Numerous studies show an important indicator of developing disease can be seen through changes in tumor marker values. These changes can be seen 2-3 years before diagnosis giving an important opportunity for early intervention.

Testing on a regular basis allows monitoring tumor marker values where a pattern of increasing score may indicate appropriate action be taken to significantly improved chances for a successful outcome.

OneTest is a new cancer risk test using a panel of tumor biomarker tests with an algorithm to help identify cancer risk in generally healthy people.  Depending on gender, a blood specimen is analyzed for six or seven FDA approved tumor biomarkers which may be elevated when cancer is present, even in early stages of development.  To calculate an individual’s cancer risk, OneTest employs machine-learning algorithms which incorporate the results of these blood tests along with your individual information, such as age and gender, which can further improve test accuracy compared to testing tumor biomarkers individually.

Testing of tumor markers is performed at the 20/20 GeneSystems’ CLIA laboratory using Roche Cobas e411 platform. FDA approved Roche’s IVD reagent kits are used according to manufacturer specifications. Results of individual tumor marker tests are used to calculate a risk score using machine learning proprietary algorithm and generate a patient report. The OneTest report provides the following information to the patient: 1) the individual 5 (males) or 6 (females) biomarker levels, an overall cancer risk score calculated using a machine learning algorithm, and for those at high risk, the organ systems (e.g. gastrointestinal, genitourinary, etc.) in which primary tumor is likely originated.

“Machine learning” algorithms

Machine Learning is a form of artificial intelligence (AI) in which a computer system has the ability to continually process and incorporate new data and thus fine tune its output over time. Because the algorithm used to combine and interpret patient biomarker levels with relevant clinical factor data is derived from machine learning, this algorithm is amenable to periodic review and redefinition. While the current algorithm is fixed on the basis of rigorous studies performed to date, 20/20 GeneSystems, Inc. is committed to the performance of regular review of the algorithm as the existing patient dataset grows. By providing 20/20 GeneSystems, Inc. with outcome data from patient follow-up subsequent to the OneTest, real-world experiences can inform further development and fine-tuning of the OneTest algorithm and continuously improve the accuracy of the test.