Scientist pipetting sample into a vial for DNA testing
Credit: Cavan Images / iStock / Getty Images Plus

Machine learning and transcriptomics have been combined to create a computational pipeline to systematically predict patient response to cancer drugs, including resistance emergence, at single-cell resolution. The researchers who built the pipeline, PERCEPTION, say this is the first of its kind. It is not yet “clinic ready,” but they hope that by encouraging more groups to use it, the model will become more robust and better validated.

Their study was published in Nature Cancer. The first author is Sanju Sinha, PhD, of the Cancer Molecular Therapeutics Program at Sanford Burnham Prebys. The senior authors are Eytan Ruppin, MD, PhD, and Alejandro Schaffer, PhD, at the National Cancer Institute. PERCEPTION is accessible at

Currently, few cancer patients benefit from early targeted therapy. One recent report found that by December 2020, 89 small-molecule targeted antitumor drugs were approved by the FDA and the National Medical Products Administration (NMPA) of China (Zhong et al., Signal Transduction and Targeted Therapy, 2021). As those authors wrote, “Despite great progress, small-molecule targeted anti-cancer drugs still face many challenges, such as a low response rate and drug resistance.”

The PERsonalized Single-Cell Expression-Based Planning for Treatments in Oncology (PERCEPTION) uses artificial intelligence and single-cell transcriptomics to match patients to cancer drugs. The transcriptome is the complete set of RNA transcripts produced by the genome.  

“A tumor is a complex and evolving beast. Using single-cell resolution can allow us to tackle both of these challenges,” said Sinha. “PERCEPTION allows for the use of rich information within single-cell omics to understand the clonal architecture of the tumor and monitor the emergence of resistance.” 

Sinha added, “The ability to monitor the emergence of resistance is the most exciting part for me. It has the potential to allow us to adapt to the evolution of cancer cells and even modify our treatment strategy.”

Sinha and colleagues used a branch of AI called transfer learning to build PERCEPTION. Transfer learning is the reuse of a pre-trained model on a new problem

PERCEPTION uses published bulk-gene expression from tumors to pre-train its models. Then, single-cell data from cell lines and patients, even though limited, is used to tune the models.

The model was validated by predicting the response to monotherapy and combination treatment in three independent, recently published clinical trials for multiple myeloma, breast, and lung cancer. In each case, PERCEPTION correctly stratified patients into responder and non-responder categories. In lung cancer, it even captured the development of drug resistance as the disease progressed, a notable discovery with great potential.

Sinha said PERCEPTION is not yet ready for the clinic, but the team’s approach shows that single-cell information can be used to guide treatment. He hopes to encourage the adoption of this technology in clinics to generate more data, which can help refine the technology for clinical use.

“The quality of the prediction rises with the quality and quantity of the data serving as its foundation,” said Sinha. “Our goal is to create a clinical tool that can predict the treatment response of individual cancer patients in a systematic, data-driven manner.” 

Also of Interest