Sven V, 48, doesn't look like
your typical struggling artist. When he was diagnosed
with lung cancer a year ago, Sven sold his thriving internet
business to concentrate, as he says, on the bigger picture.
Fortunately for Sven, the tumour that the computerized
tomography (CT) scan picked up on is more than likely
non-lethal but the surgery to confirm this diagnosis has
its own attendant risks. Sven's case raises an important
question: is more detail in imaging always better? The
answer, a somewhat counterintuitive negative, comes from
a new study comparing CT scanning with X-rays for lung
cancer screening, published in the April issue of Radiology.
"What we're finding with CT screening
are more early-stage cancers. That could be good news
finding cancers that we'd otherwise find at a
late stage. However, a number of these are probably
non-lethal or slow-growing cancers that the patient
would likely have died with and not from. Other cancers
were so aggressive that early detection with CT did
not make a difference," says lead author Dr Stephen
J Swensen, professor and chair of the Department of
Radiology at Rochester, Minnesota's Mayo Clinic, in
a published interview.
WHAT'S
THE USE?
While the value of early cancer detection is considered
self-evident, Dr Swensen and his colleagues observed
many negative outcomes associated with using the new
technology. They studied 1,520 current and former smokers
over age 50 a group considered to be at high
risk of lung cancer. Sixty-one percent of the subjects,
who were almost evenly split between men and women,
were current smokers, while the remaining 39% of subjects
had kicked the habit. As part of the study, each patient
received a low-dose helical CT scan initially, and then
annual scans over the following four years of followup.
BIOPSIES
ARE NOT BENIGN
CT scans revealed 3,356 nodules in 1,118 of the 1,520
patients. Only 68 of these nodules proved to be primary
lung cancers in 66 patients. The initial screen found
31 of these tumours, while the rest were detected on
subsequent scans. False-positives, on the other hand,
occurred in 69% of the subjects. This was revealed when
the uncalcified nodules detected were established as
benign by biopsy or observation.
"Currently, half of lung nodules
suspected of being cancerous that go to surgery outside
of research study centres turn out to be benign," Dr
Swensen points out. Diagnosis of these nodules requires
surgical intervention, which is not only expensive but
can also affect patient quality of life and mortality.
Some patients would suffer chronic pain from the surgery,
he observes. Moreover, average mortality from lung cancer
surgery itself ranges from 3-5%. "That's a big price
to pay if it's a benign nodule," he says.
To top it all off, the mortality
rate in this study was not significantly different from
that observed during the Mayo Lung Project of the 1970s
a trial that used the humble chest X-ray as a
screening tool.
The researchers suggest that while
CT helps radiologists find more early-stage cancers,
many of these tumours are slow growing and probably
would not have caused the patient's death. "Our results
lead us to be very cautious, because there's a chance
that we may be doing more harm than good," Dr Swensen
concludes.
Radiology Apr 2005, 235:259-65
|