PD-L1 testing in skin cancer
PD-L1 IHC; CD274; B7-H1; programmed death-ligand 1
PD-L1 (programmed death-ligand 1) immunohistochemistry has become a near-ubiquitous companion diagnostic for immune checkpoint inhibitor therapy across many cancers — though its role in cutaneous malignancies remains less rigid than in lung, head-and-neck or gastric cancers. Different assays (Dako 22C3, Dako 28-8, Ventana SP263, Ventana SP142) and different scoring systems (tumour proportion score TPS, combined positive score CPS, immune-cell score IC) are not interchangeable. In skin oncology, PD-L1 testing has variable utility — broadly informative in cSCC (cemiplimab), MCC (avelumab) and melanoma (most ICI commissioned irrespective of PD-L1 status). The UK pathway typically does not require pre-treatment PD-L1 testing for routine melanoma or MCC ICI initiation, but PD-L1 status is increasingly used in clinical trials and may inform later-line decisions.
Principles
- PD-L1 (CD274) is expressed on tumour cells and tumour-infiltrating immune cells; binds PD-1 on activated T cells, delivering an inhibitory signal that dampens anti-tumour immunity.
- Anti-PD-1 (pembrolizumab, nivolumab, cemiplimab) and anti-PD-L1 (avelumab, atezolizumab, durvalumab) drugs block this axis to restore T-cell activity.
- PD-L1 IHC is a snapshot of expression at a single time point in a single biopsy — heterogeneous across tumours and within a tumour.
Available assays and scoring systems
- Assays are not interchangeable — each clone has a specific staining pattern and threshold:
- Dako 22C3 — companion diagnostic for pembrolizumab in many indications.
- Dako 28-8 — companion for nivolumab.
- Ventana SP263 — companion for durvalumab; analytically harmonised with 22C3 in many studies.
- Ventana SP142 — companion for atezolizumab; quite different scoring (immune-cell focus).
- Scoring systems:
- Tumour proportion score (TPS) — % of tumour cells with membranous PD-L1 staining.
- Combined positive score (CPS) — (PD-L1+ tumour cells + PD-L1+ lymphocytes + PD-L1+ macrophages) / total viable tumour cells × 100, capped at 100.
- Immune-cell score (IC) — % of tumour area occupied by PD-L1+ immune cells.
- UK pathology labs typically use one or two clones across all tumour types; choice is institutional.
Skin-cancer context
- Melanoma — PD-L1 status is prognostic and predictive in many studies but UK ICI commissioning (pembrolizumab, nivolumab, ipi+nivo, nivo+rela) does not require PD-L1 testing. Most metastatic / adjuvant melanoma patients are treated regardless of PD-L1 status.
- cSCC — cemiplimab (NICE TA802) is offered for metastatic / locally advanced cSCC without PD-L1 selection. PD-L1 high-expression predicts better response in some studies but is not a gating criterion.
- MCC — avelumab is offered without PD-L1 selection (NICE TA691 for untreated metastatic MCC; TA517 history for metastatic MCC after chemotherapy). Checkpoint inhibitors are active irrespective of MCPyV status; MCPyV-negative tumours have higher TMB, but comparative response differences by viral status remain mixed.
- BCC — limited data; not commissioned for routine ICI use.
- Trials are exploring PD-L1 selection more rigorously — particularly in neoadjuvant settings (NADINA / S1801).
Practical pitfalls
- Inter-assay variability — switching clones can change the result.
- Inter-observer variability — moderate; pathology training matters.
- Biopsy site — primary tumour vs metastasis may have different PD-L1 expression.
- Timing — PD-L1 expression can change with prior treatment.
- Heterogeneity within a tumour — small biopsy may not be representative.
- Negative PD-L1 does not predict non-response in melanoma — patients with PD-L1-negative tumours still benefit from ICI in ~ 30%.
- Document the assay and scoring system on the pathology report.
References
- Tsao MS et al. PD-L1 immunohistochemistry comparability study in real-life clinical samples — four clinical PD-L1 assays. J Thorac Oncol; 2018.
- Robert C et al. Five-year outcomes with pembrolizumab versus chemotherapy for metastatic non-small cell lung cancer. J Clin Oncol; 2021.
- NICE technology appraisals for PD-1 / PD-L1 inhibitors in skin oncology: TA802 (cemiplimab cSCC), TA691 (avelumab MCC), TA366 / TA766 / TA837 (pembrolizumab melanoma), TA384 / TA400 / TA684 / TA950 (nivolumab-containing melanoma regimens); accessed 18 May 2026.
Spot a correction?
If any clinical statement, citation or link on this page needs updating, please email admin@skinoncology.net with the page name, the proposed correction and the supporting source.