Development and Evaluation of a Multimodal Approach to Predict Lung Cancer Risk and Determine EGFR Mutation Profile in a Lung Cancer Screening Population

---

---

Ongoing Project of $3,397,255 over 3 years

• Supported by CQDM through:
  • Pfizer Canada
  • MEIE
• And by a co-funding partner:
  • Fondation IUCPQ
  • BioMark Diagnostic Solutions
  • AstraZeneca

---

Challenge

Lung cancer is the most lethal cancer in industrialized countries, with different studies showing that its high mortality can be reduced through an early diagnosis. Implementation of a lung cancer screening program for patients at high risk of developing the disease has been recently recommended by the Quebec National Institute of Excellence in Health and Social Services (INESSS). This program targets about 380,000 individuals in the province of Quebec. Potential harms associated with the implementation of a lung cancer screening program include false-positive results leading to unnecessary tests and invasive procedures. The additional clinical resources (access to physician, radiology, pathology) to obtain a reasonable balance of benefits over harms represent a significant challenge in the context of a public health system.

Solution

The project responses to this new clinical reality with a focus to ensure the benefits of lung cancer screening in reducing lung cancer mortality and at the same time alleviate the extra pressure on the health care system. More specifically, in this research proposal, the team aims to improve the efficiency of lung cancer screening program by integrating genetic score, metabolic profile and radiomics (i.e., high-throughput analysis of medical imaging) by artificial intelligence to better select the patients that should undergo additional imaging and invasive procedures such as lung biopsy and surgery.  The work will be based on data and samples from IUCPQ-UL’s biobank and the Lung cancer screening demonstration project.

Expected achievements/Impact

This project will develop and assess the clinical performance of predictive biomarkers to complement and enhance the outcomes of a lung cancer screening program. It will be the first time that a multimodal predictive model that integrates radiomics, liquid biopsy and genetic score to improve patient management and reduce costs will be developed and applied in the context of a screening cohort for this disease. By working with private biotech companies well established in the province of Quebec and worldwide leaders in their field, it will be possible to rapidly scale up the technologies to be applied in our population.