News

Cancer is personal. We all know somebody who detected it early and somebody who found it too late.

That cancer is picked up by screening - the earlier and more often the better, the standard line goes - seems to make so much sense. And research shows that some interventions, particularly for high-risk individuals, are in fact effective. But it's not as simple as that. Sometimes, science contradicts common sense.

If you have or had cancer, you'll know that the journey and the science are rarely black and white. In our new world of patients driving the decision-making, which I favour, that choice is now shared between health-care provider and patient. Part of this shared responsibility is to look at all the science, including the data that don't necessarily jibe with your intuition.

A study published last week in the Journal of the National Cancer Institute reflects how research results can be counterintuitive. It would seem to make sense that if we screen people at more regular intervals, then we are more likely to find more cancers. Yet in this study, in which more than 17,000 men were followed for up to 10 years, and either screened every two years or every four years for prostate cancer with the prostate-specific antigen (PSA) blood test, screening more frequently did not reduce the number of aggressive tumours.

Then there is the PSA test itself. In an accompanying editorial, David Crawford, a professor at the University of Colorado Health Sciences Center, points out we still don't have high-quality screening trials telling us whether PSA testing saves lives. Some medical groups support it while others are skeptical. Meanwhile, almost half of Canadian men have been screened.

PSA testing is but one example of research ambiguity. Much of our time in front-line medicine is spent answering common-sense questions from our patients, spurred by research findings that are either counter-intuitive or open to interpretation: Why not start mammograms at 40 instead of 50? Why are you stopping at 70? You told me before to do a breast self-exam, and now you're saying not to?

Patients have a reasonable desire to do all they can - for fear of the disease - and are guided by the perception that more testing is better than less.

When there is an available test and a vacuum of evidence to guide practice, both patients and doctors tend toward over-testing, as we rarely get negative feedback for this, while we often do for under-testing.

But more testing isn't necessarily better.

The Prostate, Lung, Colorectal, and Ovarian cancer trial (PLCO) is a good example. Ovarian cancer is very tricky to diagnose early: The symptoms are fatigue, bloating and some stomach problems, which basically describes every woman in my practice.

In the trial, women had extra testing: an ultrasound of the uterus and ovaries, and a special blood test called CA-125. In the 28,816 healthy women who underwent the initial screening, 29 tumours were detected, 20 of which were invasive. The most interesting part of the study was that the investigations triggered 541 unnecessary surgeries.

It's possible that more testing is saving you, but it is also possible that it could be harming you.

I wish the message were simple - everybody should be screened - as this would save me time, be easier for you to understand and be on-message with the powerful social networks of cancer patients and cancer organizations. Creating a conflicting message is less effective, takes much more time and can leave both doctors and patients confused.

Even when we know that cancer is present, the answers still aren't always clear. For example, surgery is not always critical in prostate cancer. Research on older men with prostate cancer described as having less than 10 years of life left (which meant they were usually over 70) compared two groups: those who had surgery to remove the tumour and those who were simply watched. When the two groups were followed up, there was no difference in lifespan, and the group that received surgery had a poorer quality of life.

For some men in my practice, it is very difficult to initiate an active surveillance of prostate cancer, as opposed to acting on their core belief that cancer must be removed.

The somewhat disconcerting fact is that cancer rates skyrocket after 75, and if we look hard enough we can find cancer, but many people have no symptoms and die of other causes.

What tends to be a more exact indicator of cancer occurrence is your own personal risk, be it family history, poor lifestyle habits or other exposures. And the best predictor of sensible cancer screening has nothing to do with testing - it's having a good relationship with a family doctor. Maintaining this connection and knowing your values can be a beacon for navigating the uncertain waters of cancer detection.

Michael Evans is an associate professor at the University of

Toronto and staff physician at

Toronto Western Hospital. mevans@globeandmail.com