Indian scientists have developed an artificial intelligence (AI) framework that could change how cancer is understood and treated. The framework provides a new lens to look at cancer — not by its size or spread alone, but by its molecular personality.
Cancer is not just a disease of growing tumours, but is powered by a set of hidden biological programs called the ‘hallmarks of cancer’. These hallmarks explain how healthy cells turn malignant and how they spread, evade the immune system and resist treatment.
For decades, doctors have relied on staging systems like TNM, which stands for tumour, nodes and metastasis to describe the size and spread of tumours. But such systems often miss the deeper molecular story as to why two patients with the “same” cancer stage can have very different outcomes.
Scientists of SN Bose National Centre for Basic Sciences and Ashoka University have introduced the first AI framework that can read the ‘molecular mind’ of cancer and predict its behaviour, according to the Ministry of Science and Technology.
The team led the framework titled OncoMark to analyse 3.1 million single cells across 941 tumours spanning 14 cancer types, creating synthetic “pseudo-biopsies” that represent hallmark-driven tumour states. This huge dataset allowed AI to learn how hallmarks like metastasis, immune evasion and genomic instability work together to fuel tumour growth and therapy resistance.
“Quantifying the biological processes that drive cancer progression remains a key challenge in oncology. Although the hallmarks of cancer provide a foundational framework for understanding tumour behaviour, existing diagnostic tools rarely measure these hallmarks directly. Here we present a neural multi-task learning-based framework that estimates hallmark activity using gene expression data from tumour biopsies,” the researchers said in their study published recently in Communications Biology, a peer-reviewed journal by the Nature Publishing Group.
OncoMark achieved over 99 per cent accuracy in internal testing and remained above 96 per cent across five independent cohorts. It was validated on 20,000 real-world patient samples from eight major datasets, showing broad applicability. For the first time, scientists could actually visualise how hallmark activity rises with advancing cancer stage, the Ministry claimed.
The new framework can reveal which hallmarks are active in a patient’s tumour, pointing doctors towards drugs that directly target those processes. It can also help identify aggressive cancers that might look less harmful under standard staging, supporting earlier intervention.
“It predicts the activity of 10 cancer hallmarks simultaneously and with high accuracy and enables efficient analysis of transcriptomic data to inform understanding of tumour biology and support individualised treatment strategies,” the researchers said.
