AI-Assisted LC-OCT Detects Subclinical Basal Cell Carcinoma
Key Takeaways
- AI-assisted line-field confocal optical coherence tomography (LC-OCT) identified histologically confirmed subclinical basal cell carcinoma (BCC) in a prospective feasibility study of 150 patients at increased BCC risk, according to results from a new feasibility study in JAMA Dermatology.
- Investigators identified 17 subclinical BCCs in 14 patients (9.3%); the primary analysis yielded a positive predictive value (PPV) of 83.3% (95% CI, 58.1% to 96.4%).
- The study did not establish screening sensitivity, cost-effectiveness, or clinical benefit..
Artificial intelligence (AI)-assisted line-field confocal optical coherence tomography (LC-OCT) identified subclinical basal cell carcinoma (BCC) in clinically inconspicuous facial skin among patients at increased risk for BCC, according to results from a prospective cross-sectional feasibility study.
Researchers evaluated 150 consecutive patients receiving inpatient care at a single study center in Germany. Eligible patients had at least 2 established BCC risk factors. Lesions that were macroscopically suggestive of BCC were excluded before systematic facial screening with AI-assisted LC-OCT.
AI-Assisted LC-OCT Identifies BCC Before Clinical Signs
The study’s prespecified primary outcome was the positive predictive value of AI-assisted LC-OCT for histologically confirmed subclinical BCC. Participants had a mean age of 72.9 years; 38.0% were female and 62.0% were male.
Investigators identified 17 subclinical BCCs in 14 patients, representing 9.3% of the study population. Among 18 lesions classified as BCC using AI-assisted LC-OCT, 15 were confirmed histologically, producing a PPV of 83.3% (95% CI, 58.1% to 96.4%). One lesion was a false-positive result identified as actinic keratosis, while 2 patients declined biopsy.
Sensitivty analysis showed PPV reached 94.4% (95% CI, 72.7% to 99.9%). Of the 17 BCCs identified, 13 (76.5%) were superficial, 3 (17.6%) were nodular, and 1 (5.9%) was infiltrative. Subtype classification was accurate in 13 of 15 histologically confirmed cases (86.7%).
The feasibility design did not establish screening sensitivity, cost-effectiveness, or whether earlier detection improves clinical outcomes.
“In this cross-sectional feasibility study, the findings suggest that systematic screening with AI-assisted LC-OCT may be a feasible approach for BCC detection at a preclinical stage in populations at high risk for BCC,” the authors wrote. “Further prospective studies are needed to assess the sensitivity and cost-effectiveness of this method, and whether early detection translates to tangible clinical benefit before implementation in routine care can be recommended.”
Source
Ronicke M, et al. JAMA Dermatology. 2026. Doi:10.1001/jamadermatol.2026.2992