Skip to Content

Translational Epidemiology: Targeting Lung Cancer Screening

by Shelia Hoar Zahm, Sc.D.

Tenure-track investigators Hormuzd A. Katki, Ph.D., and Anil K. Chaturvedi, Ph.D., share a passion for applying epidemiologic discoveries to improve public health. Using their expertise in risk-based screening, the investigators demonstrated that the benefits and harms of low-dose computed tomography (LDCT) screening strongly depended on an individual’s pre-screening risk of lung cancer death, as detailed in their recent publication* in the New England Journal of Medicine. This finding is a proof-of-principle that targeting lung cancer screening efforts based on an individual’s risk of lung cancer death, rather than on coarse age/smoking criteria, could improve the efficiency of LDCT screening programs, resulting in more cases identified and fewer false positives.

The investigators’ interest in this issue began in 2011, when NCI reported the results of the National Lung Screening Trial (NLST). The results showed that screening with LDCT resulted in 20 percent lower lung cancer mortality compared to screening with chest radiography. This landmark finding led to a recommendation by the U.S. Preventive Services Task Force (USPSTF) for LDCT screening of persons aged 55 to 74 years who meet the NLST entry criteria for smoking: having a minimum of 30 pack-years smoking history and, if former smokers, had quit no more than 15 years before screening. Other expert bodies recommended LDCT screening programs using similar combinations of criteria.

However, LDCT screening of the millions of eligible U.S. smokers would entail a dramatic increase in use of LDCT resources, and many more smokers would endure further scans, invasive testing, and/or surgery as a result of false-positive findings. Drs. Katki and Chaturvedi recognized an opportunity to show that risk-based screening could be used to refine the selection criteria and thereby improve the efficiency of LDCT lung screening.

In collaboration with Dr. Katki’s postdoctoral fellow Stephanie Kovalchik, Ph.D. (now at the RAND Corporation in Santa Monica, California), the investigators used epidemiologic data to identify the persons at highest risk of lung cancer death among the already high-risk trial participants. The researchers categorized the NLST participants into risk quintiles using a lung cancer death risk calculator that they developed and validated with data from the NCI Prostate, Lung, Colorectal and Ovarian Screening Trial. The risk model included age, body mass index, family history of lung cancer, pack-years of smoking, years since smoking cessation, and emphysema diagnosis.

The authors reported that in the NLST, 88 percent of the prevented lung cancer deaths occurred in the 60 percent of participants in the top three quintiles of risk. These participants accounted for 64 percent of the false-positive screens. In contrast, only 1 percent of the prevented lung cancer deaths occurred in the 20 percent of participants in the lowest risk quintile.

“Screening only those with the best benefit-to-harm profile might result in preventing nearly all of the preventable cancer deaths with considerable cost savings, as well as reductions in complications from clinical work-ups of smokers whose findings turn out to be negative,” Dr. Katki said.

The new USPSTF lung screening recommendations cited this paper, as did the accompanying editorial by Dr. Peter Bach of the Memorial Sloan Kettering Cancer Center, who cited data from it to opine that screening guidelines should focus on screening only those people at sufficiently high risk of lung cancer death.

Work is ongoing to validate the risk model in additional study populations, including the NIH-AARP Diet and Health Study, the American Cancer Society Cancer Prevention Study II, and the Centers for Disease Control and Prevention National Health Interview Study. Once validated, the model will be made publicly available on the NCI web site. Drs. Chaturvedi and Katki are also evaluating whether biomarkers associated with lung cancer risk provide additional risk stratification.

“In earlier work, we identified circulating immune/inflammation markers that predict risk of lung cancer,” Dr. Chaturvedi explained. “We plan to evaluate if incorporating these biomarkers into eligibility criteria will aid in identifying smokers most likely to benefit and less likely to be harmed from LDCT screening.”

Ideally, lung cancer screening guidelines would reflect the principle of “equal management of people at equal risk of disease,” akin to that used in cervical cancer screening guidelines (discussed in the July 2013 issue of Linkage). Doing so would ensure simplified and consistent management of people with equal risk of lung cancer death, regardless of the specific combinations of factors leading to that risk.

“Many of the organizations who issue cancer screening guidelines have encouraged establishment of screening registries, which are needed to provide the data we would use to develop risk-based clinical management guidelines in the future,” Dr. Katki noted.

As exciting as the prospect of lung cancer screening is, however, “the most important thing for smokers to do to lower their risk of lung cancer, and all other tobacco-related diseases, is to quit smoking,” Drs. Chaturvedi and Katki stated.

* Kovalchik SA, Tammemagi M, Berg CD, Caporaso NE, Riley TL, Korch M, Silvestri GA, Chaturvedi AK, Katki HA. Targeting of low-dose CT screening according to the risk of lung-cancer death. N Engl J Med 2013 Jul 18;369:245-254.

Back to the Spring 2014 issue of Linkage