Medscape Medical News from the:

• Radiological Society of North America (RSNA) 96th Scientific Assembly and Annual Meeting

This coverage is not sanctioned by, nor a part of, the Radiological Society of North America.

From Medscape Medical News

False Positives Plague CT Lung Cancer Screening

Fran Lowry

Authors and Disclosures

December 8, 2010 (Chicago, Illinois ) — Some 23% of all computed tomography (CT) screens that were done in the National Lung Screening Trial (NLST) for lung cancer were categorized as being positive for lung cancer, when in fact they were not.

The relatively high rate of false positives is something that proponents of lung cancer screening grappled with here at the Radiological Society of North America 96th Scientific Assembly and Annual Meeting.

Dr. David Gierada

David S. Gierada, MD, professor of radiology at Washington University School of Medicine, St. Louis, Missouri, and NLST coprincipal investigator, told Medscape Medical News that just 649 of the 75,136 screens led to a diagnosis of lung cancer.

Findings that warranted further investigation were the presence of a noncalcified nodule or nodules 4 mm or more, and any other suspicious finding such as lobar collapse, enlarged hilar or mediastinal lymph nodes, and any endobronchial lesions.

The NLST screened its participants at 3 different time points. In round 1, of 26,314 people who were screened, 7,193 (27.3%) were deemed to be positive. In round 2, of 24,718 people who were screened, 6,902 (27.9%) were deemed to be positive. And in round 3, of 24,104 people who were screened, 4,054 (16.8%) were deemed to be positive for lung cancer.

"The false-positive rate is high, but it's not surprising," Dr. Gierada told Medscape Medical News. "We expected that. But we didn't know how that rate would compare with the false-positive rates that occur when screening for other forms of cancer, such as breast cancer, colon cancer, or prostate cancer."

In general, the false-positive rates for other forms of cancer screening are "quite a bit lower than for CT screening," he said.

The high false-positive rate notwithstanding, Dr. Gierada said that it is very encouraging to know that there is a test that will have some impact on the high mortality rates seen with lung cancer.

But, as did the other NLST investigators, he reiterated that the 20% reduction in mortality from screening, although significant, is still not going to be a cure for the high prevalence of lung cancer mortality in the United States today.

Variability among radiologists reading the scans might be an issue at first, but Dr. Gierada believes that this will lessen as radiologists become more comfortable with screening.

"For screening radiologists, it can be a bit of a subjective process in interpretation. They may differ to some degree in how they measure a nodule, and this may affect how some nodules are classified," he said.

However, the amount of variability among readers in the NLST trial was very modest "and similar to the variability that has been seen in other CT screening tasks, such as assessment for pulmonary embolism or even mammography interpretations," Dr. Gierada said.

He said that if future research helps to improve the specificity of interpretation of certain types of lesions — perhaps on the basis of their location within the lung or their shape — that "will be extremely helpful and may get people to be more consistent in their readings."

Computer-aided diagnostic technology is another promising avenue that might eventually help characterize suspicious nodules and improve standardization of the measurement of nodules, he said.

It is important to recognize that CT scanning is not perfect, Dr. Gierada said. "It's possible that a fast-growing aggressive lesion could develop in the year between screens. It's also possible that there will be small lesions within the airways that may be easy to miss on CT. And certainly readers can miss things; detection rates are not 100%.

Dr. Gierada stressed that when a policy for lung cancer screening is eventually instituted, it will be important for those being screened and their clinicians to be aware of the issue of false positives, and recognize that it is one of the limitations of the screening examination.

"The high positivity rate is an issue that is very difficult for people undergoing screening because once they are labeled as having a positive screen or an abnormality in their lung that might be cancer, this generates a lot of anxiety," he said. "People will benefit more from seeing the screening procedure in different terms. Rather than a one-time 'do I have cancer or not' screen, they should look upon it as more of a lung cancer surveillance process in which an abnormality, if found, most likely will turn out to have a very low likelihood of being cancer. But if any change in that suspicious lesion is noted, it will be treatable when it is still very small and before it causes symptoms and spreads to a higher stage."

By far the largest proportion of people will fall into that category, he said.

Dr. Gierada has disclosed no relevant financial relationships.

Radiological Society of North America (RSNA) 96th Scientific Assembly and Annual Meeting: NLST Panel Presentation. November 29, 2010.

• Print This
• Share
  Facebook Twitter

Medscape Medical News © 2010 WebMD, LLC
Send comments and news tips to news@medscape.net.

Receive an email from Medscape whenever new articles on this topic are available.

• 
  https://profreg.medscape.com/px/updateTopicAlertSubscription.do?action=subscribeFromViewArticle&conceptId;=5000186&callback;=?
  Add Lung Cancer to My Topic Alert

• Lung Cancer Screening Tests May Yield False-Positive Results
• Risk for False-positive Lung Cancer Screening is Substantial

• Lung Cancer