Imagine a routine blood test that could detect cancer in its earliest stages—before it overwhelms the body’s ability to control it.
New research published in Science Translational Medicine suggests that reality may not be far off.
The research uses machine learning—specifically an algorithm called Alu Profile Learning Using Sequencing, or A-Plus, to detect Alu elements in the blood—a type of repetitive DNA.
People with “solid” cancers (those in organs such as the breast or prostate as opposed to those of the blood, such as leukemia, which are considered “liquid”) tend to have fewer Alu elements in their blood than people without cancer. Researchers used these findings to improve a test that finds cancer early. According to a news release, researchers reproduced their results and validated their findings using a sample size ten times larger than those usually used for these types of studies.
Blood Testing and Artificial Intelligence
Using blood tests to detect cancer is called liquid biopsy—which uses body fluids (usually blood) to detect cancer instead of having to do a standard biopsy which removes tissue from a tumor to look for cancer cells. Liquid biopsies are more convenient for patients as they are less invasive, less painful, and have a lower risk of complications.
Christopher Douville is an assistant professor of oncology at Johns Hopkins Medicine and the study’s lead author.
“Blood testing holds great promise for the earlier detection of cancers before people exhibit any symptoms,” Mr. Douville said in a news release. “However, analyzing results with machine learning has not necessarily translated into long-term success for patients when minor fluctuations produce widely different predictions in these complex models. To have a long-term impact on patient care, physicians and patients must have confidence that models consistently and reproducibly classify cancer status. In our manuscript, we evaluated 1,686 individuals multiple times to assess whether our machine learning model consistently delivers the same answer.”
When asked how the addition of machine learning, or AI (artificial intelligence) is beneficial in detecting Alu elements in the blood, Mr. Douville told The Epoch Times via email, “AI can integrate hundreds of thousands of predictive features to detect complex patterns. The underlying patterns can often be missed using conventional approaches.”
Another benefit of using AI is reducing the risk of false positive results, Mr. Douville explained, saying, “Machine learning can identify complex patterns often missed using more conventional approaches. Given how difficult it is to identify trace amounts of cancer in the blood, machine learning offers a way to increase sensitivity.”
Researchers collected 3,105 blood samples from people with solid cancers and 2,073 from people without cancer—but used a total of 7,615 samples so that duplicates could be used to test the model’s function and precision. The study screened for 11 types of cancer, including breast, colon and rectum, esophagus, lung, liver, pancreas, ovary, and stomach cancer.
Notably, most of the samples from those with cancer had the disease in its early stages and had either few or no metastases at the time they were diagnosed.
After testing, researchers reached 98.9 percent specificity. “This is crucial when screening asymptomatic patients, so people aren’t told incorrectly that they have cancer,” Mr. Douville noted in the news release.
A press release on the City of Hope website explains how this DNA gets into the bloodstream:
“When a cell dies, it breaks down and some of the DNA material of the cell leeches into the bloodstream. Cancer signals can be found in this cell-free DNA (cfDNA). The cfDNA of normal cells breaks down at a typical size, but cancer cfDNA fragments break down at altered spots. This alteration is hypothesized to be more present in repetitive regions of the genome.”
Fragmentomics
This allowed researchers at City of Hope and Johns Hopkins University to come up with a new way to “detect the difference in fragmentation patterns in repetitive regions of cancer and normal cfDNA [cell-free DNA].”
The technique uses something called fragmentomics, which “looks at the pattern of the amount and sizes of DNA fragments in the blood,” according to the National Cancer Institute. “Fragmentomics requires about eight times less blood than required by whole genome sequencing,” Cristian Tomasetti, a corresponding author of the study and director of City of Hope’s Center for Cancer Prevention and Early Detection explained in the press release.
Kamel Lahouel is the study’s co-first author and an assistant professor in TGen’s Integrated Cancer Genomics Division. “Our technique is more practical for clinical applications as it requires smaller quantities of genomic material from a blood sample. Continued success in this area and clinical validation opens the door for the introduction of routine tests to detect cancer in its earliest stages.”
Could searching for Alu’s potentially lead to detecting all types of cancer in the future? Mr. Douville says, “In our study, we only evaluated 11 different types of solid cancers but have reason to believe this could generalize to additional cancers.”
Using Alu elements represents a new way to detect cancer earlier and potentially improve patient outcomes.
“Alu elements are frequently overlooked as a possible cancer biomarker due to the technical challenges associated with their analysis. Our study shows that Alu element representations are altered in the cfDNA of patients with many different cancer types and can be used to enhance methods designed for the earlier detection of cancer.” Mr. Douville said.
Collaborative Effort
The study was a massive collaborative effort, with 38 contributing authors, 28 of whom were from Johns Hopkins University School of Medicine and City of Hope, a cancer research and treatment organization.
Additional co-authors were from the Department of Medicine and Department of Epidemiology at the University of Pittsburgh, the Department of Surgery at NYU Langone Health, Pham Ngoc Thach University of Medicine and Saigon Precision Medicine Research Center in Vietnam, and the Walter and Eliza Hall Institute of Medical Research, the University of Melbourne, the University of Technology Sydney, and the University of New South Wales in Australia according to the news release.
As for when the test might be available to practitioners and the public, a representative from the City of Hope told The Epoch Times in an email:
“This summer, City of Hope is poised to test a novel screening method developed by City of Hope and TGen for the early detection of all cancers. Phase A aims to enroll 30,000 non-cancer volunteers 65-75 years of age, of which 15,000 will be randomized to the control group, and 15,000 will be in the screening group.”