# PD-L1 assay discordance (SP142, SP263 and 22C3 in breast cancer)

Source: https://onco.cc/terms/pd-l1-assay-discordance/  
OnCo record `pd-l1-assay-discordance` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Three commercial PD-L1 stains give different answers on the same breast tumour: the SP142 assay used with atezolizumab called fewer than half of tumours positive while the 22C3 and SP263 assays called about three quarters. Because the pembrolizumab licence rests on 22C3 combined positive score 10, which test a laboratory runs, and how it scores it, decides who is offered immunotherapy.

## Summary

In a central re-analysis of 614 IMpassion130 tumours (68.1 percent of the trial population), PD-L1 on immune cells at 1 percent or more was present in 46.4 percent by VENTANA SP142, 74.9 percent by SP263 and 73.1 percent by Dako 22C3, and 80.9 percent had a 22C3 combined positive score of 1 or more; analytical concordance of SP263 and 22C3 with SP142 was 69.2 and 68.7 percent; almost all SP142-positive cases were positive on the other assays, but 29.6 percent of SP263-positive and 29.0 percent of 22C3-positive cases were SP142-negative; the atezolizumab benefit in SP263- and 22C3-positive patients (progression-free survival hazard ratios 0.64 to 0.68, overall survival 0.75 to 0.79) was driven by the double-positive cases, and single-positive cases showed overall survival hazard ratios of 0.87 to 0.95 (Rugo 2021, the paper record on this site). The assays also answer different questions: SP142 in IMpassion130 scored PD-L1 on tumour-infiltrating immune cells as a percentage of tumour area at a 1 percent cut-off (Schmid 2018), whereas KEYNOTE-355 used the 22C3 combined positive score, which counts staining tumour cells, lymphocytes and macrophages against all viable tumour cells, with the benefit confined to CPS 10 or more (Cortes 2020; Cortes 2022). Only the 22C3 CPS 10 threshold has an approved and NICE-recommended drug attached in triple-negative disease (TA801); the parent page records that the atezolizumab US indication was withdrawn and lists assay harmonisation among its open problems. The biomarker readouts pd-l1-cps and pd-l1-ic-score carry the label wording for each score.

## Fields

- Kind: Term
- Last checked: 2026-09-24
- Also known as: PD-L1 assay comparison; SP142 versus 22C3; PD-L1 antibody clones; PD-L1 test harmonisation; Non-interchangeable PD-L1 assays
- Tags: breast; tnbc

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/PD-L1
- Rugo, JNCI 2021: PD-L1 immunohistochemistry assay comparison (SP142, SP263, 22C3) in IMpassion130: https://doi.org/10.1093/jnci/djab108
- Schmid, N Engl J Med 2018: IMpassion130, atezolizumab and nab-paclitaxel by PD-L1 immune-cell status: https://doi.org/10.1056/nejmoa1809615
- Cortes, Lancet 2020: KEYNOTE-355, pembrolizumab plus chemotherapy in untreated metastatic triple-negative breast cancer: https://doi.org/10.1016/s0140-6736(20)32531-9
- NICE TA801: pembrolizumab plus chemotherapy for untreated triple-negative locally recurrent unresectable or metastatic breast cancer (June 2022): https://www.nice.org.uk/guidance/ta801

## Connected records

- cancers: [Metastatic triple-negative breast cancer](https://onco.cc/cancers/tnbc-metastatic/), [Triple-negative breast cancer (TNBC)](https://onco.cc/cancers/tnbc/)
- targets: [PD-L1](https://onco.cc/targets/pdl1/)
- drugs: [Atezolizumab](https://onco.cc/drugs/atezolizumab/), [Pembrolizumab](https://onco.cc/drugs/pembrolizumab/)
- terms: [Combined positive score (CPS)](https://onco.cc/terms/cps/), [PD-L1 combined positive score 10 in triple-negative breast cancer](https://onco.cc/terms/pd-l1-cps-10-tnbc/)
- trials: [IMpassion130](https://onco.cc/trials/impassion130/), [KEYNOTE-355](https://onco.cc/trials/keynote-355/)
- key papers: [PD-L1 Immunohistochemistry Assay Comparison in Atezolizumab Plus nab-Paclitaxel-Treated Advanced Triple-Negative Breast Cancer](https://onco.cc/key-papers/paper-rugo-pd-l1-assay-comparison-impassion130-jnci-2021/)

---
JSON: https://onco.cc/api/v1/entities/pd-l1-assay-discordance.json