CT screening overdiagnoses lung cancer
CT screening overdiagnoses lung cancer
9/12/2013
About 18% of lung cancers identified by low-dose Computed Tomography (CT) screening were slow-growing tumors that would not have affected patients during their lifetime. This is based on an analysis of the US National Lung Screening Trial (NLST) showed.
That trial showed a mortality advantage to screening, but for every one lung cancer death prevented per 320 patients with screening in the trial, 1.38 cases of overdiagnosis would be expected, Edward F. Patz Jr., MD, of Duke University Medical Center, and colleagues found.
"These overdiagnosis cases represent an important potential harm of screening because they incur additional cost, anxiety, and morbidity associated with cancer treatment," they wrote online in JAMA Internal Medicine.
Patz and colleagues recommended that physicians include the risk of overdiagnosis in describing the risks of low-dose CT screening for lung cancer to patients.
While the findings may help shape patient expectations, they wouldn't likely shift the risk-benefit ratio much for individual patients, Patz suggested in an interview with MedPage Today.
Nor is the recent U.S. Preventive Services Task Force recommendation to screen high-risk patients annually likely to be affected, he argued.
"I don't think this will shift recommendations at all," he said. "It's just part of this entire puzzle we're trying to piece together, how we can best offer a mass screening program as public policy."
The American College of Radiology agreed, in a statement calling the overdiagnosis rate "modest" and in line with the projected rate with other types of cancer screening.
"Lung cancer screening using low-dose CT is the only test ever shown to reduce mortality in high-risk smokers, the leading cause of cancer death in the U.S. It does so cost effectively compared to other screening tests," the statement said. "Overdiagnosis is an expected part of any screening program and does not alter these facts."
Preparations for the lung cancer screening programs rolling out across the country should proceed as the medical community continues to address the issue of overdiagnosis, the ACR recommended.
The organization said it plans to proceed with its efforts to support those programs, which include forming appropriateness criteria and making a structured reporting and data collecting system to standardize methods.
Most programs appear to be following the NLST or modified versions of its criteria.
The trial randomized 53,454 men and women ages 55 to 74 with at least a 30 pack-year history of smoking to screening using low-dose CT or chest radiography.
During the median 6.4 years of follow-up, CT-based screening picked up 1,089 lung cancers compared with 969 in the chest x-ray arm.
Since the actual cancer rate was likely the same between the two well-matched groups, those extra cancers detected could have represented overdiagnosis, the researchers explained.
The excess cancer rates were 18.5% when calculated as a probability that the CT-detected cancer wouldn't have become clinically apparent during the screening phase if CT wouldn't have been done and 11% when calculated more from a public health perspective as the fraction of all lung cancer cases diagnosed in the study that wouldn't have been diagnosed then without CT screening.
The overdiagnosis rate was 31% when compared with no screening.
The probability that a tumor represented an overdiagnosis versus chest x-ray screening was also higher at 22.5% for non-small cell lung cancer and at 78.9% for bronchoalveolar lung cancer.
"These data raise the question as to the necessity and type of therapy required if a diagnosis of minimally invasive adenocarcinoma is established and challenge the diagnostic community to develop a classification scheme that could accurately phenotype all lung tumors," Patz's group wrote.
The 4 to 5 years of follow-up after screening "may not have been long enough to account for the lead time of all low-dose CT-detected cancers, particularly because tumor growth rates are quite variable and do not consistently follow classical expected exponential growth curves," the researchers cautioned.
Because CT screening found smaller, earlier stage tumors than chest x-ray, that arm would likely have had additional cancer rate "catch up" over time, so the overdiagnosis estimates "provide an upper bound on the true overdiagnosis rate associated with low-dose CT screening relative to chest radiology screening," they explained.
The key to reducing the harm of overdiagnosis will be to find biomarkers to separate out the indolent lung cancers, Patz suggested.
Source: MedPage Today: http://www.medpagetoday.com/Pulmonology/LungCancer/43327