In addition, the largest difference between expected and observed metastatic cancer is in older age groups. "The headline seems to be coming from patients who are greater than 80 years old," Garnick says. But those patients often have many other health problems that lead to death before even aggressive prostate cancer.
Otis Brawley, a medical oncologist and chief medical officer of the American Cancer Society who was not involved with the study, says, "It certainly adds to the body of evidence, but it's by no means definitive." One problem is that many men whose initial diagnosis is localized, not metastasized, prostate cancer, end up relapsing even after having their prostate is removed. This means that the cancer had already metastasized by the time of the first diagnosis, but it could not be detected at that time. Those patients already had metastatic disease when screened, but they were not diagnosed at the time, and so they are not counted in Messing's study.
Another problem is that modeling studies, like this one, are inherently less powerful than other kinds of medical research: randomized clinical trials, the gold standard in medical research, in which patients are randomly assigned different treatments or no treatment; case-control studies, which compare patients who have a condition with those who do not; or cohort studies, which determine the risk of contracting a disease by studying a group of people with similar demographics. So the results from a modeling study, no matter how compelling, are not definitive. They should lead to other studies, and eventually clinical trials.
On both sides of the debate researchers agree that men should be informed of the risks and possible benefits of PSA screening and subsequent treatment before they start the process, a challenge when contradictory reports abound. Garnick says, "My heart goes out to both patients and physicians because the whole concept of screening, diagnosing and offering treatments is among the most complicated in medicine."
"What we need is a better biomarker that not only tells us that the patient has cancer but what the behavior of that cancer is likely to be," Garnick continues. "It becomes a circular argument until we get better biomarkers." For now, men and their families have to make tough choices about whether knowledge is power or ignorance is bliss.