AI Test for Lung Cancer Risk

This trial explores how Artificial Intelligence (AI) can predict the risk of developing lung cancer over the next three years. It involves two groups: one that has already undergone lung cancer screening and another that has not. Participants will receive either an AI-based prediction (AI-Inferred Lung Cancer Risk Prediction) using past CT scans and health data or a lab test using DNA from their blood. The trial is best suited for individuals who have smoked for 20 years or more and either currently smoke or quit within the last 15 years. As an unphased study, this trial offers a unique opportunity to contribute to groundbreaking research that could revolutionize lung cancer prediction and prevention.

The trial information does not specify if you need to stop taking your current medications. However, it mentions that adjuvant hormone therapy for cancer is allowed, suggesting some medications might be permitted.

Research has shown that AI tools for predicting lung cancer risk are promising. Researchers have tested these AI models to assess their ability to predict lung cancer development by analyzing data from scans and other medical tests. They have found these models to be more accurate than older methods in identifying at-risk individuals.

Regarding safety, the AI tool poses no physical side effects because it only analyzes existing medical data like CT scans. It does not involve taking a drug or undergoing a procedure. The AI tool assists doctors in determining who might need further testing or treatment.

Researchers are excited about this trial because it introduces a novel way to assess lung cancer risk using artificial intelligence (AI). Unlike the standard low-dose CT (LDCT) scans, which are the current go-to method for screening, the AI approach analyzes CT imaging alongside clinical features to predict risk more precisely. Additionally, for those who've never had an LDCT scan, there's an innovative test analyzing circulating DNA fragmentomics, offering a new avenue for early detection. These methods promise to enhance accuracy and potentially catch lung cancer earlier, which could be a game-changer in treatment outcomes.

Research has shown that AI tools hold promise in predicting lung cancer risk. In this trial, participants in the screen-established cohort will receive a research-use-only (RUO) multimodal artificial intelligence risk prediction based on lung screening CT imaging and clinical features. Studies have demonstrated that an AI model called Sybil is highly accurate, identifying high-risk and low-risk patients with accuracy rates between 86% and 94%. This model can predict lung cancer risk from just one low-dose CT scan and has proven effective up to six years in advance. Meanwhile, participants in the screen-naïve cohort will receive a regulatory-cleared laboratory-developed test for lung cancer screening, circulating DNA fragmentomics. Another study found that when the AI correctly identified all cancer cases within one year, it also classified 68.1% of non-cancer cases as low risk, reducing unnecessary worry. These findings suggest that AI could be a powerful tool for early lung cancer detection and personalized screening.⁶⁷⁸⁹¹⁰