Lung cancer detection trials and methods under scrutiny
CHICAGO, Nov. 26—Two prominent radiologists analyzed the varying methods of lung cancer clinical trials, as well as scrutinized the current protocols for lung cancer detection to a standing-room only audience at the “Update on Lung Cancer: Screening and Staging” refresher course at the 93rd annual meeting of the Radiological Society of North America (RSNA).

Caroline Chiles, MD, professor of radiology at Wake Forest University Baptist Medical Center in Winston-Salem, N.C., reviewed the effectiveness of various clinical trials. She first acknowledged that screening is not always effective, because “screening tends to detect slowly growing cancer rather than rapidly growing cancers,” like lung cancer. She also pointed to overdiagnosis as an additional problem.

Chiles stressed that the strongest study design are randomized controlled trials. She highlighted the comprehensive Mayo Clinic trial that took place between 1971 and 1983. The researchers examined 9,211 male smokers, and also monitored a control arm. In early diagnosis, the prognosis proved to be effective – 48 percent in the Stages I and II cancer in the interventional arm compared to 32 percent in the control arm. But the long-term diagnosis surprised the researchers because the lung cancer mortality rate in the interventional arm was 3.2 and 3 in the control arm. There was also no difference in advanced cancer. As a result, Chiles believes that “overdiagnosis wastes healthcare services.” She stated that overdiagnosis might be more of an issue in women than men.

Chiles said that CT gives “us an opportunity to recognize lung cancer in Stage 1,” but she also questioned if it represented a stage shift. She concluded that it is difficult to calculate without a control arm.

For further evidence, Chiles referenced the International-ELCAP trial that was designed to analyze the survival rate of patients with Stage I lung cancer detected on CT screening, and published in the New England Journal of Medicine in 2006. The I-ELCAP examined 31,567 people, who were screened from 1993 to 2005, and 27,456 repeat screens were performed 7 to 18 months after the initial screening. A total of 412 participants (85 percent) were diagnosed at baseline screening, and 74 were diagnosed at their annual repeat. The Kaplan-Meier estimate of lung-cancer-specific survival rate for all patients with lung cancer is 80 percent, according to Chiles. Of the 412 participants of the I-ELCAP trial, 302 underwent surgical resection within one month of diagnosis; and for this subset, the estimated 10-year survival rate was 92 percent. 

Chiles concluded by highlighting two on-going randomized controlled clinical trials that she thinks will eventually shed light on how helpful imaging is in the detection of lung cancer. The U.S.-based National Lung Screening Trial (NLST) and the Belgium-based NELSON Trial are long-term trials on lung cancer screening in high-risk subjects.

The NLST, sponsored by the National Cancer Institute and launched in 2002, is comparing two ways of detecting lung cancer: spiral CT and standard chest x-ray. Researchers are following 53,964 cases, ages 55-74, including former smokers. Participants are randomly assigned to have either chest x-rays or CT scans, and will receive the same test each year for three years, with a follow-up through 2009.

The NELSON trial, launched in 2003, is monitoring 15,428 cases, ages 50 to 75, including former smokers. The researchers wish to establish if screening for lung cancer by multi-slice low-dose CT in high-risk subjects truly lead to a 25 percent decrease in lung cancer mortality; to estimate the impact of lung cancer screening on health related quality of life and smoking cessation; to estimate cost-effectiveness and help policy making. Overall, the primary aim of the trial is a reduction in lung cancer, according to the NELSON trial website. For interventions, the screen arm will undergo 16-multislice CT of the chest in year one, two and four of the study; a pulmonary function test; blood sampling; questionnaires; and smoking cessation advice for smokers. In the control arm, participants will simply endure smoking cessation advice for current smokers with no imaging screening. This trial slated to conclude on Dec. 31, 2015.

Chiles is convinced that the conclusions of these two trials will shed particular light on the effective of imaging on the detection of lung cancer.

David Yankelevitz, MD, radiologist at New York-Presbyterian and Weill Cornell Medical College, both located in New York City, spoke about the need to standardize the terminology of the clinical trial. For instance, in the a positive result in the I-ECLAP trial is defined as a nodule 5mm or greater, whereas in the NLST trial, it is considered 4mm or greater. These small differences can produce very different results, he said.

Yankelevitz mentioned the technology drift that occurs while the time of the trial has elapsed, “screening trials are typically quite long, and technology changes rapidly.” He said there is a potential for the imaging method to become obsolete before the trial is completed, and it can be difficult to adjust for changes in technology mid-trial.

“Technology can really potentially change results. I don’t know where it will be five years from now, ten years from now. Technology is changing just so rapidly,” he said.

Yankelevitz also pointed to regimen drifts, which he deemed as analogous to technology drifts because they can affect the trial outcomes. He said that “our understanding of techniques, methods and protocols change amidst trials,” which complicates the process.

Both Chiles and Yankelevitz reminded the audience that no radiological method or trial is currently considered the gold-standard in lung cancer detection, but the field still is subject to many years of clinical trials that need to be considered when approaching their patients.