomly assigned 1:1 to receive either the pembrolizumab regimen or SOC. In the pembrolizumab arm, patients were treated with neoadjuvant pembrolizumab in 2 cycles, surgery, then adjuvant pembrolizumab in 3 cycles plus radiotherapy with or without cisplatin, and 12 cycles of pembrolizumab as maintenance therapy. Notably, pembrolizumab was administered at 200 mg once every 3 weeks. Those treated in the SOC arm received surgery and adjuvant radiotherapy with or without cisplatin. Radiation dose depended on pathological risk, which included low risk, high risk, and gross residual disease. Patients from the pembrolizumab and SOC arms received 60 Gy in 30 fractions if they were low risk, 66 Gy in 33 fractions if high risk, and 50 Gy in 35 fractions if they had gross residual disease. Similarly, the 3 cycles of cisplatin were administered at 100 mg/m2 once every 3 weeks, regardless of pathological risk in either arm.

The primary end point was EFS per RECIST 1.1 criteria by blinded independent central review (BICR). Secondary end points included major pathological response (mPR), defined as 10% or less residual invasive squamous cell carcinoma in resected primary tumor and all sampled regional lymph nodes by blinded independent pathologist review (BIPR), OS, and safety.

Baseline Patient Characteristics

Among the pembrolizumab/SOC and SOC arms, respectively, the median age was 60.0 years (range, 29-82) and 61.0 years (range, 22-87), and the majority of patients were male (78.8%; 78.9%). Of note, most were reported to be current or former smokers (80.7%; 76.1%), and currently consume alcohol (68.9%; 67.8%). The primary tumor sites included the oral cavity (pembrolizumab/SOC, 60.3%; SOC, 60.7%), larynx (22.3%; 20.8%), hypopharynx (7.7%; 7.4%), oropharynx (9.6%; 10.8%), and missing (0%; 0.3%). Positive HPV status was reported in 3.3% vs 4.3% of patients, respectively. Additionally, 64.5% and 65.8% of patients had a PD-L1 CPS of 10 or greater in the respective arms; 95.6% and 95.4% had a PD-L1 CPS of 1 or greater.

Each subgroup demonstrated favorability towards the pembrolizumab combination compared with the SOC.

Additional Efficacy Data

At the data cutoff, the 12-month EFS rate in the population with a PD-L1 combined positive score (CPS) of 10 or greater was 74.0% vs 60.0% in the pembrolizumab plus SOC (n = 234) and SOC alone (n = 231) arms, respectively. The respective EFS rates at 24 and 36 months were 65.1% vs 53.2% and 59.8% vs 45.9%. Of note, the median EFS was 59.7 months (95% CI, 41.1-NR) in the pembrolizumab/SOC arm compared with 26.9 months (95% CI, 18.3-51.5) in the SOC alone arm (HR, 0.66; 95% CI, 0.49-0.88; P = .0022). EFS events occurred in 36.3% and 46.3% of patients from the respective arms, including death (pembrolizumab/SOC, 18.4%; SOC, 16.0%), distant progressive disease (6.4%; 16.9%), local and distant progressive disease (0.9%; 0.9%), and local progressive disease with recurrence (10.7%; 12.6%).

Among patients with a CPS of 1 or greater, the 12-, 24-, and 36-month EFS in the pembrolizumab/SOC (n = 347) and SOC (n = 335) arms were 74.8% vs 61.3%, 64.6% vs 53.4%, and 58.2% vs 44.9%, respectively. The median EFS was 59.7 months (95% CI, 37.9-NR) compared with 29.6 months (95% CI, 19.5-41.9), respectively (HR, 0.70; 95% CI, 0.55-0.89; P = .0014). EFS events in this population occurred in 36.9% and 46.6% of patients from the respective arms, including death (pembrolizumab/SOC, 18.2%; SOC, 18.5%), distant progressive disease (6.9%; 15.2%), local and distant progressive disease (1.2%; 1.8%), and local progressive disease with recurrence (10.7%; 11.0%).

The mPR rates in the CPS of 10 or greater population were 13.7% (95% CI, 9.5%-18.8%) vs 0% (95% CI, 0%-1.6%), in the pembrolizumab/SOC and SOC arms, respectively, with an estimated difference of 13.7% (95% CI, 9.7%-18.7%; P < .00001). In the CPS of 1 or greater population, the mPR rates were 9.8% (95% CI, 6.9%-13.4%) vs 0% (95% CI, 0-1.1%). The estimated difference was 9.8% (95% CI, 7.0%-13.3%; P < .00001). Among all patients from both arms, the mPR rates were 9.4% (95% CI, 6.6%-12.8%) vs 0% (95% CI, 0-1.0), with an estimated difference of 9.3% (95% CI, 6.7%-12.8%; P < .00001).

Regarding OS in the CPS of 10 or greater population, the 12-, 24-, and 36-month OS rates in the pembrolizumab/SOC and SOC arms, respectively, were 85.8% vs 77.8%, 74.8% vs 67.1%, and 68.2% vs 59.2%. OS events were reported in 31.2% vs 38.5% of patients, and the median OS was NR (95% CI, NR-NR) vs 61.8 months (95% CI, 49.2-NR; HR, 0.72; 95% CI, 0.52-0.98; P = .02).

Safety Data

The median duration of therapy in the as-treated population was 9.1 months (range, 0.03-22.3) vs 2.9 months (range, 0.03-7.2) in the pembrolizumab/SOC (n = 361) and SOC (n = 315) arms, respectively. Treatment-related adverse effects (TRAEs) included those that led to discontinuation (pembrolizumab/SOC, 17.7%; SOC, 12.4%), serious TRAEs (19.1%; 10.5%), and death (1.1%; 0.3%). Any-grade TRAEs were reported in 81.4% and 81.9% of patients from the respective arms. The most common any-grade TRAEs occurring in at least 10% of patients included radiation skin injury (39.3%; 47.0%), stomatitis (38.8%; 52.1%), hypothyroidism (19.4%; 1.9%), fatigue (18.0%; 13.0%), nausea (17.7%; 21.3%), dry mouth (17.5%; 21.6%), dysgeusia (12.7%; 17.8%), decreased neutrophil count (11.6%; 18.7%), decreased lymphocyte count (11.4%; 8.9%), vomiting (10.8%; 7.9%), decreased white blood cell count (10.8%; 15.9%), dysphagia (10.2%; 15.6%), decreased weight (9.7%; 10.8%), and anemia (8.9%; 10.8%).

Disclosures: Uppaluri reported research funding from Merck Sharp & Dohme LLC (MSD); research funding from Daiichi Sankyo; and consulting fees from Daiichi Sankyo, MSD, and Regeneron.

References

1. Uppaluri R. Neoadjuvant and adjuvant pembrolizumab plus standard of care in resectable locally advanced head and neck squamous cell carcinoma: phase 3 KEYNOTE-689 study. Presented at: 2025 AACR Annual Meeting; April 25-30, 2025; Chicago, IL. Abstract CT001.
2. Merck’s Keytruda (pembrolizumab) met primary endpoint of event-free survival (EFS) as perioperative treatment regimen in patients with resected, locally advanced head and neck squamous cell carcinoma. News release. Merck. October 8, 2024. Accessed April 27, 2025. https://www.merck.com/news/mercks-keytruda-pembrolizumab-met-primary-endpoint-of-event-free-survival-efs-as-perioperative-treatment-regimen-in-patients-with-resected-locally-advanced-head-and-neck-squamous-c/
3. Study of pembrolizumab given prior to surgery in combination with radiotherapy given post-surgery for advanced head and neck squamous cell carcinoma (MK-3475-689). ClinicalTrials.gov. Updated February 7, 2025. Accessed April 27, 2025. https://clinicaltrials.gov/study/NCT03765918