Skip to content
Bending the Curve Trial by Trial
?

Head and neck: oral cavity, pharynx, larynx and nasopharynx (upper aerodigestive tract)

Head and neck cancer

How tobacco, alcohol and HPV research, organ preservation, chemoradiotherapy and immunotherapy shaped care for head and neck cancers.

Oral cavity and pharynx cancer only · 5-year relative survival · United States, SEER 8 registries

54.32%→73.6%

1975 (observed value; first year in the table) to 2018 (observed value; last observed year in the table). Read these with the caveats under “Then and now”.Source

25 milestones, by decade

  • A case-control study in four areas of the United States used the tobacco and alcohol histories of 1,114 people with oral or pharyngeal cancer and 1,268 population-based controls. From attributable-risk calculations, its authors estimated that smoking and drinking together account for approximately three-fourths of oral and pharyngeal cancers in the United States. Among people using both, risk of oropharyngeal cancer was increased more than 35-fold at the highest intakes (two or more packs of cigarettes and more than four alcoholic drinks a day), and the abstract reports no excess among those who had stopped smoking for 10 or more years. The study was published in 1988 and its abstract does not mention HPV.
  • In a study of 271 oropharyngeal cancers from three SEER registries (1984-2004), HPV prevalence by one assay rose from 16.3% in 1984-1989 to 71.7% in 2000-2004. The authors' reweighted estimates put population-level incidence of HPV-positive oropharyngeal cancer up 225% from 1988 to 2004 (0.8 to 2.6 per 100,000) and HPV-negative incidence down 50% (2.0 to 1.0 per 100,000). Median survival was 131 months for HPV-positive and 20 months for HPV-negative cases (adjusted hazard ratio 0.31, 95% CI 0.21 to 0.46). These are model-based estimates from a tested sample of registry specimens, and the abstract's projection to 2020 is not repeated here.
  • Two randomised trials in HPV-positive oropharyngeal cancer, both reported online in 2018, tested replacing cisplatin with cetuximab alongside radiotherapy. In RTOG 1016, cetuximab did not meet the non-inferiority criterion for overall survival (hazard ratio 1.45, one-sided 95% upper CI 1.94), with estimated 5-year overall survival of 77.9% versus 84.6% with cisplatin. In De-ESCALaTE HPV, severe toxicity at 24 months did not differ between arms (mean 4.8 events per patient in each), while 2-year overall survival was 97.5% with cisplatin and 89.4% with cetuximab. The trials enrolled different populations and used different primary endpoints, so their figures are not directly comparable.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 25
Treatment type
Setting

Molecular origins

  1. 1988 · Understanding · UnderstandingTobacco and alcohol reported to combine in head and neck cancer riskStudies of people with and without mouth, throat and laryngeal cancer reported that smoking and drinking together raise risk far more than either alone: the two effects multiply rather than add. A 1988 US study estimated that the two habits together account for about three-quarters of oral and pharyngeal cancers there.Case-control and pooled data. Blot 1988 (1114 cases, 1268 controls): oral and pharyngeal cancer risks combined more multiplicatively than additively, over 35-fold at two or more packs and more than four drinks daily. Tuyns 1988 (1,147 male cases, 3,057 controls; larynx/hypopharynx): joint exposure consistent with a multiplicative model. Hashibe 2009 (INHANCE, 11,221 cases, 16,168 controls): greater than multiplicative, psi 2.15 (95% CI 1.53 to 3.04); PAR for tobacco or alcohol 72% (61 to 79%).
  2. 1991 · De-escalation · ChemotherapyInduction chemotherapy plus radiation preserved the larynx (VA)In a Veterans Affairs study, people with advanced voice-box cancer were randomly assigned to chemotherapy followed by radiation, or to surgery followed by radiation. About two in three people were alive at two years in both groups, and about two in three of those given chemotherapy kept their voice box. The authors called the results preliminary.Randomised study, 332 patients with untreated stage III-IV laryngeal squamous carcinoma: induction cisplatin-fluorouracil then radiation (salvage laryngectomy if no response or recurrence) versus laryngectomy plus postoperative radiation. Median follow-up 33 months; estimated 2-year survival 68% in both groups (P = 0.9846). More local recurrences (P = 0.0005) and fewer distant metastases (P = 0.016) with chemotherapy. Larynx preserved in 64%; 59 patients (36%) needed laryngectomy. Authors called the results preliminary.

Targeted therapy

  1. 2000 · Understanding · UnderstandingHPV linked to a distinct subset of head and neck cancersFrom 2000, studies reported that human papillomavirus (HPV) is found in a subset of head and neck cancers, especially in the oropharynx, and is likely a cause. In US registry samples, the share of oropharyngeal cancers with HPV rose from about one in six in the 1980s to about seven in ten by 2000 to 2004.Gillison 2000 (253 HNSCC): HPV in 25% (95% CI 19 to 30%); oropharyngeal site OR 6.2 for HPV presence; HPV-positive tumours HR for death 0.41 (0.20 to 0.88). D'Souza 2007 (100 cases, 200 controls): oral HPV-16 infection OR 14.6 (6.3 to 36.6). Chaturvedi 2011 (SEER, 271 tumours): HPV prevalence 16.3% (1984 to 1989) to 71.7% (2000 to 2004); HPV-positive incidence up 225% (1988 to 2004).
  2. 2000 · Understanding · ChemotherapyMACH-NC: chemotherapy added to locoregional treatment (2000)Researchers pooled individual patient data from many randomised trials of adding chemotherapy to surgery or radiation for head and neck cancer. The first report found only a small survival gain. Later updates reported benefit for chemotherapy given together with radiation, but not for chemotherapy given before or after.Individual-patient-data meta-analyses (MACH-NC). 2000: 63 trials, 10,741 patients, hazard ratio of death 0.90 (95% CI 0.85-0.94; P<0.0001), absolute benefit 4% at 2 and 5 years; three larynx-preservation trials hazard ratio 1.19 (0.97-1.46). 2009 update, 50 concomitant trials: hazard ratio 0.81, absolute benefit 6.5% at 5 years. 2021 update: concomitant 0.83 (0.79-0.86); induction 0.96 (0.90-1.01); adjuvant 1.02 (0.92-1.13).
  3. 2003 · Treatment · ChemotherapyRTOG 91-11: concurrent cisplatin for larynx preservationIn a three-way trial for locally advanced voice-box cancer, radiation with cisplatin given at the same time kept the larynx in place more often at two years than chemotherapy first or radiation alone. Survival was similar across the groups, and serious side effects were more frequent with the chemotherapy-based treatments.Intergroup RTOG 91-11, 547 randomised patients with locally advanced laryngeal cancer: induction cisplatin-fluorouracil then radiotherapy, radiotherapy with concurrent cisplatin, or radiotherapy alone. At 2 years, intact larynx 88% (concurrent) vs 75% (induction; P=0.005) and 70% (radiotherapy alone; P<0.001). Long-term report (520 analysed, median follow-up 10.8 years): laryngectomy-free survival HR 0.75 (induction) and 0.78 (concurrent) vs radiotherapy alone; overall survival concurrent vs induction HR 1.25 (95% CI 0.98-1.61).
  4. 2004 · Treatment · ChemotherapyAdding cisplatin to postoperative radiotherapy for high-risk diseaseTwo large trials gave people with high-risk head and neck cancer radiation after surgery, with or without cisplatin. Cisplatin improved control of the cancer in the head and neck area in both trials and survival in one, with more severe side effects. Long-term follow-up of one trial found no overall difference.RTOG 9501 (N=459): local or regional recurrence HR 0.61 (95% CI 0.41-0.91); DFS HR 0.78; OS HR 0.84 (0.65-1.09; P=0.19). EORTC 22931 (N=334): 5-year PFS 47% vs 36% (HR 0.75); OS 53% vs 40% (HR 0.70; 0.52-0.95). Cisplatin vs radiotherapy alone throughout. Grade 3 or higher adverse effects 77% vs 34% (RTOG 9501), 41% vs 21% (EORTC 22931). RTOG 9501 at 10 years: no significant overall differences; benefit in an unplanned subset.
  5. 2006 · Treatment · TargetedCetuximab with radiotherapy improves locoregional control and survivalIn this trial, people with advanced head and neck cancer had radiation alone or radiation with the antibody drug cetuximab. Cancer stayed controlled in the head and neck area about ten months longer with cetuximab, and median survival was about twenty months longer, with no more of the common radiation side effects apart from rash and infusion reactions.Randomised study, radiotherapy alone (n=213) vs plus weekly cetuximab (n=211), locoregionally advanced head and neck squamous-cell carcinoma. Cetuximab vs radiotherapy alone: median locoregional control 24.4 vs 14.9 months (HR 0.68; P=0.005). At median follow-up 54.0 months, median OS 49.0 vs 29.3 months (HR 0.74; P=0.03). Five-year report: OS HR 0.73 (95% CI 0.56-0.95), 5-year OS 45.6% vs 36.4%. Cetuximab was not compared with cisplatin-based chemoradiotherapy.
  6. 2008 · Treatment · TargetedEXTREME: cetuximab added to platinum-fluorouracil, first linePeople with head and neck squamous cell cancer that had returned or spread, and who had not yet had treatment for that stage, were given chemotherapy with or without the antibody cetuximab. Those who also received cetuximab lived longer on average, about ten months compared with seven, and the cancer took longer to grow.Randomised phase 3 trial, untreated recurrent or metastatic SCCHN (N=442): cisplatin or carboplatin plus fluorouracil, with or without weekly cetuximab. Median OS 10.1 vs 7.4 months (HR 0.80; 95% CI 0.64-0.99; P=0.04). Median PFS 5.6 vs 3.3 months (HR 0.54; P<0.001); response rate 36% vs 20%. Sepsis occurred in 9 patients with cetuximab and 1 without. FDA approved this first-line indication on 7 November 2011.
  7. 2010 · Understanding · UnderstandingTumour HPV status found to be a strong, independent prognostic factorA retrospective analysis of people with stage III or IV oropharyngeal cancer treated in a chemoradiotherapy trial found that those with HPV-positive tumours did much better: about eight in ten were alive at three years, against about six in ten with HPV-negative tumours. The analysis was not a test of treating them less.Retrospective analysis by HPV status of stage III or IV oropharyngeal SCC in RTOG 0129 (accelerated vs standard fractionation radiotherapy, each with cisplatin). HPV-positive 206 of 323 (63.8%). 3-year OS 82.4% vs 57.1% (P<0.001); adjusted HR for death 0.42 (95% CI 0.27 to 0.66); median follow-up 4.8 years. Recursive partitioning (HPV status, pack-years, T and N stage) defined low, intermediate and high risk.

Immunotherapy

  1. 2015 · Treatment · SurgeryElective vs therapeutic neck dissection in early oral cancer (Tata)In early oral cancer with no sign of spread to the neck, 500 patients were included in a prespecified report comparing removal of the neck lymph nodes during the first operation with waiting and operating only if cancer appeared in the neck. At three years, survival was about 80 percent with early neck surgery and about 68 percent with waiting.Tata Memorial randomised trial, lateralised T1-T2 oral squamous-cell carcinoma; prespecified report of the first 500 of 596 patients (245 elective, 255 therapeutic neck dissection), median follow-up 39 months. 3-year OS 80.0% (95% CI 74.1-85.8) vs 67.5% (61.0-73.9); HR for death 0.64 (95% CI 0.45-0.92; P=0.01). 3-year DFS 69.5% vs 45.9% (P<0.001). Adverse events 6.6% vs 3.6%.
  2. 2016 · Treatment · ImmunotherapyCheckMate 141: nivolumab versus standard therapy after platinumPeople whose head and neck squamous cell cancer had grown within six months of platinum chemotherapy received either the immune drug nivolumab or one standard single drug. Median survival was 7.5 months with nivolumab and 5.1 months with standard treatment, and fewer people had severe (grade 3 or 4) treatment-related side effects with nivolumab.Randomised open-label phase 3 (2:1; N=361): recurrent SCCHN progressing within 6 months of platinum-based chemotherapy; nivolumab 3 mg per kilogram every 2 weeks vs methotrexate, docetaxel or cetuximab. Median OS 7.5 vs 5.1 months (HR 0.70; 97.73% CI 0.51-0.96; P=0.01); 1-year OS 36.0% vs 16.6%. Median PFS 2.0 vs 2.3 months (HR 0.89; P=0.32). Response 13.3% vs 5.8%. Grade 3-4 treatment-related adverse events 13.1% vs 35.1%.
  3. 2016 · De-escalation · RadiationReduced-dose radiotherapy tested for good-risk HPV-positive oropharynxThree phase 2 studies tested giving less radiation to people with HPV-related throat cancer who had a favourable outlook. Most people in the reduced-dose groups had no progression at two years, and swallowing was better with lower doses in one study. These studies could not show that less treatment is as effective.Phase 2 studies in HPV-associated oropharyngeal cancer. E1308 (single arm, 80 evaluable): after induction chemotherapy and primary-site complete response, 54 Gy plus cetuximab, 2-year PFS 80% (n=51). NRG-HN002 (306 randomly assigned and eligible): 2-year PFS 90.5% with 60 Gy plus cisplatin and 87.6% with 60 Gy over 5 weeks (historical benchmark 85%). E3311 (359 evaluable): 2-year PFS 94.9% (50 Gy) and 96.0% (60 Gy) after transoral surgery.
  4. 2016 · Treatment · Chemotherapy⚑ needs reviewGemcitabine-cisplatin versus fluorouracil-cisplatin in metastatic NPCIn China, people with nasopharyngeal cancer that had returned or spread received either gemcitabine with cisplatin or fluorouracil with cisplatin. The time before the cancer grew or the person died was longer with gemcitabine: a median of 7.0 months, compared with 5.6 months.Multicentre, randomised, open-label phase 3 (N=362; 22 hospitals in China): recurrent or metastatic NPC; gemcitabine plus cisplatin vs fluorouracil plus cisplatin every 3 weeks for up to 6 cycles. Median PFS (independent committee) 7.0 vs 5.6 months (HR 0.55; 95% CI 0.44-0.68; P<0.0001). Grade 3 or 4 leucopenia 29% vs 9%; mucosal inflammation 0 vs 14%.
  5. 2016 · Treatment · Immunotherapy⚑ needs reviewPembrolizumab after platinum: accelerated approval, then KEYNOTE-040Pembrolizumab, another immune drug, was approved in the United States in 2016 for head and neck cancer that had progressed on platinum chemotherapy, with approval linked to how many tumours shrank. A later randomised trial found median survival of 8.4 months with pembrolizumab and 6.9 months with standard single drugs.KEYNOTE-012 (phase 1b, single-arm): response rate 18% (95% CI 8-32) in 45 evaluable PD-L1-positive patients and 18% (95% CI 12-26) in a 132-patient unselected expansion cohort. FDA accelerated approval 5 August 2016. KEYNOTE-040 (phase 3, N=495, vs methotrexate, docetaxel or cetuximab): median OS 8.4 vs 6.9 months (HR 0.80; 95% CI 0.65-0.98; nominal P=0.0161). FDA fulfilled the postmarketing requirement 10 June 2019.
  6. 2017 · Understanding · UnderstandingAJCC 8th edition stages HPV-associated oropharyngeal cancer separatelyIn its eighth edition, published as a 2017 summary by the task force, the American Joint Committee on Cancer introduced a separate staging scheme for oropharyngeal cancer associated with high-risk HPV, apart from oropharyngeal cancer with other causes. The edition also reorganised chapters on the pharynx and skin.Lydiatt 2017 summarises the AJCC eighth edition head and neck section. Most significant change: a separate staging algorithm for high-risk HPV-associated oropharyngeal cancer. Other changes: pharynx divided into 3 chapters; cutaneous malignancies moved to a head and neck chapter; revised T categories for oral cavity, skin and nasopharynx; extranodal extension added to N category except in viral-related cancers and mucosal melanoma.
  7. 2017 · Detection · Detection & prevention⚑ needs reviewPlasma EBV DNA tested as a screening test for nasopharyngeal cancerIn a 2017 Hong Kong study, more than 20,000 people with no symptoms had a blood test for Epstein-Barr virus (EBV) DNA, with repeat testing, endoscopy and MRI if it stayed positive. Thirty-four were found to have nasopharyngeal cancer, more often at an early stage than in a historical comparison group. There was no randomised control group.Prospective study of plasma EBV DNA screening in 20,174 participants without symptoms (registry: ethnic Chinese men aged 40 to 62). Persistently positive in 309; 34 nasopharyngeal carcinomas. Stage I or II in 71% vs 20% of a historical cohort (P<0.001); 3-year PFS 97% vs 70% (HR 0.10; 95% CI 0.05 to 0.18). Sensitivity 97.1%, specificity 98.6%. No mortality endpoint in the abstract.
  8. 2018 · Reversal · TargetedCetuximab worse than cisplatin with radiotherapy, HPV-positive cancerTwo trials in HPV-related throat cancer asked whether cetuximab could replace cisplatin alongside radiation to reduce side effects. In both, people given cetuximab did worse on survival or cancer control, and neither trial found a clear side-effect advantage for cetuximab. Both teams recommended cisplatin as the standard of care.Two phase 3 trials in HPV-positive oropharyngeal cancer compared radiotherapy with cetuximab vs cisplatin. RTOG 1016 (849 randomised; non-inferiority margin 1.45): OS HR 1.45 (one-sided 95% upper CI 1.94), 5-year OS 77.9% (cetuximab) vs 84.6% (cisplatin); PFS HR 1.72 (95% CI 1.29-2.29). De-ESCALaTE (N=334, low risk): severe toxicity events did not differ (4.8 vs 4.8 per patient); 2-year OS 89.4% (cetuximab) vs 97.5% (cisplatin).
  9. 2019 · Treatment · Immunotherapy⚑ needs reviewKEYNOTE-048: pembrolizumab first line, recurrent or metastaticAdults whose head and neck cancer had returned or spread, and who had not yet been treated for it, received pembrolizumab alone, pembrolizumab with chemotherapy, or cetuximab with chemotherapy. Median survival was 13.0 months with pembrolizumab plus chemotherapy against 10.7 months with cetuximab plus chemotherapy; pembrolizumab alone improved survival in people whose tumours expressed PD-L1.Randomised open-label phase 3 (1:1:1; N=882): untreated recurrent or metastatic HNSCC; pembrolizumab, pembrolizumab plus platinum-5-FU, or cetuximab plus platinum-5-FU. Median OS, pembrolizumab plus chemotherapy vs control, total population: 13.0 vs 10.7 months (HR 0.77; 95% CI 0.63-0.93). Pembrolizumab alone, CPS 1 or more: 12.3 vs 10.3 months (HR 0.78; 95% CI 0.64-0.96). Neither regimen improved PFS at the second interim analysis.
  10. 2019 · Understanding · SurgeryORATOR: transoral robotic surgery vs radiotherapy, oropharynxA small randomised trial compared radiation with transoral robotic surgery plus neck dissection for early throat cancer. Swallowing-related quality of life one year later was somewhat better after radiation, though the authors said the difference was not clinically meaningful. At five years there were no differences in survival, and side effects differed in pattern.ORATOR (phase 2, N=68, 34 per arm; T1-T2, N0-2 oropharyngeal squamous cell carcinoma): radiotherapy (with chemotherapy if N1-2) vs transoral robotic surgery plus neck dissection. Primary endpoint, swallowing quality of life (MDADI) at 1 year: mean 86.9 (SD 11.4) vs 80.1 (13.0), P = 0.042; authors judged the difference not clinically meaningful. Final results (median follow-up 5.1 years): MDADI scores converged, grade 2-5 adverse events 91% (radiotherapy) vs 97% (TORS plus neck dissection), P = 0.61, no differences in OS or PFS.

Precision & engineered

  1. 2020 · De-escalation · SurgerySentinel-node strategy vs neck dissection in early oral cancerIn operable early oral and oropharyngeal cancer, patients were randomly assigned to neck dissection or to a sentinel node biopsy, with neck dissection only if the sentinel node held cancer. Neck recurrence-free survival at two years was about 90 percent with both approaches, and functional outcomes were better with the sentinel approach during the first six months.Phase 3 equivalence trial, 307 randomised, T1-T2 N0 operable oral or oropharyngeal cancer; 279 analysable (139 neck dissection, 140 sentinel node). Neck node recurrence-free survival at 2 years 89.6% with dissection vs 90.7% with the sentinel strategy (equivalence P<.01). 5-year RFS and 2- and 5-year DSS and OS were not significantly different. Median hospital stay 8 vs 7 days (P<.01); functional outcomes worse after dissection up to 6 months.
  2. 2021 · Reversal · ImmunotherapyFour add-ons to definitive head and neck treatment show no benefitFour large trials added a new drug to standard treatment for head and neck cancer that had not spread widely: avelumab, pembrolizumab or xevinapant during chemoradiotherapy, or atezolizumab afterwards. None improved the main outcome, and with xevinapant the results were worse than with placebo.Four randomised phase 3 trials (N=697, 804, 406, 730) in locally advanced SCCHN. Avelumab plus chemoradiotherapy: PFS HR 1.21 (95% CI 0.93-1.57). Pembrolizumab plus chemoradiotherapy: EFS HR 0.83 (95% CI 0.68-1.03). Adjuvant atezolizumab: EFS HR 0.94 (95% CI 0.70-1.26). Xevinapant plus chemoradiotherapy: EFS HR 1.33 (95% CI 1.05-1.67); OS HR 1.39 (95% CI 1.04-1.86). Results are not pooled.
  3. 2021 · Treatment · Immunotherapy⚑ needs reviewPD-1 antibodies added to gemcitabine-cisplatin in first-line NPCIn three trials in Asia, people with nasopharyngeal cancer that had returned or spread received gemcitabine and cisplatin with either a PD-1 immune drug or placebo. In each trial the cancer took longer to grow with the immune drug; in the toripalimab trial the median was 21.4 months at final analysis, compared with 8.2 months with placebo.Three double-blind phase 3 trials, same gemcitabine-cisplatin backbone vs placebo, first-line recurrent or metastatic NPC. Primary-analysis PFS HR (independent review): toripalimab 0.52 (95% CI 0.36-0.74; N=289), camrelizumab 0.54 (95% CI 0.39-0.76; N=263), tislelizumab 0.52 (95% CI 0.38-0.73; N=263). JUPITER-02 final OS HR 0.63 (95% CI 0.45-0.89). CAPTAIN-1st 5-year OS 37.8% vs 24.2% (HR 0.74). FDA approved toripalimab 27 October 2023 and penpulimab 23 April 2025.
  4. 2022 · Reversal · ImmunotherapyFirst-line immunotherapy combinations: no overall survival gainFour randomised trials tested a second drug alongside immunotherapy as first treatment for head and neck cancer that had returned or spread. The combinations were durvalumab with tremelimumab, nivolumab with ipilimumab, lenvatinib with pembrolizumab, and feladilimab with pembrolizumab. None improved overall survival over its comparator.Four randomised first-line R/M SCCHN trials, not pooled. KESTREL (PD-L1-high): durvalumab vs EXTREME, OS HR 0.96 (95% CI 0.69-1.32); durvalumab plus tremelimumab vs EXTREME, HR 1.05 (0.80-1.39). CheckMate 651: nivolumab plus ipilimumab vs EXTREME, OS HR 0.95 (97.9% CI 0.80-1.13). LEAP-010: lenvatinib plus pembrolizumab vs placebo plus pembrolizumab, PFS HR 0.64, OS HR 1.15 (95% CI 0.91-1.45). INDUCE-3: OS HR 1.51 (95% CI 0.99-2.29).
  5. 2025 · Treatment · ImmunotherapyKEYNOTE-689: perioperative pembrolizumab for resectable diseaseFor people with head and neck cancer that could be removed by surgery, two doses of pembrolizumab before surgery and fifteen afterwards, alongside the usual radiotherapy, were compared with usual care alone. At three years, about 58 in 100 people were free of progression, recurrence or death with pembrolizumab, compared with about 46 in 100 with usual care.Open-label phase 3 (N=714): resectable locally advanced HNSCC; 2 cycles of neoadjuvant and 15 of adjuvant pembrolizumab plus standard care (surgery, adjuvant radiotherapy with or without cisplatin) vs standard care. First interim analysis, median follow-up 38.3 months: 36-month EFS 59.8% vs 45.9% in CPS 10 or more (HR 0.66; 95% CI 0.49-0.88), 58.2% vs 44.9% in CPS 1 or more (HR 0.70), 57.6% vs 46.4% in the total population (HR 0.73; 95% CI 0.58-0.92).
  6. 2025 · Treatment · ImmunotherapyNIVOPOSTOP: nivolumab added to postoperative cisplatin-radiotherapyAfter surgery for head and neck cancer with features that raise the risk of relapse, people had cisplatin chemotherapy with radiotherapy, with or without the immune drug nivolumab. Relapse or death was less likely with nivolumab, and grade 4 side effects related to treatment were more common (about 10 in 100 people compared with 5 in 100).Open-label phase 3, N=680 randomised 1:1 (efficacy analysis of 666 assigned up to cutoff): resected SCCHN with at least one high-risk pathological feature; cisplatin 100 mg per square metre and radiotherapy 66 Gy with or without nivolumab. Median follow-up 30.3 months. Investigator-assessed DFS HR 0.76 (95% CI 0.60-0.98; stratified log-rank P=0.034). Treatment-related grade 4 adverse events 10% vs 5%.

Then and now

  • 5-year relative survival · United States, SEER 8 registries

    • 54.32%

      1975 (observed value; first year in the table)

    All races, both sexes; SEER Cancer Stat Facts category 'oral cavity and pharynx cancer' (the page does not list the sites included and does not report oropharyngeal cancer by HPV status). SEER 8 five-year relative survival series, 1975-2018. Age range is not stated for this series.

    Modeled trend value for 1975: 54.21%. The observed series moves from year to year (lowest 52.21% in 1982, highest 74.22% in 2017). Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column. The page headings and chart note ('SEER 8 5-Year Relative Survival Percent from 1975-2018') do not label the 'Year' column as year of diagnosis. SEER 8 covers a limited set of registries. The page does not describe changes over time in site mix, HPV status, staging or treatment, does not apportion any change and attributes it to no cause. Read beside the mortality values. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 73.6%

      2018 (observed value; last observed year in the table)

    All races, both sexes; SEER Cancer Stat Facts category 'oral cavity and pharynx cancer' (the page does not list the sites included and does not report oropharyngeal cancer by HPV status). SEER 8 five-year relative survival series, 1975-2018. Age range is not stated for this series.

    Modeled trend value for 2018: 73.30%; the table shows no observed five-year value for 2019 onward. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column. The page headings and chart note ('SEER 8 5-Year Relative Survival Percent from 1975-2018') do not label the 'Year' column as year of diagnosis. SEER 8 covers a limited set of registries. The page does not describe changes over time in site mix, HPV status, staging or treatment, does not apportion any change and attributes it to no cause. Read beside the mortality values. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 69.9%

      2016-2022

    All races, both sexes; oral cavity and pharynx cancer, all stages combined (SEER 21 excluding IL, 2016-2022, as shown in the page headline).

    A different registry set and period from the SEER 8 series above; do not chain or subtract them. Relative survival excludes deaths from other causes.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 88.7%

      2016-2022; localized stage at diagnosis (26% of cases)

    All races, both sexes; oral cavity and pharynx cancer, SEER Combined Summary Stage 'localized' (confined to the primary site).

    SEER summary stage, not AJCC TNM or HPV/p16 status. Same SEER 21 series as the all-stages value; the stage percentages (26, 55, 12 and 6 per cent for localized, regional, distant and unknown) are as displayed on the page, which gives 25.9% in its text for localized.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 69.7%

      2016-2022; regional stage at diagnosis (55% of cases)

    All races, both sexes; oral cavity and pharynx cancer, SEER Combined Summary Stage 'regional' (spread to regional lymph nodes).

    SEER summary stage, not AJCC TNM or HPV/p16 status. Same SEER 21 series as the all-stages value.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 36%

      2016-2022; distant stage at diagnosis (12% of cases)

    All races, both sexes; oral cavity and pharynx cancer, SEER Combined Summary Stage 'distant' (cancer has metastasized).

    SEER summary stage, not AJCC TNM or HPV/p16 status. Same SEER 21 series as the all-stages value; unknown stage (6% of cases) is shown by the page as 62.3%.

    Source

  • Age-standardised mortality · United States (U.S. Mortality data shown on the SEER page; not a SEER registry series)

    • 4.3per 100,000

      1975 (observed value; first year in the table)

    All races, both sexes; oral cavity and pharynx cancer deaths; age-adjusted rate per 100,000 (the page text read does not name the standard population).

    Death-rate series 1975-2024, observed. The 1977 value is 4.32, the highest in the table. Modeled trend value for 1975: 4.40. Mortality depends on incidence as well as survival: SEER 8 incidence per 100,000 was 13.70 in 1975 and 12.06 in 2023 (observed). Shown beside the survival series.

    Source

  • Age-standardised mortality · United States (U.S. Mortality data shown on the SEER page; not a SEER registry series)

    • 2.37per 100,000

      2009 (observed value; lowest value in the 1975-2024 table)

    All races, both sexes; oral cavity and pharynx cancer deaths; age-adjusted rate per 100,000 (the page text read does not name the standard population).

    Modeled trend value for 2009: 2.42. Observed values after 2009 are higher (2.74 in 2024); the page text reports that age-adjusted death rates have been rising on average 0.8% each year over 2015-2024.

    Source

  • Age-standardised mortality · United States (U.S. Mortality data shown on the SEER page; not a SEER registry series)

    • 2.74per 100,000

      2024 (observed value; last year in the table)

    All races, both sexes; oral cavity and pharynx cancer deaths; age-adjusted rate per 100,000 (the page text read does not name the standard population).

    Modeled trend value for 2024: 2.72. The page headline gives 2.7 per 100,000 (2020-2024 deaths) and 11.7 per 100,000 for new cases (2019-2023, SEER 21). Deaths by site depend on cause-of-death coding, which the page does not discuss.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 62.54%

      1975 (observed value; first year in the table)

    All races, both sexes; laryngeal cancer. SEER 8 five-year relative survival series, 1975-2018. Age range is not stated for this series.

    Modeled trend value for 1975: 67.75%, so the observed 1975 value sits below the fitted line. The observed series moves widely from year to year (lowest 60.70% in 2003, highest 72.96% in 1977) and the modeled trend declines from 67.75% (1975) to 63.33% (2018), so the two observed endpoints do not describe a steady change. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column. The page headings and chart note ('SEER 8 5-Year Relative Survival Percent from 1975-2018') do not label the 'Year' column as year of diagnosis. SEER 8 covers a limited set of registries. The page does not describe changes over time in site mix, HPV status, staging or treatment, does not apportion any change and attributes it to no cause. Read beside the mortality values. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 67.27%

      2018 (observed value; last observed year in the table)

    All races, both sexes; laryngeal cancer. SEER 8 five-year relative survival series, 1975-2018. Age range is not stated for this series.

    Modeled trend value for 2018: 63.33%, so the observed 2018 value sits above the fitted line; the table shows no observed five-year value for 2019 onward. Relative survival excludes deaths from other causes. 'Observed' is the table's label for the actual yearly value, as opposed to the modeled trend value in the adjacent column. The page headings and chart note ('SEER 8 5-Year Relative Survival Percent from 1975-2018') do not label the 'Year' column as year of diagnosis. SEER 8 covers a limited set of registries. The page does not describe changes over time in site mix, HPV status, staging or treatment, does not apportion any change and attributes it to no cause. Read beside the mortality values. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 62.5%

      2016-2022

    All races, both sexes; laryngeal cancer, all stages combined (SEER 21 excluding IL, 2016-2022, as shown in the page headline).

    A different registry set and period from the SEER 8 series above; do not chain or subtract them. Relative survival excludes deaths from other causes.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 79.8%

      2016-2022; localized stage at diagnosis (49% of cases)

    All races, both sexes; laryngeal cancer, SEER Combined Summary Stage 'localized' (confined to the primary site).

    SEER summary stage, not AJCC TNM. Same SEER 21 series as the all-stages value; percentages of cases (49, 30, 15 and 6 per cent for localized, regional, distant and unknown) are as displayed on the page, which gives 48.7% in its text for localized.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 49.4%

      2016-2022; regional stage at diagnosis (30% of cases)

    All races, both sexes; laryngeal cancer, SEER Combined Summary Stage 'regional' (spread to regional lymph nodes).

    SEER summary stage, not AJCC TNM. Same SEER 21 series as the all-stages value.

    Source

  • 5-year relative survival · United States, SEER 21 (excluding IL)

    • 36%

      2016-2022; distant stage at diagnosis (15% of cases)

    All races, both sexes; laryngeal cancer, SEER Combined Summary Stage 'distant' (cancer has metastasized).

    SEER summary stage, not AJCC TNM. Same SEER 21 series as the all-stages value; unknown stage (6% of cases) is shown by the page as 57.0%.

    Source

  • Age-standardised mortality · United States (U.S. Mortality data shown on the SEER page; not a SEER registry series)

    • 1.68per 100,000

      1975 (observed value; first year in the table)

    All races, both sexes; laryngeal cancer deaths; age-adjusted rate per 100,000 (the page text read does not name the standard population).

    Death-rate series 1975-2024, observed; the highest table value is 1.71 (1976). Modeled trend value for 1975: 1.71. SEER 8 incidence per 100,000 was 4.66 in 1975 and 1.82 in 2023 (observed). The page text reports that smoking is a major risk factor for this cancer and that falling smoking rates have led to a downturn in incidence and mortality; it does not quantify that attribution.

    Source

  • Age-standardised mortality · United States (U.S. Mortality data shown on the SEER page; not a SEER registry series)

    • 0.88per 100,000

      2024 (observed value; last year in the table)

    All races, both sexes; laryngeal cancer deaths; age-adjusted rate per 100,000 (the page text read does not name the standard population).

    The lowest observed value in the table is 0.86 (2023). Modeled trend value for 2024: 0.87. The page text reports that age-adjusted death rates have been falling on average 1.3% each year over 2015-2024 and new cases 2.5% each year over 2014-2023.

    Source

  • 10-year relative survival · United Kingdom

    • 54.2%

      1970s (page text: 'more than 1 in 2')

    Laryngeal cancer; the page's trend sentence gives this value for males diagnosed in the 1970s.

    The page says 'survive their disease' and does not name the measure (net or relative survival) behind these ten-year values. Cohorts from different decades; not comparable with the SEER values, which are five-year relative survival in the United States.

    Source

  • 10-year relative survival · United Kingdom

    • 45.5%

      2018 (the page says 'predicted'; page text: 'almost 1 in 2')

    Laryngeal cancer; the page's trend sentence gives this value for males in 2018, and its opening sentence gives the same 45.5% for people diagnosed with laryngeal cancer in the UK without separating sex.

    Predicted, not observed follow-up, per the page. The page states that laryngeal cancer survival in the UK has decreased in the last 50 years and that ten-year survival has decreased in men in all broad age groups since the early 1970s. The measure (net or relative) is not named on the page. A lower later value than in the 1970s does not by itself show worse treatment: the page says survival trends reflect changes in treatment and stage distribution, which themselves vary by age, sex and deprivation.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Larynx preservation in locally advanced laryngeal cancer

Locally advanced squamous cell carcinoma of the larynx treated with a larynx-preserving strategy · Locally advanced · endpoint: Larynx preservation

Results are not comparable across trials.

The two entries report different endpoints at the same two-year landmark. RTOG 91-11 reports the proportion with an intact larynx; the VA trial (1991, a preliminary interim report) is drawn with a distinct marker because its abstract gives two-year overall survival, not a preservation rate for the laryngectomy comparator, and gives one estimate for both groups. RTOG 91-11 had three arms: only concurrent cisplatin plus radiotherapy versus induction cisplatin-fluorouracil then radiotherapy is drawn here; the radiotherapy-alone arm is in the trial record. Larynx preservation is not the same as preserved function. Eras, eligibility and chemotherapy regimens differ. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100Larynx preservation at 24 months (%)19912003va-laryngeal-cancer-studyRTOG 91-11
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
Larynx preservation at 24 months, rate, across 2 trials in this lane. va-laryngeal-cancer-study (1991): experimental 68 % vs control 68 %; RTOG 91-11 (2003): experimental 88 % vs control 75 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
va-laryngeal-cancer-study1991OS (differs from lane)68 %68 %
RTOG 91-112003Larynx preservation75 %88 %

Postoperative treatment of high-risk resected head and neck cancer

Resected head and neck squamous cell carcinoma with high-risk features (eligibility differs by trial) · Adjuvant · endpoint: OS

Results are not comparable across trials.

EORTC 22931 and RTOG 9501 used different eligibility criteria and report overall survival at different times: a five-year value for EORTC 22931 and a ten-year value from the later RTOG 9501 report. Only the five-year value shares the axis; the RTOG 9501 ten-year value is listed but not drawn, and the two trials are not like-for-like. In RTOG 9501 the overall ten-year differences were not significant. NIVOPOSTOP (2025) added nivolumab to cisplatin and radiotherapy with open-label cisplatin and radiotherapy as the control; its primary result is disease-free survival reported as a hazard ratio only, so it is not drawn. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100OS at 60 months (%)2004EORTC 22931
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS at 60 months, rate, across 1 trial in this lane. EORTC 22931 (2004): experimental 53 % vs control 40 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
EORTC 229312004OS40 %53 %
RTOG 95012004OS (not plotted: different measure or time point)27 %29.1 %

De-escalation in HPV-positive oropharyngeal cancer

HPV-positive oropharyngeal squamous cell carcinoma treated with definitive radiotherapy-based therapy · Locally advanced · endpoint: OS

Results are not comparable across trials.

The 2016 phase 2 studies (E1308, HN002, E3311) have no standard-treatment control arm, or compare each arm with a historical benchmark, so they are listed but not drawn. RTOG 1016 (HPV-positive oropharyngeal cancer; primary endpoint overall survival with a non-inferiority margin; five-year value) and De-ESCALaTE (low-risk HPV-positive; primary outcome overall severe toxicity, with overall survival a secondary outcome; two-year value) differ in population, endpoint hierarchy and landmark time. Only the RTOG 1016 five-year value is drawn; the De-ESCALaTE two-year value is listed. De-ESCALaTE's abstract does not label the hazard-ratio direction, so no hazard ratios are quoted. In both trials the experimental arm replaced cisplatin with cetuximab. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100OS at 60 months (%)2018RTOG 1016
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS at 60 months, rate, across 1 trial in this lane. RTOG 1016 (2018): experimental 77.9 % vs control 84.6 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
RTOG 10162018OS84.6 %77.9 %
De-ESCALaTE HPV2018OS (not plotted: different measure or time point)97.5 %89.4 %

Recurrent or metastatic head and neck cancer, first line

Recurrent or metastatic head and neck squamous cell carcinoma not previously treated for that stage · Metastatic · endpoint: OS

Results are not comparable across trials.

Median overall survival in months, each trial's own arms. Comparators and populations differ. EXTREME compared cetuximab added to platinum-fluorouracil with platinum-fluorouracil alone. KEYNOTE-048 (total population, second interim analysis; the abstract's other populations and the later updates are not drawn), KESTREL (PD-L1-high only; durvalumab alone) and CheckMate 651 (all randomly assigned patients; nivolumab plus ipilimumab) each used the EXTREME regimen as the control. LEAP-010 (PD-L1 CPS 1 or more) and INDUCE-3 (PD-L1-positive; stopped early, no medians) used placebo plus pembrolizumab as the control, a different kind of comparator, so they are part of the milestone but are not drawn. KESTREL's durvalumab-tremelimumab arm is in the trial record but not drawn. Eras and follow-up differ. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

05101520OS, median (months)2008201920222022EXTREMEKEYNOTE-048KESTRELCheckMate 651
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 4 trials in this lane. EXTREME (2008): experimental 10.1 months vs control 7.4 months; KEYNOTE-048 (2019): experimental 13 months vs control 10.7 months; KESTREL (2022): experimental 10.9 months vs control 10.9 months; CheckMate 651 (2022): experimental 13.9 months vs control 13.5 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
EXTREME2008OS7.4 months10.1 months
KEYNOTE-0482019OS10.7 months13 months
KESTREL2022OS10.9 months10.9 months
CheckMate 6512022OS13.5 months13.9 months

Recurrent or metastatic head and neck cancer after platinum

Recurrent or metastatic head and neck squamous cell carcinoma that progressed on or after platinum-containing therapy · Metastatic · endpoint: OS

Results are not comparable across trials.

Median overall survival in months, each trial's own arms. In both trials the control was an investigator's choice of single-agent methotrexate, docetaxel or cetuximab, so the control is not one regimen. Eligibility differs: CheckMate 141 enrolled people whose cancer progressed within six months of platinum; KEYNOTE-040 also included progression after platinum-containing treatment for locally advanced disease at three to six months. CheckMate 141 was open-label, and KEYNOTE-040's overall survival p value is described as nominal. KEYNOTE-012 is a single-arm study and is not drawn. The two trials are not compared with each other. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

0246810OS, median (months)20162016CheckMate 141KEYNOTE-040
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
OS, median, across 2 trials in this lane. CheckMate 141 (2016): experimental 7.5 months vs control 5.1 months; KEYNOTE-040 (2016): experimental 8.4 months vs control 6.9 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
CheckMate 1412016OS5.1 months7.5 months
KEYNOTE-0402016OS6.9 months8.4 months

Recurrent or metastatic nasopharyngeal carcinoma, first line

Recurrent or metastatic nasopharyngeal carcinoma receiving first-line platinum-based chemotherapy · Metastatic · endpoint: PFS

Results are not comparable across trials.

Median progression-free survival in months, each trial's own arms, from trials run in Asia. The 2016 trial compared gemcitabine-cisplatin with fluorouracil-cisplatin (independent image committee assessment); the three later trials compared a PD-1 antibody plus gemcitabine-cisplatin with placebo plus gemcitabine-cisplatin. JUPITER-02 and CAPTAIN-1st are interim analyses (JUPITER-02 by blinded independent review; its final analysis median is not drawn), whereas RATIONALE-309 gives medians only in a roughly three-year follow-up update, with median follow-up of 27.5 months, which is a different analysis time. Overall survival results exist for the later trials but are not drawn. Compare only within each trial's own arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

02468101214PFS, median (months)2016202120212021gp-vs-fp-npcJUPITER-02CAPTAIN-1stRATIONALE-309
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS, median, across 4 trials in this lane. gp-vs-fp-npc (2016): experimental 7 months vs control 5.6 months; JUPITER-02 (2021): experimental 11.7 months vs control 8 months; CAPTAIN-1st (2021): experimental 9.7 months vs control 6.9 months; RATIONALE-309 (2021): experimental 9.6 months vs control 7.4 months.
Data table
TrialYearEndpointControl (months)Experimental (months)
gp-vs-fp-npc2016PFS5.6 months7 months
JUPITER-022021PFS8 months11.7 months
CAPTAIN-1st2021PFS6.9 months9.7 months
RATIONALE-3092021PFS7.4 months9.6 months

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Locally advanced laryngeal cancer: larynx-preserving treatment

Locally advanced squamous cell carcinoma of the larynx · Locally advanced

  1. Chemotherapy1Radiotherapy with concurrent cisplatinRTOG 91-11: concurrent cisplatin for larynx preservation (2003)
2003. Based only on the trials cited in each step. RTOG 91-11 compared concurrent cisplatin and radiotherapy with induction chemotherapy then radiotherapy and with radiotherapy alone; preserved larynx is not preserved function.
Text version
  • 1991: Induction chemotherapy, then radiotherapy
  • 2003: Radiotherapy with concurrent cisplatin

Recurrent or metastatic head and neck cancer: lines of treatment

Recurrent or metastatic head and neck squamous cell carcinoma · Metastatic

  1. Immunotherapy11st line: pembrolizumab ± chemotherapyKEYNOTE-048: pembrolizumab first line, recurrent or metastatic (2019)
  2. Targeted21st line: cetuximab + platinum-fluorouracilEXTREME: cetuximab added to platinum-fluorouracil, first line (2008)
  3. Immunotherapy3After platinum: nivolumabCheckMate 141: nivolumab versus standard therapy after platinum (2016)
  4. Immunotherapy4After platinum: pembrolizumabPembrolizumab after platinum: accelerated approval, then KEYNOTE-040 (2016)
2019. Based only on the trials cited in each step; guidelines and local practice vary. KEYNOTE-048 compared pembrolizumab with or without chemotherapy against the EXTREME regimen; pembrolizumab alone applies to PD-L1 CPS 1 or more.
Text version
  • 2008: 1st line: cetuximab + platinum-fluorouracil
  • 2016: 1st line: cetuximab + platinum-fluorouracil → After platinum: nivolumab → After platinum: pembrolizumab
  • 2019: 1st line: pembrolizumab ± chemotherapy → 1st line: cetuximab + platinum-fluorouracil → After platinum: nivolumab → After platinum: pembrolizumab

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

  • 2018

    Cetuximab worse than cisplatin with radiotherapy, HPV-positive cancer

    Two trials in HPV-related throat cancer asked whether cetuximab could replace cisplatin alongside radiation to reduce side effects. In both, people given cetuximab did worse on survival or cancer control, and neither trial found a clear side-effect advantage for cetuximab. Both teams recommended cisplatin as the standard of care.

    Two phase 3 trials in HPV-positive oropharyngeal cancer compared radiotherapy with cetuximab vs cisplatin. RTOG 1016 (849 randomised; non-inferiority margin 1.45): OS HR 1.45 (one-sided 95% upper CI 1.94), 5-year OS 77.9% (cetuximab) vs 84.6% (cisplatin); PFS HR 1.72 (95% CI 1.29-2.29). De-ESCALaTE (N=334, low risk): severe toxicity events did not differ (4.8 vs 4.8 per patient); 2-year OS 89.4% (cetuximab) vs 97.5% (cisplatin).

  • 2021

    Four add-ons to definitive head and neck treatment show no benefit

    Four large trials added a new drug to standard treatment for head and neck cancer that had not spread widely: avelumab, pembrolizumab or xevinapant during chemoradiotherapy, or atezolizumab afterwards. None improved the main outcome, and with xevinapant the results were worse than with placebo.

    Four randomised phase 3 trials (N=697, 804, 406, 730) in locally advanced SCCHN. Avelumab plus chemoradiotherapy: PFS HR 1.21 (95% CI 0.93-1.57). Pembrolizumab plus chemoradiotherapy: EFS HR 0.83 (95% CI 0.68-1.03). Adjuvant atezolizumab: EFS HR 0.94 (95% CI 0.70-1.26). Xevinapant plus chemoradiotherapy: EFS HR 1.33 (95% CI 1.05-1.67); OS HR 1.39 (95% CI 1.04-1.86). Results are not pooled.

  • 2022

    First-line immunotherapy combinations: no overall survival gain

    Four randomised trials tested a second drug alongside immunotherapy as first treatment for head and neck cancer that had returned or spread. The combinations were durvalumab with tremelimumab, nivolumab with ipilimumab, lenvatinib with pembrolizumab, and feladilimab with pembrolizumab. None improved overall survival over its comparator.

    Four randomised first-line R/M SCCHN trials, not pooled. KESTREL (PD-L1-high): durvalumab vs EXTREME, OS HR 0.96 (95% CI 0.69-1.32); durvalumab plus tremelimumab vs EXTREME, HR 1.05 (0.80-1.39). CheckMate 651: nivolumab plus ipilimumab vs EXTREME, OS HR 0.95 (97.9% CI 0.80-1.13). LEAP-010: lenvatinib plus pembrolizumab vs placebo plus pembrolizumab, PFS HR 0.64, OS HR 1.15 (95% CI 0.91-1.45). INDUCE-3: OS HR 1.51 (95% CI 0.99-2.29).

Today and tomorrow

Standard of care today

Prevention

FDA regulatory action read:

  • on 12 June 2020 the FDA approved, under its accelerated approval regulations (21 CFR 601.41), a supplement (BL 125508/868) adding prevention of oropharyngeal and other head and neck cancers caused by the HPV types targeted by the vaccine to the indication of GARDASIL 9. The letter states that clinical benefit is evidenced by effects such as prevention of oral persistent infection with HPV types 16, 18, 31, 33, 45, 52 or 58, and lists a required confirmatory study (V503-049, a three-dose regimen in men 20 through 45 years of age, with prevention of oral persistent infection as the endpoint) with planned study completion on 31 December 2025 and final report submission on 30 September 2026. Four later letters on the FDA product page (20 August 2020, 28 April 2023, 14 March 2024, 25 March 2025) were opened
  • the text extracted from them does not mention that study or the head and neck indication. The current prescribing information (revised 03/2025
  • DailyMed label version 31, effective 13 July 2026) still states that the oropharyngeal and head and neck cancer indication is approved under accelerated approval based on effectiveness in preventing HPV-related anogenital disease and that continued approval may be contingent upon verification in a confirmatory trial, so none of the documents read shows a conversion
  • whether the confirmatory study has reported was not determined. Vaccination policy documents were not read
FDA, Center for Biologics Evaluation and Research 2020FDA 2026National Library of Medicine, DailyMed 2026
Locally advanced
Squamous cell carcinoma of the oral cavity, oropharynx, larynx and hypopharynx. FDA regulatory action read: on 1 March 2006 the FDA approved a cetuximab supplement (BL 125084/46) to include treatment of locally or regionally advanced squamous cell carcinoma of the head and neck in combination with radiation therapy, and single-agent use after platinum-based therapy had failed in recurrent or metastatic disease (the letter is a scanned image; its three letter pages were read and contain no accelerated-approval wording, and the two further pages of the file are redacted). In HPV-positive oropharyngeal cancer, two randomised trials reported online in 2018 compared cetuximab with cisplatin alongside radiotherapy and favoured cisplatin on survival or tumour control (see the milestone timeline). Guideline documents are cited by identifier only and their text was not read, so no recommendation is attributed to them (whether these are the current editions was not checked): EHNS-ESMO-ESTRO 2020 (Ann Oncol, PMID 33239190) and NCCN Guidelines Insights, Head and Neck Cancers, version 2.2025 (J Natl Compr Canc Netw, PMID 39938471).
FDA 2006Gillison ML, Trotti AM, Harris J, Eisbruch A, Harari PM, Adelstein DJ, et al 2019Mehanna H, Robinson M, Hartley A, Kong A, Foran B, Fulton-Lieuw T, et al 2019Guideline: Machiels JP, René Leemans C, Golusinski W, Grau C, Licitra L, Gregoire V 2020Guideline: Colevas AD, Cmelak AJ, Pfister DG, Spencer S, Adkins D, Birkeland AC, et al 2025
Perioperative
Resectable locally advanced head and neck squamous cell carcinoma, regulatory documents read. FDA, 12 June 2025: pembrolizumab (BLA 125514/S-180) for tumours with PD-L1 CPS of 1 or more by an FDA-approved test, as neoadjuvant treatment, continued as adjuvant treatment with radiotherapy with or without cisplatin and then as a single agent. EMA: the Keytruda procedural-steps document lists an accepted extension of indication to the same population (CPS 1 or more), with CHMP opinion on 18 September 2025 and Commission decision on 24 October 2025. KEYNOTE-689 and NIVOPOSTOP (nivolumab added to postoperative cisplatin-radiotherapy) are covered in the milestone timeline; no regulatory document was read for nivolumab in this setting. No guideline text was read for this setting.
FDA 2025EMA 2026
Metastatic
Recurrent or metastatic head and neck squamous cell carcinoma, FDA letters read. Cetuximab, 7 November 2011: first-line treatment of recurrent locoregional disease or metastatic disease in combination with platinum-based therapy with 5-FU. Pembrolizumab, 5 August 2016 (accelerated approval): disease progressing on or after platinum-containing chemotherapy. Nivolumab, 10 November 2016: single agent after progression on or after platinum-based therapy. Pembrolizumab, 10 June 2019: the 2016 indication's accelerated-approval language was removed in a supplement approved as fulfilment of a postmarketing requirement, and new first-line indications were added, with platinum and fluorouracil and, for tumours with PD-L1 CPS of 1 or more by an FDA-approved test, as a single agent. The ASCO guideline on immunotherapy and biomarker testing in recurrent and metastatic head and neck cancers (J Clin Oncol 2023, PMID 36521102) is cited by identifier only; its text was not read, so no recommendation is attributed to it. EU documents for these indications were not read.
FDA 2011FDA 2016FDA 2016FDA 2019Guideline: Yilmaz E, Ismaila N, Bauman JE, Dabney R, Gan G, Jordan R, et al 2023
Metastatic

Nasopharyngeal carcinoma, recurrent or metastatic, regulatory documents read. FDA, 27 October 2023:

  • toripalimab-tpzi (BLA 761240) with cisplatin and gemcitabine for first-line treatment of metastatic or recurrent locally advanced disease, and as a single agent after progression on or after platinum-containing chemotherapy. FDA, 23 April 2025: penpulimab-kcqx (BLA 761258
  • the corrected letter was signed 29 April 2025 and keeps the 23 April effective date) with cisplatin or carboplatin and gemcitabine for first-line treatment of recurrent or metastatic non-keratinizing disease, and as a single agent for metastatic non-keratinizing disease after platinum-based chemotherapy and at least one other prior line. EMA product pages read on 20 October 2026: Loqtorzi (toripalimab) with cisplatin and gemcitabine for first-line treatment of recurrent (not amenable to surgery or radiotherapy) or metastatic disease, opinion adopted 25 July 2024 and marketing authorisation issued 19 September 2024
  • Tevimbra (tislelizumab) lists first-line use with gemcitabine and cisplatin among its indications, and its EPAR procedural-steps document records the extension of indication (II/0017) with CHMP opinion on 22 May 2025 and Commission decision on 9 July 2025. ESMO-EURACAN nasopharyngeal carcinoma guidelines (Ann Oncol 2021, PMID 33358989) and the 2023 update on adjuvant therapy and first-line treatment of recurrent or metastatic disease (PMID 36529446) are cited by identifier only
  • their text was not read
FDA 2023FDA 2025EMA 2026EMA 2026EMA 2025Guideline: Bossi P, Chan AT, Licitra L, Trama A, Orlandi E, Hui EP, et al 2021Guideline: Bossi P, Chan AT, Even C, Machiels JP 2023

Trials to watch

  • LiGeR-HN1 NCT06525220

    In first-line recurrent or metastatic head and neck squamous cell carcinoma with PD-L1-positive tumours, does petosemtamab (an EGFR and LGR5 bispecific antibody) added to pembrolizumab improve outcomes compared with pembrolizumab alone? Registry record: randomised, open-label phase 3; status active, not recruiting; enrolment 699 (actual); primary outcomes listed are overall survival and objective response rate; estimated primary completion January 2028; no results posted.

  • LiGeR-HN2 NCT06496178

    In previously treated recurrent or metastatic head and neck squamous cell carcinoma, does petosemtamab compared with investigator's choice of monotherapy differ in overall survival? Registry record: randomised, open-label phase 3; status active, not recruiting; enrolment 621 (actual); the registry lists overall survival as the primary outcome; estimated primary completion February 2028; no results posted.

  • NANORAY-312 NCT04892173

    In older people with locally advanced head and neck squamous cell carcinoma who cannot receive platinum-based chemotherapy, does intratumoural NBTXR3 (JNJ-90301900), activated by radiotherapy with or without cetuximab, differ in progression-free survival from radiotherapy with or without cetuximab? Registry record: phase 3; status recruiting; enrolment 500 (estimated); primary outcome listed is progression-free survival by independent central review; estimated primary completion June 2028; no results posted.

  • EA3161 NCT03811015

    In locally advanced, intermediate-risk HPV-positive oropharyngeal squamous cell carcinoma, does maintenance nivolumab after cisplatin-radiotherapy differ from observation? Registry record: randomised phase 3 (NCI); status active, not recruiting; enrolment 636 (estimated); overall survival is among the listed primary outcomes; estimated primary completion January 2030; no results posted.

  • PATHOS NCT02215265

    After transoral surgery for HPV-positive oropharyngeal cancer, can risk-stratified, reduced-intensity postoperative treatment (registry arms: no adjuvant treatment, radiotherapy at 60 Gy or 50 Gy, and 60 Gy radiotherapy with or without cisplatin) be used? Registry record: randomised phase 3; status active, not recruiting; enrolment 1,269 (estimated); a co-primary endpoint of swallowing function (MDADI) and overall survival; estimated primary completion October 2027; no results posted.

  • TAISHAN-301 NCT06629597

    In recurrent or metastatic nasopharyngeal carcinoma after prior PD-(L)1 inhibitor treatment and at least two lines of chemotherapy, does YL201 differ from investigator's choice of chemotherapy (docetaxel, capecitabine or gemcitabine) in objective response rate or overall survival? Registry record: randomised, open-label phase 3; status active, not recruiting; enrolment 400 (estimated); estimated primary completion December 2027; no results posted.

References

Show all 93 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2010UnderstandingUnderstanding

landmarkLocally advanced

Tumour HPV status found to be a strong, independent prognostic factor

A retrospective analysis of people with stage III or IV oropharyngeal cancer treated in a chemoradiotherapy trial found that those with HPV-positive tumours did much better: about eight in ten were alive at three years, against about six in ten with HPV-negative tumours. The analysis was not a test of treating them less.

Retrospective analysis by HPV status of stage III or IV oropharyngeal SCC in RTOG 0129 (accelerated vs standard fractionation radiotherapy, each with cisplatin). HPV-positive 206 of 323 (63.8%). 3-year OS 82.4% vs 57.1% (P<0.001); adjusted HR for death 0.42 (95% CI 0.27 to 0.66); median follow-up 4.8 years. Recursive partitioning (HPV status, pack-years, T and N stage) defined low, intermediate and high risk.

Oropharyngeal cancers caused by HPV had been associated with favourable survival, but whether tumour HPV status was an independent prognostic factor was not known.[1]

Investigators analysed, retrospectively and post hoc, patients with stage III or IV oropharyngeal squamous-cell carcinoma who had been enrolled in a randomised trial of accelerated-fractionation versus standard-fractionation radiotherapy, each given with cisplatin. The radiotherapy schedules themselves gave similar 3-year overall survival (70.3% versus 64.3%).[1]

HPV status was determined in 323 of 433 patients with oropharyngeal cancer in the trial; HPV-positive tumours were found in 206 of 323 (63.8%) and HPV-negative tumours in 117. Three-year overall survival was 82.4% for HPV-positive and 57.1% for HPV-negative disease. After adjustment for age, race, tumour and nodal stage, tobacco exposure and treatment assignment, the hazard ratio for death was 0.42 (95% CI 0.27 to 0.66). Risk of death rose with each additional pack-year of smoking, and a recursive-partitioning analysis based on HPV status, pack-years, tumour stage and nodal stage sorted patients into low, intermediate and high risk of death.[1]

Other cohorts pointed the same way. In an ECOG phase II trial of 96 patients, HPV-positive tumours had higher response rates and a 2-year overall survival of 95% versus 62%.[2] In a retrospective analysis of the cetuximab registration trial, p16-positive oropharyngeal cancer had longer overall survival than p16-negative disease with radiotherapy alone (HR 0.40, 95% CI 0.21 to 0.74).[3]

These are prognostic findings, not tests of treatment intensity.

Standard before

HPV-related oropharyngeal cancers were associated with favourable survival, but the independent prognostic significance of tumour HPV status was unknown (Ang 2010 background).

Why it mattered

The analysis showed that HPV status predicted survival after adjustment for stage, smoking and treatment, and proposed a risk classification combining HPV status, smoking, tumour stage and nodal stage.

Caveats

  • Retrospective, post hoc analysis of patients in a randomised trial of radiotherapy fractionation; the paper states it was not part of the study protocol. HPV status was available for 323 of 433 patients with oropharyngeal cancer, and the HPV comparison is not randomised. The randomised comparison itself found no significant difference between fractionation schedules in 3-year overall survival.
  • Better prognosis does not by itself show that treatment can safely be reduced; de-escalation is addressed by separate trials in the local and regional treatment area.
  • Supporting cohorts use different designs: Fakhry 2008 (ECOG phase II, 96 patients with oropharynx or larynx tumours; 2-year OS 95% vs 62%) and Rosenthal 2016 (182 p16-evaluable patients in the cetuximab registration trial; p16 rather than HPV testing, HR 0.40 with radiotherapy alone).
  • HPV status was defined by in situ hybridisation (HPV-16 DNA, then a probe cocktail for other oncogenic types); the authors note that sensitivity of the cocktail for non-HPV-16 types is unknown. Their text was read in the open author manuscript (PMC2943767); the Fakhry 2008 and Rosenthal 2016 papers were checked against abstracts only.