Few prospective studies on outcomes in CSCC have been performed, and outcomes for patients treated specifically by Mohs micrographic surgery (MS) have not been described in the US. To address this research gap, we conducted a 5-year, multicenter, prospective cohort study on the long-term clinical outcomes of patients with invasive CSCC treated with MS and published our results in the Journal of the American Academy of Dermatology. Our goal was to prospectively quantify outcomes by tumor stage (T stage) using the Brigham and Women’s Hospital (BWH) and the American Joint Committee on Cancer Staging Manuel, Eighth Edition (AJCC8) T staging systems, as well as to verify historically high-risk features in MS-treated CSCC.
Fifteen US institutions participated to obtain data on 745 patients with invasive CSCC, which showed a 1.6% local recurrence (LR) rate after treatment with MS, the lowest 5-year LR for any CSCC treatment modality reported to date. Increasing Breslow tumor thickness (BT) was the primary predictor of poor outcomes, with an increased tumor thickness showing an increased risk of LR, nodal metastasis (NM), and disease-specific death (DSD) on multivariate modeling. Location on the lip also increased NM risk. Incidental perineural invasion (IPNI) did not show statistically worse LR, NM, nor DSD on multivariate modeling, contrary to previous findings. Whereas this population often receives adjuvant postoperative radiation therapy, no tumors with IPNI were adjuvantly radiated. Histologic differentiation, tumor size, and immunosuppressed status also showed no significant effect on outcomes, suggesting a mitigating effect of histologically tumor-free margins provided by MS. Neither the BWH nor AJCC8 T staging systems were predictive of LR; however, both were predictive of NM and DSD.
Our study validates the assertion that MS provides the lowest LR rate possible and supports its use, even in high T stage CSCC and allows clinicians to pause and consider close clinical monitoring over referring for radiation therapy when clear margins have been achieved via MS. The lack of significance of all traditionally high-risk characteristics other than BT suggests that consideration should be made for reporting Breslow thickness on pathology reports, as it was predictive of all outcomes of interest.