Julia Draznin Maltzman, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 30, 2003
Screening means looking for the cancer before it had the chance to produce symptoms and come to medical attention. Finding a cancer before it gives symptoms increases the likelihood of finding it at a very early stage. When cancer is found at an early stage it is often easier to treat and there is a higher potential for cure.
Screening high-risk populations for lung cancer has been a fiercely debated issue. Many experts question the benefits of screening in general, while others, may agree on the need for screening; they disagree about the modality of choice. Over the years, three possible screening modalities for lung cancer have emerged: sputum cytology, a chest x-ray (CXR), and a spiral computerized tomography (CT) scan.
Some patients may be able to produce mucus with deep coughing. This mucus may be sent to the pathologist for review. In some cases, cancer cells may be expectorated within the mucus. A careful microscopic investigation may be able to identify those cancer cells and make the diagnosis without invasive procedures. Unfortunately, only a few cancers are shed into the mucous of a deep cough. Most cancers do not directly communicate with the trachea (wind pipe) and therefore do not shed their cells into the mucous.
A CXR is an x-ray of all of the organs within the chest. These include the lungs, the heart, the great vessels, and the ribs. A CXR has been considered the standard screening mode for lung cancer as it is fast, relatively inexpensive, and non-invasive. Although a CXR can easily identify large abnormalities, some more subtle presentations, lesions less than one or so centimeter, can be missed. Since a CXR gives a two-dimensional photograph of the chest, even medium size lesions behind the heart may not be readily visible.
A spiral CT uses sophisticated x-ray machines that scan in a rotating (spiral) pattern to produce multiple, detailed, pictures of the body. The entire procedure can take as little as 20 seconds -- a single large-breath hold. Spiral CT scans can identify lung lesions less than one centimeter in size. A spiral CT can visualize everything that is seen on CXR including soft tissues (muscle, fat and lymph nodes) and the thoracic spine. The most commonly sited disadvantage of this technique is cost. Also, spiral CT may pick up numerous benign abnormalities that can lead to invasive evaluations that were not needed.
The Early Lung Cancer Action Project (ELCAP) study, published in the Lancet in July 1999 analyzed the effect of screening high-risk individuals with CT scans. One thousand high-risk individuals, defined by a history of smoking, asbestos exposure and emphysema, were enrolled. Nodules were detected in 233 participants (23%) by CT scan versus 68 participants (7%) by CXR. Of all the positive CT scan results, 27 cancers were diagnosed. However, of the twenty-seven cancers, twenty-six were completely respectable with surgery. Twenty-three participants had stage I lung cancer. The study concluded that of all the cancers found by CT scanning, more than eighty-five percent were stage I. The study admitted to many false positives, however, noted that these maybe managed with few invasive diagnostic procedures.
In follow up of this paper, the same author, in the August 2003 issue of the journal Chest, published an article reviewing the financial benefit of their previous clinical findings. This publication is the result of a collaborative effort between New York Weil Cornell Medical Center, Mount Sinai School of Medicine, and Columbia University Graduate School of Business. This was a unique study in that it used already published data by the ELCAP and offered a financial analysis showing a benefit to screening. The study concluded that the yearly cost of saving one life using a single CT scan could be as low as $2,500. Supporters compared lung cancer screening to other well-accepted screening strategies such as Pap smears and mammography. The cost per life saved can be as $50,000 and $24,000 respectively.
The authors contend that lung cancer screening by CT is cost effective because the chances of finding curable disease is quite high. The cost of a surgical treatment for cure is less than half of the cost of late stage care. The authors further argue that as this screening technique becomes more widely available and used, the price should decrease commensurately and the cost-effectiveness ration becomes even more attractive.
In another study simultaneously published in Lancet, researchers found that spiral CT with selective follow up by Positron Emission Tomography (PET) scan was a more reliable screening technique. A PET scan was added to improve diagnostic accuracy for suspicious lesions and avoid false positive findings. Over one thousand patients had spiral CT performed at baseline annually for five years, with or without PET. Calcified nodules with a maximum diameter of 5mm were deemed not to be worrisome and were scheduled to repeat a spiral CT at one year. Lesions larger than 5mm underwent a high resolution CT assessment. PET scan was used to better define smaller non-calcified lesions. All PET positive lesions were taken to biopsy as well as all non-calcified lesions larger than 20mm. In addition, patients had multiple annual tests such as urine, blood and sputum samples as well as basic spirometry. All patients were smokers over the age of fifty. By the second year of data accrual, twenty-two cases of lung cancer were detected – eleven at baseline evaluation and eleven by the repeat CT scan done one year later. The authors concluded that the combined use of spiral CT and selective PET could effectively detect early stage lung cancer.
A surprising twist in this study was that six of the eleven cancers diagnosed at the one year follow up exam, albeit small, were present on the baseline CT scan. The delay in diagnosis did not seem to change outcome, as all six cases were still stage I lung cancers. However, this certainly implies that spiral CT scans have serious limitations. The authors quickly admonished that a conservative approach to such small lesions seen on CT scan is justified and maybe followed up at one year with impunity.
The national cancer institute (NCI) is currently conducting a randomized, controlled trial entitled the National Lung Screening Trial (NEST). This trial will compare CXR versus spiral CT as methods for early lung cancer detection and screening. The risks and benefits of spiral CT scans compared to CXR will also be analyzed. In addition, the study hopes to show if either test is better at reducing deaths from this often-fatal disease. The study opened a year ago and is planned to accrue 50,000 participants in a period of two years. Upon entry to the study, each individual will be randomized to receive either a spiral CT or a CXR. They will have the same screening procedure again one and two years later. Annual monitoring of the participants health status will continue until 2009. Multiple centers throughout the country are enrolling participants in this trial, including the University of Pennsylvania.
There is no question that screening modalities such as CXR, CT scans and PET can detect small, early stage lung cancer lesions. The debate revolves around cost and the number of false positive results that can cause much fear and anxiety. False positive results also lead to invasive and potentially unnecessary testing that carry their own morbidities. It is not altogether uncommon, for example, that a scar from tobacco smoking, infection, or other lung inflammation may be mistaken for an early cancer. Unnecessary biopsies or worse yet, open chest surgeries (thoracotomy), can put the patient through much stress both physically and psychologically.
At this time, CT screening for lung cancer is not covered by most medical insurance. However, the exam costs about $350, and proponents argue, is about the cost of 1/3 yearly tobacco habit of one pack per day.
Another criticism is over treatment. Chemotherapy or other treatments have the potential to be given at the first hint of lung cancer. However, some small tumors may never become life threatening and treating these lesions only exposes the patient to the harmful side effects of therapy. One of the weaknesses of modern medicine is that it cannot distinguish which lesion is problematic and which is not.
The argument is complicated and often visceral. All these screening studies, however, take only high-risk patients – long-term smokers. Perhaps, the best solution is to avoid becoming a high-risk patient.