NEW YORK – A group of researchers from institutions including Tufts University, Johns Hopkins University, and Brown University are utilizing a $2.5 million grant from the National Institutes of Health to investigate the use of DNA methylation for lung cancer screening.
Building on previous research using the CLUE II cohort of archived lung cancer patient blood samples that found certain methylation markers that could be used to identify risk of lung cancer, the researchers are now using another dataset to further expand their research.
In their prior research, the team looked at methylation levels in blood leukocytes to determine if they could identify DNA patterns that might be linked to future risk of lung cancer, said Dominique Michaud, a professor at Tufts University School of Medicine and one of the researchers.
In a paper published in Epigenetics in 2022, the researchers said that they measured blood leukocyte DNA methylation levels in 430 patients from the CLUE II cohort and identified 16 single sites and 40 differentially methylated regions that were significantly associated with lung cancer risk and not primarily driven by smoking history. At the time of the sample collection, none of the patients had been diagnosed with lung cancer. The researchers compared samples from people who went on to develop lung cancer with those who didn't to determine the risk-related methylation markers.
After publishing its findings, the team decided to apply them to another dataset: the National Cancer Institute-sponsored National Lung Screening Trial, which includes archived blood collected from smokers before undergoing low-dose CT scans — the current standard of care for lung cancer screening. The goal of this new research is to determine whether some methylation markers can be used to better stratify patients at higher risk for lung cancer and potentially improve recommendations for who should be screened, Michaud said.
The team is asking, "Can we reduce the number of people who are being screened by finding markers that indicate which people are at higher risk?" she said, adding she and her colleagues intend to look at 1,500 samples from the new cohort with the hope that they will have results within the next three to four years.
Because the samples are from smokers, the team is trying to understand the epigenetic changes that occur in this group, and how they affect risk, she said.
The researchers also hope to replicate what they found in the CLUE II cohort in this new cohort and develop a more predictive model to see if their associations hold up, while they look for new methylation markers that they haven't yet discovered in the NSLT cohort, Michaud said.
She noted that the methylation markers the team has already found may not be able to discriminate between inflammation from lung disease and inflammation from something else, so the team intends to find new markers that can distinguish between sources of inflammation.
However, the researchers have additional plans beyond this cohort. Because the NSLT cohort is limited because most participants were White, Michaud and her team would like to look for potential markers in more diverse populations and see if they can replicate their findings in different patient populations, Michaud said.
Right now, the researchers are using an Illumina methylation array to conduct their studies, but Michaud said that if the markers were to be implemented in a clinical setting, a new test would have to be developed to focus only on the specific markers indicating risk.
Any test that is developed will likely not be a lung cancer diagnostic, but rather a tool to reduce the number of low-dose CT scans needed to identify lung cancer patients, she added. The blood test would likely be performed before a CT scan to identify patients that are at higher risk and rule out those that are low risk and don't need a scan.
Michaud noted that CT scans are expensive, hard to access, increase exposure to radiation, and can produce false positives. By using a blood test, such as the one being developed by her and her colleagues, to rule out some patients, those issues could be reduced.
DNA methylation to detect or diagnose cancer has been an area of interest for many researchers and diagnostic firms in recent years. Companies such as Precision Epigenomics, which is developing a methylation-based test for multiple solid tumors, and Grail, which offers the multi-cancer detection Galleri test based on assessing methylation patterns in circulating cell-free DNA, have targeted methylation as a key component of their commercial technologies.
Meantime, researchers from a broad range of institutions have also used methylation for cancer detection, including a team from MD Anderson that recently developed a method to classify subtypes of small cell lung cancer via methylation and machine learning and a group of researchers from the University of Southern California that has created an ovarian cancer test that uses methylation for early detection.
"I'm hopeful about methylation because it just really tells you a lot about exposures that are going on throughout your life course and things that have changed and impact your immune response," Michaud said. "I think it's going to be really useful."