Jury still out on chest x-ray CAD

Chest x-rays coupled with computer-aided detection (CAD) technology may have a higher identification rate for actionable lung nodules in a lung cancer screening population than unaided x-rays, though the ultimate role of CAD chest radiography remains unclear, according to a study published March 20 in PLoS ONE.

Lung cancer is an ideal target for a screening program due to the fact that it’s detectable in a preclinical phase and treatment is more effective when the disease is found earlier, explained Peter Mazzone, MD, of the Cleveland Clinic, and colleagues. However, chest x-ray screening without CAD has failed to show a reduction in mortality, while CT screening, which is more sensitive, comes at a higher cost and also results in a high number of false positives. Searching for the best of both worlds, the authors turned to chest x-ray CAD, but their findings were limited.

“Chest x-ray CAD systems may improve upon the sensitivity of standard chest x-rays for the detection of early stage lung cancer, yet are unlikely to identify the very small benign lung nodules found on chest CT imaging,” wrote the authors.

Mazzone and colleagues began a placebo-controlled trial featuring over 1,400 subjects, age 40-75 years, with 10 or more pack-years of smoking and/or additional risk factors for developing lung cancer. A total of 710 were randomized to undergo chest x-ray CAD, while the control group went through the process of being imaged, but were not actually imaged. Radiologists first reviewed images without the assistance of CAD.

Of those who received a chest x-ray CAD, 29 were found to have an actionable lung nodule on prevalence screening. Seven of the nodules were first discovered using CAD assistance. Fifteen subjects had a chest CT performed for additional evaluation and a lung nodule was confirmed in four, with two representing lung cancer. Both cancers were identified by the CAD chest radiograph, but also were seen by the unaided radiologist.

“The number of lung cancers identified was too small to make any judgments about the intended primary endpoint,” wrote Mazzone. “[These data suggest] that advances in the CAD technology, beyond the version used, are required before an impact of the technology could be expected, and that these advances could be of significant benefit to chest x-ray interpretation.”

The authors noted the main limitation of the study was a slower than expected overall recruitment, and with the announcement of the National Lung Screening Trial (NLST) results, which confirmed a reduction in mortality from CT screening, a decision was made to “terminate and evolve” the study.

“It was apparent that recruiting the number of subjects we required to adequately assess our primary outcome was difficult at baseline, and that it would become much more difficult to enroll subjects to a placebo controlled trial of CAD chest x-ray screening given the positive findings of the NLST. It was also apparent that all other screening modalities would need to be compared to the benefit shown by chest CT screening, and that this would not be possible with the resources available to complete our trial,” wrote the authors.

The new study protocol will directly compare chest x-ray CAD to reduced dose chest CT, according to Mazzone and colleagues.