A panel of 45 US specialists has agreed on which cutaneous squamous cell carcinomas need staging and surveillance imaging, citing bone invasion and tumours larger than 4cm as key risk features.

Writing in JAMA Dermatology [link here], the panel recommended surveillance for lesions with at least a 15% risk of metastasis. Two risk-prediction tools, riSCC and the Erasmus MC model, had been developed to help identify these cases, the authors noted.

The group, drawn from dermatology, medical oncology and surgery, reached consensus or near consensus that staging and surveillance imaging was warranted for lesions with:

• bone invasion
• deep invasion
• features suggestive of metastatic disease
• a diameter greater than 4cm
• poor differentiation with a diameter greater than 2cm
• poor differentiation with lymphovascular and subcutaneous fat invasion
• poor differentiation with lymphovascular and perineural invasion

CT was the panel’s preferred method for initial staging and ongoing surveillance. Panel members recommended imaging every six months in the first year, continuing for up to two years.

Consensus was not reached on how long surveillance should continue after that. Three-quarters of the panel favoured monitoring for up to three years, while one quarter said tracking lesions for up to five years was appropriate.

The panel also agreed that imaging should be guided by tumour stage, recommending it for BWH T2b/T3 tumours, most AJCC8 T3 tumours, and AJCC8 T4a/b lesions.

Less clarity on intermediate-risk lesions

Dr Emily Ruiz.

The panel found the picture less clear for intermediate-risk tumours. Dermatologist Dr Emily Ruiz, of Brigham and Women’s Faulkner Hospital, and colleagues found immunosuppression, in-transit metastasis, or nodal or distant metastases could prompt clinicians to consider imaging.

“Life expectancy and treatment-related factors, including receipt of adjuvant radiotherapy, immunotherapy, or surgical margin status, did not reach consensus as factors that were associated with imaging,” the authors wrote.

The panel did not endorse molecular tools, such as the gene expression profiling test Castle GEP, to guide decisions on extended surveillance, citing a lack of data showing patient benefit.

“The lack of consensus may reflect limited data on how the results of GEP are associated with the use of imaging or the high cost of GEP testing,” the panel wrote. “Additionally, the current use of GEP tests is practitioner- and institution-dependent, which may contribute to the lack of consensus.”