Implementing perioperative immunotherapy for head and neck cancer after KEYNOTE-689: questions and challenges

Neoadjuvant immunotherapy is a recent advance in the care of multiple solid tumors, including melanoma, lung cancer, and head and neck cancers. Until recently, clinical trials of neoadjuvant immunotherapy for head and neck squamous cell carcinoma (HNSCC) were limited to phase II studies, with promising results but small sample sizes, as recently reviewed by Contrera and colleagues (1). However, KEYNOTE-689 (K-689), the first randomized, phase III clinical trial of neoadjuvant immunotherapy for HNSCC, met its primary endpoint of event-free survival (EFS). Neoadjuvant and adjuvant pembrolizumab is now approved by the United States Food and Drug Administration for surgically resectable HNSCC with positive staining for programmed cell death ligand 1 (PD-L1), defined as a combined positive score (CPS) ≥1. While this is major progress in the field of head and neck cancer, there are still several questions left unanswered and challenges to address for full implementation. In this commentary, we summarize the design and results of the trial, discuss some of its more unexpected findings, and elaborate on likely anti-tumor mechanisms of neoadjuvant immunotherapy. Lastly, we highlight some of the ongoing challenges with implementation of this new approach and areas of research that are still needed.

Summary of K-689

The K-689 clinical trial sought to determine whether the addition of programmed cell death 1 (PD-1) directed immunotherapy to the definitive surgical management would extend EFS in HNSCC patients with resectable, locally advanced disease (2). The study included patients with newly diagnosed, nonmetastatic locally advanced HNSCC. CPS scores were determined for each patient as the percentage of PD-L1 staining cells within a tumor section out of the number of viable tumor cells, with subgroups defined as CPS ≥10, CPS ≥1, and total population (including those without CPS scores). Patients were randomized to either standard of care treatment or neoadjuvant pembrolizumab followed by surgery and post operative radiation with pembrolizumab followed by maintenance pembrolizumab (Figure 1). The primary endpoint of EFS was defined as the time from randomization to progression during the neoadjuvant phase, locoregional or distant disease recurrence identified by imaging or biopsy, or death from any cause. Secondary endpoints included overall survival (OS) and major pathologic response (mPR, defined as ≤10% remaining viable tumor) and pathologic complete response. The addition of pembrolizumab to standard of care significantly improved EFS time in all subgroups (Figure 1). Within the pembrolizumab treated group, mPR at time of surgery occurred in 32 patients within the CPS >10 subgroup (9.4%), 34 in the CPS ≥1 subgroup, and 34 in the total population, indicating that most mPRs occurred with the CPS ≥10 group. Interestingly, when stratified by age, there was a lack of benefit from the experimental arm in patients over 65 years of age. When stratified by disease site, patients with hypopharyngeal tumors had worse EFS on the pembrolizumab arm; larynx cancers showed better EFS versus oral cancers only in the CPS ≥10 group, which may be related to smoking and tumor mutational burden. Stage III patients appeared to have more benefit versus stage IV. Adverse event rates were similar between the pembrolizumab and control arms, except for immune-related adverse events (irAEs), which were significantly more common in the pembrolizumab treatment group.

Figure 1 Overview of KEYNOTE-689 study design and results. CPS, combined positive score; CRT, chemoradiation therapy; HNSCC, head and neck squamous cell carcinoma; RT, radiation therapy.

Questions and challenges

The observed results from K-689 are worth discussing. For one, the mPR rate was only 9.4%, and was almost exclusively seen in patients with CPS ≥10. However, mPR was stringently defined as <10% remaining viable tumor tissue. The proportion of patients showing some degree of pathologic response was likely much greater, but this detail was not included in the manuscript.

Another interesting finding was the difference in EFS based on age, with no significant benefit seen in patients over 65 years of age. Our prior work has shown that a subset of surgically treated HNSCC patients, particularly oral cancer patients, have high numbers of memory T cells in circulation that are associated with aging and inferior responses to immunotherapy (3-5); we have previously shown increased risk of locoregional disease relapse in patients with high numbers of these T cells in the blood at the time of surgery (5). It is unclear how much this form of “immuno-aging” might be contributing to the reduced benefit from pembrolizumab in patients over 65 years of age. Other possible contributing factors could be failure to complete adjuvant therapy or less radiation or chemotherapy dose intensity in the older population (results not clear in the K-689 report).

We also look forward with great expectations to the OS data, which remain immature at the moment.

Neoadjuvant immunotherapy: mechanisms and possible advantages

Upon exposure to antigen, T cells increase expression of the immune checkpoint PD-1. Expression of PD-1 increases with continued antigen exposure in order to limit excessive T cell activity. This process is co-opted in the setting of cancer, where many tumors increase expression of PD-L1 to drive T cells towards an exhausted and dysfunctional state (6-8). Through PD-1 directed immunotherapy such as pembrolizumab, these inhibitory signals are removed, reinvigorating T cell activity, increasing the proliferation of a subset of exhausted cells, and enhancing egress from the tumor bed into peripheral circulation (Figure 2A). Treatment with immunotherapy has been shown to induce proliferation of tumor-specific T cells, which is especially important in controlling distant disease and limiting recurrence and metastasis (9,10). Neoadjuvant immunotherapy effectively “vaccinates” the patient against their tumor, resulting in tumor-specific T cells surveilling the body after completion of definitive therapy. The fact that abrogation of distant metastatic disease was the main driver for the EFS improvement supports this notion.

Figure 2 Schematic of pembrolizumab mechanism of action. Created with Biorender.com. (A) Schematic of pembrolizumab mechanisms of action in head and neck cancer; (B) pembrolizumab re-invigorates T cell activation and effector functions; (C) treatment with pembrolizumab causes a proliferative burst of stem-like T cells. HNSCC, head and neck squamous cell carcinoma; IFN-γ, interferon gamma; MHC, major histocompatibility complex; PD-1, programmed cell death 1; PD-L1, programmed cell death ligand 1.

As an immune checkpoint molecule, PD-1 signaling limits T cell response and activity. Blocking PD-1 with immunotherapy removes this inhibition, reinvigorating T cell effector functions including production of cytolytic enzymes and cytokines (e.g., interferon gamma) (11). This restored effector function permits CD8⁺ T cells to effectively induce tumor killing, leading to better tumor control (Figure 2B).

Exhausted T cells exist in multiple stages along a continuum, identified by expression of key transcription factors such as T cell factor 1 (TCF-1) and T-bet. Stem-like exhausted cells are identified by TCF-1 positivity and low expression of T-bet, as the least terminally exhausted cells in this continuum (12,13). Importantly, these exhausted stem-like T cells have the capacity for proliferation upon anti-PD-1 treatment (13,14) (Figure 2C). The proliferative burst from exhausted stem-like T cells in response to anti-PD-1 immunotherapy has been shown to be crucial in immunotherapy response across multiple tumor types (15,16).

Implementation challenges—the actual patient experience

We and others have already noticed nuances, challenges, and barriers in implementing this new treatment paradigm outside of a clinical trial.

(I) Logistics of treatment coordination and cost

Traditionally, patients with surgically treated disease could be seen by a surgeon, then presented at a multidisciplinary tumor board, with additional radiation oncology and medical oncology visits when feasible; this rarely resulted in delays in scheduling surgery. With the K-689 treatment paradigm, the patient must be seen by a surgeon and medical oncologist in a timely fashion, followed by insurance authorization for both pembrolizumab and surgery. Collaboration with a pathologist is also needed to obtain PD-L1 staining in a timely fashion. This requires effective communication among multidisciplinary teams. At our institution, all new head and neck cancer patients are seen initially by the entire multidisciplinary team, as previously described (17); this has greatly facilitated our ability to offer perioperative immunotherapy in a timely fashion. It has also increased the need for infusion space, which is an additional logistic challenge.

These logistic coordination needs may be particularly challenging outside of high-resource settings. Further, the cost of pembrolizumab itself presents additional challenges in some countries and health systems. We note that in some national health systems drug approval is contingent upon cost-effectiveness, which is not a requirement in the United States.

(II) Prolonged total treatment time

Unlike cytotoxic chemotherapy, immune checkpoint blockade requires weeks or months for responses to occur. Although the regimen in K-689 consisted of only 2 doses of pembrolizumab, with surgery scheduled within 6 weeks of randomization, this may double the amount of time during which a patient is waiting for surgery. As noted above, scheduling the patient to see a head and neck surgeon and medical oncologist, then securing insurance authorization, might incur additional delays. It is worth noting that delays to surgery occurred in 12% of patients in the neoadjuvant arm, versus only 3.3% in the standard-of-care arm. During this time, patients may experience more anxiety, endure more pain, and risk tumor progression prior to surgical resection. Pain often requires opioid medication, which has been associated with inferior responses to immune checkpoint blockade in head and neck cancer patients (18). Larynx cancer patients may be at increased risk of aspiration pneumonia while awaiting surgery. Finally, the adjuvant and maintenance doses of pembrolizumab also significantly prolong the total treatment time (by approximately 42 weeks).

(III) Patient selection

It may be tempting to offer neoadjuvant therapy to patients with very advanced local or regional disease, such as T4b oral tumors or N3 nodal disease. However, these stage IVB patients were not included in K-689, and it remains unclear whether K-689 results could be plausibly extrapolated to these cases. These more aggressive and advanced tumors pose the risk of progression on single agent anti-PD-1 therapy to involve an adjacent subsite or become unresectable when surgery is delayed for immunotherapy without a tangible response. In the trial, 45 patients in the neoadjuvant arm demonstrated clinical progression by the time of surgery, compared to 9 in the control arm. In more advanced cases, a chemo-immunotherapy approach could be more plausible, otherwise it may make more sense to proceed directly to surgery.

As noted above, less benefit from pembrolizumab was noted in patients over 65 years of age after subgroup analysis. There also appeared to be less benefit in women and never-smokers, though fewer of those patients were included versus men and smokers, respectively. We have certainly seen women, nonsmokers, and older patients (even octogenarians) with impressive responses, but these patients should be counseled that they may be less likely to respond.

Few patients with human papillomavirus (HPV)-driven oropharyngeal cancer were included, and all had T4 tumors. Thus, neoadjuvant pembrolizumab for HPV-associated should be limited to carefully selected patients with T4 tumors when given outside of a clinical trial. Similarly, the study included few patients with hypopharyngeal cancer (7%), few patients with CPS scores <1 (4%), and zero patients with poor performance status [Eastern Cooperative Oncology Group (ECOG) status >1]; thus, it is unlikely that the results of K-689 can be extrapolated broadly to these groups.

(IV) irAEs

Although pembrolizumab overall is well tolerated, potentially irAEs occurred in almost half of the patients randomized to pembrolizumab, with grade 3 or higher events noted in 10%. Fortunately, life-threatening irAEs, such as grade 4–5 pneumonitis, were rare.

(V) Response-adapted surgery

It is often tempting to proceed with response-adapted surgery, which involves reducing the surgical resection to accommodate a smaller tumor after response to immunotherapy, potentially decreasing morbidity for the patient. This paradigm has been incorporated into the phase III HN014 trial with NRG Oncology, using neoadjuvant cemiplimab for high-risk cutaneous squamous cell carcinoma, a disease with much higher clinical and pathologic response rates to neoadjuvant immunotherapy. However, for K-689, the protocol stipulated that surgeons should not change the surgical plan based on response, resecting the amount of tissue that was estimated upon initial presentation (and sometimes more, in the event of tumor progression). Thus, surgeons should continue to comply with initial surgical plans, since response-adapted surgery has not been established in a phase III study for mucosal HNSCC.

In summary, approval of neoadjuvant and adjuvant pembrolizumab for patients with stage III/IVA opens an exciting chapter in the surgical management of these patients. Nevertheless, patients ought to be carefully selected and counseled on the risks of irAEs, tumor progression, and failure to respond to pembrolizumab. Patients should also be counseled about a longer total treatment time, with the potential for increased morbidity. Lastly, although there is ample evidence to support the idea that patients will have long-term improvements in anti-tumor immunity with a lower risk of tumor recurrence, an OS benefit has not yet been demonstrated, as those data have not yet matured. Further research and experience are needed to refine the neoadjuvant paradigm in real practice, not the least of which is understanding the contribution of specific components, namely the neoadjuvant versus adjuvant benefits of immunotherapy. Although preclinical studies in HNSCC and clinical studies in melanoma suggest the neoadjuvant (preoperative) component may be more important (19,20), the recently completed phase III NIVOPOSTOP study does suggest that the adjuvant/postoperative component does bring some benefit in high-risk patients (21). The K-689 study is therefore a significant yet first step forward. Integration of immune checkpoint blockade with radiation, chemotherapy, and next-generation immunotherapies in the neoadjuvant setting are additional exciting avenues currently under investigation.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the Editorial Office, Translational Cancer Research. The article has undergone external peer review.

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2042/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2042/coif). J.H.G. is a consultant for Acera Surgical and has given lectures on their behalf; and was an institutional PI for a multi-institutional clinical trial sponsored by the University of Arizona. N.F.S. reports paid or unpaid consulting for Astra Zeneca, Eisai Medical, Exelixis, Merck, Merck EMD Serono, Pfizer, Kura, Vaccinex, CUE, BionTech, GSK, TOSK, Seagen, Flamingo, Infinity, Inovio, Aveo, Medscape, Onclive, Uptodate, BMS, J&J, JohnsonCornerstone, Celldex, Surface Oncology, Urogen, Summit, Guidepoints, Astex, Imugene, Faron Pharmaceutical, Coherus, Adagene, Fulgent, Reddy Laboratories, Springer, Nanobiotix, and Taiho, as well as research funding from National Institutes of Health, Exelixis and Bristol Myers Squibb. N.C.S. reports paid consulting for GeoVax, Aspargo Labs, Inc., Johnson & Johnson, and Regeneron, as well as funding from National Institutes of Health and Taiho Oncology. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Contrera KJ, Kansara S, Goyal N, et al. Neoadjuvant Therapy for Mucosal Head and Neck Squamous Cell Carcinoma: A Review From the American Head and Neck Society. JAMA Otolaryngol Head Neck Surg 2025;151:615-25. [Crossref] [PubMed]
2. Uppaluri R, Haddad RI, Tao Y, et al. Neoadjuvant and Adjuvant Pembrolizumab in Locally Advanced Head and Neck Cancer. N Engl J Med 2025;393:37-50. [Crossref] [PubMed]
3. Alpert A, Pickman Y, Leipold M, et al. A clinically meaningful metric of immune age derived from high-dimensional longitudinal monitoring. Nat Med 2019;25:487-95. [Crossref] [PubMed]
4. Ferrara R, Naigeon M, Auclin E, et al. Circulating T-cell Immunosenescence in Patients with Advanced Non-small Cell Lung Cancer Treated with Single-agent PD-1/PD-L1 Inhibitors or Platinum-based Chemotherapy. Clin Cancer Res 2021;27:492-503. [Crossref] [PubMed]
5. Kinney BLC, Brammer B, Kansal V, et al. CD28-CD57+ T cells from head and neck cancer patients produce high levels of cytotoxic granules and type II interferon but are not senescent. Oncoimmunology 2024;13:2367777. [Crossref] [PubMed]
6. Ahmadzadeh M, Johnson LA, Heemskerk B, et al. Tumor antigen-specific CD8 T cells infiltrating the tumor express high levels of PD-1 and are functionally impaired. Blood 2009;114:1537-44. [Crossref] [PubMed]
7. Iwai Y, Ishida M, Tanaka Y, et al. Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade. Proc Natl Acad Sci U S A 2002;99:12293-7. [Crossref] [PubMed]
8. Taube JM, Anders RA, Young GD, et al. Colocalization of inflammatory response with B7-h1 expression in human melanocytic lesions supports an adaptive resistance mechanism of immune escape. Sci Transl Med 2012;4:127ra37. [Crossref] [PubMed]
9. Luoma AM, Suo S, Wang Y, et al. Tissue-resident memory and circulating T cells are early responders to pre-surgical cancer immunotherapy. Cell 2022;185:2918-2935.e29. [Crossref] [PubMed]
10. Sievers C, Craveiro M, Friedman J, et al. Phenotypic plasticity and reduced tissue retention of exhausted tumor-infiltrating T cells following neoadjuvant immunotherapy in head and neck cancer. Cancer Cell 2023;41:887-902.e5. [Crossref] [PubMed]
11. Barber DL, Wherry EJ, Masopust D, et al. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 2006;439:682-7. [Crossref] [PubMed]
12. Hudson WH, Gensheimer J, Hashimoto M, et al. Proliferating Transitory T Cells with an Effector-like Transcriptional Signature Emerge from PD-1(+) Stem-like CD8(+) T Cells during Chronic Infection. Immunity 2019;51:1043-1058.e4. [Crossref] [PubMed]
13. Miller BC, Sen DR, Al Abosy R, et al. Subsets of exhausted CD8(+) T cells differentially mediate tumor control and respond to checkpoint blockade. Nat Immunol 2019;20:326-36. [Crossref] [PubMed]
14. Im SJ, Hashimoto M, Gerner MY, et al. Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy. Nature 2016;537:417-21. [Crossref] [PubMed]
15. Sade-Feldman M, Yizhak K, Bjorgaard SL, et al. Defining T Cell States Associated with Response to Checkpoint Immunotherapy in Melanoma. Cell 2018;175:998-1013.e20. [Crossref] [PubMed]
16. Fang X, Wu G, Hua J, et al. TCF-1(+) PD-1(+) CD8(+)T cells are associated with the response to PD-1 blockade in non-small cell lung cancer patients. J Cancer Res Clin Oncol 2022;148:2653-60. [Crossref] [PubMed]
17. Halle T, Ryan M, Schmitt NC, et al. Designing and Building a Colocated Multidisciplinary Head and Neck Cancer Clinic. NEJM Catal Innov Car NEJM Catal Innov Care Deliv 2023. doi: 10.1056/CAT.22.0235.10.1056/CAT.22.0235
18. Scheff NN, Nilsen ML, Li J, et al. The effect of opioids on the efficacy of immunotherapy in recurrent/metastatic squamous cell carcinoma of the head and neck. Oral Oncol 2023;140:106363. [Crossref] [PubMed]
19. Friedman J, Moore EC, Zolkind P, et al. Neoadjuvant PD-1 Immune Checkpoint Blockade Reverses Functional Immunodominance among Tumor Antigen-Specific T Cells. Clin Cancer Res 2020;26:679-89. [Crossref] [PubMed]
20. Patel SP, Othus M, Chen Y, et al. Neoadjuvant-Adjuvant or Adjuvant-Only Pembrolizumab in Advanced Melanoma. N Engl J Med 2023;388:813-23. [Crossref] [PubMed]
21. Bourhuis J, Auperin A, Borel C, et al. NIVOPOSTOP (GORTEC 2018-01): A phase III randomized trial of adjuvant nivolumab added to radio-chemotherapy in patients with resected head and neck squamous cell carcinoma at high risk of relapse. J Clin Oncol 2025;43:LBA2. [Crossref]