iBiopsy

Median Technologies Continues Its Strong Positioning in Early-Stage Diagnosis with a New iBiopsy Clinical Development Plan on Lung Cancer Screening (LCS)

Regulatory News:

Median Technologies (ALMDT:PA) (Paris:ALMDT) unveils today a new iBiopsy Clinical Development Plan (CDP) for the diagnosis of early-stage lung cancer in high-risk populations based on Low Dose Computed Tomography (LDCT) scans, strengthening its positioning in early-stage diagnosis, such as early-stage liver cancer diagnosis and non-alcoholic steatohepatitis (NASH) identification.

Lung cancer is today the number one cancer killer globally and is responsible for approximatively 22% of cancer deaths¹. The 5-year survival rate for lung cancer is only 18,6%² because the majority of lung cancers are diagnosed late in the disease, when cure is no longer possible. In the past years, a significant number of clinical trials on very large patient cohorts have been conducted to determine the impact of diagnosing lung cancer at an early stage on lung cancer mortality. Results from these multicentric randomized trials have shown that lung cancer screening using LDCT reduces lung cancer mortality by 20–26% in high-risk populations³. In the US, LDCT lung cancer screening for high-risk individuals is included in the United States Preventive Services Task Force (USPSTF) screening guidelines and reimbursed by CMS⁴. Lung cancer screening programs are under discussion to be implemented in Europe. In France, the recently announced 2021-2030 Cancer Plan includes new measures for the early diagnosis of cancers associated with poor prognosis (5-year survival rate < 20%) including lung cancer⁵.

Deep learning models have the potential to significantly increase the accuracy, consistency and adoption of lung cancer screening worldwide through rapid diagnosis support, risk stratification and reduced need for costly and invasive interventions (biopsies, resections). Median will use its iBiopsy platform to develop AI-based algorithms enabling the identification and characterization (i.e., discrimination between malignant and benign) of lesions in thoracic LDCT images, decreasing the number of false positives and overdiagnosis.

Median has been working for many months on a LCS solution and will benefit from its years of experience and knowledge in the domain as well as of its existing relationships with some of the leading clinical sites in the world. Median has already collected over 17,000 cases of LCS from multiple databases and is currently curating and annotating a preliminary set of 1,800 data to train and validate its algorithms. The company expects to deliver a proof of concept in H2, 2021 based on this subset.