Elsevier

Urology

Volume 65, Issue 3, March 2005, Pages 549-553
Urology

Adult urology
CME article
Evidence suggesting PSA cutpoint of 2.5 ng/mL for prompting prostate biopsy: Review of 36,316 biopsies

https://doi.org/10.1016/j.urology.2004.10.064Get rights and content

Abstract

Objectives

To determine whether a prostate-specific antigen (PSA) level of 2.0, 2.5, or 4.0 ng/mL is the most appropriate cutpoint for determining the need for prostate biopsy. It has been suggested that the PSA cutpoint of 4.0 ng/mL is inappropriate because the rate of prostate cancer detection is similar in patients with lower PSA values. Some investigators have recommended a 2.6-ng/mL cutpoint. Others have recommended a cutpoint of 2.0 ng/mL.

Methods

A total of 36,316 prostate biopsies submitted to DIANON Systems from January 1, 1997 through December 31, 2001 were reviewed. These biopsy specimens also had DIANON PSA test results available that had been performed within 6 months of the biopsy date. These biopsies were stratified according to the PSA level within the 6 months before the time of biopsy, and the prostate cancer detection rate was determined for the stratified PSA levels.

Results

The detection rate of prostate cancer varied according to the PSA level. The incidence of prostate cancer was similar for the groups with less than 2.0 ng/mL and 2.0 to 2.5 ng/mL (18.67% and 21.89%, respectively). Also, the groups with 2.5 to 4.0 ng/mL and 4.0 to 10.0 ng/mL had similar cancer detection rates (27.48% and 30.08%, respectively).

Conclusions

The prostate cancer detection rate for a PSA level between 2.5 and 4.0 ng/mL was similar (27.48%) to that for the PSA range of 4.0 to 10.0 ng/mL (30.08%). The absolute cutpoint used to determine the need to evaluate a patient for prostate cancer by biopsy is not clear; however, many studies have suggested that 2.5 ng/mL may be a more appropriate cutpoint than 4.0 ng/mL.

Section snippets

Material and methods

From 1997 and 2001, all 233,889 prostate biopsy specimens submitted to DIANON Systems (Stratford, Conn) were retrospectively reviewed and analyzed for patient age and the presence of prostate adenocarcinoma. Because the patients were not individually identified, institutional review board approval was not indicated. The overall prostate cancer detection rates were calculated. A total of 36,316 cases were identified with available DIANON PSA tests performed within 6 months of prostate biopsy.

Results

The detection rate for prostate cancer varied according to the PSA level, with a greater percentage of cancers detected in direct correlation with rising PSA levels. The number of submitted biopsies also increased with rising PSA levels, except for the PSA group of greater than 10.0 ng/mL, for which fewer biopsies were submitted compared with the 4.1 to 10.0 ng/mL range (8802 versus 20,984 biopsies). The largest number of biopsies submitted was for elevated PSA levels in the 4.1 to 10.0 ng/mL

Comment

In the current study, 36,316 of 233,889 biopsies submitted to DIANON Systems were reviewed and stratified according to the PSA level. The detection rates were similar for the PSA subgroups of less than 2.0 and 2.1 to 2.5 ng/mL (18.67% and 21.89%, respectively). The cancer detection rate for the PSA subgroups 2.6 to 4.0 and 4.1 to 10.0 ng/mL was 27.48% and 30.08%, respectively. A total of 2940 biopsies were submitted in the lower PSA range between 2.6 and 4.0 ng/mL and 808 cancers diagnosed. As

Conclusions

In an analysis of 36,316 biopsies submitted to a reference laboratory with prebiopsy PSA determinations available, the detection rate of prostate cancer for a PSA level of 2.5 to 4.0 ng/mL was similar (27.48%) to that for a PSA level of 4.0 to 10.0 ng/mL (30.08%). Recently, the standard cutpoint of 4.0 ng/mL has been critically assessed, and several studies have supported a lower absolute PSA threshold for recommending prostate biopsy. Our results indicate that the prevalence of prostate cancer

References (22)

  • S.H. Woolf

    Screening for prostate cancer with prostate-specific antigen

    N Engl J Med

    (1995)
  • Cited by (0)

    View full text