Less-heavy smokers with certain genetic characteristics have a risk of developing lung cancer comparable to that of heavy smokers, a recent study published in Carcinogenesis has found. The study’s results bolster the decision made by the U.S. Preventive Services Task Force (USPSTF) to expand the pool of patients who qualify for low-dose CT screening.
“CT screening is costly, so there have to be cutoffs,” said Wei Zheng, M.D., a professor of medicine at Vanderbilt University Medical Center and director of the Vanderbilt Epidemiology Center. “Our study had two aims: to provide additional evidence in support of the new guidelines, and to see if we could use a score based on genetic factors to help identify high-risk populations.”
“It’s useful to combine genetic history with smoking history to identify the population likely to benefit from screening.”
Reducing Disparities, Increasing Lung Cancer Screening
In early March, the USPSTF lowered the level of smoking required for lung cancer screening from 30 to 20 pack-years. The revised guidelines address disparities. Previously, nearly half of Black people diagnosed with lung cancer had not qualified for lung cancer screening, compared to only 16 percent of white lung cancer patients.
Regardless of guidelines, though, most patients who qualify for lung cancer screening do not get tested, although the American Lung Association and the American Thoracic society have been working to increase screening rates.
The two groups have issued a Lung Cancer Screening Implementation Guide, developed with support from other Vanderbilt researchers including Kim Sandler, M.D., a cardiothoracic radiologist and co-director of the Vanderbilt Lung Screening Program.
Support for a Lower Pack-year Cutoff
The new study’s findings confirm and support the guideline expansion, Zheng explained. “We found that current smokers with a 20-to-29 pack-year history had risks similar to those experienced by heavy smokers, the ones who had been recommended for screening under the 2014 USPSTF guidelines,” he said. Previously, the USPSTF had recommended lung cancer screening for people who had a 30 pack-year history, whether they were current or former smokers who had quit within the previous 15 years.
“We also saw that it’s useful to combine genetic history with smoking history to identify the population likely to benefit from screening,” Zheng said, adding that this was a novel discovery.
Risk Score Based on GWAS Studies
Zheng and colleagues previously created a polygenic risk score (PRS) that used lung cancer-related risk variants derived from genome-wide association studies (GWAS). They demonstrated that this score was strongly associated with lung cancer risk, adhering to a dose-response pattern.
For the most recent research, Zheng used UK Biobank genotyping data from a cohort of 308,490 English, Scottish, and Welsh adults, aged 40 to 69 (144,173 men and 164,317 women). The dataset included information on smoking status (current, former, never), pack-years, and age at cessation for ex-smokers. To construct a PRS, the team used 19 variants from 14 genetic loci.
“Current smokers with a 20-to-29 pack-year history had risks similar to those who had been recommended for screening under the 2014 USPSTF guidelines.”
During a median follow-up of 5.8 years, 1,449 members of the cohort developed lung cancer. Heavy smokers, including those who had smoked for 30 or more pack-years and had quit within the previous 15 years, had a hazard ratio (HR) of 19.9, which was close to the HR of 20.7 observed for current smokers with 20 to 29 pack-years.
Combining History and Genetic Risk
The PRS proved to have a statistically significant association with lung cancer risk, adhering to a dose-response pattern in all three groups (current, former and never smoked).
The researchers classified cohort members jointly by both PRS and smoking status. Compared with never smokers with low genetic risk, the greatest risk of lung cancer proved to be experienced by current smokers with 20 or more pack-years and the highest genetic risk; their estimated HR was more than 36.
Future Directions
Zheng would like to perform a larger study of this type. “Although our initial sample size of 300,000 people sounds quite large, when you get down to specific categories like smokers ages 20 to 29 and then stratify by genetic risk, the numbers drop quickly,” he said. To develop a clearer picture of how genetic factors affect lung cancer risk, a bigger patient pool is needed.
In addition, UK Biobank data is overwhelmingly based on people of European descent, Zheng explained, with very few Black people included. “This is another limitation that we would like to address going forward,” Zheng said.
