ZERO - The Project to End Prostate Cancer
ABOUT PROSTATE CANCERGET INVOLVEDPROGRAMSRESEARCHPARTNERSSTOREABOUT US
Risk Factors
Prevention
Detection
Diagnosis
Treatment
New Prostate Cancer Research
Prostate Cancer Resources
Managing Bone Metastases and Pain
Side Effects of Prostate Cancer Treatment

Faulty Risk Prediction of Prostate Cancer by Nomograms Without Considering PSA Inter-Assay Variability

UroToday.com | 06.04.2010

PSA variability exists between commercial assays and as such it is recommended that patients have their PSA tests performed by the same assay over time. Nomograms to determine CaP detection, risk, progression and outcomes incorporate PSA into their calculations.

The group of Dr. Kurt Miller evaluated the impact of PSA inter-assay variability on numerous prostate biopsy nomograms for CaP detection. Total (tPSA) and free PSA were determined with five different assays in 780 biopsy-referred men.

Age, prostate volume and digital rectal examination (DRE) status along with the PSA results were applied to five established online available nomograms for CaP detection and the criteria calibration and discrimination were used to characterize the usefulness of the nomogram models under these conditions.

The investigators found CaP in 455 (58%) men and 325 had no evidence of CaP. Median tPSA concentrations ranged from 5.5 to 7.04ng/mL, while the median percent-free PSA ranged from 10.6% to 16.4%.

Both the calibration and discrimination of the nomograms significantly varied when using different PSA assays.

Median nomogram probabilities, which indicate the CaP risk, ranged from 0.59 to 0.76 when using the same nomogram, but different PSA assays.

They also found that various nomograms resulted in different CaP probabilities with the same PSA assay.

Comparable areas under the curves of the receiver-operating characteristics were noted, but considerable differences between the five assays were also seen when analyzing the sensitivities and specificities at various nomogram probability cutoffs.

They concluded that the accuracy of predicting CaP probabilities by different nomograms is limited by insufficient interchangeability of the PSA values measured by various PSA assays.

Presented by Carsten Stephan, Kerstin Siemßen, Frank Friedersdorff, Florian Jentzmik, Henning Cammann, Serdar Deger, Mark Schrader, Michael Lein, Klaus Jung, Hellmuth Meyer, and Kurt Miller at the American Urological Association (AUA) Annual Meeting - May 29 - June 3, 2010 - Moscone Center, San Francisco, CA USA

Copyright UroToday.com 2010

 

 

All active news articles
About Prostate Cancer  |  Donate  |  Get Involved  |  Programs  |  Research   |  Partners  |  Store  |  About Us  |  Site Map  |  Contact Us  |  Privacy PolicySubscribe to the ZERO Prostate Cancer News Feed  

© Copyright 2010, ZERO - The Project to End Prostate Cancer. All rights reserved.

ZERO does not endorse any course of treatment for men with prostate cancer or other conditions. Treatment decisions should be made by a patient and his health
service provider. If you have problems with the site or need more information, please call us at 1.888.245.9455 or e-mail your comments. All names, logos and articles
are the copyrights of the respective owners.