A simple blood test could one day help doctors detect lung cancer without using expensive DNA sequencing.
Researchers at Tel Aviv University have developed a method that looks for chemical patterns on tiny pieces of DNA circulating in the blood. In an early study, the test identified many people with lung cancer while correctly ruling out most healthy participants.
Lung cancer is one of the world’s deadliest cancers, largely because it is often discovered after it has already grown or spread. Early-stage lung cancer may cause few or no obvious symptoms. Finding tumors earlier can give patients more treatment options and improve the chance of successful treatment.
Doctors currently use low-dose CT scans to screen some people who have a high risk of lung cancer, particularly those with a substantial history of smoking. CT scans can find small spots in the lungs, but many suspicious findings turn out not to be cancer. This can lead to repeated scans and, in some cases, invasive tests such as biopsies.
Scientists have therefore been trying to develop reliable blood tests for cancer. Tumors release small fragments of their DNA into the bloodstream as cancer cells die and break apart. These fragments are mixed with much larger amounts of DNA released from healthy cells.
Many existing approaches try to identify cancer-related genetic changes by sequencing DNA. Sequencing can provide detailed information, but it can also be expensive and requires specialized equipment and computer analysis. The Tel Aviv University team took a different approach.
Instead of reading the sequence of DNA letters, the new test examines chemical marks attached to DNA. These marks can change when a cell becomes cancerous, creating patterns that may reveal where the DNA came from. The researchers describe this pattern as a chemical fingerprint of the tumor.
The study was led by Professor Yuval Ebenstein of Tel Aviv University in collaboration with JaxBio Technologies, Bnai Zion Medical Center and Sheba Medical Center. The research was published in npj Precision Oncology. The team tested the approach in 103 people, including 51 patients with lung cancer and 52 healthy participants.
After cell-free DNA was collected from blood, the researchers attached light-emitting markers to it. The DNA was then placed on a specially developed chip and examined with an optical scanner. Patterns of light allowed the scientists to measure chemical differences across selected parts of the genome without sequencing the DNA.
The researchers first used some samples to teach their computer model which patterns were associated with cancer. They developed a signature based on 170 regions of the genome. They then tested the signature on a separate group of samples without knowing in advance which belonged to cancer patients.
For patients with stage 2 to stage 4 lung cancer, the test had a sensitivity of 93.1%. Sensitivity describes how well a test identifies people who actually have the disease. In practical terms, a higher sensitivity means fewer cancers are missed.
The test also had a specificity of 90.3%. Specificity measures how well a test correctly identifies people who do not have the disease. This is particularly important in cancer screening because false alarms can lead to anxiety, additional scans and unnecessary medical procedures.
The chemical patterns also provided information about the type of lung cancer. Researchers were able to distinguish between adenocarcinoma and squamous cell carcinoma, two major forms of the disease. These cancers arise from different types of cells and can require different treatment decisions.
The team also explored whether the blood test could help doctors follow patients during treatment. In people whose tumors responded to therapy, the DNA pattern moved closer to the pattern seen in healthy people. Patients whose disease did not respond showed little comparable change.
That treatment-monitoring result is preliminary and needs to be tested in much larger groups. If confirmed, however, a blood test could potentially give doctors another way to see whether treatment is working between imaging appointments. It might also help researchers follow changes in cancer over time.
Another possible advantage is cost and speed. According to the researchers, the current test can be completed in about two to three days and costs roughly $60 per sample. Because it does not depend on DNA sequencing, the technology may eventually be easier to introduce into standard clinical laboratories.
The results are promising, but the study remains small. Only 103 people participated, and the reported performance for stage 2–4 disease does not by itself establish how well the test will detect the earliest stage of lung cancer, which is where a screening test could have its greatest value. Larger studies must also include people with other lung diseases that could potentially resemble cancer.
Another important issue is how the test would perform in real screening populations, where most people do not have cancer. Even a test with good sensitivity and specificity can produce substantial numbers of false-positive results when the disease being tested for is uncommon. For this reason, the method is more likely to complement CT imaging than replace it if future trials are successful.
Overall, the study offers an inventive way to search for cancer signals in blood without reading the entire DNA sequence. Its combination of speed, relatively low cost and encouraging accuracy makes it worth further investigation. The next challenge is to prove that the test works reliably in much larger and more diverse groups, especially people with very early lung cancer.
Source: Tel Aviv University
