Risk-based model better for selecting ever-smokers for CT lung cancer screening

By Will Boggs MD

A risk-based model for selecting ever-smokers for computed tomography (CT) lung cancer screening may prevent more cancer deaths than screening according to U.S. Preventive Services Task Force (USPSTF) criteria, researchers say.

"Selecting smokers for CT lung cancer screening by using a validated lung cancer risk calculation might lead to more effective and efficient screening than current subgroup-based selection guidelines," Dr. Hormuzd A. Katki from the National Cancer Institute in Bethesda, Maryland, told Reuters Health by email.

USPSTF recommends annual CT screening for a risk factor-based subgroup of smokers - current and former smokers aged 55 to 80 years and 55 to 77 years, respectively, with at least 30 pack-years of smoking and, for former smokers, no more than 15 years since quitting.

Dr. Katki and colleagues sought to develop and validate empirical lung cancer incidence and death risk models generalizable to U.S. smokers using data from the Prostate, Long, Colorectal, and Ovarian (PLCO) cancer screening trials and the National Lung Screening Trial (NLST).

Under USPSTF recommendations, the models estimated that 9 million U.S. ever-smokers would qualify for lung cancer screening, and this would avert nearly 46,500 lung cancer deaths over five years.

Risk-based selection screening of the same number of ever-smokers at the highest five-year lung cancer risk, however, would avert 20% more deaths (55,717) and would reduce the estimated number needed to screen (NNS) to prevent one death from 194 to 162, the researchers report in JAMA, online May 15.

"Compared with USPSTF eligibility, risk-based screening strategies preferentially include more current smokers overall, more low-intensity long-term current smokers, and more high-intensity former smokers who have quit for more than 15 years," the researchers note.

"Implementing risk-based precision screening is a big challenge," Dr. Katki said. "Although researchers have developed many risk tools, they require further validation, especially to ensure they are accurate for U.S.-population risks."

"Much research remains to be conducted to develop and evaluate risk-based decision aids to communicate risk information to doctors and patients," he said. "The goal is to develop a proven shared decision-making process that ensures that doctors and patients make good decisions about screening."

"The majority of lung cancer deaths may not be screen-preventable," added coauthor Dr. Anil Chaturvedi, also at NCI. "The best way for smokers to reduce their risk of death from lung cancer and other smoking-associated diseases is to quit smoking as soon as possible. Resources to quit smoking are available at NCI's smokefree.gov."

Dr. Michael K. Gould from Kaiser Permanente Southern California in Pasadena, who wrote an accompanying editorial, told Reuters Health by email, "I was intrigued to see that the enhanced risk-based approach favored screening in populations that tend to have more limited access to health care - current smokers, African Americans and individuals with lower educational attainment. Providing these groups with access to screening is going to be challenging."

"Lung cancer screening is a personal decision and we should avoid using a one-size-fits-all approach," he said. "While the net benefit of screening is least uncertain for individuals who meet NLST/USPSTF/Medicare criteria, the baseline risk of lung cancer death varies widely within this group, including some for whom the risk is fairly low. In contrast, there are many high-risk individuals who do not belong to this group."

"Virtually every decision in medical care involves weighing tradeoffs between potential benefits and harms," Dr. Gould said. "Having better information about the size of the risks should help patients and clinicians to make better decisions."

"Lung cancer screening is the right choice for many high-risk smokers and former smokers but not all of them," he concluded. "Clinicians need to help individuals make the best decision possible."

Dr. Eoin Gray from the University of Sheffield in the UK, who recently reviewed risk-prediction models for lung cancer, told Reuters Health by email, "Prediction models can identify a better target population for screening than targeting ever-smokers aged 50-80. Many models have been shown to improve screening recommendations. This should be preferred in subsequent screening trials."

"However, first the leading model and optimal risk threshold needs to be identified and validated in multiple external populations to assess if the promising results can be replicated," said Dr. Gray, who was not involved in the study. "If this can be observed, then physicians should consider screening guidelines that capture a high proportion of lung cancer cases but reduce the false positives rates as screening too many controls will create unnecessary costs and undue worry."

The National Cancer Institute supported this research. One coauthor reported receiving support from Medial EasySign.

SOURCE: http://bit.ly/1W0KR0l and http://bit.ly/24XGbKe

JAMA 2016.

(c) Copyright Thomson Reuters 2016. Click For Restrictions - http://about.reuters.com/fulllegal.asp