Perioperative pembrolizumab may shift treatment paradigm in resectable HNSCC

In the first interim analysis of the phase III KEYNOTE-689 trial, pembrolizumab improves event-free survival (EFS) and major pathological response (mPR) rate, independent of CPS*, when added to standard-of-care (SoC) treatment for resectable locally advanced head and neck squamous cell carcinoma (LA HNSCC).

“Neoadjuvant pembrolizumab followed by surgery and adjuvant pembrolizumab concurrent with and after postoperative (chemo)radiotherapy [(C)RT] represents a new SoC in [this patient setting],” noted Dr Ravindra Uppaluri from the Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, US, at AACR 2025.

After a median follow-up of 38.3 months, EFS was significantly longer with pembrolizumab + SoC vs SoC only in the CPS ≥10 (median 59.7 vs 26.9 months, hazard ratio [HR], 0.66; p=0.0022), CPS ≥1 (median 59.7 vs 29.6 months, HR, 0.70; p=0.0014), and overall (median 51.8 vs 30.4 months, HR, 0.73; p=0.0041) populations. This treatment effect was consistent across most subgroups.

Pembrolizumab + SoC also trumped SoC only in terms of mPR rate estimated difference (13.7, 9.8, and 9.3 percent for the respective CPS ≥10, CPS ≥1, and overall cohorts; p<0.00001 for all). [AACR 2025, abstract CT001]

The significance boundary for overall survival was not met, but the trend appears to favour pembrolizumab + SoC than SoC only (median not reached vs 61.8 months; HR, 0.72; p=0.02).

Pembrolizumab + SoC was tied to higher rates of serious treatment-related adverse events (TRAEs; 19.1 percent vs 10.5 percent), TRAES that led to discontinuation (17.7 percent vs 12.4 percent) and death (1.1 percent vs 0.3 percent), and grade ≥3 immune-mediated AEs (10 percent vs 0.6 percent) than SoC only. The most common grade 1–2 immune-mediated AEs were thyroid disorders.

SoC yields unsatisfactory outcomes

SoC for resectable HNSCC is surgery, followed by adjuvant radiation with or without chemotherapy. “This paradigm has been in place for over 2 decades; unfortunately, outcomes for many patients continue to be unsatisfactory,” Uppaluri noted in the AACR press release.

“Before surgery, when tumour burden and antigen load are high, immunotherapy can enhance immune responses and start to act on tumour killing. After SoC treatment, immunotherapy can address any leftover cancer cells that may still be present,” he said.

“[W]e had previously shown in a phase II trial that perioperative pembrolizumab was safe and led to intriguing clinical responses in surgically treated head and neck cancer patients,” he added.

To validate these responses, the team sought to evaluate perioperative pembrolizumab in 714 patients (median age 60 years, 78.8 percent men) with resectable LA HNSCC**. The primary tumour site was the oral cavity (60.5 percent), followed by the larynx (21.55 percent). The CPS ≥10 and ≥1 subgroups included 465 and 682 patients, respectively.

The participants were randomized 1:1 to SoC alone or with pembrolizumab*** 200 mg IV Q3W. SoC included surgery (all participants) and postoperative (C)RT (60 Gy in 30 fractions for low-risk, 66 Gy in 33 fractions ± chemo^# for high-risk, or 70 Gy in 35 fractions ± chemo for gross residual disease).

A new SoC?

“Hundreds of patients all over the world participated in this important trial, and implementing their contribution to changing the current SoC is a major goal,” Uppaluri said. “This new information supports changing the current SoC to include neoadjuvant and adjuvant pembrolizumab. For the first time in over 20 years, patients with this challenging disease have a new therapeutic approach.”

“Now that we know this is a safe approach, we can start thinking about how we can potentially modify surgery and/or adjuvant therapy to try to reduce the challenges patients face from the side effects of treatments,” he added.