Screening Reduces Relative Risk for Metastatic Prostate Cancer

Becky McCall

March 30, 2011

March 30, 2011 (Vienna, Austria) ( UPDATED March 31, 2011 ) — A new study has shown that screening for prostate cancer using prostate-specific antigen (PSA) levels reduces the risk of developing metastatic prostate cancer by 31% compared with not screening, at an average follow-up of 12 years, according to the most recent data to emerge from 4 of the European Randomized Study of Screening for Prostate Cancer (ERSPC) centers.

Results from this substudy were presented here at the European Association of Urology (EAU) 26th Annual Congress, providing an analysis of the prevention of metastatic disease by screening for prostate cancer.

The ERSPC began in the early 1990s and aims to determine whether a reduction of 25% in prostate cancer mortality could be achieved by PSA-based screening. Preliminary results were published in 2009 in The New England Journal of Medicine.

"The purpose of this particular analysis was to study the effect of screening on progression to metastatic disease. Here, we report on the overall effect of screening on metastatic rates, which is different from The New England Journal of Medicine report on metastases in 2009. Furthermore, an analysis of metastatic disease identified 'at diagnosis' and 'during follow-up' [is] presented for comparison," Fritz H. Schröder, MD, from the Department of Urology, Erasmus MC, Rotterdam, the Netherlands, told a packed conference hall. Morbidity from prostate cancer is an endpoint of the ERSPC study, and metastatic disease is an important indicator of prostate cancer morbidity and patient quality of life, he noted.

Patient age at entry ranged from 55 to 69 years. Screening intervals ranged from 2 to 4 years, depending on the center, which included the Netherlands, Sweden, Finland, and Switzerland. Some statistical analyses used cumulative hazard ratios calculated by the Nelson-Aalen method, and others used the SPSS 17.0 and STATA software packages.

The researchers used a definition of metastatic prostate cancer involving:

  • a positive bone scan with or without X-ray confirmation, or a positive computed tomography or magnetic resonance imaging study;

  • a PSA level higher than 100 ng/mL;

  • metastatic disease at diagnosis or within 3 months of diagnosis of prostate cancer; or

  • metastatic disease during follow-up, meaning more than 3 months after diagnosis.

All participating centers showed similar rates of metastatic prostate cancer identification both at diagnosis and during follow-up stages in screened patients, despite center-related differences in intervals and procedures.

After a 12-year analysis, the cumulative rate of metastatic disease in the screening group was 0.71%, whereas the control group showed a rate of 1.03%. This is a relative risk reduction of about 31% (P < .001), Dr. Schröder explained.

Dr. Schröder said metastatic disease, if unrecognized at entry into this study, was usually diagnosed at the first screen.

The next step, Dr. Schröder said, is to evaluate the effect of screening on metastatic prostate cancer during follow-up. Discovering the reasons for this difference would be crucial for understanding the screening process and comprehending its failures, he added.

The number needed to screen was 318, and the number needed to treat was 16. "These numbers are an important step forward if compared to 1410 and 49 in our 2009 [The New England Journal of Medicine] paper to prevent 1 prostate cancer death."

Similar rates of metastatic disease at diagnosis and during follow-up with screening were observed in all 4 centers in spite of differences in intervals and procedures. Screening led to a decrease in progression of metastatic disease for approximately 7 years, after which the rates went up.

At the first screen (the prevalence screen), many men have quite advanced disease, and many of these illnesses progress early during follow-up, Dr. Schröder commented. "Many of the metastatic cases at diagnosis were found during the first year. Then the numbers decreased until after year 7, when another wave of metastatic cases started. We think that for some reason, cancer cases with metastatic status found very soon after the diagnosis of cancer has been made, at any time during the study, and by whatever mechanism may get quicker and more effective care than those who develop during follow-up," pointed out Dr. Schröder.

A high rate of metastatic cases at diagnosis in the control group can best be explained by "contamination" — also termed "opportunistic screening" — or the use of PSA testing outside the screening protocol. "This is understandable as an early reaction of men randomized to the control arm who were hoping to be screened," said Dr. Schröder.

Dr. Schröder commented to Medscape Medical News: "We feel that this is clinically relevant information because of the very traumatic nature of metastatic prostate cancer. We found lower numbers needed to treat to prevent 1 metastatic case. Also, we present findings which are important for a better understanding and future improvement of the screening procedures used within the ERSPC study."

Next, Dr. Schröder evaluated the effect of screening on the incidence of metastatic prostate cancer found during follow-up. The high rate of disease progression during the first 7 years "explain[s] the long time it takes to achieve a significant prostate cancer mortality benefit," Dr. Schröder told meeting attendees. Further studies are necessary to analyze which factors influence the high progression rate in screen-detected PCa. "The figures show that no difference was found for progression to metastatic disease between screening and control groups during follow-up," he added.

Dr. Schröder also pointed out that similar numbers in both groups of the study had relatively advanced disease at randomization, and that these cases tend to progress more rapidly to metastasis, which might explain the similar rates of identifying metastasis during the first 5 years. He added that the lack of a difference "during follow-up" might be explained by preliminary data from the Rotterdam part of the study, which show that about one third of these metastatic cases occurred in men who did not follow the screening protocol, whereas the majority were classified as cases that might have been missed by screening or that developed so rapidly that any early attempt at early diagnosis in a curable stage would fail. Only about 10% of the metastatic cases at follow-up were cancers detected on screening.

Identifying factors that might play a role in making some screening centers more effective than others could be important for understanding the new findings. Specifically, the shorter screening interval of 2 years in Sweden, as opposed to 4 years in the other centers, could be important. "We've looked at whether the center, the screen interval, the screening procedure, or the age at randomization plays a role, but could not establish significant differences for either predictor," Dr. Schröder commented.

Age distribution for metastatic disease progression during follow-up does not make a significant difference in relative risk reduction, according to Dr. Schröder. "For total overall metastatic disease risk reduction...the age group of 65 - 69 seems to do best, according to data from the Netherlands, but this is in contrast to the Swedish study, where we saw the biggest difference in the lower age groups."

In conclusion, Dr. Schröder said that screening decreases the risk of developing metastatic disease by about 31% during an average follow-up of 12 years. Additional factors such as missed prostate cancers at screening, rapid progression, and no shows have been identified as potential predictors of poorer outcome during follow-up, but more data are needed to further explore this important issue.

Dr. Christopher Chapple, MD, consultant urological surgeon at the Royal Hallamshire Hospital, Sheffield, United Kingdom, commented on the results. "The presentation from Professor Schröder emphasizes the importance of picking up metastatic prostate disease at the initial diagnosis of prostate cancer, where there is an increased relative risk reduction of 52.5% as compared to an overall 31% reduction in the subset analysis for ERSPC centers."

He added: "It will be interesting to see what more detailed analysis of the whole project demonstrates in due course. I look forward to seeing further information from this extremely elegant and important prospective study tackling the important issue of screening prostate cancer."

Commenting on the results, Michael Marberger, MD, from the Department of Urology, University of Vienna, Austria, said: "The ERSPC results have changed our understanding of screening/early diagnosis of prostate cancer. With more data reaching maturity and longer follow-up becoming available, the results are even more convincing: Early diagnosis is well worth the effort."

Dr. Schröder and Dr. Marberger have disclosed no relevant financial relationships.

European Association of Urology (EAU) 26th Annual Congress; Plenary session. Presented March 22, 2011.