Blood test detects cancer DNA 3 years before diagnosis in early study

Researchers examining blood samples collected decades ago have found traces of cancer DNA floating in the plasma of patients years before they received a diagnosis. In a study published in May in Cancer Discovery, scientists detected free-floating genetic material from dying precancerous or cancerous cells in blood drawn three years before tumors were identified through conventional means. The finding suggests that a blood test might one day catch cancer in its earliest stages, when treatment is most likely to succeed.

Dr. Yuxuan Wang and her team at Johns Hopkins University worked backward through time, studying plasma that had been stored since the 1980s. They focused on 26 people who developed cancer within six months of donating blood, comparing them to 26 control subjects who remained cancer-free for at least 17 years after their donation. When the researchers screened the more recent samples—those taken just before diagnosis—they found between one and three common cancer-related mutations in seven of the plasma samples. Six of those patients had also given blood between 3.1 and 3.5 years earlier. When Wang's team examined those older samples for the same mutations, they found the warning signs in two of them, confirming that detectable cancer DNA was present years before any clinical diagnosis.

The researchers then took a more granular approach, sequencing the plasma DNA to hunt for mutations unique to each patient. Using white blood cell genomes as a reference, they identified between four and 90 distinct mutations in the plasma samples from three patients. Across the five early samples they examined in detail, they found hints of cancer in three of them. The patients in the study had been diagnosed with breast, colon, liver, lung, pancreas, and rectal cancer—a range that raises questions about whether the test works equally well across all tumor types. Wang noted that some organs shed tumor DNA more readily than others, and that the blood-brain barrier might prevent brain cancer DNA from entering the bloodstream.

The limitations are substantial. The test failed to detect any cancer DNA in 18 of the 26 participants who later developed tumors within months of providing their samples. For a clinical screening tool, that represents a significant gap. Wang suggested that using larger volumes of plasma might improve detection rates, but that remains untested. There is also the risk of false positives—results that alarm patients and potentially lead to unnecessary biopsies or treatments. Catherine Alix-Panabières, a cancer researcher at the University of Montpellier in France who was not involved in the work, emphasized that implementing such tests would require clear ethical guidelines for handling incidental findings.

The financial barrier is another hurdle. Sequencing a patient's DNA to identify personalized mutations can cost several hundred to thousands of dollars. Even if the test proves reliable, Wang said, it may not be feasible to offer it to everyone seeking screening. Instead, it might be reserved for people with family histories of cancer or other known risk factors. The study itself was small—just 52 plasma samples from Black and white men and women aged 45 to 64 from four U.S. states. Larger investigations involving hundreds or thousands of participants from genetically diverse backgrounds would be needed to validate the approach before doctors could use it with confidence.

Alix-Panabières estimated that widespread clinical adoption is realistically five to ten years away. The promise is real: earlier detection of cancer generally correlates with better outcomes, since treatment can begin before tumors grow and spread. But the path from a promising early study to a tool doctors can reliably use in the clinic remains long. The research opens a door, but many questions about accuracy, cost, and equity still stand in the way.