The Canadian Cancer Society (CCS)
has this month called for more organized screening for
breast, colorectal and cervical cancer. The Society's
arguments are backed by many supporters, and questioned
by precious few detractors. The logic behind extending
cancer screening programs seems unanswerable. Early
stage disease is typically far more treatable than advanced
tumours. Survival rates would be vastly higher if we
could detect every tumour when it was still small, localized
and asymptomatic.
These three cancers, breast, colorectal
and cervical, are the three that regularly emerge from
studies of screening as the most practicable. The benefits
of cervical screening were obvious from the moment the
Pap smear was invented. Breast cancer screening was
an argument that was won a long time ago. Colorectal
cancer screening has been a little more controversial,
but with Canadians experiencing one of the highest rates
in the world of this disease, the risk-benefit ratio
of screening is better here than elsewhere.
All of these screening strategies
have been reviewed by groups like the US Preventive
Services Task Force and the Canadian Task Force on the
Periodic Health Examination. A hard look at the evidence
for screening generally shows a clear benefit in the
three cancer types identified as priorities by the CCS.
But when the evidence is reviewed
for other cancers, there's often a fly in the ointment
that makes the whole thing impracticable. Sometimes
the early cancers are hard to find, as in ovarian cancer,
or finding them can involve risky or invasive methods,
like carcinogenic CT scans for lung cancer. Or the benefits
of early detection are questionable, as with prostate
cancer, where many of the detected patients go untreated,
and most end up dying of something else.
In many cases, the false positive
rate is far higher than the actual cancer detection
rate. A false positive can set off a chain of events
whose costs and associated morbidity must be factored
into any realistic assessment of screening's benefits.
Throw in the danger of false negatives, which can delay
vital treatment, and the risk-benefit ratio of some
forms of screening can be very doubtful indeed.
We aren't helped by the frankly
dismal methodology of so much screening research. Many
studies involved patients who were already high-risk
in one way or another.
But the commonest flaw in trials
of screening programs is a failure to look at actual
health outcomes rather than just test results. Screening
can't be evaluated in isolation from treatment and prognosis.
A screening method now judged impracticable could become
worthwhile overnight, not because the screening technology
has changed, but because a new treatment is available.
While there are technical difficulties
with many forms of screening, the three programs advocated
by the CCS are hindered only by a lack of will and of
money. We already have organized breast screening programs
in every province and territory, but the uptake, estimated
at 34% nationally, is far below the 70% that the CCS
considers a reasonable target. The CCS estimates such
an uptake would reduce breast cancer deaths by a quarter.
Cervical screening is widespread
but disorganized. We could do better with an organized
system that seeks out those most at risk. So many of
the Pap smears conducted in Canada today are of women
who have already tested negative on previous smears.
More first tests on high-risk women, especially underserved
recent immigrants, would do more to reduce disease burden.
Colorectal screening by fecal occult
blood testing has proven benefits in patients who are
regularly screened over at least a decade, but the Cancer
Society's recommendation for biannual testing seems
unnecessarily modest. They estimate that colorectal
cancer deaths could be reduced by 17% if 70% of Canadians
between the ages of 50 and 74 had a fecal occult blood
test every two years. Most existing research suggests
that annual testing is the way to catch significant
numbers of early tumours.
But on the whole these guidelines
are sound. Provinces should implement these recommendations,
and should think big in doing so. Half-hearted approaches
would be wasteful. There's good reason to believe that
big centralized labs will produce more accurate results
with fewer false negatives. Cancer screening could end
up saving money as well as lives, but that will only
happen if really significant proportions of cancers
are detected at early stages.
Perhaps the key recommendation
of the Cancer Society is their injunction not to give
up hope on screening for more difficult cancers. There
are promising new ways to screen for oral cancer, and
the potential contribution of dentists shouldn't be
overlooked. We must improve our understanding of risk
factors, because screening that is extravagantly wasteful
in the general population can make good economic sense
in a targeted group.
Above all, we should never give
up looking for an effective screening test for the number
one killer, lung cancer. Even if sensitivity and specificity
were less than ideal, being singled out as a high-risk
patient needing lung cancer screening could well have
a salutary effect on health by driving people to quit.
So many smokers avoid thinking about risk. That kind
of escapism just isn't possible when waiting for test
results.
|
