A Prospective, Observational Study Evaluating the Diagnostic Performance of a Newly FDA-Approved Line-Field Optical...

We prospectively enrolled 131 lesions of 104 patients to confirm or rule out the diagnosis of BCC. Dermoscopy, line-field optical coherence tomography (LC-OCT), and Reflectance confocal microscopy 1500 (RCM) images were evaluated by one blinded expert (OM).

The primary objective of this study was to assess the diagnostic performance of LC-OCT in lesions clinically suspicious for basal cell carcinoma (BCC), by determining the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of this technique. Diagnostic performances were evaluated in two distinct real-life settings: lesions for which tissue-coupled reflectance confocal microscopy (RCM) served as the reference standard, and lesions for which histopathological examination following biopsy was used as the reference standard. The secondary objectives were to assess clinician confidence when using dermoscopy, LC-OCT, and RCM, and to estimate the agreement rate between LC-OCT and RCM findings.

Design: Single-center, nonrandomized, blinded, prospective, observational study.

OptiSkin.

Patients with Fitzpatrick Skin Types I-IV between the ages of 29-92 years from OptiSkin are eligible for this study.

Clinical, dermoscopic, and LC-OCT images were obtained for all subjects. Further, measurements on the lesion included either an RCM examination or a traditional cutting biopsy, depending on body location, insurance coverage, or patient preference. At each stage, the clinician provided her confidence level in her diagnosis.

A total of 131 lesions clinically suspicious for BCC were imaged with LC-OCT in 104 patients. LC-OCT successfully imaged all 131 lesions, including those in challenging anatomical regions inaccessible to RCM, such as the nose.

Tissue-coupled RCM was performed on 63 lesions, for which LC-OCT achieved 95.9% sensitivity, 85.7% specificity, and 93.7% accuracy (PPV 95.9%, NPV 85.7%). This analysis revealed four disagreements: two FN and two FP. The remaining 68 lesions, diagnosed with histology, showed 86.4% sensitivity, 79.2% specificity, and 83.8% accuracy (PPV 88.4%, NPV 76.0%). This analysis revealed 11 disagreements, 6 FN, and 5 FP. Using RCM or histology when available as the gold standard, LC-OCT had an overall sensitivity of 91.2%, specificity of 77.50%, and accuracy of 86.92%.

For the 63 lesions evaluated with both LC-OCT and RCM, the agreement rate was 93.6% (Cohen's Kappa 0.81), with only 4 diagnostic discrepancies (2 FN and 2 FP). High-confidence LC-OCT diagnoses (≥ 8/10) achieved 100% sensitivity.

Our findings demonstrate that LC-OCT provides high sensitivity, specificity, accuracy, PPV, and NPV when tested against both tissue-coupled RCM and histopathology. LC-OCT was easy to use in challenging facial regions where the RCM device can be difficult to operate. This high diagnostic accuracy suggests that LC-OCT may reduce the need for immediate biopsy, streamline clinical decision-making, and minimize scarring, thereby improving cosmetic outcomes. Although the widespread adoption of LC-OCT in the United States remains limited by current reimbursement barriers, this technology has the potential to become an essential tool in modern dermatologic oncology.