Bolton (MA), San Antonio (TX)
According to the American Cancer Society, more people die from lung cancer than any other type of cancer. Lung cancer accounts for 27% of cancer deaths in the United States, and more people will die from lung cancer than from colon, breast, and prostate cancers combined. This is true for both men and women. There will be an estimated 224,210 new cases of lung cancer in the United States in 2014 (116,000 men and 108,210 women). In addition, there will be an estimated 159,260 deaths from lung cancer in 2014 (86,930 men and 72,330 women).
There has long been interest in screening and early detection of lung cancer as patients with a neoplasm found at earlier stages have far better outcomes than those diagnosed with advanced disease. The disease is aggressive and by the time patients present clinically, the prognosis is usually poor.
Screening imaging methods include chest radiography and computed tomography (CT). Either of these modalities can be used with or without computer-assisted detection or diagnosis (CAD). Computer-assisted detection points out possible findings to the radiologist who then decides if the finding is abnormal. Computer-assisted diagnosis uses a computer algorithm to analyze features of a lesion to determine the level of suspicion and is intended to enhance the reader’s diagnostic performance.
Both of these technologies may be expected to offer more benefit when used by relatively inexperienced readers and may help to standardize diagnostic performance. At present though, there are no randomized or observational clinical trials that have been reported evaluating chest radiographs with CAD for screening for lung cancer. There is also insufficient evidence to determine whether CAD technology may improve the accuracy of scanning interpretations using CT.
However, a retrospective non-screening study in lung cancer patients did show modest results with CAD, correctly identifying cancerous nodules in many patients although there were a high number of false positive results.
Biases in Lung Cancer Screening
Screening studies are typically limited by biases that may erroneously give positive results such as improved survival. These include lead-time, length-time, and overdiagnosis biases.
A lead-time bias causes survival to seem improved when the difference is only on the basis of earlier detection of disease and not effective treatment. Finding indolent, less aggressive cancers can result in a length-time bias, which gives the appearance of longer survival as the tumors are found well before they become symptomatic. This is similar to overdiagnosis bias that occurs when cancers are found and treated that would not have resulted in mortality (indolent or clinically insignificant tumors).
These biases are a very real problem for determining truly positive results of screening. Carefully constructed clinical trials should hopefully be able to overcome these screening biases and show real improved outcomes.
High-quality, randomized clinical trials examining the effect of screening on lung cancer morbidity and mortality are necessary to determine the true impact of this screening on patient outcomes. Previous randomized clinical trials of chest radiography and serial sputum samples have not shown improved patient outcomes. For example, randomized screening clinical trials utilizing chest radiography published in the 1980s found that, although patients had a higher incidence of earlier stage and resectable lung cancers, there were no statistically significant differences in mortality attributable to lung cancer between the two groups.
More recently, there has been interest in low-dose computed tomography (LDCT) scanning as a screening technique. The radiation exposure for this examination is greater than for that of a chest radiograph, but far less than for a conventional CT scan. Earlier European studies found that screening CT will find early stage lung cancers but is hindered by a relatively high rate of false positive scans (detection of benign nodules). Subsequent work-up of benign nodules leads to repeat imaging and/or invasive procedures (biopsy or surgery) with added expense and unnecessary risk to the patient. Most importantly, the screening CT clinical trials, until data published in 2011, have not shown an improvement in lung cancer morbidity or mortality.
In August 2011, the results of the National Lung Screening Trial (NLST) were published in The New England Journal of Medicine. The aim of the study was to determine if LDCT could reduce mortality from lung cancer in high-risk patients. The study randomized over 53,000 patients who had three annual screening exams with either LDCT or posteroanterior chest radiography. The patients were followed for over 7 years. Similar to prior studies, there were high numbers of false positive results. However, the relative reduction in mortality from lung cancer with LDCT screening compared with chest radiography was 20.0%. The authors concluded that screening high-risk patients with LDCT reduces mortality from lung cancer.
This NLST was the first major clinical trial to show improved mortality in lung cancer screening. Subsequently and largely based on the NLST findings, the U.S. Preventive Services Task Force (USPSTF) began recommending LDCT for lung screening in December 2013. They specifically recommended LDCT for patients age 55 to 80 who have a 50 pack per year smoking history and currently smoke or have quit smoking within the past 15 years. Excluded are those patients that are non-surgical candidates and those with other medical problems that substantially limit life expectancy. In the short time period since that recommendation, screening LDCT has quickly entered clinical practice.
Smaller trials using low-dose computed tomography scanning as a screening technique are under way in Europe and may provide additional beneficial data.
To download a PDF version of this article please click here
Many questions remain to be answered now that lung cancer screening has been shown to be effective. Cost-effectiveness, optimization of screening protocols and regimens, and the hazards of screening (radiation risk, risk of overdiagnosis, risk of invasive procedures, etc.) will need to be addressed.
Recently, NLST data were reexamined with results published in the Journal of the American Medical Association highlighting the problem of overdiagnosis due to screening. The paper reports an 18% rate of indolent or clinically insignificant tumors. The role of LDCT with CAD and chest radiography with CAD in the future of lung cancer screening is yet to be determined.
Additional clinical trials are needed in all of these areas to optimize lung cancer screening recommendations.
About the Author
Michael Orsi, M.D., completed a fellowship in Cross Sectional Imaging in both Thoracic and Abdominal Imaging at the world renowned Johns Hopkins Hospital in Baltimore, MD. He completed his residency in Diagnostic Radiology at the University of Texas Health Science Center in San Antonio where he served as Chief Resident. He is board-certified in Diagnostic Radiology and specializes in imaging of the Thorax, Abdomen and Pelvis. He serves as a reviewer for the Journal of Radiology and has published original research related to minimally invasive surgical techniques such as liver radiofrequency ablation and advanced imaging as it applies to MRI techniques for the liver.
About Intrinsic Imaging LLC
Located in Bolton, Massachusetts and San Antonio, Texas, Intrinsic Imaging is an FDA audited, ISO 9001:2008 and ISO 13485:2003 certified, GAMP® 5 compliant medical imaging core lab specializing in providing imaging core lab services for clinical trials. Its comprehensive medical imaging core lab services include, but are not limited to, expert radiologist consultation, protocol and charter development, site qualification, training and management, as well as image acquisition, processing and detailed radiologic analysis.
Intrinsic Imaging has more than sixty full-time, board-certified diagnostic radiologists on staff that have sub-specialization in all therapeutic areas including, but not limited to, Cardiovascular, Central Nervous System, Gastrointestinal & Genitourinary, Medical Device, Musculoskeletal and Oncology.