Cathepsin detection to identify malignant cells during robotic pulmonary resection.

Background: Intraoperative molecular imaging (IMI) uses a fluorescent probe to identify occult cancers. VGT-309 is a quenched activity-based probe that is activated in the presence of cathepsins, enzymes overexpressed in cancer cells, and detected by near-infrared (NIR) light. This study aims to evaluate the sensitivity and the positive predictive value (PPV) of robotic-assisted thoracic surgery (RATS) with intraoperative molecular imaging (RIMI) using VGT-309 to localize tumors using NIR light to detect areas with increased cathepsin activity. Our secondary outcome was to compare RIMI to video-assisted thoracic surgery (VATS) with intraoperative molecular imaging (VIMI).

Methods: In a phase 2 clinical trial at the University of Pennsylvania, patients (n=10) with suspicious pulmonary lesions underwent RATS. First, white light was used followed by RIMI to identify tissues with increased cathepsin activity. Then, VIMI was performed to compare the sensitivity and PPV in identifying the cathepsin activity. The resected specimens were then evaluated for fluorescence and underwent histopathological analysis for cathepsin expression. Image analysis was performed using ImageJ software. Statistical analysis was conducted using IBM SPSS Statistics software. A P value of 0.05 or less was considered significant.

Results: RATS with white light identified 6 out of the 10 pulmonary nodules, whereas adding RIMI identified an additional 4 more pulmonary nodules. RIMI and VIMI were able to detect the same 8/10 (80%) nodules. The addition of VIMI did not identify any lesions that RIMI may have missed. The mean fluorescence intensity of tumors visualized by RIMI was 115.81 A.U. [standard deviation (SD) =58.57] compared to 95.6 A.U. (SD =14.81) by VIMI (P=0.41). The mean tumor-to-background ratios (TBR) of tumors visualized by RIMI was 9.20 (SD =9.12) compared to 2.29 A.U. (SD =1.11) using VIMI (P=0.1). The sensitivity of RIMI and VIMI was 88.9% which was superior to that of RATS (55.6%). The PPV of RATS was 83.3% compared to 100% in RIMI and VIMI.

Conclusions: RIMI is a valuable option for visualization of occult disease using VGT-309-guided IMI through identifying areas of increased cathepsin activity. In this small series, RIMI and VIMI showed clinical equivalence in sensitivity and PPV of detecting cathepsin activity.