This Journal of Investigative Dermatology study, published on 22 September, used mass spectrometry to profile archived primary cutaneous squamous cell carcinomas that did and did not go on to metastasise. Of 4,819 protein groups, 284 differed between the two groups, in pathways linked to reduced cell death and adhesion and increased blood vessel growth and cell migration.
A machine-learning classifier built on these proteins predicted metastatic potential with 88.7% accuracy. Some of the differences were supported in independent gene-expression datasets.
The unmet need is real: a share of tumours that metastasise have no high-risk clinical or histological features. But this is a discovery study without prospective validation, and accuracy in a selected archival sample often falls in routine use. Staging and margins remain the tools to use today.
- Keep using clinical and histological high-risk features to stage cSCC.
- This protein signature is a research tool, not an available test.
- Remember that some metastasising tumours have no obvious high-risk features.
- Keep follow-up for high-risk cSCC focused on regional nodes.
Why it matters
It targets the tumours that look low-risk but are not, the gap current staging misses.
Don't overread it
Accuracy came from archival samples without prospective validation.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for dermatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free