Tailoring the best treatments to cancer patients remains a highly important endeavor in the oncology field. However, personalized treatment courses are challenging to determine, and technologies or methods that can successfully be employed for precision oncology are lacking.
Researchers at the NCI have built a new method for guiding cancer patient therapy based on single cell transcriptomics data of their tumors. This precision oncology data science and software framework is termed PERsonalized single-Cell Expression-based Planning for Treatments In ONcology (PERCEPTION). It capitalizes on recent, matched bulk and single-cell transcriptome profiles sourced from large-scale cell-line drug screenings and builds treatment response models from patient single-cell (SC) tumor transcriptomes. As a proof-of-concept, PERCEPTION has been shown to successfully predict response to monotherapy and combination therapy based on SC-expression profiles, in screenings of standard cancer cell lines and patient-tumor-derived cell lines. It has also successfully identified responders to a combination therapy based on tumor SC-expression data from recent multiple myeloma and breast cancer clinical trials. Further, it has discovered the development of treatment resistance to five standard tyrosine kinase inhibitors in a recent SC dataset obtained from lung cancer patients undergoing treatment. PERCEPTION is the first technology to demonstrate that SC gene expression can provide a framework to predict effective targeted therapies for individual cancer patients in a data-driven manner.
The NCI is currently seeking research co-development partners for this first-in-kind computational method that is predictive of therapeutic response based on clonal SC gene expression of tumors
- First software of its kind that aims to tailor the best treatments to cancer patients based on single cell transcriptomics data of their tumors
- Has proof-of-concept retrospective validations in patient clinical trial data where PERCEPTION performed better than published state-of-the-art single-cell-based and bulk-based predictors
- Useful for companies focused on single cell sequencing or interested in the single cell domain
- Tool for personalized identification of effective patient therapies (precision oncology) based on single cell transcriptomics data of their tumors
- Tool for discovery of new drug combinations for specific cancer types and sub-types