Chapter 15 of 25
- Some tumours use PD-1 / PD-L1 signals to suppress immune attack
- Checkpoint inhibitors block this inhibitory signal
- Immune cells may attack cancer; healthy organs can also be inflamed
| Term or setting | What it means |
|---|---|
| First-line recurrent/metastatic: alone | PD-L1 CPS at least 1; clinical urgency and fitness also matter. |
| First-line recurrent/metastatic: with platinum and 5-FU | No minimum CPS under the US indication; this does not imply equal benefit in every CPS group. |
| Perioperative resectable locally advanced disease | CPS at least 1; a planned course before and after surgery, with indicated postoperative treatment. |
The immune system and its brakes
The immune system can recognize abnormal cells, but tumours can use immune-control pathways to avoid attack. PD-1 is a receptor on immune cells; PD-L1 is one of its partner proteins. Their interaction can reduce immune activity. Pembrolizumab and nivolumab block PD-1, releasing part of this brake. They do not simply “boost immunity” in the way a supplement claims to do.
Some cancers respond for a long time, while others do not respond at all. Immunotherapy may take time to work and can sometimes allow disease to progress before benefit is apparent. It should not be chosen solely because it sounds gentler than chemotherapy, particularly when a tumour is causing rapidly worsening symptoms or threatening breathing.
PD-L1 testing: CPS is not TPS
The Combined Positive Score, or CPS, considers PD-L1 staining in tumour cells and certain nearby immune cells relative to the number of viable tumour cells. Tumour Proportion Score, or TPS, measures the percentage of tumour cells staining for PD-L1. A TPS result cannot simply be substituted for a CPS result. The laboratory method and the treatment indication matter.
A higher CPS can make pembrolizumab alone more attractive in selected recurrent or metastatic squamous cancer, but CPS is not a guarantee of response. Other considerations include tumour growth, symptom burden, previous platinum and immunotherapy, autoimmune disease and organ function. PD-L1 rules for nasopharyngeal or salivary cancers are not automatically those used for ordinary mucosal squamous cancer.
Recurrent or metastatic squamous cancer
KEYNOTE-048 (2019) studied previously untreated recurrent or metastatic squamous cancers that were not curable with local treatment. It compared pembrolizumab alone, pembrolizumab with platinum and 5-FU, and cetuximab with platinum and 5-FU. It established pembrolizumab-based first-line treatment, with selection by PD-L1 and clinical need. Monotherapy is appropriate for eligible PD-L1-positive patients, while combination therapy is often considered when a higher probability of early shrinkage is needed.
CheckMate-141 (2016) studied disease progressing within six months of platinum treatment and compared nivolumab with the clinician’s choice of methotrexate, docetaxel or cetuximab. Survival improved with nivolumab. This is evidence for a specific platinum-refractory setting; it does not establish that changing from one PD-1 drug to another works after resistance. Regulatory labels differ by country, and eligibility must be checked locally.
What the CPS thresholds mean
For first-line recurrent or metastatic mucosal squamous cancer that cannot be cured locally, pembrolizumab alone requires PD-L1 CPS of at least 1 under the US indication. CPS 20 or more can support the discussion of monotherapy, but even a high score does not guarantee response. Rapidly worsening, bulky or symptomatic disease may favour a combination when the person can tolerate it.
Pembrolizumab with platinum and 5-FU has a US first-line indication without a minimum CPS requirement. This does not make benefit identical across CPS groups, and labels differ internationally. CPS below 1 should not be used to justify first-line pembrolizumab alone. The perioperative CPS threshold of at least 1 belongs to a different indication and cannot be substituted for these advanced-disease rules.
Immunotherapy around surgery
KEYNOTE-689 (2025) tested pembrolizumab before surgery and continued afterward with risk-adapted standard treatment, followed by pembrolizumab alone, against surgery and standard postoperative treatment. The trial showed improved event-free survival in eligible resectable locally advanced disease. In the United States, FDA approval on 12 June 2025 specifies adults with PD-L1 CPS at least 1. This approval does not mean that every head and neck tumour or every early-stage cancer should receive the regimen.
This is a planned perioperative pathway, not a replacement for surgery or indicated postoperative radiation and cisplatin. It should be discussed before the operation. Country-specific approval, current guideline eligibility, access and potential immune toxicity need confirmation. Postoperative nivolumab evidence from NIVOPOST-OP is a separate question, discussed in Chapter 18.
Immune-related side effects and urgent symptoms
Checkpoint drugs can cause inflammation in healthy organs, including lungs, bowel, liver, skin and hormone-producing glands. Pneumonitis may cause cough or breathlessness; colitis may cause diarrhoea, blood in stool or abdominal pain. Hepatitis may cause jaundice or may first appear on blood tests. Thyroid or adrenal problems can cause fatigue, dizziness, vomiting, weakness or unusual weight change.
Report new symptoms early, including after treatment has stopped. Seek urgent care for breathlessness, chest pain, severe diarrhoea, fainting, confusion, severe weakness or a widespread blistering rash. Tell emergency staff the exact immunotherapy drug and date of treatment. The team may hold treatment, prescribe steroids or replace missing hormones; do not start or stop these treatments without advice. Some hormone effects require long-term replacement.
References and evidence review
- Burtness et al. KEYNOTE-048, Lancet (2019)
- Ferris et al. CheckMate-141, NEJM (2016)
- FDA: perioperative pembrolizumab approval, United States (12 June 2025)
- Uppaluri et al. KEYNOTE-689, NEJM (2025)
- ASCO guideline: Immunotherapy and Biomarker Testing in Recurrent and Metastatic Head and Neck Cancers (2023)
Evidence/editorial check: 25 October 2026. AI-assisted preparation is distinct from individual physician review. See the linked review policy for source-access limits and review status. Regulatory indications, reimbursement and availability vary by country. Trial results alone do not establish approval everywhere; your treating team should confirm the applicable current guidance.