In the Journals

Model identifies patients who would benefit from annual lung cancer screening

Incorporating data from low-dose CT lung cancer screening into a prediction model helped identify patients who would benefit from annual lung cancer screening, according to findings recently published in JAMA Network Open.

Researchers used data from the National Lung Screening Trial Lung Screening Study (LSS) to create a model that included results from the initial screening — negative or positive — as a predictor to the PLCOm2012 risk model. They analyzed incident lung cancers that occurred 1 to 4 years after the third screen and validated the model in the National Lung Screening Trial American College of Radiology Imaging Network (ACRIN) study.

According to Martin C. Tammemägi, PhD, of the department of health sciences at Brock University, St. Catharines, Ontario, Canada and colleagues, 298 lung cancers were diagnosed in 22,229 participants at follow-up. When the data was adjusted for PLCOm2012 risks, participants with one positive screen and the last one negative had higher odds for lung cancer vs. those with three negative screens (OR = 1.93; 95% CI, 1.34-2.76). Participants with two positive screens and the last one negative or two negative screens and the last one positive also had higher odds (OR = 2.66; 95% CI, 1.6-4.43) and when two or more screens, including the last one, were positive (OR = 8.97; 95% CI, 5.76-13.97).

In ACRIN validation data, the model that included PLCOm2012 scores and the LSS results (PLCO2012results) showed significantly greater discrimination (area under the curve = 0.761; 95% CI, 0.716-0.799) than when PLCOm2012 screening results were excluded (AUC = 0.687; 95% CI, 0.645-0.728). This same validation data showed good calibration, according to researchers.

“In a cost-effectiveness analysis of lung cancer screening in Ontario, Canada, the primary driver of costs was the computed tomography examinations,” Tammemägi and colleagues wrote. “The use of PLCO2012results may lead to fewer screens, which may improve cost-effectiveness and reduce radiation exposure and other screening harms, such as false-positive findings and overdiagnosis.” – by Janel Miller

Disclosures: Tammemägi reports developing the PLCOm2012 risk prediction model (free to all noncommercial users) and reported that Brock University owns the rights to sublicense use of the PLCOm2012 to commercial users who financially profit from the use of the PLCOm2012 model (parts of those proceeds are to come to Tammemägi; to date, zero financial returns have been made to Tammemägi regarding any use of the PLCOm2012). Please see the study for all other authors’ relevant financial disclosures.

 

 

Incorporating data from low-dose CT lung cancer screening into a prediction model helped identify patients who would benefit from annual lung cancer screening, according to findings recently published in JAMA Network Open.

Researchers used data from the National Lung Screening Trial Lung Screening Study (LSS) to create a model that included results from the initial screening — negative or positive — as a predictor to the PLCOm2012 risk model. They analyzed incident lung cancers that occurred 1 to 4 years after the third screen and validated the model in the National Lung Screening Trial American College of Radiology Imaging Network (ACRIN) study.

According to Martin C. Tammemägi, PhD, of the department of health sciences at Brock University, St. Catharines, Ontario, Canada and colleagues, 298 lung cancers were diagnosed in 22,229 participants at follow-up. When the data was adjusted for PLCOm2012 risks, participants with one positive screen and the last one negative had higher odds for lung cancer vs. those with three negative screens (OR = 1.93; 95% CI, 1.34-2.76). Participants with two positive screens and the last one negative or two negative screens and the last one positive also had higher odds (OR = 2.66; 95% CI, 1.6-4.43) and when two or more screens, including the last one, were positive (OR = 8.97; 95% CI, 5.76-13.97).

In ACRIN validation data, the model that included PLCOm2012 scores and the LSS results (PLCO2012results) showed significantly greater discrimination (area under the curve = 0.761; 95% CI, 0.716-0.799) than when PLCOm2012 screening results were excluded (AUC = 0.687; 95% CI, 0.645-0.728). This same validation data showed good calibration, according to researchers.

“In a cost-effectiveness analysis of lung cancer screening in Ontario, Canada, the primary driver of costs was the computed tomography examinations,” Tammemägi and colleagues wrote. “The use of PLCO2012results may lead to fewer screens, which may improve cost-effectiveness and reduce radiation exposure and other screening harms, such as false-positive findings and overdiagnosis.” – by Janel Miller

Disclosures: Tammemägi reports developing the PLCOm2012 risk prediction model (free to all noncommercial users) and reported that Brock University owns the rights to sublicense use of the PLCOm2012 to commercial users who financially profit from the use of the PLCOm2012 model (parts of those proceeds are to come to Tammemägi; to date, zero financial returns have been made to Tammemägi regarding any use of the PLCOm2012). Please see the study for all other authors’ relevant financial disclosures.