http://bmj.com/cgi/content/full/326/7401/1231

BMJ  2003;326:1231 (7 June)
 

PDF of this article Email this article to a friend Respond to this article Read responses to this article PubMed citation Related articles in PubMed Download to Citation Manager Search Medline for articles by: Gottlieb, S. Alert me when: New articles cite this article

Collections under which this article appears: Cancer: prostate Screening

News

Prostate specific antigen test must be repeated before biopsy

Scott Gottlieb

New York

The blood test that measures prostate specific antigen (PSA) to detect prostate cancer often produces false positive results and should be repeated at least once before an invasive biopsy is performed, a new study says.

"We recommend having the findings confirmed by repeating the test after waiting at least six weeks," said Dr James Eastham, a surgeon in the department of urology at the Memorial Sloan-Kettering Cancer Center in New York and lead author of the study.

"Even if the repeat test shows an elevated level, prostate cancer will only be discovered in about one quarter of men who undergo biopsy... A single, elevated PSA level does not automatically warrant a prostate biopsy," he said.

In the study Dr Eastham and colleagues tested blood samples from 972 men, aged 35 to 89, who had originally enrolled in a study on preventing colon polyps. A total of five consecutive blood samples collected over four years from each man were tested ( JAMA 2003;289: 2695-700[Abstract/Free Full Text]).

Depending on the threshold, between 15% and 37% of the men had at least one high result. A threshold of 2.5 ng/ml yielded the highest number of positive results. A threshold of 4.0 ng/ml resulted in high scores in 21% of the men.

The percentage of men whose high results returned to normal in a subsequent test ranged from 44% to 55%, depending on the threshold. Of those who could be tested again after their level had dropped down to normal, a majority, 65% to 83%, remained normal. Thus some men may undergo an unnecessary biopsy.

Concentrations above 4.0 ng/ml can signal prostate cancer, but not always—sometimes a rise in PSA is due to another cause, and sometimes cancer can occur without a rise. About a fifth of aggressive prostate tumours are found in men with normal PSA concentrations.

The test, which has been in wide use since 1990, has enabled doctors to detect prostate cancer much earlier than they could when they relied just on physical examination. Yet false positive results are common, because cancer is not the only cause of raised PSA concentrations. Inflammation can increase them, as can recent ejaculation. And the enlargement of the prostate that occurs normally with age brings a gradual increase. The new study indicates that natural fluctuation can also lead to high results of the test.

The researchers said their study is limited because PSA concentrations may have been falling in some men as a result of treatment for prostate cancer. However, they said this is unlikely, because the men filled out healthy surveys annually that asked about newly diagnosed conditions.

Experts continue to argue over whether PSA testing offers any advantage over other screening methods for prostate cancer. The test has not yet been shown to increase life span.

Although the blood samples in the study were taken at one year intervals, the researchers suggested that follow up tests be given after four to six weeks. Testing sooner might not reflect natural fluctuations. "Four to six weeks seems like a reasonable timeframe, considering the anxiety of the patient and the doctor," Dr Eastham said.

Following the recommendation might mean missing some cancers, he added, but only those that are at an early stage. Two large scale studies are under way to determine whether the PSA test reduces the number of deaths from prostate cancer, but their results are not expected until at least next year.

Rapid responses:

Read all Rapid responses

Urologists in "regression to the mean" shock!

Paul W Masters

bmj.com, 9 Jun 2003 [Full text]

repeat PSA before biopsy

Stefano Ciatto

bmj.com, 9 Jun 2003 [Full text]

PDF of this article Email this article to a friend Respond to this article Read responses to this article PubMed citation Related articles in PubMed Download to Citation Manager Search Medline for articles by: Gottlieb, S. Alert me when: New articles cite this article

Collections under which this article appears: Cancer: prostate Screening

© 2003 BMJ Publishing Group Ltd