Scientists have developed a new artificial intelligence (AI) framework capable of identifying hidden cancer stem-like cell populations that may be responsible for tumour recurrence, metastasis and treatment resistance, potentially opening new avenues for precision cancer medicine.
Researchers from the S. N. Bose National Centre for Basic Sciences (SNBNCBS), an autonomous institute under the Department of Science and Technology (DST), in collaboration with Ashoka University, have developed the AI-based framework to detect rare cancer stem-like cell states from tumour gene-expression data.
The technology could be particularly useful in analysing large numbers of patient samples in settings where high-resolution single-cell sequencing facilities are not readily available.
Cancer treatments can eliminate large numbers of tumour cells, but a small population of cells may survive treatment and contribute to the return of tumours, their spread to other organs and the development of resistance to therapy. Scientists have long suspected that rare cancer stem-like cells play a major role in these processes.
However, identifying these cells has remained difficult because they are rare and can change their biological identity over time.
Three cancer stem-like cell states identified
Led by Dr Shubhasis Haldar, the research team developed a framework called ACSCeND (AI-based Cancer Stem-like Cell Profiler and Neoplasm Deconvoluter).
Unlike conventional computational approaches that assign a tumour a single “stemness” score, ACSCeND identifies three distinct developmental states of cancer stem-like cells – pluripotent-like, multipotent-like and unipotent-like.
The system combines information learned from high-resolution single-cell sequencing with deep learning to analyse conventional bulk tumour RNA sequencing data. This enables researchers to identify hidden cell populations in thousands of patient samples where single-cell sequencing may not be available.
Analysed more than 25,000 tumour samples
The researchers tested ACSCeND against existing computational methods and found that it consistently performed better across independent datasets and different sequencing platforms.
The framework was subsequently used to analyse more than 25,000 tumour samples from major international cancer databases, including The Cancer Genome Atlas (TCGA) and PRECOG.
The analysis found that tumours containing higher levels of highly potent, pluripotent-like cancer stem cells were associated with poorer patient survival, greater likelihood of tumour recurrence and reduced response to modern immunotherapies.
The framework also helped identify molecular programmes that allow these cancer stem-like cells to survive, adapt and evade the immune system.
Builds on earlier AI platform
The new work builds on the research team’s earlier AI platform, OncoMark, which was designed to decode biological characteristics driving cancer progression across millions of cells and achieved more than 99 per cent predictive accuracy.
While OncoMark focused on understanding the biological processes underlying cancer progression, the new ACSCeND framework addresses one of the more difficult challenges in cancer biology – identifying the rare stem-like cells that contribute to tumour evolution and treatment failure.
Researchers believe the findings could eventually help in identifying patients at higher risk of relapse, discovering potential drug targets and developing more effective personalised cancer treatment strategies.
The study also highlights the growing role of artificial intelligence in biomedical research, where AI tools can identify complex biological patterns in massive genomic datasets that would be difficult to detect through conventional analysis.
According to the researchers, technologies such as OncoMark and ACSCeND demonstrate the potential of AI to accelerate advances in cancer diagnosis, treatment-response prediction and precision medicine, while enabling analysis of patient data even in settings with limited access to advanced sequencing technologies.
The post AI framework identifies hidden cancer stem cells, offers new path to precision medicine appeared first on DD India.
