NEW YORK—A new screening procedure to detect early lung cancer has the potential to dramatically reduce mortality, Claudia Henschke, MD, PhD, chief, Division of Chest Imaging, Weill Medical College of Cornell University, said at a media briefing.
Lung cancer is the number one cancer killer in the United States, accounting for more than 160,000 deaths annually, more than from breast, prostate, cervical, and colorectal cancer combined, she said.
“Early lung cancers are curable, but early detection is rare,” Dr. Hen-schke said. Unlike colon, breast, cervical, and prostate cancers, there are currently no recommendations for screening for lung cancer. The technology for detecting early lung cancers is less than 10 years old and is rapidly evolving, she added.
In an effort to develop an effective lung cancer screening procedure, Dr. Henschke and a team of researchers are investigating the use of helical low-dose CT imaging, which is capable of spotting very small changes characteristic of early lung cancers.
The scan takes about 20 seconds and can cover the entire lung in a single breath hold.
It delivers only slightly more radiation than a chest x-ray, Dr. Henschke said. It is, however, significantly more sensitive than chest x-rays, which can, at best, spot malignancies at 30 doubling times, when they are about the size of a 25 cent coin. This is shortly before the cancer becomes symptomatic and the chance of cure is small, she said.
CT scans can spot malignancies at 20 doubling times, when they are about the size of a grain of rice. As CT technology advances, even smaller lesions will be detectable, she said.
The ELCAP Study
A pilot study conducted by the Early Lung Cancer Action Project (ELCAP) was designed to evaluate baseline and annual repeat screening with the low-dose helical scan. ELCAP is a multicenter, multi-disciplinary group of oncologists, epidemiologists, radiologists, pulmonologists, and pathologists headed by Dr. Henschke.
The overall design and findings from baseline screening of the ELCAP study were published in The Lancet last year (354:99-105, 1999). The study enrolled 1,000 symptom-free, high-risk subjects, smokers and previous smokers aged 60 years and older with at least 10 pack-years of smoking history.
Participants received both chest x-rays and CT scans. Of the 27 lung cancers found, chest x-ray detected only 7, while CT found all 27. Of those cancers found by CT scan, 23 were stage I disease, compared with only 4 found by x-ray.
Dr. Henschke’s group has proposed the single-arm study design of ELCAP as the basis of a new NIH study of the technology. Discussions are currently in process about the final study design. The NIH favors a randomized trial using chest x-ray in the control arm, but Dr. Henschke believes such a design may be neither ethical nor practical.
“Randomization is not suited to a study of a diagnostic procedure,” she said, observing that it would clearly be unethical to have a screening arm and a no-screening arm or a control arm using a less effective screening method. Furthermore, individuals in the study who are randomized to x-ray may seek CT screening on their own, thus contaminating the results.
Dr. Henschke also addressed concerns that early detection of lung abnormalities may lead to unnecessary surgery. She said that in the ELCAP study, all abnormalities were scanned again 3 months after they were initially detected, which she termed “a reasonable amount of time to wait.” Thanks to engineered three-dimensional imaging, even minute changes could be detected.
Biopsies were performed only on tumors exhibiting change in subsequent scans, resulting in fewer biopsies than when conventional chest x-rays are used, she said.
The CT screening procedure is also controversial because of the costs involved. The procedure is available at only a few centers and currently costs $300 and up per scan. Before insurance companies can be expected to cover this cost, they will have to be convinced of its cost-effectiveness.
Dr. Henschke, however, needs no convincing. “It will be very cost-effective because the high-risk individuals can be so readily identified,” she said.
She cited an article in the Canadian Medical Journal that put the cost of annual CT screening for lung cancer in high-risk individuals at $5,000 to $10,000 per life-year saved. This is much lower than the cost of mammography screening for breast cancer (about $230,000 per life-year saved) and Pap tests for cervical cancer screening (about $50,000 per life-year saved). CT screening for lung cancer may even be cost saving, Dr. Henschke said, because of differences in the cost of early-stage vs late-stage treatment.
The Bottom Line
“The bottom line is that this procedure is highly cost-effective,” Dr. Hen-schke said, adding that what is needed now is a definitive study demonstrating that identifying early lung cancers through CT screening translates into the saving of lives. “New screening techniques are constantly being developed, and they need to be rapidly assessed and implemented,” she said.
