Abstract

Lung cancer, particularly non-small cell lung cancer (NSCLC), represents a major clinical burden in India. Although EGFR tyrosine kinase inhibitors (EGFR-TKIs) have substantially improved outcomes for patients with EGFR-mutated NSCLC, considerable heterogeneity exists in treatment response, duration of benefit, and development of acquired resistance. Current clinical decision-making relies predominantly on limited genomic biomarkers, clinical characteristics, and physician experience, which may not adequately capture the complex molecular determinants of treatment response.
A major unmet need is therefore the development of robust, population-relevant approaches for predicting EGFR-TKI response and identifying patients who are likely to derive durable benefit from a given treatment strategy. This gap is particularly important for Indian patients, as most existing genomic and AI/ML models have been developed using predominantly non-Indian populations and may not fully reflect the molecular and clinical characteristics of Indian patients.
This talk addresses this unmet need by generating and integrating multi-omics data from Indian NSCLC patients and systematically comparing these profiles with relevant Caucasian datasets. The study will identify population-specific and shared molecular features associated with EGFR-TKI response and resistance and use these features to develop a causal learning framework. In the longer term, the framework can be extended to incorporate additional clinical information, patient history, and imaging data, thereby supporting comprehensive patient stratification and precision treatment selection.
The primary beneficiaries will be Indian patients with EGFR-mutated NSCLC, who may benefit from more accurate treatment-response prediction and more informed therapeutic decision-making. Clinicians and healthcare systems may also benefit through improved patient stratification and the potential to reduce ineffective treatment and associated costs.