Evaluation of parietal pleural invasion in subpleural lung cancer by VEGFR2 targeted microbubble-based ultrasound molecular imaging.

Background: The parietal pleural invasion in subpleural lung cancer is associated with poor prognosis and surgical difficulty. The target protein of vascular endothelial growth factor receptor 2 (VEGFR2) is highly expressed in invasive lesions. This study preliminarily investigated the value of ultrasound (US) molecular imaging based on microbubbles targeted to VEGFR2 in early diagnosis of parietal pleural invasion of subpleural lung cancer. Methods: Animal protocols were approved by the Institutional Administrative Panel on Laboratory Animal Care. The microbubbles targeted specifically to VEGFR2 was developed by using anti-VEGFR2 antibodies and biotin-streptavidin (SA) interaction. The shape and size of the microbubbles were detected using a microscope and dynamic light scattering (DLS). The binding specificity was tested on human umbilical vein endothelial cells (HUVECs). Expression of VEGFR2 on HUVECs was tested with immunocytochemical method. A rabbit subpleural lung cancer model was constructed by injecting VX2 tumor tissue with US guidance. Two weeks after subpleural lesion formation, the subpleural lesion and the parietal pleural invasion were evaluated with contrast-enhanced ultrasound (CEUS) based on VEGFR2-targeted microbubbles (VEGFR2-MBs) and control microbubbles (Con-MBs) by two sonographers. The parietal pleural invasion was ultimately determined using the gold standard method of findings from anatomical and pathological examination. Results: The microbubbles of VEGFR2-MBs and Con-MBs visually appeared as a uniform spherical shape under the optical microscope, with an average particle size of 1.7±0.51 and 1.5±0.54 μm, respectively, without a significant statistical difference. In cell culture, adherence of VEGFR2-MBs on HUVECs (7.6±1.6 microbubbles per HUVECs) was significantly higher than adherence of Con-MBs (0.1±0.04 microbubble per HUVECs; P<0.001). When the HUVECs were preincubated with VEGFR2 antibody, adherence of VEGFR2-MBs on HUVECs was significantly weakened (2.1±0.8 microbubbles per HUVECs; P<0.001). Ultimately, 17 rabbits were subjected to complete ultrasonic evaluation. There were 4, 3, 2 and 6, 0, 0 cases with abnormally enhanced areas observed between the lesions and the parietal pleura, on the parietal pleura and both in the VEGFR2-MBs and Con-MBs groups, respectively, as confirmed by anatomical and pathological evaluations. The sensitivity, specificity, and accuracy of the VEGFR2-MBs and Con-MBs groups for diagnosing parietal pleura invasion were 81.8%, 100.0%, and 88.2% and 54.5%, 100.0%, and 70.6%, respectively. There was a statistically significant difference between the two groups in terms of accuracy (P=0.15) and specificity (P=0.11) according to Fisher's exact test. Conclusions: VEGFR2-MBs-based US molecular imaging has great potential for clinical application in early detection of parietal pleura invasion in peripheral lung cancer patients.