PSA screening failed to provide mortality benefit

Andriole GL. J Natl Cancer Inst. 2012;doi:10.1093/jnci/djr500.

Annual prostate screening led to more diagnoses of tumors but not to fewer deaths from the disease, according to study results recently published online in the Journal of the National Cancer Institute.

To determine whether there was a reduction in prostate cancer mortality from screening using PSA testing, Gerald L. Andriole, MD, chief urologic surgeon at the Siteman Cancer Center at Barnes-Jewish Hospital, Washington University School of Medicine in St. Louis, and colleagues examined the prostate component of the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial.

In this trial, 76,685 men, aged 55 to 74 years, were enrolled at 10 screening centers between November 1993 and July 2001 and randomly assigned to intervention or control arms of the study: 38,340 men were assigned to the intervention arm, which included organized screening of annual PSA testing for 6 years and annual digital rectal examination for 4 years, and 38,345 men were assigned to the control arm, which consisted of routine care and sometimes included opportunistic screening. Researchers concluded screening in October 2006, after which all incident prostate cancers and deaths from prostate cancer through 13 years of follow-up or through Dec. 31, 2009, were established.

At the conclusion of the study, approximately 92% of the study participants were followed to 10 years and 57% followed to 13 years — at which point 4,250 participants had been diagnosed with prostate cancer in the intervention arm compared with 3,815 in the control arm. Cumulative incidence rates for prostate cancer in the intervention and control arms were 108.4 and 97.1 per 10,000 person-years, respectively, resulting in a relative increase of 12% in the intervention arm (RR=1.12; 95% CI, 1.07-1.17).

After 13 years of follow-up, the cumulative mortality rates from prostate cancer in the intervention and control arms were 3.7 and 3.4 deaths per 10,000 person-years, respectively, resulting in a nonstatistically significant difference between the two arms (RR=1.09; 95% CI, 0.87-1.36).

“Based on our updated results with nearly all men followed for 10 years and more than half for 13 years, we are learning that only the youngest men — those with the longest life expectancy — are apt to benefit from screening. We need to modify our current practices and stop screening elderly men and those with a limited life expectancy,” Andriole said in a press release. “Instead, we need to take a more targeted approach and selectively screen men who are young and healthy and particularly those at high risk for prostate cancer, including African Americans and those with a family history of the disease.”

The study also demonstrated that of the 4,250 patients with prostate cancer diagnosed in the intervention arm, 455 (10.7%) died of causes other than the cancer types examined, whereas in the control arm, 377 (9.9%) of the 3,815 patients diagnosed with prostate cancer died of other causes. Based on this finding, the researchers said increased screening identifies multiple tumors that will not necessarily cause harm.

“Mass screening of all men on the basis of age alone is not the way to go, but screening can still be useful in select men,” Andriole said. “We have to take a more nuanced approach to determine which men should be screened with PSA in the first place, how frequently they should be tested, the PSA level at which they should be biopsied and whether their cancer warrants aggressive therapy.”

Disclosure: The researchers report consulting, contractual and funding support from Amgen, Augmenix, Bayer, Cambridge Endo, Caris, France Foundation, GenProbe, GlaxoSmithKline, Myriad Genetics, the NCI, the National Institute of Diabetes and Digestive and Kidney Disease, and Steba Biotech.

Mark Stein

Andriole and colleagues have updated the initial report of the PLCO Cancer Screening Trial, with 3 additional years of followup. In the current report 92% of subjects were followed for 10 years and 57% were followed for 13 years. It is important to realize that this is not a trial of screening versus observation. Rather it is trial of organized yearly screening versus "opportunistic screening" — obtaining PSA at the discretion of the physician. In actuality this lead to an estimated 52% of patients in the control arm receiving PSA testing. A separate study (Pinsky PF et. al. Clinical Trials 2010, 303) estimated that the mean number of screening PSAs over 6 years for subjects in the control arm was 2.7 compared to 5.0 in the screened arm. More intense screening in the experimental arm lead to a 12% increase in the detected incidence of detected prostate cancer, but did not translate into an improvement in overall survival. This trial does not answer the more fundamental question of a survival benefit from opportunistic screening versus no screening at all.

Mark Stein, MD
HemOnc Today Editorial Board member

Disclosure: Dr. Stein reports no relevant financial disclosures.

The recent update from Dr. Andriole and colleagues is important.

In recent months, the utility of PSA tests for prostate cancer screening have come under heavy attack, particularly from the US Preventive Services Task Force (USPSTF), which recently published draft guidelines in Annals of Internal Medicine suggesting that prostate screening should be dropped from prime time.

Although the approach to public communication of its opinion was disappointing, it is clear the USPSTF made an important point — in order to judge screening for any disease as being effective, it has to increase OS or reduce overall morbidity. To date, no randomized study has demonstrated either of those elements for PSA screening.

Derek Raghavan

In fact, for patients aged older than 75 years, it appears that PSA screening actually may reduce survival. In a world facing upwardly spiraling health care costs, we need to ensure we are getting true value from our interventions, in addition to creating no harm.

There are well-published limitations to one of the three studies USPSTF cited in its recommendation — the PLCO study — including lack of compliance, variable follow-up and treatment, and group contamination (eg non-screened patients seeking to be screened outside the trial), and these have been acknowledged by its authors. That said, this report takes the follow-up period up to 13 years, a pretty significant duration for a male who presents in his mid-60s.

Once again, there has been no evidence of an OS benefit, notwithstanding a series of these studies that have demonstrated a reduction of deaths from prostate cancer.

I am reminded of studies performed by the Veterans Administration more than 50 years ago in which they assessed the utility of high-dose estrogens for advanced prostate cancer. While it was clear that lives were saved by reducing deaths from prostate cancer, these benefits were offset by an equivalent number of intercurrent deaths from cardiac and cerebrovascular complications of estrogens, and thus high-dose estrogens were removed from our “routine” therapeutic armamentarium. This analogy should not be lost by the urology community today.

To my knowledge, no randomized trial — including the European studies — has demonstrated an overall survival benefit from PSA screening, in contrast to well-defined sets of data that support mammography, colorectal screening and even a pivotal trial of spiral CT scanning to reduce overall deaths among smokers.

It also is important to understand that the extant trials simply have not addressed the status pertaining to African Americans, nor to the male population with significant family history of prostate cancer. More work needs to be done to define with certainty who should be screened, how they should be screened, how to use PSA in an optimal fashion and who really does not require screening.

The argument of physicians and advocates who blindly support PSA screening as a right for men at risk needs to be tempered by rational review of the facts and of the data. Although many have claimed that the death “rate” from prostate cancer has declined dramatically since the introduction of PSA screening, the reality is that — according to the well-known ACS annual review of cancer statistics in CA-A Cancer Journal for Clinicians — there were 25,943 deaths from prostate cancer projected in 1989 and 28,170 projected for 2012.

While the denominator has increased, it is not at all clear what that means. Are we finding a reservoir of cases that would never present a life threat to the community, or are we facing a new and frightening epidemic of this disease?

Whatever the explanation, we clearly cannot claim that PSA screening has made deaths from prostate cancer a thing of the past. Although some experts claim that we have the key answers, and that further studies should not be done, the reality is we simply cannot frame rational health policy on prostate screening today because we just don’t know enough about the disease.

Until that time, we need to ensure that the population is educated about the disease, understands that it is treatable (with several options) and recognizes that not all cases require urgent therapy, and that some cases don’t need treatment at all. In a period of genuine equipoise, a “reasonable” option is to encourage men to discuss this illness with their own physicians in the hope that they will achieve a rational personal plan to deal with it.

– Derek Raghavan, MD, PhD, FACP, FRACP

HemOnc Today Editorial Board member

Disclosure: Dr. Raghavan reports no relevant financial disclosures.

Follow HemOncToday.com on Twitter.