Pathological and Oncological Sciences II.
Introduction: Basal cell carcinoma (BCC) stands as the predominant form of skin malignancy. Presently, the histological subtype (HST) of BCC, pivotal for treatment stratification, necessitates invasive biopsy for definitive diagnosis.
Aims: This study endeavors to assess the potential of optically guided high-frequency ultrasound (OG-HFUS) imaging in discerning aggressive HST BCCs from their low-risk counterparts.
Method: A prospective investigation entailed clinical and dermoscopic evaluations of BCCs, followed by OG-HFUS imaging at 33 MHz, subsequent surgical excision, and histopathological analysis. 75 patients harboring 78 BCCs were recruited, with 63 lesions forming the basis for developing a novel OG-HFUS risk classification algorithm, and 15 utilized for algorithm validation. The cohort exhibited a mean age of 72.9 ± 11.2 years, with histological delineation identifying 16 lesions as aggressive HST (infiltrative or micronodular subtypes) and 47 as low-risk HST (superficial or nodular subtypes). Statistical analysis comprised a one-sided Fisher’s exact test for categorical assessment and Receiver Operating Characteristic (ROC) curve analysis for diagnostic accuracy evaluation.
Result: OG-HFUS discriminated aggressive BCC HSTs effectively, identifying irregular shape (p < 0.0001), ill-defined margins (p < 0.0001), and non-homogeneous internal echoes (p = 0.004). A risk-categorizing algorithm showed higher sensitivity (82.4%) and specificity (91.3%) than traditional evaluations (sensitivity: 40.1%, specificity: 73.1%). Dermoscopic assessment had PPV and NPV of 30.2% and 76.8%, while OG-HFUS-based algorithm showed PPV of 94.7% and NPV of 78.6%. Validation using independent images (n = 15) by blinded evaluators revealed sensitivity of 83.33% and specificity of 91.66%.
Conclusion: The study underscores OG-HFUS as a promising tool for identifying aggressive BCC HSTs, leveraging readily discernible morphological parameters to inform early therapeutic decision-making.
Funding: This research was funded by ÚNKP-23-4-II-SE-8 (N.K.), New National Excellence Program of the Ministry for Innovation and Technology (N.K.) and Semmelweis University: SE 250 + Excellence PhD Scholarship (M.B.).