Researchers Estimate Overdiagnosis Rate for CT Lung Cancer Screening

A study published in JAMA Internal Medicine finds that 18.4% of the lung cancers detected with low-dose CT in the National Lung Cancer Screening Trial (NLST) may have been indolent. The researchers looked at what became of 53,452 patients in both the control and research arms of the NLST study over 6.4 years to try to determine as accurately as possible the degree of overdiagnosis of lung cancer in the patients that were screened with low-dose CT compared to the patients screened with chest x-rays. The researchers noted that a considerable limitation of research is that since we currently know relatively little about which cancers are dangerous and which are indolent, opting to simply observe detected cancers puts patients at serious risk. Therefore, the next best way to get at the overdiagnosis rate is to assume the two groups of patients — those screened with low-dose CT and those screened with x-rays — would have developed roughly the same number of cancers and the higher rate of cancer in the low-dose CT group must therefore be the upper limit of the overdiagnosis rate. The ACR, commenting on the study in an official press statement, noted that while more than 18% overdiagnosis may sound high, it is relatively modest compared to the overdiagnosis rate for other cancer screening tests. “Physicians should certainly discuss the risk and benefits of CT lung cancer screening with patients – including that of overdiagnosis. However, for high-risk patients, the group in which CT lung cancer screening is proposed, the lifesaving benefit outweighs the risks. It is now a matter of incorporating the available information – including this JAMA article – and adjusting protocols to minimize those risks as we move forward,” stated Paul Ellenbogen, M.D., FACR, chair of the ACR Board of Chancellors, in the press release. A significant factor in favor of CT lung cancer screening is that although it is not a perfect test, it is the only one currently available to patients that meet the NSLT criteria for being at high risk for lung cancer (e.g., those 55 through 79 years old and have a 30 pack year or greater history of smoking.) The ACR also noted in its statement that it is getting close to completing its own practice guidelines, appropriateness criteria, and structured reporting and data collecting system. These have been eagerly anticipated as many imaging providers began offering the test to high risk patients after the NSLT study was published in 2011 ahead of an official consensus on best practices for doing so. Not surprisingly, this lead to criticism and concern by some industry watchers. For example, in August of the same year that the NSLT findings were published, Kaiser Health News ran an article critical of the practice of some hospitals offering discount low-dose CT lung cancer screening as a loss leader that brings in potential new cancer treatment patients.