Methylation analysis improves lung cancer risk prediction
Whether lung cancer is detected at an early stage is crucial for the survival of the patients: While 55 percent of patients with stage I lung cancer survive the first five years after diagnosis, the figure is less than ten percent among those with advanced stage IV tumors.
Heavy smokers are at particularly high risk of developing lung cancer. Numerous studies have already shown that multi-slice computed tomography (CT) scans with low radiation doses are capable of detecting carcinomas in the lungs at early stages, reducing lung cancer mortality by up to 30 percent among heavy smokers at high risk.
"However, such examinations can have harmful side effects, such as psychological stress and unnecessary follow-up examinations in the case of false positive findings, or exposure to radiation," explains Hermann Brenner, an epidemiologist at the German Cancer Research Center (DKFZ) and the National Center for Tumor Diseases (NCT) in Heidelberg. "That's why it would be very desirable to be able to offer CT as an early detection method in an even more targeted way to those people who are at particularly high risk and who could particularly benefit from it."
Hermann Brenner's team has therefore investigated numerous biomarkers associated with lung cancer risks in recent years. Smoking causes strong epigenetic changes to the genome that also affect numerous genes associated with cancer development. Several studies have already provided evidence that methylation of two genes, AHRR as well as F2RL3, is associated with increased lung cancer risk.
This evidence has now been confirmed by first author Megha Bhardwaj using data from participants in the ESTHER* study. This cohort study, led by Brenner and conducted in collaboration with the Saarland Cancer Registry, has been running since 2000. For the current analysis, the research team considered data from162 ESTHER participants (smokers or former smokers) who developed lung cancer during the long-term follow-up and from 721 study participants (also smokers or former smokers) who did not develop lung cancer.
Both AHRR methylation and F2RL3 methylation exceeded the power of the various lung cancer risk models previously used, which were based solely on obtaining smoking history and other risk factors from the participants. Combining the methylation analysis with the conventional risk models further improved the informative value. This was true for all lung cancer types, all age groups, and both former smokers and participants who were still smoking at diagnosis.
It is particularly crucial that a risk model does not exclude anyone from early detection who will nevertheless later develop lung cancer. Also with regard to this criterion, methylation analysis in combination with conventional risk assessment models proved to be helpful. Analysis of methylation markers in combination with the various risk prediction models reduced the number of individuals falsely excluded by up to 68 percent, without increasing the rate of false positives.
"Identifying those smokers who will benefit most from screening is one of the biggest challenges in implementing such a lung cancer screening program," said Hermann Brenner. "Therefore, we are pursuing multiple approaches to enable targeted and effective screening strategies." For example, his team has already been able to develop a risk assessment based on inflammatory markers and on microRNA markers.
"Probably the best significance would be achieved with a combination of different approaches," Megha Bhardwaj explains. For such biomarker signatures to be used in practice, they must not only be informative, but also cost-effective and capable of being determined in a high-throughput format. A number of development steps are still needed before that point can be reached.
Megha Bhardwaj, Ben Schöttker, Bernd Holleczek, Hermann Brenner: Enhanced selection of people for lung cancer screening using AHRR (cg05575921) or F2RL3 (cg03636183) methylation as biological markers of smoking exposure
Cancer Communication 2023, DOI: 10.1002/cac2.12450
* ESTHER: Epidemiological Study on Opportunities for Prevention, Early Detection and Optimized Therapy of Chronic Diseases in the Elderly Population
Here you can find the original report.