In vivo molecular skin fluorescence imaging for noninvasive assessment of atypical nevi and melanoma: A validation trial.
Summary
In a multicenter prospective validation of 240 pigmented lesions, skin fluorescence imaging targeting αvβ3 integrin achieved sensitivities of 93% and 87% and specificities of 77% and 91% at score cutoffs of 5 and 7, respectively (AUC 0.907). SFI effectively discriminated low-risk from high-risk lesions and may reduce unnecessary biopsies.
Key Findings
- Prospective, multicenter validation on 240 lesions prior to biopsy
- At SFI cutoff 5: sensitivity 93%, specificity 77%; at cutoff 7: sensitivity 87%, specificity 91%
- AUC of ROC was 0.907 (95% CI 0.864–0.951)
- Lesion distribution included 99 without dysplasia, 110 dysplastic nevi (low/high grade), and 31 melanomas (in situ/invasive)
Clinical Implications
SFI can be incorporated as an adjunct triage tool to identify high-risk melanocytic lesions for biopsy while safely deferring low-risk lesions, potentially reducing patient morbidity and healthcare costs.
Why It Matters
Introduces a noninvasive, molecularly targeted diagnostic that demonstrates high accuracy for melanoma risk stratification. This could shift initial evaluation workflows and reduce invasive procedures.
Limitations
- Single-arm validation without comparator devices
- Generalizability beyond participating clinics and across skin phototypes requires further study
Future Directions
Conduct head-to-head trials versus dermoscopy/other assistive tools, assess performance across Fitzpatrick types, and evaluate impact on biopsy rates and cost-effectiveness in real-world pathways.
Study Information
- Study Type
- Cohort
- Research Domain
- Diagnosis
- Evidence Level
- II - Prospective multicenter diagnostic validation using histopathology as reference standard
- Study Design
- OTHER