Objectives: The objectives of this study are to describe the standardization and dissemination of dynamic contrast-enhanced ultrasound (DCE-US) for the evaluation of antiangiogenic treatments in solid tumors across 19 oncology centers in France and to define a quality score to account for the variability of the evaluation criteria used to collect DCE-US data.
Materials and methods: This prospective Soutien aux Techniques Innovantes Coûteuses (Support for Innovative and Expensive Techniques) DCE-US study included patients with metastatic breast cancer, melanoma, colon cancer, gastrointestinal stromal tumors, renal cell carcinoma and patients with primary hepatocellular carcinoma tumors treated with antiangiogenic therapy. The DCE-US method was made available across 19 oncology centers in France. Overall, 2339 DCE-US examinations were performed by 65 radiologists in 539 patients.One target site per patient was studied. Standardized DCE-US examinations were performed before treatment (day 0) and at days 7, 15, 30, and 60. Dynamic contrast-enhanced ultrasound data were transferred from the different sites to the main study center at the Institut Gustave-Roussy for analysis. Quantitative analyses were performed with a mathematical model to determine 7 DCE-US functional parameters using raw linear data. Radiologists had to evaluate 6 criteria that were potentially linked to the precision of the evaluation of these parameters: lesion size, target motion, loss of target, clear borders, total acquisition of wash-in, and vascular recognition imaging window adapted to the lesion size.Eighteen DCE-US examinations were randomly selected from the Soutien aux Techniques Innovantes Coûteuses (Support for Innovative and Expensive Techniques) database. Each examination was quantified twice by 8 engineers/radiologists trained to evaluate the perfusion parameters. The intraobserver variability was estimated on the basis of differences between examinations performed by the same radiologist. The mean coefficient of variability associated with each quality criterion was estimated. The final quality score, ranging from 0 to 5, was defined according to the value of coefficient of variability for each criterion.
Results: A total of 2062 examinations were stored with raw linear data. Five criteria were found to have a major impact on quality: lesion size, motion, loss of target, borders, and total acquisition of wash-in. Only 3% of the examinations were of poor quality (quality of 0); quality was correlated with the radiologists' experience, such that it was significantly higher for radiologists who had performed more than 60 DCE-US examinations (P < 0.0001).
Conclusions: The DCE-US methodology has been successfully provided to several centers across France together with strict rules for quality assessment. Only 3% of examinations carried out at these centers were considered not interpretable.