Freenome has filed a patent for a method of detecting and treating colorectal cancer using a computer programmed with a trained machine learning model. The method involves obtaining an autoantibody profile from a biological sample, processing it using the machine learning model, and identifying the presence of colorectal cancer. This could potentially enable early detection and treatment of the disease. GlobalData’s report on Freenome gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Freenome, AI-assisted genome sequencing was a key innovation area identified from patents. Freenome's grant share as of September 2023 was 6%. Grant share is based on the ratio of number of grants to total number of patents.
Early detection of colorectal cancer using autoantibody biomarkers
A recently filed patent (Publication Number: US20230243830A1) describes a method for detecting and treating colorectal cancer using a computer programmed with specific instructions. The method involves obtaining an autoantibody profile from a biological sample of the subject, processing the profile using a trained machine learning model, and identifying the presence of colorectal cancer based on the output value provided by the model. The computer is further programmed to detect or treat the cancer based on the identification.
The method utilizes a pre-determined autoantibody panel comprising autoantibodies to at least three antigens selected from a group of antigens including NME5, USP16, UBE2S, RNF41, CD20, ANKHD1, TXNL1, NAT6, Supt6h, PRDM8, OTUD5, PNKP, SRSF7, ASB9, NXN, ZBTB21, EYA1, GSPT1, MLIP, RBM38, ARMC5, TP53, BRD9, CDK4, PRMT6, PCOLCE, and SDCBP. The autoantibody profile can be obtained from various biological samples such as body fluids, stool, urine, blood, tissue biopsy, and more.
The trained machine learning model is specifically designed to distinguish between subjects with colorectal cancer and those without, providing an output value associated with the presence of colorectal cancer. This allows for the identification of colorectal cancer in the subject. The method can be used for both detecting and treating colorectal cancer, providing potential benefits in early diagnosis and personalized treatment.
The patent also mentions the possibility of incorporating additional analyses into the method. For example, the computer can be programmed to determine the methylation status of nucleic acid molecules in the biological sample, providing a methylation profile of the subject. This profile can be further processed using the trained machine learning model. Similarly, the computer can measure the amount of proteins in the sample, providing a protein profile that can also be processed using the machine learning model.
Overall, this patent describes a method for detecting and treating colorectal cancer using a computer programmed with specific instructions and a trained machine learning model. By analyzing the autoantibody profile of a subject, the method can accurately identify the presence of colorectal cancer and potentially guide personalized treatment decisions. The incorporation of additional analyses, such as methylation and protein profiling, further enhances the potential of this method in improving cancer diagnosis and treatment.