Lung cancer remains one of the world’s deadliest cancers, and non-small cell lung cancer (NSCLC) represents about 85% of all lung cancers. Detection of the disease at an early stage and tailoring the treatment to the individual patient are crucial to improving survival rates—but this requires accurate “biomarkers,” biological signposts that reflect the presence, type, or stage of disease.
Traditionally, scientists have searched for biomarkers by analyzing one type of biological data at a time—genes (genomics), for instance, or proteins (proteomics). But diseases like cancer are complex and are fueled by many interrelated biological processes. To capture such complexity, scientists now use a multi-omics approach, combining multiple types of biological data—genomics, proteomics, metabolomics (metabolites), and transcriptomics (RNA expression), for instance—to get a more integrated view of what’s happening in the body.
The issue is that multi-omics generates enormous and complex datasets. That is where artificial intelligence (AI) comes in. AI, especially machine learning / deep learning algorithms, can analyze large volumes of data quickly, identify patterns not discernible to the human eye, and identify possible biomarkers for diagnosis, prognosis, and treatment planning.
This article explains how AI is being applied in multi-omics research in NSCLC. It identifies several areas of promise:
- Earlier detection – AI can detect subtle changes in biological data before symptoms arise.
- Personalized treatment – By associating biomarker profiles with treatment response, AI can help clinicians choose the optimal therapy for a particular patient.
- Better disease monitoring – AI algorithms can track changes over time to predict relapse or progression.
However, the paper also points to challenges:
- Data quality & consistency – AI is only as good as the data it is learning from, and inconsistencies can reduce accuracy.
- Privacy & ethics – Working with sensitive health data, strict safeguards are required.
- Clinical integration – Translating AI findings into everyday hospital practice is still a major challenge.
Overall, the combination of AI and multi-omics has the potential to revolutionize NSCLC therapy—from earlier and more accurate diagnoses to truly personalized treatment strategies. While there is still some distance to cover before these methods enter the mainstream, the technology is evolving rapidly and could change how doctors diagnose and manage lung cancer in the not-so-distant future.
Full text: Brandon Wilkins, Emily Hartman, Blake Kelley, Pranali Pachika, Joshua Bradley, James Bradley, The impact of artificial intelligence on a multi-omics approach toward predictive biomarkers for non-small cell lung cancer, Explor Digit Health Technol. 2025;3:101153 DOl: https://doi.org/10.37349/edht.2025.101153.