June 29, 2019 02:07 pm Chris Crawford – A research letter published June 4 in Annals of Internal Medicine discussed a study that analyzed how new risk thresholds could identify more current and former smokers who are eligible for lung cancer screening, potentially saving more lives.
In its final recommendation statement from 2013, the U.S. Preventive Services Task Force recommended annual low-dose CT screening for lung cancer in adults ages 55-80 who have a smoking history of at least 30 pack-years and who currently smoke or have quit within the past 15 years; this is a "B" recommendation.
AAFP Commission on Health of the Public and Science member James Stevermer, M.D., of Fulton, Mo., told AAFP News that this earlier task force recommendation (which is now in the process of being updated) identified a patient population in which the USPSTF agreed that the benefits of screening exceeded the risks based on population risk data from before 2001.
"However, population demographics have changed (i.e., fewer people smoke, meaning there are more former smokers), and this risk model may not work as well now," Stevermer said.
Notably, the AAFP's lung cancer screening recommendation at the time differed from that of the USPSTF, saying the evidence was insufficient to recommend for or against lung cancer screening with low-dose CT in this patient population. The Medicare Evidence Development and Coverage Advisory Committee agreed with the Academy's assessment in its recommendation on the subject.
To update its 2013 recommendation, the USPSTF has indicated it may consider other risk models if they can be shown to improve the balance of harms and benefits of screening, Stevermer said.
The current modeling study evaluatied three proposed models, using 2015 National Health Interview Survey data for more current population demographics, according to the research letter.
"Based on these data, selection of one of these models could make screening more efficient; they reported improvement both in terms of number needed to screen (to prevent one lung cancer death), as well as rates of false-positives per prevented lung cancer death," Stevermer said.
Changing the risk threshold in these models, he added, could increase or decrease the size of the population eligible for screening (along with increasing or decreasing the number of cancer deaths prevented).
"However, the downsides to this approach may include the need to use more complex rules to identify which patients' risks are high enough to potentially benefit from screening," Stevermer said.
If the eligible population is increased, the potential for increased benefit comes with potential for increased harm, including expensive/invasive follow-up procedures and imaging without improved outcomes for patients. Increased radiation exposure from repeat screening, for example, is part of this concern.
The three proposed risk thresholds include the National Comprehensive Cancer Network guidelines (version 1.2019) recommendation to screen ever-smokers with a six-year lung cancer risk of 1.3% or higher according to the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial Model 2012.
The research letter said this threshold of 1.3% or higher selects the same proportion of ever-smokers as does application of the USPSTF criteria to the PLCO trial.
The same authors found that applying a six-year lung cancer risk of 1.51% or higher according to the PLCOM2012 was more efficient than the USPSTF criteria because the method limits screening to at least the 65th percentile of risk in PLCO ever-smokers and would screen fewer PLCO participants than would the USPSTF criteria.
The third proposed risk threshold found that five-year risk for lung cancer death of at least 1.2% according to the Lung Cancer Death Risk Assessment Tool would allow screening of the same number of ever-smokers as applying the USPSTF guideline to the U.S.-representative 2010-2012 National Health Interview Survey, yet potentially save more lives.
Using the USPSTF guidelines, in 2015, 8 million U.S. ever-smokers were eligible for screening. An additional 4.6 million, 3.3 million and 1 million people were eligible using the NCCN, PLCOM2012 and LCDRAT risk thresholds, respectively.
Stevermer said it's possible, but certainly not guaranteed, that EHRs could help with the task of screening this patient population.
"Another challenge is that it's unlikely this will ever be tested in prospective fashion, so these global decisions will likely be based primarily on modeling data," he said.
Recalibrating the PLCOM2012 and LCDRAT models to screen the same 8 million people as the USPSTF guideline improved screening efficiency by 13% to 20% and reduced false-positive results per death prevented by 10% to 15%, Stevermer noted. These adjusted models also reduced cancer deaths by 14% to 24%.
"I'm not surprised that models based on newer data might improve efficiency and effectiveness, but I wouldn't have anticipated as large an improvement," he said. "Again, these data are based on modeling approaches, so there's no guarantee that real-world results will be as good."
Stevermer emphasized that the AAFP will review the new USPSTF guideline for endorsement whenever it is finalized.
"My hope is that if some of these newer risk thresholds are included in the upcoming USPSTF guideline on lung cancer screening, they are easy to implement in a busy practice."
More From AAFP
Familydoctor.org: Lung Cancer