Potential Complications of Lung Cancer Screening: False Positives
The basic problem in lung cancer screening is that cancers and benign nodules look the same. With some exceptions, a radiologist cannot tell a cancer from a benign nodule just by looking at them.
A false positive test is a positive test report in a patient who does not have lung cancer. Such a report can cause anxiety, unnecessary testing, inconvenience, radiation exposure and/or even unnecessary biopsy or surgical removal of a lung nodule.
Such an operation, in turn, can have no benefit, only harms such as loss of lung function, post-surgical complications or even death.
How often do false positive results occur?
This is a difficult to answer. Estimates of false positives in lung cancer CT screening range between 98% and 5%.
How can such a wide discrepancy of results be possible?
The 98% figure – that is still being quoted today – is from a faulty interpretation of the results of the National Lung Screen Trial (NLST), a randomized control trial (RCT) conducted by the National Cancer Institute (NCI) between 2001-2011.
NLST contained a number of major flaws in research design.
- NLST criterion for a positive test was too loose. It included nodules with a largest diameter of 4 millimeters (for example, a nodule 4 mm. X 2 mm.)
- In addition, NLST considered the same nodule as positive even if it had been seen on a previous CT scan, in effect counting it as false positive twice.
- NLST used no diagnostic plan after a positive result was reported.
- Many nodules that were at first considered as false positive later turned out to be lung cancers.
International Early Lung Cancer Action Program (IELCAP) researchers use a different criterion for a positive result
– an average diameter 6 mm. or greater (for example 6×6 mm, but not 6×5 mm. in diameter). Using this criterion, false positives fall to 12% on the first CT screen. This standard has been adopted by most modern screening programs.
False positives are much less common on annual-repeat CT scans – approximately 5% in IELCAP.
Are all patients with false positive CT scans harmed?
No.
It is very important to re-emphasize that lung cancer screening is not a test; it is a process. The process includes an organized plan (algorithm) designed to prevent complications, when a benign nodule is detected by CT screening. The IELCAP group uses a diagnostic algorithm, first introduced around 2000 that is reviewed and revised each year based on analysis of growing data on more than 80,000 individuals screened at more than 70 centers worldwide.
Other methods include the clinical practice guideline for lung cancer screening of the National Comprehensive Cancer Network (NCCN) and the LungRads method of the American College of Radiology.
It is important to stress again that the NLST study did not include a diagnostic plan. As a result, when a study member had a positive test result, the next step was decided as an individual choice by his doctors. This resulted in substantial harm, as many patients had diagnostic procedures, like bronchoscopy (looking into the lungs using a flexible scope and taking biopsies) that had little chance of diagnosing small cancers in the periphery of the lung. Worse, many people had surgical resection of lung nodules that were benign.
The lesson learned from the NLST experience is that a diagnostic plan (algorithm) is necessary to prevent unnecessary invasive tests, while at the same time not allowing a delay that will allow lung cancers to progress and reach higher stages where cure is less likely.
How do diagnostic algorithms prevent unnecessary invasive tests and operations?
The plan currently used is not complicated or difficult to understand.
First, nodules are put into one of four categories
1. Calcified nodules
2. Solid nodules
3. Non-solid nodules
4. Part-solid nodules.
and then size is carefully measured.
A 1. calcified nodule is very rarely cancer and no further action is required until the next annual CT scan.
A 3. Non-solid nodule may be a cancer, but is typically non-invasive (in-situ) and so slowly-growing, that no action is required until the next annual CT scan. The risk that such a cancer will become incurable in a year is close to zero.
A 2. Or 4. Solid or part solid nodules have a risk of being a faster growing, invasive cancer.
If the solid nodule is less than 6 mm. in average diameter, it is safe to wait 12 months until the next annual CT scan to re-measure the size of the nodule.
If the solid nodule is larger than 6 mm. Then it is not safe to wait a year.
If the nodule is between 6-8 mm. A CT scan is done again in 3 months. If significant growth is seen, the nodule is probably a cancer and further investigation is required.
If the nodule is larger than 8 mm. A PET scan is recommended.
If the PET scan is negative, the nodule is probably not a cancer and a repeat CT is ordered to check for growth in a few months.
If the PET scan is positive, the nodule is probably a cancer or an infection. Biopsy is indicated. This is preferably by trans thoracic needle biopsy, done under CT scan control. Alternatively, the nodule can be surgically removed.
If the nodule is 1.5 cm. or larger, immediate action is necessary. Lung cancers of this size are dangerous and may spread. Delay to observe for possible growth is not safe. Biopsy is indicated.
Non-solid nodules can be observed at one year intervals by CT scan. If a solid nodular component arises within the non-solid nodule, it probably represents development of an invasive cancer.
If these basic rules are followed, false positive test results seldom result in unnecessary needle biopsy, bronchoscopy or surgical resection.
On the flip side of the coin, if the rules are followed, it is uncommon for a lung cancer to reach higher stage or become incurable.
The bottom line is that when lung cancer screening is carried out in an organized and responsible manner, patients with benign nodules in their lungs are seldom harmed.
Despite careful guideline based management, a small percentage of screened patients will undergo biopsy or resection of benign nodules.
A recent report from the NELSON study from Scandinavia, which used a carefully organized diagnostic plan, shows a false positive rate of only 5%.
Complications of lung cancer screening: Unnecessary operations, complications, deaths