Doctors Recommend Patience While Researchers Examine Results of Lung Cancer Screening Trial

Vanderbilt University Medical Center was one of 33 sites involved in the National Lung Screening Trial, which was halted prematurely late last year after the study showed that a low-dose CT scan can detect lung cancer at much earlier stages than a conventional chest X-ray and thus reduce deaths by as much as 20 percent. (See related story below.)

That’s promising news indeed, yet John A. Worrell, MD, Vanderbilt professor of radiology, cautioned that it’s much too early to draw conclusions about future screening protocols.

Lung cancer is the No. 1 cancer killer in America, claiming more lives than breast, colon and prostate cancer combined. Lung cancer is typically diagnosed late in the disease process because there is no uniform screening test for early detection. That’s why this massive study, funded by the National Cancer Institute, was such a critical investigation. “The basic study design was a very strong one – a double-blinded, randomized clinical trial, which is the gold standard of trials,” said Worrell, who added that the participant population was geographically and ethnically diverse.

Researchers enrolled more than 53,000 current or former smokers, ages 55 to 74, without signs, symptoms or a history of lung cancer. The participants were randomly assigned to receive three annual screens using either low-dose CT or a standard chest X-ray. They were then followed for up to another five years. On Oct. 20, the study’s independent Data and Safety Monitoring Board voted unanimously to stop the trial. By that time, there were 354 deaths from lung cancer among participants screened by low-dose CT as compared to a significantly larger 442 lung cancer deaths in the chest X-ray group. “It was thought ethical to tell the people who were in the chest X-ray limb of the study about the results,” Worrell said. “Because of that, the news was made public.” That process bypassed the traditional analysis, peer review and publication stages; thus, months of data crunching are still ahead.

“Is there a difference women versus men? A difference in ethnic groups? A difference in geography? All those questions are pretty important when trying to advise real patients in front of you whether they should do this or not,” Worrell explained. “There’s no official position that has been taken by any learned society, and there’s no third party that I know of that is reimbursing for this.”

The trial culminated just as the nation is working to reel in out-of-control healthcare expenditures, and policymakers are weighing the costs and benefits of expensive procedures. Researchers gathered cost information as part of the study, but it’s yet to be reviewed. CT scans are much more costly than X-rays, yet there’s no evidence that standard chest X-rays reduce lung cancer mortality.

A further complication is that the trial’s false positive rate was 25 to 33 percent – and Worrell said at Vanderbilt, the false positives were running about 50 percent. Of course, a positive result, false or not, prompts additional testing, biopsy and, in a rare instance, even surgery, as well as anxiety for the patients and their loved ones.

Gregg P. Allen, MD, is the chief medical officer for MedSolutions, a company founded to help third-party payers manage costs without sacrificing quality care. One of MedSolutions’ specialties is diagnostic imaging. “We just don’t know what the cost/benefit ratio is yet. In our view, there’s probably something very good that’s going to come out of this study in terms of enhancing our knowledge of when to screen high-risk patients,” Allen said.

Predictably, MedSolutions’ clients have been asking for the company’s input on the study’s findings, and Allen authored a white paper for payers to outline the issues. He noted that it’s important to let researchers “analyze the other pieces of the puzzle” before any decisions are made.

Allen expressed concern that the study may prompt physicians to start ordering CT screens, for example, for 45-year-olds who have smoked for 10 years and exhibit no symptoms. “There’s no evidence that that is going to be effective, but that’s what will happen,” he predicted. “I suspect that there are doctors today who are ordering screening chest CTs now on the basis of this result in people who don’t fit the criteria of the study.” He cautioned, as well, that doctors might order a diagnostic rather than a screening CT, thus exposing the patient to five times more radiation than necessary.

Allen called for “rational use” of high-technology imaging, and Worrell said he’s telling physicians who are asking him for advice to take a wait-and-see attitude. “Wait a couple of months until the data are analyzed and see which groups are going to benefit most,” he said, adding, “Don’t smoke is really the message. We spent $250 million – actually a little more – on this study so far, and it’s a self-inflicted wound. I’m not down on smokers – people who try to quit and can’t, I have every sympathy for. What I am saying is that if you can quit, do it, and if you don’t smoke, don’t start.”


Related Story

Vanderbilt Wins Cancer Detection Grants

Vanderbilt-Ingram Cancer Center researchers have nabbed $6 million over five years to support early detection of lung and colon cancers. The awards are from the National Cancer Institute’s Early Detection Research Network (EDRN).

Pierre Massion, MD, was awarded $3 million for the creation of the Vanderbilt Clinical Validation Center, while Daniel Liebler and David Tabb, both with PhDs, were awarded $3 million to create the Vanderbilt Biomarker Development Laboratory. 

“I am very excited about the opportunity to validate the biomarker candidates that we have developed at Vanderbilt for the past ten years,” Massion said. “I am convinced if we want to detect lung cancer early we need to use multiple modalities, including identifying the highest-risk patient population, analyzing the data on chest imaging tests and determining which biomarkers have predictive value for early detection. If we can find the patients with the highest risk of disease, we have an opportunity to detect the disease before it is symptomatic and give those patients a chance for a cure.”

Massion and his colleagues will recruit Nashville-area patients at high risk for lung cancer for a screening trial called the “Nashville Early Diagnosis Lung Cancer Project,” and will evaluate a set of biomarkers to determine whether the molecular signatures are helpful in early diagnosis of lung cancer. Biomarkers may be found in blood, urine or tissue samples from patients. 

Massion’s group will collaborate with Liebler and Tabb, who will develop and apply new proteomics methods and informatics tools to identify proteins that might be useful as biomarkers for lung and colon cancer.