I3LUNG: Integrative science, Intelligent data platform for Individualized LUNG cancer care with Immunotherapy

Grant agreement ID: 101057695

AI-based platform to assist in lung cancer immunotherapy

Although immunotherapy is the new standard of care for advanced non-small cell lung cancer (NSCLC), less than 50 % of patients benefit from this treatment in the long term. Programmed death-ligand 1 biomarker is used to predict immunotherapy outcomes with limited efficacy, and other potential biomarkers have not yet been validated in randomised clinical trials. The EU-funded I3LUNG project aims to develop AI-based tools for improving patients’ survival and quality of life. The project will set up a global platform comprising data from 2 000 patients for AI models validation; moreover, it will collect multi-omics data from 200 NSCLC patients for information integration and application in leading immunotherapy decisions.

Objective

Immunotherapy (IO) is the new standard of care for many patients with advanced Non-Small Cell Lung Cancer (aNSCLC), yet only around 30-50% of treated patients benefit from IO in the long term. Programmed Death-Ligand 1 (PD-L1) remains the only biomarker used to predict patient outcome to IO, though its efficacy is limited. Other potential biomarkers have been identified, yet not validated in prospective randomized clinical trials, providing only partial evidence. Due to the dynamic complexity of the immune system-tumour microenvironment, its interaction with the host and patient behaviour, it?s unlikely for a single biomarker to accurately predict patient outcome. Artificial Intelligence (AI) and machine learning (ML) frameworks, that synthetize and correlate information from multiple sources, are essential to develop powerful decision-making tools able to deal with this highly complex context and provide individualized predictions to improve patient outcomes reducing the economic burden of health care systems in NSCLC.

The aim of the I3LUNG project is to develop such AI-based tools to assist in improving survival and quality of life, preventing undue toxicity, and reducing treatment costs. I3LUNG adopts a two-pronged approach: setting up a transnational platform of available data from 2000 patients in order to validate the AI models, and generating a multi-omics prospective data collection in 200 NSCLC patients integrating diverse -omic information then validate its usefulness in leading IO therapeutic decisions. A psychological study will help in defining the impact of AI-guided decisions on patients, eliciting their preference, and physicians comparing AI with Human Intuition. The final goal is the construction of a novel integrated AI-assisted Data Storage and Elaboration Platform backed up by Trustworthy Explainable AI methodology, ensuring its accessibility and ease of use by healthcare providers and patients alike.

Patients and Treatment

I3LUNG will enroll more than 2000 patients with metastatic non-small cell lung cancer (mNSCLC), the most common subtype of lung cancer, with the aim of investigating their individualized response to immunotherapy. This is a cancer treatment that uses the power of the body’s own immune system to prevent, control, and eliminate cancer. In our specific case, immunotherapy used for the treatment of mNSCLC are molecules called checkpoint inhibitors, that act by boosting immune cells to better recognize cancer cells and consequently eliminate tumor.

Immunotherapy is now the standard of care for this class of mNSCLC patients, administered in combination or not with chemotherapy, and often presenting with strong toxicity, unfortunately not always leading to a sustained clinical efficacy and moreover being very costly for the health system.

The biomarkers Issue

Biomarkers are biological molecules found in blood, other body fluids or samples, or tumor tissues which can be used to see how well the body responds to a treatment for a disease or condition. In mNSCLC, programmed Death-Ligand 1 (PD-L1) ( a protein expressed in the tumor cell) still remains the only biomarker used to predict patient outcome to immunotherapy, despite its less-than-ideal predictive performance, highlighting how more comprehensive molecular/translational analysis are desperately needed in this setting of patients.

Today, there is not a way to predict response to immunotherapy, outside PD-L1 and biomarkers remain a critical missing link in attempting to identify appropriate candidates for immunotherapy and tailoring most effective treatment regimens.

I3LUNG Solution to the Problem

In recent years, the explosion of Artificial Intelligence (AI) and machine learning (ML) has created the exciting opportunity to use a new set of tools to assess the vast amounts of data generated from clinical trials and research.

Thanks to the collection of biological, molecular, radiological, and clinical data from more than 2000 mNSCLC patients, I3LUNG will integrate all the collected information and thanks to the power of AI, generate a ML algorithm that will be able to predict the individual response to immunotherapy regimens. This tool will help to properly stratify mNSCLC patients and create a tailored treatment for each case, moving lung cancer care away from a “one-size-fits-all” approach to more of a personalized treatment plan.

This individualized patient selection strategy will also help to reduce the European economic burden and improve patient outcomes by better matching available treatments to patients.

I3LUNG project will cover a timeframe of 5 years. It started on 1st June 2022 until May 2027

Grant agreement ID: 101057695