Prospective subtyping of head and neck basal cell carcinoma by dermoscopy and handheld <scp>RCM</scp>

BACKGROUND: The management of head and neck basal cell carcinoma (BCC) relies on diagnostic biopsies. However, substantial discordance exists between biopsy and excisional specimens. Handheld reflectance confocal microscopy (HH-RCM) could reduce invasive biopsies. OBJECTIVES: The primary objective was to assess the diagnostic accuracy of combined dermoscopy and HH-RCM for diagnosing and subtyping primary head and neck BCC. The secondary objective was to determine the outcome when applying international Mohs micrographic surgery (MMS) criteria to this diagnostic pathway. METHODS: We conducted a prospective, single-centre, paired diagnostic comparison study. Diagnostic confidence was recorded using a 3-point Likert scale for all outcomes. RESULTS: A total of 340 consecutive lesions suspected of being BCC by the naked eye were included. Biopsy subtypes were found to be inconsistent with excisional specimens in 33% of cases. Combined dermoscopy-HH-RCM demonstrated an overall BCC diagnostic sensitivity of 97.5% (95% CI: 95.0-99.0) and a specificity of 85.2% (95% CI: 72.9-93.4). In cases where diagnostic confidence was high, sensitivity improved to 98.8% (95% CI: 96.5-99.7) and specificity to 94.1% (95% CI: 80.3-99.3). For BCC subtyping, the sensitivity and specificity for detecting non-superficial BCC were 95.5% (95% CI: 92.9-98.1) and 44.8% (95% CI: 26.7-62.9), respectively. For detecting infiltrative BCC, sensitivity was 58.8% (95% CI: 45.3-72.3), and specificity was 79.6% (95% CI: 74.3-84.9). Hypothetically applying international MMS criteria to the subset of our cohort with high diagnostic confidence (n = 281) would have resulted in the excision of ≤0.7% (n = 2) of benign lesions. CONCLUSIONS: Dermoscopy HH-RCM diagnosis significantly reduces unnecessary diagnostic biopsies. It could potentially replace confirming biopsies in managing head and neck BCC when considering micrographically controlled surgery.