Researchers Develop Cost-Effective Early-Cancer Detection Method

Detecting cancer earlier, especially before it metastasizes, is the key to fighting it successfully. Recently, cell-free DNA (cfDNA) circulating in the bloodstream has drawn attention as a potential early cancer detection tool.

Researchers from the University of California at Los Angeles reported successful results from an experimental cancer-detection system. Nature Communications published the article on September 29. 

Related Article: ‘Junk DNA’ May Provide New Treatment for Neurological Disorders

The Pros and Cons of Using cfDNA

According to previous studies, cfDNA methylation is a promising biomarker capable of detecting cancer and locating its tissue of origin. However, cfDNA-based cancer detection faces some obstacles, including low tumor concentrations in DNA blood fragments of early-stage cancer patients, the genetic diversity of cancer, and small sample sizes available to investigate the diversity of diseases.

Profiling cfDNA methylome can address this challenge, as it retains the genome-wide epigenetic profiles of cancer abnormalities. Thus, the classification models can learn and exploit significant features newly as training cohorts grow, as well as expand their scope to more cancer types. However, the conventional way of profiling the cfDNA methylome is cost-prohibitive for clinical use.

Here, researchers introduce a cost-effective, integrated experimental and computational system named cell-free DNA Methylome Sequencing (cfMethyl-Seq).

Powerful Detection System with Increasing Sample Sizes

Researchers used the novel approach to test if it could accurately detect four commonly diagnosed cancers, colon, liver, lung, and stomach cancer at early stages, and the specific location of tumors.

The study collected blood samples from 408 study participants and applied them to a methylome-based blood test. Of those, 217 were cancer patients, and 191 were cancer-free control subjects. Cross-batch validations, age-matched validations, and independent validations were performed to prevent bias in the study.      

Their model was 80.7% accurate in detecting cancers across all stages and about 74.5% in detecting early-stage cancer, with just under 98% specificity, meaning there was only one false positive case.  For tissue-of-origin accuracy, the model correctly identified tumor location with an average accuracy of 89.1% percent for all cancer stages and about 85% percent in early-stage patients.  

The cfDNA methylome approach allows the inclusion of new markers and the better weighting of existing markers as training cohorts grow. With the promising results, the team is pursuing funding for large clinical trials to validate the technology and make it available to benefit patients.

Author

Nai Ye Yeat