Freenome Unveils Promising Early Data on Colorectal Cancer Screening Test at American College of Gastroenterology Annual Meeting

PHILADELPHIA--()--South San Francisco–based Freenome, the leading artificial intelligence (AI) genomics company, is presenting promising early data on its efforts to utilize machine learning to detect colorectal cancer (CRC) at its earliest stages, when treatment is most effective.

The current lack of an effective, blood-based test may be a contributing factor to low rates of CRC screening today, with nearly 1 in 3 adults aged 50-75 years non-adherent with screening guidelines. Existing methods, such as colonoscopy and stool-based testing, are often seen as inconvenient and uncomfortable by patients, leading many to postpone or forgo recommended screenings.

Freenome’s poster, Early Stage Colorectal Cancer Detection Using Artificial Intelligence and Whole-genome Sequencing of Cell-free DNA (P1997), suggests that an AI-powered blood test may soon be a viable alternative.

The poster will be presented on Tuesday, October 9th at 10:30 am EDT at the Annual Meeting of the American College of Gastroenterology (ACG) in Philadelphia.

Freenome’s poster represents the first readout from the largest international, multicenter study to date analyzing cell-free DNA (cfDNA) in early-stage retrospective CRC patient samples. The study evaluated 1,253 samples in total, including 797 samples from patients with CRC, 82% of whom had early-stage cancer (stages I and II). Efficacy in early-stage patients is necessary to establish the clinical utility of any cancer screening test.

According to Girish Putcha, MD, PhD, Chief Medical Officer at Freenome, “We have shown that it’s possible to take a machine learning–based approach to decode the relationship between a patient’s cell-free DNA profile and his or her cancer status, detecting CRC in our dataset with a top performance of 82% sensitivity at 85% specificity. These are very encouraging data that support the continued development of a CRC screening test that includes cfDNA and machine learning as key components.”

The work is a key aspect of Freenome’s CRC development program, which also includes AI-EMERGE, the first prospective clinical validation study of an AI-genomics blood test, and evaluation of several other blood-based analytes, such as proteins. (Click here to see select data from Freenome’s multi-analyte proof-of-concept study, presented earlier in 2018.)

“Adherence is critical to the success of any screening program. Having a blood-based option would increase the number of people getting screened by helping us reach those uninterested in currently available options,” said Douglas J. Robertson, MD, MPH, Professor at the Geisel School of Medicine at Dartmouth. “These early results are encouraging, and I look forward to subsequent prospective validation studies of the technology.”

Earlier efforts to develop blood-based cancer tests have focused on DNA shed by the tumor itself, so-called circulating tumor DNA (ctDNA). Sometimes referred to as “liquid biopsy,” this tumor-centric approach has shown limited effectiveness as a cancer-detection tool, due in part to the extremely low levels of tumor-derived material circulating in the blood during early-stage disease.

Freenome’s AI-based approach goes beyond the tumor to examine other biological signals in the blood, identifying complex patterns associated with the body’s response to specific tumor types, which may include signals from the immune system.

Further work extending this approach to other cancer types is currently underway.

About Freenome’s AI-PATTERNS Oncology Studies

By using AI to recognize disease-associated patterns among billions of circulating, cell-free biomarkers, Freenome is developing simple, accurate, and noninvasive blood tests for early-cancer screening and treatment selection. Freenome’s AI-PATTERNS studies are a series of rigorous clinical development and validation studies across a diverse range of cancer types. The first in the series, AI-EMERGE, is focused on colorectal cancer.


Lena Cheng, (650) 822-7962


Lena Cheng, (650) 822-7962