Improving hepatocellular carcinoma screening in Switzerland: new strategies.
Description: Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related death worldwide and the fifth in Switzerland. The stage of disease at time of diagnosis determines prognosis and treatment effectiveness. To enable early detection of potentially curable HCC, clinical practice guidelines recommend semi-annual surveillance with liver ultrasound (US) with or without serum alfa-feto protein (AFP). However, US and AFP are known for limited HCC diagnostic accuracy.In this context, in order to achieve a higher diagnostic performance of HCC screening, different alternative strategies need to be tested. First, the use of CEUS could be a valuable tool in case of lesion found on US in order to reduce false positive rates allowing diagnostic of benign lesions during the exam. Thus, this will increase US specificity and decrease the rate of US-related unnecessary complementary investigations. However, it was never tested in the context of HCC screening. Alternatively, a novel concept of abbreviated MRI (AMRI) screening strategy has been conceived. AMRI consists in the selection of sequences designed to answer the targeted question of HCC detection, thus reducing time and cost while maintaining high diagnostic accuracy. Multiple AMRI protocols have been proposed, with retrospective investigation allowing to the conclusion that Dynamic (Dyn)-AMRI is the most accurate protocol for HCC screening, with higher diagnostic performance than non-contrast (NC)-AMRI and gadoxetate-enhanced hepatobiliary phase (HBP)-AMRI. However, these results require prospective validation in screening conditions, particularly regarding the head-to-head comparison of US to each HCC screening alternative that is missing. Moreover, Dyn-AMRI was never tested prospectively with the proper intravenous (IV) contrast (extracellular contrast agent instead of gadoxetate acid), with expected higher diagnostic performance. Moreover, NC-AMRI presents the advantage of not requiring IV contrast injection that - despite its lower sensitivity compared to Dyn-AMRI - needs to be consider in the context of screening. In complement to AMRI, recent progress in deep learning (DL) for medical imaging evidenced that it is now robust enough for routine clinical use including for cancer detection. DL could be used as an additional tool for radiologists, for detection sensitivity improvement and/or for reduction of interpretation time, both crucial in a screening context. Lately, as any screening test needs to be cost-effective in the cost-health system where it is intent to be used, a cost-effectiveness analysis integrating each potential strategy within the particular Swiss HCC screening population and health system is required. Even if preliminary results evidenced the cost-effectiveness of AMRI in North America, it was never tested in Europe and cost effectiveness of CEUS and DL for HCC screening are currently unknown.The ultimate objective of the present project is to develop and validate a novel screening strategy for HCC screening suited to at-risk patients in Switzerland. This precise, safe, available and cost-effective method will help detecting asymptomatic HCC at an early, curable stage, which ultimately should reduce death burden from HCC. In order to achieve this objective, the project will comprise three work packages (WP):- WP 1: Compare the diagnostic performance of US, CEUS and AMRI for HCC detection in an at-risk population. - WP 2: Exploratory aim: Develop and test a DL model for HCC detection based on AMRI.- WP 3: Perform a cost-effectiveness analysis comparing US to different alternative screening strategies for HCC in Switzerland.A prospective observational study comparing US, CEUS and AMRI head-to-head for HCC screening in patients at risk followed in Lausanne University Hospital is planned. Imaging follow-up will continue as long as the patient does not develop HCC. Data from WP 1 will be used to train the DL model and to perform the cost-effectiveness analysis comparing each alternative strategies for HCC screening. Target enrollment for this study is 330 patients, as this is expected to yield 30 patients with HCC.
Amount: CHF 581,100
Funding type: research
Funder scheme: Project funding
Start date: 2023-01-01
End date: 2026-12-31
Investigator: Vietti Violi, Naïk
Institution: University of Lausanne – LA
Subfield: Hepatology
Field: Medicine
Domain: Health Sciences
Funded outputs count: 4
Funder award ID: 207944
OpenAlex ID: G8472082444