The magazine of UC Davis Comprehensive Cancer Center

Spring/Summer 2014

Dr. Dall'Era

To screen or not to screen

Can the PSA test diagnose prostate cancer without leading to unnecessary treatments?

Medical choices can be confusing. For years, men have relied on the prostate-specific antigen (PSA) test to warn them if they are at higher than normal risk for cancer. It’s become a rite of passage: as men enter middle age, the PSA is added to their annual test regimen. An elevated reading might indicate cancer, leading to a biopsy and possibly treatment.

But in recent years, some physicians have questioned the benefits of the PSA test. While it can signal cancer, an elevated PSA also might indicate nothing more than prostatitis or benign prostatic hyperplasia — serious conditions, but hardly lethal. And even when the PSA does indicate the potential presence of cancer, it cannot determine if the disease is aggressive, requiring immediate treatment, or slow-growing.

In 2012, the United States Preventive Services Task Force (USPSTF) advised against routine PSA testing. That independent panel, consisting of experts in prevention and evidence-based medicine, determined that the test’s imprecision was leading to unnecessary prostate biopsies, prostate cancer over-diagnosis and over-treatment. The panel advised that the serious and unnecessary side effects, such as pain, urinary incontinence or sexual dysfunction some men experience following treatment might outweigh their risks of dying from prostate cancer.

But do the task force recommendations go too far? UC Davis researchers have entered the fray, and their findings may indicate an important role for the PSA.

PSA pros and cons

All men with a prostate have some PSA in their blood. However, those levels tend to rise when disease hits. In 1994, the U.S. Food and Drug Administration (FDA) approved the PSA test to measure these levels in men who, due to age, ethnicity or both, are at higher risk for prostate cancer. There was much enthusiasm for the test — perhaps too much.

“When the PSA first came out, it wasn’t studied in the way it should have been,” says Marc Dall’Era, assistant professor and vice chair of urology at the UC Davis Comprehensive Cancer Center. “A big group of cancers are extremely slow growing and are better off undiagnosed. However, we do diagnose them through PSA screening and biopsies. As a result, men get treated aggressively, regardless of the risk their cancer may pose.”

These and other concerns helped lead to the USPSTF’s decision to recommend against widespread PSA testing. But Dall’Era and others believe the PSA can still be a valid diagnostic tool. While the test is clearly imperfect, the task force may have relied on bad evidence to make their decision.

“If you’re designing a good study, you have a group that gets screened for prostate cancer and a group that doesn’t,” says Dall’Era. “But in the U.S., the vast majority of men get screened at some point. The American study, which led to the [USPSTF’s] PSA test recommendations, had a group that got a lot of screening and a control group that was screened less often.”

As a result, more than half the men in the control group were still getting screened. So it’s no surprise that the test showed only marginal differences in survival rates between these two samples. In addition, other research has shown that prostate cancer deaths have declined by as much as 42 percent since the PSA’s introduction. What could be driving these declines?

Doing the research

One explanation for reduced prostate cancer mortality could be more effective treatments. Dall’Era and colleagues decided to investigate whether survival rates for men initially diagnosed with metastatic prostate cancer have improved.

“Let’s assume PSA screening isn’t working. But if that’s not reducing prostate cancer mortality, what is?” asks Dall’Era. “What if we’re doing a lot better treating men with advanced disease?”

Fortunately, they have an amazing resource to answer that question: the California Cancer Registry. Managed by UC Davis, the database is a veritable gold mine for researchers, containing information on more than 2.5 million cancer cases. The registry includes data on demographics, cancer types, disease status, patient diagnosis dates, treatments and survival. Dall’Era’s team dove deep into the registry to determine if improved treatments had led to reduced mortality. Unfortunately, they hadn’t.

“There was no survival improvement for men with metastatic disease to account for the overall mortality decline,” says Dall’Era.

Instead, the study confirmed what previous research had shown: over the past 20 years, 65 percent fewer men have been initially diagnosed with metastatic prostate cancer. PSA screening has caused a major shift in diagnostic staging. That is, cancers are now being diagnosed at a much earlier stage.

“As far as I’m concerned, a major portion of this decline is coming from the PSA,” says Ralph de Vere White, UC Davis urologist, director of the UC Davis Comprehensive Cancer Center and a co-author on the Dall’Era paper. “Imperfect as it is, the PSA is likely detecting cancers that, if left untreated, would go on to metastasize and kill people.”

What’s next?

While these findings do not prove the PSA is lowering cancer mortality, they constitute powerful evidence to support that hypothesis. But the question remains: Should men get the PSA test?

One train of thought suggests the problem may not be the test but rather how physicians and patients respond to the results. Men with elevated PSA levels often get a biopsy. If the biopsy indicates cancer, they often receive aggressive treatment, whether they need it or not.

“Men who are at low to medium risk may be good candidates for surveillance rather than treatment,” says Dall’Era. “If they have a slow-growing cancer, we may not need to intervene at all. Right now, only about 20 percent of men who are candidates for surveillance are getting it.”

This nuanced approach offers many advantages, such as reducing unnecessary surgeries and radiation treatments, but it can be a hard sell for a patient who only wants to be cancer-free.

“If we tell patients they need early detection, but then tell them we’re not going to treat their cancer — it seems counterintuitive,” says de Vere White. “We have to talk to patients before the biopsy. They need to know that if we find prostate cancer and that cancer does not seem to threaten their life, we will suggest they go on active surveillance.”

In the meantime, new genetic tests that are emerging could help physicians and patients decide whether to choose treatment or surveillance. These diagnostics could potentially highlight tough cancers that require an immediate response.

“This is the direction we want to go,” says Dall’Era. “I don’t think we should abandon the PSA; we just need to rethink which patients should receive aggressive treatment.”