More detail on the squamous cell carcinoma report does not predict metastasis better

Design

Two nested case-control studies within a national registry cohort, conditional logistic regression

Population

19,120 patients with cutaneous squamous cell carcinoma, Netherlands, over 10 years of follow-up; 195+195 and 250+250 matched sets

Primary outcome

Association between refined histopathological variables and metastasis

Effect

No significant associations in the population-based set; solar elastosis inversely associated in the risk-matched set (extensive OR 0.28, 95% CI 0.09-0.87; moderate OR 0.20, 0.07-0.56)

A steady stream of refined histopathological variables has been proposed for risk-stratifying cutaneous squamous cell carcinoma — morphological subtype, tumour budding, peritumoral infiltrate, mitotic count, solar elastosis. Whether any of them adds anything to depth of invasion and differentiation grade has largely gone untested.

This study tested it in two nested case-control studies drawn from the Netherlands Cancer Registry and the national pathology registry, with over ten years of follow-up across 19,120 patients. The first set matched 195 metastatic cases to 195 non-metastatic controls on follow-up time; the second matched 250 cases to 250 controls on calculated metastatic risk. Pathologists reviewed every primary tumour, and analysis used conditional logistic regression.

In the population-based set, multivariable analysis found no significant association between any refined variable and metastasis. In the risk-matched set, the only signal ran the other way: extensive solar elastosis (OR 0.28, 95% CI 0.09-0.87) and moderate solar elastosis (OR 0.20, 95% CI 0.07-0.56) were inversely associated with metastasis.

The useful conclusion is negative and worth acting on. If a request for extended histopathological reporting is being made to sharpen prognosis, this is evidence it will not, and the conventional variables remain what stratification should rest on.

• Depth of invasion and differentiation grade remain the variables to act on
• Do not ask for tumour budding or mitotic counts expecting better prognostic information
• The solar elastosis signal is hypothesis-generating — do not read it as protective
• Registry-based, so reporting quality varied; the negative result is the robust part
• Useful when a patient or colleague asks whether a more detailed report would help

Why it matters

It removes a reason to request extended pathology reporting that has been spreading without evidence.

Don't overread it

The inverse association with solar elastosis appeared in one analysis set only and should not be read as a protective effect.

The statistics, in plain English

A null result in a well-sized study is informative, but 'no significant association' is not proof of no association — small effects could be missed. The solar elastosis odds ratios have wide intervals (0.09 to 0.87) and appeared in only one of two analysis sets, which is exactly the pattern that turns out to be chance often enough to be treated with suspicion. The authors say as much.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.