Risk-Based CT Lung Cancer Screening Could Avoid More Deaths

Using a risk-based model for computed tomography (CT) lung cancer screening could prevent more deaths and lower the number needed to screen (NNS) in order to prevent one death compared with the US Preventive Services Task Force (USPSTF) screening recommendations, according to results of a new study.

The USPSTF screening guidelines are based on results of the National Lung Screening Trial (NLST); the guidelines recommend that any current and former smokers aged 55 to 80 years (55 to 77 years for former smokers), with no more than 15 years since quitting among former smokers, should receive annual CT screening for lung cancer.

“In the NLST, 88% of CT-prevented lung cancer deaths occurred in the 60% of participants at highest risk, whereas the 20% of participants at lowest risk accounted for only 1% of CT-prevented lung cancer deaths,” wrote study authors led by Hormuzd A. Katki, PhD, of the National Cancer Institute in Bethesda, Maryland.

For the new study, the researchers used data from the NLST; the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO); and the National Health Interview Survey (NHIS) to develop and validate lung cancer incidence and death models. Instead of just using age and smoking history, the risk models incorporated a variety of individual variables, including body mass index, education, more detailed smoking status, and others. These models were then applied to a cohort of ever-smokers aged 50 to 80 years to compare screening outcomes using the new risk-based models and USPSTF recommendations. The results were published in JAMA.

Under the USPSTF guidelines, a total of 9.0 million US individuals would be eligible for screening, and the CT screening program would prevent an estimated 46,488 lung cancer deaths over 5 years. Using the risk-based models, screening the 9.0 million people at highest risk for lung cancer death would prevent 55,717 deaths over that 5-year period. This represented a 20% relative increase in CT-preventable deaths, and an 11% absolute increase (P < .001).

The risk-based approach also had a greater screening effectiveness. With the USPSTF approach, the NNS to prevent one death was 194; with the risk-based model, the NNS was 162 (P < .001). Also, with USPSTF guidelines, there were 133 false-positive CT exams per prevented death; with the risk-based models, there were 116 false positives (P < .001).

In an accompanying editorial, Michael K. Gould, MD, of Kaiser Permanente Southern California in Pasadena, wrote, “The findings are provocative and support the notion that an enhanced risk-based approach to screening is potentially more effective and more efficient than performing a risk assessment based on age and smoking history.” He noted that implementing risk-based screening across a variety of types of practices would present significant challenges.

“In clinical practice, the decision to screen is very personal and should be individualized for each patient,” Gould wrote. “The challenge for clinicians is to make sure that individual patients receive the information they need to make the best decision possible about whether screening is the right choice for them.”