LAUSR

Most cases of cutaneous squamous cell carcinoma (cSCC) are successfully removed by surgery and do not require further investigations or clinical follow-up. However, some patients should be considered for more active interventions and surveillance due to the risk of regional and distant metastasis. Patients with advanced cSCC may be candidates for immune checkpoint inhibitors such as anti-programmed death 1 monoclonal antibodies. The challenge for the clinician, then, is to identify patients with low-risk tumours and those with high-risk tumours in order to decide on appropriate treatment and follow-up. In this issue of the BJD, Venables and coauthors report the results from a nested case–control study to validate the performance of four staging systems for cSCC in predicting regional and distant metastasis. Although one staging system, the Brigham and Women’s Hospital staging system, showed the highest overall discriminative ability, the positive predictive value was low for all four staging systems. This is an important study, as previous validation studies on staging systems for cSCC have been performed on selected patient cohorts from larger hospitals, except for one smaller study from Norway. Using a large and population-based cancer registry dataset from England, the authors confirm the need for a better staging system or improvement and refining of current staging systems. The study does not include data on mortality, despite the fact that a substantial proportion of patients who die from cSCC will die from local invasive tumour growth and its complications and not from metastasis. The ideal staging system for cSCC should be easy to use in a busy everyday clinical practice, risk factors should be easy to measure, and the measurements should be robust and reliable. The authors of the present study, all leading skin cancer epidemiologists from the UK and the Netherlands, suggest putting a stronger emphasis on tumour thickness, as is done for cutaneous melanoma. Tumour thickness is relatively easy to measure and should always be included in the pathologist’s report. Differentiation grade is shown to have a low reproducibility, making the dichotomy good/moderate vs. poor/undifferentiated uncertain. Pathologists should better define, validate and report the differentiation grade of these tumours. Ideally, a less rigid categorization into limited profile groups, as suggested by Venables et al., would be desirable in order to identify those patients with close to no risk of metastasis and those with the highest risk. Staging systems for cSCC also have purposes other than predicting the prognosis for individual patients, but many clinicians do not use them regularly, partly due to their complexity. Guidelines may be helpful, but even these tend to be complicated and difficult to follow. In the absence of better staging systems, clinicians must rely on their clinical judgement, in both what to do and what to tell the patient. In addition to tumour thickness and other known risk factors, a patient’s individual circumstances and preferences should be taken into account. Both doctors and patients must accept some uncertainty. It is part of life.