Combining biomarkers to detect disease with application to prostate cancer.

Ruth Etzioni, Charles Kooperberg, Margaret Pepe, Robert Smith, Peter H. Gann

Research output: Contribution to journalArticle

68 Citations (Scopus)

Abstract

In early detection of disease, combinations of biomarkers promise improved discrimination over diagnostic tests based on single markers. An example of this is in prostate cancer screening, where additional markers have been sought to improve the specificity of the conventional Prostate-Specific Antigen (PSA) test. A marker of particular interest is the percent free PSA. Studies evaluating the benefits of percent free PSA reflect the need for a methodological approach that is statistically valid and useful in the clinical setting. This article presents methods that address this need. We focus on and-or combinations of biomarker results that we call logic rules and present novel definitions for the ROC curve and the area under the curve (AUC) that are applicable to this class of combination tests. Our estimates of the ROC and AUC are amenable to statistical inference including comparisons of tests and regression analysis. The methods are applied to data on free and total PSA levels among prostate cancer cases and matched controls enrolled in the Physicians' Health Study.

Original languageEnglish (US)
Pages (from-to)523-538
Number of pages16
JournalBiostatistics (Oxford, England)
Volume4
Issue number4
DOIs
StatePublished - Jan 1 2003
Externally publishedYes

Fingerprint

Prostate Cancer
Biomarkers
Percent
Combination Test
Curve
Diagnostic Tests
Receiver Operating Characteristic Curve
Statistical Inference
Regression Analysis
Discrimination
Specificity
Screening
Health
Valid
Logic
Estimate
Prostate cancer
Class

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Combining biomarkers to detect disease with application to prostate cancer. / Etzioni, Ruth; Kooperberg, Charles; Pepe, Margaret; Smith, Robert; Gann, Peter H.

In: Biostatistics (Oxford, England), Vol. 4, No. 4, 01.01.2003, p. 523-538.

Research output: Contribution to journalArticle

Etzioni, Ruth ; Kooperberg, Charles ; Pepe, Margaret ; Smith, Robert ; Gann, Peter H. / Combining biomarkers to detect disease with application to prostate cancer. In: Biostatistics (Oxford, England). 2003 ; Vol. 4, No. 4. pp. 523-538.
@article{1062fe5fecd14d7fa096757cec876c04,
title = "Combining biomarkers to detect disease with application to prostate cancer.",
abstract = "In early detection of disease, combinations of biomarkers promise improved discrimination over diagnostic tests based on single markers. An example of this is in prostate cancer screening, where additional markers have been sought to improve the specificity of the conventional Prostate-Specific Antigen (PSA) test. A marker of particular interest is the percent free PSA. Studies evaluating the benefits of percent free PSA reflect the need for a methodological approach that is statistically valid and useful in the clinical setting. This article presents methods that address this need. We focus on and-or combinations of biomarker results that we call logic rules and present novel definitions for the ROC curve and the area under the curve (AUC) that are applicable to this class of combination tests. Our estimates of the ROC and AUC are amenable to statistical inference including comparisons of tests and regression analysis. The methods are applied to data on free and total PSA levels among prostate cancer cases and matched controls enrolled in the Physicians' Health Study.",
author = "Ruth Etzioni and Charles Kooperberg and Margaret Pepe and Robert Smith and Gann, {Peter H.}",
year = "2003",
month = "1",
day = "1",
doi = "10.1093/biostatistics/4.4.523",
language = "English (US)",
volume = "4",
pages = "523--538",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Combining biomarkers to detect disease with application to prostate cancer.

AU - Etzioni, Ruth

AU - Kooperberg, Charles

AU - Pepe, Margaret

AU - Smith, Robert

AU - Gann, Peter H.

PY - 2003/1/1

Y1 - 2003/1/1

N2 - In early detection of disease, combinations of biomarkers promise improved discrimination over diagnostic tests based on single markers. An example of this is in prostate cancer screening, where additional markers have been sought to improve the specificity of the conventional Prostate-Specific Antigen (PSA) test. A marker of particular interest is the percent free PSA. Studies evaluating the benefits of percent free PSA reflect the need for a methodological approach that is statistically valid and useful in the clinical setting. This article presents methods that address this need. We focus on and-or combinations of biomarker results that we call logic rules and present novel definitions for the ROC curve and the area under the curve (AUC) that are applicable to this class of combination tests. Our estimates of the ROC and AUC are amenable to statistical inference including comparisons of tests and regression analysis. The methods are applied to data on free and total PSA levels among prostate cancer cases and matched controls enrolled in the Physicians' Health Study.

AB - In early detection of disease, combinations of biomarkers promise improved discrimination over diagnostic tests based on single markers. An example of this is in prostate cancer screening, where additional markers have been sought to improve the specificity of the conventional Prostate-Specific Antigen (PSA) test. A marker of particular interest is the percent free PSA. Studies evaluating the benefits of percent free PSA reflect the need for a methodological approach that is statistically valid and useful in the clinical setting. This article presents methods that address this need. We focus on and-or combinations of biomarker results that we call logic rules and present novel definitions for the ROC curve and the area under the curve (AUC) that are applicable to this class of combination tests. Our estimates of the ROC and AUC are amenable to statistical inference including comparisons of tests and regression analysis. The methods are applied to data on free and total PSA levels among prostate cancer cases and matched controls enrolled in the Physicians' Health Study.

UR - http://www.scopus.com/inward/record.url?scp=2942658010&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942658010&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/4.4.523

DO - 10.1093/biostatistics/4.4.523

M3 - Article

C2 - 14557109

AN - SCOPUS:2942658010

VL - 4

SP - 523

EP - 538

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 4

ER -