Personalizing prostate screening

Prostate cancer screening by means of PSA testing was officially marshaled into the domain of the personal recently by the US Preventive Services Task Force (USPSTF). Their draft conclusion about the overall utility of screening is that they couldn’t reach a decisive conclusion, and you and your doctor should decide.

For those of us in Preventive Medicine, the USPSTF has long been revered, and their recommendations considered the veritable bible of clinical preventive services. The Task Force is a bit “by” the government, but neither “of” nor “for” it. This multidisciplinary group of experts is convened partly at the invitation of the federal government, but operates independently. Recommendations are made for the benefit of public health, and are very strictly evidence-based.

One of the dangers in the application of very strict standards of evidence is that those items that are still subject to uncertainty will be deemed less worthy, rather than simply less mature. Since the Task Force recommendations are highly influential and routinely translate into reimbursement norms, this danger could mean: insurers will not cover what is not formally recommended. But that banishes to the realm of uncovered medical expense both some of what will prove to be baby, along with what will prove to be bathwater.

That leads us back to prostate cancer screening. The Task Force assigned a grade of “C” on its letter grade scale, indicating that net benefit for the population is almost certainly small at best, but that benefit for select individuals was both possible, and hard to predict. By deferring decisionmaking to the dialogue between patient and doctor, the Task Force is also saying that insurers should cover any testing that results from that discussion.

Not knowing that universal prostate cancer screening is a good idea is not the same as knowing it’s a bad idea. The Task Force is reminding those prone to rigid sanctimony about scientific evidence that it evolves over time, and at any given time can be murky for many reasons: it is insufficient in quantity or quality, the findings are inconsistent, or key studies have yet to be done. By leaving the door ajar for future decisions, the Task Force is reminding its constituencies of the crucial difference between absence of evidence and evidence of absence. Deferring to patients and their doctors is an implicit recommendation to respect and pay for the decisions that ensue. Perhaps, too, the Task Force is implying that when uncertainty is greatest, so, too, must be respect for patient autonomy.

Why isn’t it obviously a good idea to screen all men for prostate cancer, since the disease is prevalent, and treatment established? There are several reasons.

First, testing is prone to errors in both directions. PSA levels can be elevated in the absence of cancer (false positives), and sometimes normal in spite of it (false negatives). The former failing can lead to additional, unnecessary testing, even including biopsy. The latter can confer a false and dangerous sense of security.

But the main problem with prostate cancer screening is not difficulty in finding the cancer, but difficulty in predicting its behavior. The majority of men (roughly 80 percent ) who die after age 80 die with, but not of, prostate cancer. In other words, most guys who live long enough get this “disease.” But most of this cancer never progresses to cause symptoms or compromise overall health. The same is true at younger ages, albeit less common. When such cancers, destined to be indolent, are treated — the cure is often far worse than the disease, imposing risks from infection to erectile dysfunction to incontinence. This is why prostate cancer screening, which can save lives, can also hurt far more people than it helps.

There are dual advantages in personalizing the approach to screening.

First, even when risk factors are comparable, dispositions can vary. Some people simply much prefer knowing to not knowing, even if knowing introduces its own challenges.

Second, variation in risk factors, notably family history, can alter the ratio of potential harm to likely benefit.

Individualized decisions about prostate screening are more personal than personalized. The hope is that a truly “personalized medicine” approach will evolve, in which those decisions are informed not only by preference, but by biologic indicators that reliably point to the best choice. Ultimately, a better answer than personalizing decisions about marginally good screening is enhancements to screening. Efforts are well under way to advance that agenda. For now, prostate cancer screening isn’t formally personalized yet, but it is officially personal.