Pharmacokinetic modeling of PSMA‐targeted nanobubbles for quantification of extravasation and binding in mice models of prostate cancer. Issue 10 (12th September 2022)

• Record Type: Journal Article
• Authors: Chen, Chuan; Perera, Reshani; Kolios, Michael C.; Wijkstra, Hessel; Mischi, Massimo; Exner, Agata A.; Turco, Simona
• Is Part Of: Medical physics. Volume 49:Issue 10(2022)
• Page Start: 6547
• Page End: 6559
• Publication Date: 2022-09-12
• Subjects: contrast‐enhanced ultrasound -- molecular imaging -- nanobubbles -- pharmacokinetic modeling -- prostate cancer; Medical physics -- Periodicals; Medical physics; Geneeskunde; Natuurkunde; Toepassingen; Biophysics; Periodicals; Periodicals; Electronic journals; 610.153
• DOI: 10.1002/mp.15962
• Languages: English
• ISSNs: 0094-2405

Abstract

Abstract: Purpose: Contrast‐enhanced ultrasound (CEUS) by injection of microbubbles (MBs) has shown promise as a cost‐effective imaging modality for prostate cancer (PCa) detection. More recently, nanobubbles (NBs) have been proposed as novel ultrasound contrast agents. Unlike MBs, which are intravascular ultrasound contrast agents, the smaller diameter of NBs allows them to cross the vessel wall and target specific receptors on cancer cells such as the prostate‐specific membrane antigen (PSMA). It has been demonstrated that PSMA‐targeted NBs can bind to the receptors of PCa cells and show a prolonged retention effect in dual‐tumor mice models. However, the analysis of the prolonged retention effect has so far been limited to qualitative or semi‐quantitative approaches. Methods: This work introduces two pharmacokinetics models for quantitative analysis of time–intensity curves (TICs) obtained from the CEUS loops. The first model is based on describing the vascular input by the modified local density random walk (mLDRW) model and independently interprets TICs from each tumor lesion. Differently, the second model is based on the reference‐tissue model, previously proposed in the context of nuclear imaging, and describes the binding kinetics of an indicator in a target tissue by using a reference tissue where binding does not occur. Results: Our results show that four estimated parameters, β, β / λ $\beta /\lambda $, β + / β − ${\beta }_ + /{\beta }_ - $, for the mLDRW‐input model, and γ for the reference‐based model, were significantly different ( p ‐value <0.05) between free NBs and PSMA‐NBs. These parameters estimated by the two models demonstrate different behaviors between PSMA‐targeted and free NBs. Conclusions: These promising results encourage further quantitative analysis of targeted NBs for improved cancer diagnostics and characterization.