A study published in the journal Aging-US describes a new machine learning tool designed to identify compounds that induce senescence in cancer cells. The research, led by Ryan Wallis and corresponding author Cleo L. Bishop from Queen Mary University of London, presents SAMP-Score, a method aimed at improving drug discovery for cancers with limited treatment options such as basal-like breast cancer.
Senescence refers to the process where damaged or aged cells stop dividing. In cancer therapy, inducing this state can help control tumor growth. However, detecting true senescence is challenging in certain cancers—known as Sen-Mark+ cancers—that already display features of cellular aging and lack reliable markers for confirmation. Basal-like breast cancer is one example.
To address this challenge, the researchers developed SAMP-Score, which uses a machine learning model trained to recognize morphological patterns typical of senescent cells under a microscope. By analyzing thousands of cell images and focusing on these specific shape and structure changes—called senescence-associated morphological profiles (SAMPs)—the model can distinguish real signs of cellular aging from effects caused by toxicity or normal variation.
"To demonstrate the potential application of SAMP-Score in p16 positive cancer therapeutic discovery, we assessed a diversity screen of 10,000 novel chemical entities in MB-468 cells (p16 positive BLBC)."
Using this approach, the team screened over 10,000 experimental compounds. They identified QM5928 as a compound that consistently induced senescence across multiple types of cancer cells without causing cell death. This makes QM5928 notable for further investigation, especially since it was effective even in cancers resistant to existing drugs like palbociclib—a common issue in tumors with high p16 expression such as basal-like breast cancer.
The researchers found that QM5928 prompted the p16 protein to move into the nucleus of affected cancer cells. This relocation may indicate that p16 is actively involved in stopping cell division—a detail only observable through SAMP-Score’s detailed imaging and analysis capabilities.
By integrating machine learning with advanced imaging techniques, this work introduces a new method for identifying potential therapies targeting cellular aging mechanisms within cancer treatment strategies. The authors note that SAMP-Score is openly available on GitHub for other scientists interested in researching therapies based on cellular senescence.
