PDF herunterladen
Open Access
CC BY 4.0 · Indian J Med Paediatr Oncol
DOI: 10.1055/s-0045-1812482
Review Article

Immunotherapy Makes Inroads in Head and Neck Cancer Treatment

Autoren

  • Shruti Gohel 

    1   Department of Medical Oncology, HCG Cancer Center, Vadodara, Gujarat, India

  • Amol Akhade

    2   Department of Medical Oncology, Nair Hospital, Mumbai, Maharashtra, India
    3   Department of Medical Oncology, Bethany Hospital, Thane, Maharashtra, India
Weitere Informationen
Auch verfügbar auf
 
 als PDF herunterladen Lizenzen und Reprints

Abstract

Immunotherapy has transformed the treatment landscape of metastatic head and neck squamous cell carcinoma (HNSCC), but its role in curative-intent settings remained elusive—until now. Recent data from two pivotal phase III trials, NIVOPOSTOP and KEYNOTE-689, mark a turning point by demonstrating statistically significant improvements in disease-free and event-free survival, respectively. However, the magnitude of benefit remains limited, subgroup efficacy is unclear, and overall survival data are immature. Given the logistical complexity and potential for overtreatment, these results warrant cautious interpretation. Future strategies must prioritize biomarker-driven selection, real-world feasibility, and long-term survival outcomes before immunotherapy can claim a definitive role in curative HNSCC.


Keywords

head and neck cancer - HNSCC - immunotherapy - neoadjuvant therapy

Introduction

Head and neck squamous cell carcinoma (HNSCC) represents one of the few areas in oncology where the accelerated advancements and accompanying disappointments associated with immunotherapy have been most striking. Despite its biological plausibility and success in the metastatic disease, the integration of immune checkpoint inhibitors (ICIs) into curative-intent strategies for locally advanced HNSCC has been a frustrating journey of failed promises. After years of negative trials, recent phase III data from NIVOPOSTOP[1] and KEYNOTE-689[2] offer new hope. But hope, as we have learned repeatedly, should not substitute for rigorous evidence.

This article examines whether these trials represent a genuine inflection point in the curative treatment of HNSCC or whether we are once again overestimating modest gains, especially in the absence of mature overall survival (OS) data and realistic global applicability.


A History of Missed Opportunities: A Trial Graveyard

For nearly a decade, trials attempting to incorporate ICIs into curative HNSCC therapy have been largely unsuccessful. JAVELIN Head & Neck 100,[3] arguably the most ambitious early effort, added avelumab to chemoradiotherapy (CRT) but failed to show benefit—a hazard ratio (HR) of 1.21 for progression-free survival was observed, pointing toward potential harm. KEYNOTE-412[4] fared little better: the addition of pembrolizumab to CRT showed a nonsignificant trend (HR 0.83), narrowly missing statistical significance and falling short of the prespecified threshold for success.

Several other trials underscored the same theme. In the REACH study,[5] the cisplatin-fit group paradoxically did worse with the ICI arm (HR 1.27). The PembroRad trial,[6] replacing cetuximab with pembrolizumab in patients unfit for cisplatin, showed an HR of 1.05. Simply put, these trials did not fail because of poor design alone—they failed because the approach itself may be biologically unsound.

Even in the adjuvant setting, where one might expect better immune reconstitution after surgery, the results were disappointing. IMvoke010[7] (atezolizumab postsurgery and radiotherapy [RT]) reported an HR of 0.94, essentially a null trial. The NRG-HN004[8] study of durvalumab in cisplatin-unfit patients showed an HR of 1.33—again suggesting that simply swapping chemotherapy for immunotherapy is not a viable shortcut [Table 1]-[2].


So Why Did These Trials Fail?

Immunotherapy is Not Just Plug-and-Play

The key mistake was assuming that immunotherapy's success in lung cancer or melanoma could be imported wholesale into HNSCC. This disease is biologically distinct—characterized by intense stromal immunosuppression, T-cell exhaustion, and frequently human papillomavirus (HPV)-driven oncogenesis that does not necessarily correlate with immunogenicity.[9] [10]

Moreover, the context of delivery matters. Combining ICIs with high-dose cisplatin and radiation—both profoundly immunosuppressive interventions—may neutralize the very immune activation ICIs require. Concurrent administration may have been the wrong strategy altogether.

Another cardinal flaw was the absence of biomarker-driven stratification. Across many clinical trials, programmed death-ligand 1 (PD-L1) combined positive score (CPS) or HPV/p16 status were often not recorded, not incorporated into therapeutic strategies, or lacked adequate power to determine benefit. Consequently, any potential advantage in a subset was diluted in the broader analysis.



The Turning Point? Dissecting NIVOPOSTOP and KEYNOTE-689

NIVOPOSTOP,[1] presented at ASCO 2025, was the first study to demonstrate a statistically significant improvement in disease-free survival (DFS) with adjuvant nivolumab after surgery and postoperative CRT in patients with high-risk HNSCC. The 3-year DFS was 63.1% with nivolumab versus 52.5% with CRT alone (HR 0.76; p = 0.034). Importantly, benefits were seen regardless of p16 or PD-L1 status, though subgroup sizes were small.

Compliance was excellent: 91% completed RT, > 80% received adequate cisplatin doses, and 75% of patients belonging to the nivolumab arm completed at least 10 cycles. Fewer grade ≥ 3 treatment-related adverse events were observed in the nivolumab group (39%) compared with the control group (49%), referring to overall grade ≥ 3 adverse events (AEs), not immune-related ones. OS data remain immature.

What makes NIVOPOSTOP[1] noteworthy is not just the positive result but that it learned from prior failures. It avoided concurrent administration of ICIs with CRT and instead placed immunotherapy in the adjuvant slot—a setting where the immune system may be less suppressed and more responsive.

KEYNOTE-689,[2] now published in New England Journal of Medicine (June 2025), assessed the use of pembrolizumab in both the neoadjuvant and adjuvant settings for resectable, locally advanced HNSCC. At a median follow-up of 38.3 months, patients treated with pembrolizumab achieved a 3-year event-free survival (EFS) of 59.8%, whereas the control arm reported 45.9% (HR 0.73; 95% confidence interval, 0.58–0.92). Notably, the benefit was even more pronounced in biomarker-enriched subgroups: CPS ≥ 10 (HR 0.60), CPS ≥ 1 (HR 0.70).

Surgery completion rates were similar between the two arms (∼88%), alleviating prior concerns about neoadjuvant ICI interfering with operability. In the pembrolizumab arm, 24.5% of patients experienced grade 3 AEs, compared with 22.5% in the control arm. This robust benefit led to Food and Drug Administration approval on June 12, 2025, for perioperative pembrolizumab in CPS ≥ 1 resectable HNSCC.[11]


Should We Be Celebrating Yet? Let's Pause and Reflect

Despite the excitement surrounding NIVOPOSTOP[1] and KEYNOTE-689,[2] several cautionary issues need to be addressed before calling this a paradigm shift.

Surrogate Endpoints

Neither NIVOPOSTOP[1] nor KEYNOTE-689[2] reported mature OS data. We are again anchoring major practice changes on DFS or EFS—surrogate endpoints that often fail to predict survival in HNSCC. A 10 to 15% EFS improvement may not translate into a meaningful OS benefit, especially in a disease where salvage treatments are possible and quality of life matters.


Subgroup Ambiguity

Benefit in PD-L1–negative or HPV-positive subgroups remains unclear. These are substantial populations. Approving and funding treatment across the board risks overtreatment in many and benefit for few.


No Clarity on Timing

In KEYNOTE-689,[2] is the benefit coming from the neoadjuvant part, the adjuvant part, or both? While the trial succeeded overall, the individual contributions remain undefined, limiting the ability to streamline protocols.


Feasibility in the Real World

Weekly cisplatin, multiple infusions of ICI, and real-time PD-L1 testing—all increase complexity. In low- and middle-income countries, where most HNSCC cases occur, such logistics are often prohibitive. Unless simplified, these regimens will remain a luxury for the few.


No Quality-of-Life Data

Functional outcomes—swallowing, speech, and nutrition—are crucial in HNSCC, especially after multimodal treatment. None of these trials report validated quality-of-life outcomes, which undermines the relevance of their clinical gains.



The Way Forward—Smarter, not Just More Trials

As we plan the next wave of studies, we must prioritize strategic designs over sheer enthusiasm. The following principles should guide future research:

  • Biomarker-driven design: Trials must prospectively stratify by PD-L1 CPS, HPV status, and ideally, immune gene signatures or circulating tumor deoxyribonucleic acid.

  • Minimal residual disease (MRD): Leveraging MRD or pathologic response to guide adjuvant ICI could help limit overtreatment. Adaptive trial designs are sorely needed.[12]

  • Deescalation and organ preservation: Could ICIs allow for reduced radiation fields or avoidance of mutilating surgery in responders? Trials must test this explicitly.


Conclusion

After a long and disappointing journey, immunotherapy has finally shown signs of life in curative HNSCC. But let us be clear: these are small, cautious steps forward—not a revolution.

The survival benefits, while statistically significant, are modest and built on surrogate endpoints. The subgroups that benefit remain murky. The real-world feasibility is questionable, particularly in resource-limited settings. And the long-term impact on quality of life—perhaps the most important outcome in this disease—is still unknown.

If we are to move forward, let it be with scientific humility and practical wisdom. Let us resist the urge to rubber-stamp another expensive, complex treatment protocol based on short-term gains. Instead, we should demand clarity on who benefits, how much, and at what cost.

Immunotherapy in curative HNSCC has earned a seat at the table. But it will take more than two trials to justify putting it at the head.

Table 1

Comparison of major phase III immunotherapy trials in curative-intent HNSCC

Feature

JAVELIN 100 (Avelumab)[3]

KEYNOTE-412 (Pembrolizumab)[4]

KEYNOTE-689 (Pembrolizumab)[2]

NIVOPOSTOP (Nivolumab)[1]

Phase

Phase 3

Phase 3

Phase 3

Phase 3

Setting

Locally advanced, resectable HNSCC

Locally advanced, resectable HNSCC

Locally advanced, resectable HNSCC

Resected, high-risk HNSCC

Control arm

CRT (cisplatin + RT)

CRT (cisplatin + RT)

Surgery → adjuvant RT ± cisplatin

Surgery → CRT (cisplatin + RT)

Experimental arm

CRT + avelumab (before, during, after)

CRT + pembrolizumab (during, after)

Neoadjuvant pembrolizumab → surgery → adjuvant pembrolizumab + RT ± cisplatin

CRT → adjuvant nivolumab

IO start timing

Lead-in before CRT → concurrent CRT

Start with CRT

Neoadjuvant before surgery

Adjuvant after CRT

Primary endpoint

PFS

EFS

EFS

DFS

Primary HR (95% CI)

0.81 (0.62–1.06)

0.83 (0.68–1.03)

0.73 (0.58–0.92)

0.76 (0.60–0.98)

Primary result

Did not meet primary endpoint

Did not meet primary endpoint

Met primary endpoint

Met primary endpoint

OS HR (95% CI)

0.90 (NS)

Trend favorable (immature)

0.72 (0.52–0.98)

Immature, trend favors NIVO

PD-L1 subgroup

PFS HR 0.59 (0.34–1.02)

EFS HR 0.67 (CPS ≥ 20)

13.7% mPR improvement (CPS ≥ 10)

No clear differential by CPS

Biomarker enrichment

No

No

Stratified CPS ≥ 10

No

Safety

Higher immune AEs

Higher immune AEs

Manageable, consistent

Favorable, fewer grade ≥ 3 TRAEs

Trial status

Stopped early (futility)

Completed (failed)

Completed (positive)

Completed (positive)

Publication year

2021

2023

2025

2025

Main summary

Poor timing, no biomarker

Tight alpha, poor timing

Smarter timing, EFS success

First positive adjuvant IO trial in HNSCC

Abbreviations: AE, adverse event; CI, confidence interval; CPS, combined positive score; CRT, chemoradiotherapy; DFS, disease-free survival; EFS, event-free survival; HNSCC, head and neck squamous cell carcinoma; HR, hazard ratio; IO, immuno-oncology; mPR, major pathologic response; NS, not significant; OS, overall survival; PD-L1, programmed death-ligand 1; PFS, progression-free survival; RT, radiotherapy; TRAE, treatment-related adverse event.


Table 2

Failed immunotherapy trials in curative HNSCC

Trial name

N (patients)

Arms

EFS/DFS/PFS

OS

Remarks

REACH (cisplatin-fit group)[5]

∼300

CRT + durvalumab vs. CRT alone

HR 1.27 (PFS)

Not reported

Durvalumab added to CRT in cisplatin-fit patients showed worse outcomes (HR 1.27), raising concerns about ICI synergy with concurrent chemoradiation

PembroRad[6]

133

RT + pembrolizumab vs. RT + cetuximab

HR 1.05 (PFS)

Not reported

Replacing cetuximab with pembrolizumab in RT for cisplatin-ineligible patients did not improve outcomes, highlighting limited efficacy of ICIs in this context

IMvoke010[7]

682

Postop RT + atezolizumab vs. RT alone

HR 0.94 (DFS)

Not reported

Adjuvant atezolizumab post-surgery and RT did not improve DFS. The trial failed to demonstrate added benefit in a setting theoretically favorable for immunotherapy

NRG-HN004[8]

251

RT + durvalumab vs. RT alone (cis-unfit)

HR 1.33 (PFS)

Not reported

Durvalumab added to RT in cisplatin-unfit patients resulted in worse outcomes (HR 1.33), reinforcing the challenge of integrating ICIs in frail populations

Abbreviations: CRT, chemoradiotherapy; DFS, disease-free survival; EFS, event-free survival; HNSCC, head and neck squamous cell carcinoma; HR, hazard ratio; ICI, immune checkpoint inhibitor; OS, overall survival; PFS, progression-free survival; RT, radiotherapy.




Conflict of Interest

None declared.

Patient's Consent

Patient consent is not required.



Address for correspondence

Shruti Gohel, DrNB
Department of Medical Oncology, HCG Cancer Center
Vadodara, Gujarat
India   

Publikationsverlauf

Artikel online veröffentlicht:
03. November 2025

© 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/)

Thieme Medical and Scientific Publishers Pvt. Ltd.
A-12, 2nd Floor, Sector 2, Noida-201301 UP, India


  Lizenzen und Reprints