A. Ibrahimli1, K. Zhang2, A. Galandarova3, A. Chang2, P. Bou-Samra2, A. Chang2, Z. Karimov4, S. Singhal2, G. Kennedy2, F. Azari2, S. Singhal2 1Ankara University Faculty of Medicine, Ankara, Turkey 2University Of Pennsylvania, Philadelphia, PA, USA 3Ankara Yildirim Beyazit University Faculty of Medicine, Ankara, Turkey 4Ege University Faculty of Medicine, Izmir, Turkey
Introduction: Non-small cell lung cancers are a heterogenous group of solid organ malignancies with diverse receptor expression profiles. Therefore, fluorochrome selection for each patient's tumor is of critical importance for successful fluorescent labeling during thoracoscopy. The goal of our study was to prospectively evaluate the correlation of tumor core biopsies with intraoperative NIR fluorescence.
Methods: Data was retrospectively analyzed from a prospectively collected database. Patients underwent infusion of the study drug, Pafolacianine (OTL38), up to 24 hours prior to surgery. Core biopsies were taken intraoperatively prior to NIR camera localization. Mean fluorescence intensity of the tumors was calculated using ImageJ (National Institutes of Health) software. To evaluate the efficacy and variability of tumor-to-background ratio (TBR) calculations we analyzed patient and biologic factors, tumor characteristics and technological factors with respect to in situ TBRs.
Results: Fourteen patients were eligible for phase III pafolacianine guided resection, of whom seven consented for infusion of the tracer. The median time from infusion to resection was 18.12±3.4 hours with an average dose of 2 mg/kg of pafolacianine infused. Four of the patients were female and the median ASA score was 3. There was an average of 2.5 core biopsy passes per tumor, with a median tumor size of 0.89 ±0.45 cm determined on pathology. On histopathologic analysis by a dedicated thoracic pathologist, folate receptor alpha (FRα) greater than 1+ scored strongly correlated with observation of in situ fluorescence (p<0.01) using non-parametric analysis. The median TBR was 2.44±0.61 for lesions with FRα expression. On immunofluorescence microscopy using the NIR Pearl Image analyzer, FRα cell surface presence was associated with the highest localization of heatmap signatures with a mean fluorescence intensity of 156 a.u (p<0.01). In five out of seven patients, all core biopsies correlated with each other on IHC staining score (p<0.05). The remaining patients had at least one core biopsies that had different cell surface receptor expression on IHC.
Conclusion: Pre-operative core biopsy assessment of tumor FRα expression is associated with the detection of pafolacianine NIR fluorescence in situ. More than one core biopsy should be performed to avoid false negative results.
