Personal risk-based screening for lung cancer may be more cost-effective than recommended cutoffs
While risk-based screening strategies were robustly more cost-effective than the 2021 U.S. Preventive Services Task Force recommendation in a modeling study, whether the results will be replicable in complex real-world clinical practice remains uncertain, an editorial noted.
Personal risk-based screening for lung cancer may be more cost-effective than screening based on categorical age and smoking history cutoffs, a modeling study found.
Researchers conducted the comparative modeling analysis using four validated microsimulation models of the Cancer Intervention and Surveillance Modeling Network Lung Working Group. The microsimulation models, which informed the 2021 U.S. Preventive Services Task Force (USPSTF) recommendation on lung cancer screening, had suggested that using validated risk prediction models to estimate personal lung cancer risk and select candidates for screening would be more effective at preventing lung cancer deaths and result in greater life-year gains than screening approaches based on categorical age and smoking cutoffs. In both its 2013 and updated 2021 recommendation, the USPSTF used categorical age and smoking cutoffs for lung cancer screening, while the updated recommendation lowered the starting age from 55 to 50 years and the minimum cumulative smoking exposure from 30 to 20 pack-years.
In the current study, researchers compared the cost-effectiveness of risk model-based strategies versus the 2021 USPSTF recommendation. They simulated lung cancer-related events for 1 million men and women using smoking patterns of the 1960 U.S. birth cohort that is representative of the population targeted by screening. Simulated individuals were followed from age 45 years until age 90 years or death, whichever occurred first. Results were published Feb. 7 by Annals of Internal Medicine.
Risk model-based screening strategies that start screening at age 50 years, irrespective of whether the risk assessment was based on lung cancer incidence or mortality, were robustly more cost-effective than the USPSTF recommendation under varying modeling assumptions. The strategy with a 1.2% six-year risk threshold had the highest health benefit (incremental cost-effectiveness ratio [ICER], $94,659 per quality-adjusted life-year [QALY] gained), yielding a higher reduction in lung cancer mortality than the USPSTF recommendation (12.4% vs. 11.8%, respectively) while maintaining a similar level of screening coverage (21.7% vs. 22.6% of the general population ever screened, respectively).
Only risk model-based strategies with a six-year risk threshold of 2.0% or greater (vs. 1.2% or greater in the base case) remained cost-effective with an ICER less than $100,00 per QALY. The strategy with a 2.0% six-year risk threshold was cost-effective and yielded the highest health benefit with a mean ICER of $84,113. Compared with the 2021 USPSTF strategy, the strategy was estimated to screen fewer people (15.8% vs. 22.6%, respectively), requiring about half the screening examinations (206,000 vs. 420,000 low-dose CT scans per 100,000 people) but yielding lower lung cancer mortality reduction (9.8% vs. 11.8%).
Among other limitations, the study used simplified versions of the risk prediction models to assess lung cancer risk, the authors noted. They added that the models did not incorporate potential issues around availability of resources to satisfy the expected increase in the number of low-dose CT examinations or consider the effect of contemporary treatment modalities.
The biggest crisis facing lung cancer screening implementation is the slow uptake in practice, with only 10% to 20% of eligible Americans screened to date, an accompanying editorial noted. Integrating risk models into electronic health records would require a major investment of time and resources, further slowing uptake and posing negative implications for equity, since not all health systems would prioritize conversion to a prediction model-based approach in the setting of resource constraints, the editorialists added.
“Ultimately, we cannot assume that the results of simulation models will be replicable in complex real-world clinical practice,” they wrote. “Going forward, it will be critical to overcome barriers to widespread [lung cancer screening] implementation while balancing tradeoffs in effectiveness, efficiency, and equity.”