EAU 2009: US Study Shows No Mortality Benefit From Prostate Cancer Screening, But European Study Suggests There May Be One

Roxanne Nelson

March 18, 2009

March 18, 2009 — A huge study from the United States suggests that prostate cancer screening does not reduce deaths from the disease; however, an even larger European study suggests that it does, and that it reduces prostate cancer mortality by about 20%.

The findings come from interim data from the US Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, which was funded by the National Cancer Institute (NCI), and from the European Randomized Trial of Screening for Prostate Cancer (ERSPC).

Both studies were published online March 18 in the New England Journal of Medicine, to coincide with the presentation of the data at the European Association of Urology meeting being held in Stockholm, Sweden. The results were also discussed at a press conference held in the United States by the NCI.

The PLCO interim results show that prostate cancer screening leads to more diagnoses of the disease, but does not appear to affect prostate-cancer-related mortality. At 7 years, screening was associated with a relative increase of 22% in the rate of prostate cancer diagnosis, compared with a control group. However, a reduction in prostate-cancer-related mortality was not seen in the screened group. The authors report that, over an 11-year median follow-up, combined screening with a prostate-specific antigen (PSA) test and a digital rectal exam (DRE) did not reduce mortality.

However, results from the ERSPC show that PSA screening, without DRE, was associated with a 20% reduction in the rate of death from prostate cancer. The study showed an absolute reduction of approximately 7 prostate-cancer-related deaths per 10,000 men screened.

PSA Screening Associated With Falling Prostate Cancer Death Rate?

PLCO senior study author Christine Berg, MD, from the NCI, noted that these are not the final results from either study, and that longer follow-up is needed to determine whether the impact of more diagnoses will eventually translate into fewer deaths from prostate cancer.

"When the PLCO first began in 1992, the PSA had been in fairly wide use for about 5 years," she said. "However, we didn't know if it was saving lives or just finding prostate cancers that would never lead to death."

We didn't know if [PSA] was saving lives or just finding prostate cancers that would never lead to death.

"There are men today who were diagnosed with prostate cancer, who now suffer from the side effects of treatment, and who would never have suffered from ill health or may have never died of the disease in the first place," Dr. Berg explained during the NCI press briefing.

And although it is vital to continue to follow the cohort in the PLCO, she added, it is essential that better and more accurate methods of diagnosing prostate cancer are developed, and essential to be able to identify which prostate tumors are the most likely to be deadly.

Even though the data are preliminary, Gerald Andriole, MD, first author and principal investigator of the PLCO trial, felt that it was important to publish these results now. "Why have we been confused? Prostate cancer mortality rates have been falling since the mid-1990s and PSA testing became widespread in the early 1990s," he said.

"It seemed that the rates of death were falling from screening, but in reality, much more was happening," said Dr. Andriole, who is chief urologic surgeon at the Siteman Cancer Center at Washington University School of Medicine and Barnes-Jewish Hospital, in St. Louis, Missouri. "There was also more treatment of prostate cancer, and more interventions. Looking at the PLCO results thus far, it may be that a lot of the reduction we are observing is due to better treatment and not to screening per se."

We know now that it is going to be reasonable to stop testing at a certain age and time in a man's life, and I think we will be doing less testing in the future.

However, it is too soon to make broad screening recommendations for all men based on the study's initial findings, he added. "The take-home message at this point — in my opinion — is that the PLCO data confirm that men who have a limited life expectancy of 7 to 10 years or less probably do not need to undergo PSA testing."

It is more problematic for younger men and those at a high risk for disease, and there are not enough data yet for the youngest men in the study. "But an abnormal PSA almost automatically leads to biopsy, and cancer almost automatically leads to treatment, and we need to temper that," Dr. Andriole said. "We know now that it is going to be reasonable to stop testing at a certain age and time in a man's life, and I think we will be doing less testing in the future."

Conflict Between the Studies?

Dr. Andriole also noted that the conclusions of both trials are not really conflicting, even though they may initially appear to be.

"At first glance, it seems that they offer different conclusions about screening, but they are actually more similar than different when you scrutinize them," he explained. "There are different study designs and different populations. In the United States, we are also more likely to be more aggressive about treatment. The jury is definitely out about differences between the 2 trials."

One difference is the cutoff points used in PSA testing. In the PLCO, the cutoff point for doing a biopsy was a PSA of 4 ng/mL, whereas in the ERSPC, it was 3 ng/mL, which did not necessarily identify more aggressive disease. "The decision to use a lower cutoff point may find more tumors but it may find more tumors that are clinically insignificant," Dr. Andriole noted during the NCI press conference.

The PLCO includes an annual DRE, but in most of the centers where the European trial was conducted, this was dropped from the protocol after the first year or 2. The screening schedules also differ between the studies, Dr. Andriole pointed out. Men in the PLCO were screened every year for 6 years, whereas those in the ERSPC were screened every 4 years.

There were also racial differences in the study populations. "The patients were primarily white in the European study, whereas in the United States, we purposely included minorities that represented about 15% of the population," explained Dr. Andriole. "These differences make it very hard to stack them up side by side and compare them to each other."

Key Findings From PLCO

In the PLCO study, 76,693 men were randomized to either 6 rounds of annual screening with PSA plus 4 annual DREs (n  =38,343) or to usual care (n = 38,350). Annual random surveys of men in the usual-care group found that they were also getting screened, but to a lesser degree. However, PSA screening among controls increased from 40% at the beginning of the study to 52% in the last screening year, whereas screening with DRE ranged from 41% initially to 46% by the last screening year.

This raises an issue of contamination, said Philip Kantoff, MD, director of the Lank Center for Genitourinary Oncology at the Dana Farber Cancer Institute, in Boston, Massachusetts.

"Fifty-two percent of the men who were in the nonscreened arm had a PSA documented within the past few years in the study, as opposed to 85% in the screened arm," said Dr. Kantoff, who was not involved in the study but participated in a Perspective Roundtable discussion of both papers held by the New England Journal of Medicine. "And as a result, it's not surprising to me that there was only a modest increase in the number of cancers [only 20%] that were diagnosed in the screened arm."

The PLCO currently includes follow-up data for all participants at 7 years, and for 67% at 10 years. To date, these are the key findings of the PLCO:

  • At 7 years, 22% more prostate cancers were diagnosed in the screening group than in the usual-care group (2820 vs. 2322 men). This trend has continued in data collected up to 10 years, with 17% more prostate cancer diagnoses in the screening group.

  • The vast majority of men in both groups who developed prostate cancer were diagnosed with relatively early stage II disease, and the number of later-stage cases was similar in the 2 groups.

  • Men in the usual-care group had more prostate cancers that fell into the Gleason score range of 8 to 10; the smaller number of men with high Gleason scores in the screening group could eventually lead to a mortality difference between men in the 2 groups, the researchers note. However, data thus far have not shown such a difference.

  • At 7 years, there were 50 prostate-cancer-related deaths in the screening group and 44 deaths in the usual-care group.

  • Through year 10, there were 92 prostate cancer deaths in the screening group and 82 in the usual-care group, but the difference was not statistically significant. No detectable mortality benefit for screening vs usual-care was observed.

  • At 10 years, 312 men with prostate cancer in the screening group and 225 men with prostate cancer in the usual-care group died of other causes. The authors note that the excess number of deaths in the screening group may have been related to overdiagnosis of prostate cancer.

Dr. Kantoff agrees that that a longer follow-up is needed, and that is one of the most problematic parts of the study. The number of outcomes was also very modest. "One tenth of 1% of the entire population actually died of prostate cancer," he said. "Not surprisingly, the mortality for prostate cancer within the first 10 years is small, but with the numbers that they generated [50 in the screening group and 44 in the usual-care group], it's really hard to make a statement about differences in outcomes."

Key Findings From ERSPC

In the ERSPC, 182,000 men between the ages of 50 and 74 years who resided in 7 different European countries were randomized to either PSA screening, at an average of once every 4 years, or to a control group. The predefined core age group for the study consisted of 162,243 men aged 55 to 69 years, and the primary outcome for the study was rate of death from prostate cancer. Mortality follow-up was completed on December 31, 2006.

The average and median follow-up times were 8.8 and 9 years, respectively. These are the key findings of the ERSPC:

  • There were 214 prostate-cancer-related deaths in the screening group and 326 deaths in the control group in the core age group.

  • The cumulative incidence of prostate cancer was 8.2% in the screening group and 4.8% in the control group.

  • PSA screening was associated with a significant absolute reduction of 0.71 prostate cancer deaths per 1000 men after an average of 9 years.

  • To prevent 1 prostate cancer death, 1410 men would have to be screened and an additional 48 men would require treatment.

  • An analysis of participants who actually underwent screening during the first round provided a rate ratio for prostate-cancer-related mortality of 0.73 (95% confidence interval, 0.56 to 0.90).

This study has a number of limitations that deserve serious consideration, according to Mary McNaughton-Collins, MD, a general medicine internist and health services researcher at Massachusetts General Hospital and Harvard Medical School, in Boston, who participated in the round-table discussion but was not involved in the study.

One of the limitations is that the trial pulled together studies conducted in different countries, and since protocols were different, it wasn't a uniform study design, Dr. McNaughton-Collins noted.

In addition, this is the third interim analysis, she pointed out. So the "20% mortality reduction is only marginally statistically significant at 0.04, raising the question: Why stop now?"

The numbers needed to treat and screen are high, and this is problematic, she said. "For right now, we should probably maintain a healthy skepticism about this type of screening program," she commented. "Any effective screening program, we know, requires more than just effectiveness. We have to find out more about quality of life or cost-effectiveness."

The PLCO study was supported by the National Cancer Institute. The ERSPC study was funded by grants from Europe Against Cancer; from the fifth and sixth framework program of the European Union; from agencies or health authorities in participating countries, and from Beckman Coulter. Dr. Andriole reports receiving consulting fees from Aeterna Zentaris, Ferring Pharmaceuticals, Negma Steba, Onconome, Veridex, AstraZeneca, and Amgen, and research support from GlaxoSmithKline. Several other PLCO study authors have financial relationships; details can be found in the paper. Three of the ERSPC authors have financial relationships; details can be found in the paper.

N Engl J Med. 2009;360:1310-1319, 1320-1328


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.