PMID- 29082535
OWN - NLM
STAT- MEDLINE
DCOM- 20191004
LR  - 20200306
IS  - 1097-0258 (Electronic)
IS  - 0277-6715 (Print)
IS  - 0277-6715 (Linking)
VI  - 37
IP  - 4
DP  - 2018 Feb 20
TI  - Combining multiple biomarkers linearly to maximize the partial area under the ROC 
      curve.
PG  - 627-642
LID - 10.1002/sim.7535 [doi]
AB  - It is now common in clinical practice to make clinical decisions based on 
      combinations of multiple biomarkers. In this paper, we propose new approaches for 
      combining multiple biomarkers linearly to maximize the partial area under the 
      receiver operating characteristic curve (pAUC). The parametric and nonparametric 
      methods that have been developed for this purpose have limitations. When the 
      biomarker values for populations with and without a given disease follow a 
      multivariate normal distribution, it is easy to implement our proposed parametric 
      approach, which adopts an alternative analytic expression of the pAUC. When 
      normality assumptions are violated, a kernel-based approach is presented, which 
      handles multiple biomarkers simultaneously. We evaluated the proposed as well as 
      existing methods through simulations and discovered that when the covariance 
      matrices for the disease and nondisease samples are disproportional, traditional 
      methods (such as the logistic regression) are more likely to fail to maximize the 
      pAUC while the proposed methods are more robust. The proposed approaches are 
      illustrated through application to a prostate cancer data set, and a rank-based 
      leave-one-out cross-validation procedure is proposed to obtain a realistic 
      estimate of the pAUC when there is no independent validation set available.
CI  - Copyright (c) 2017 John Wiley & Sons, Ltd.
FAU - Yan, Qingxiang
AU  - Yan Q
AUID- ORCID: 0000-0003-3842-3933
AD  - Department of Biostatistics, The University of Texas MD Anderson Cancer Center, 
      Houston, TX, 77030, USA.
FAU - Bantis, Leonidas E
AU  - Bantis LE
AUID- ORCID: 0000-0002-2364-2562
AD  - Department of Biostatistics, The University of Texas MD Anderson Cancer Center, 
      Houston, TX, 77030, USA.
FAU - Stanford, Janet L
AU  - Stanford JL
AD  - Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, 
      Seattle, WA, 98195, USA.
AD  - Department of Epidemiology, University of Washington School of Public Health, 
      Seattle, WA, 98109, USA.
FAU - Feng, Ziding
AU  - Feng Z
AD  - Department of Biostatistics, The University of Texas MD Anderson Cancer Center, 
      Houston, TX, 77030, USA.
LA  - eng
GR  - P50 CA097186/CA/NCI NIH HHS/United States
GR  - R01 GM106177/GM/NIGMS NIH HHS/United States
GR  - U01 DK108328/DK/NIDDK NIH HHS/United States
GR  - U24 CA086368/CA/NCI NIH HHS/United States
PT  - Journal Article
PT  - Research Support, N.I.H., Extramural
DEP - 20171030
PL  - England
TA  - Stat Med
JT  - Statistics in medicine
JID - 8215016
RN  - 0 (Biomarkers)
SB  - IM
MH  - Algorithms
MH  - *Area Under Curve
MH  - Biomarkers/*analysis
MH  - Biostatistics
MH  - Computer Simulation
MH  - DNA Methylation/genetics
MH  - Disease Progression
MH  - Humans
MH  - Linear Models
MH  - Logistic Models
MH  - Male
MH  - Medical Overuse/prevention & control
MH  - Normal Distribution
MH  - Prostatic Neoplasms/diagnosis/genetics/therapy
MH  - ROC Curve
MH  - Statistics, Nonparametric
PMC - PMC6469690
MID - NIHMS914891
OTO - NOTNLM
OT  - ROC analysis
OT  - logistic regression
OT  - optimal linear combination
OT  - pAUC
OT  - parametric and nonparametric
EDAT- 2017/10/31 06:00
MHDA- 2019/10/08 06:00
PMCR- 2019/04/17
CRDT- 2017/10/31 06:00
PHST- 2016/05/27 00:00 [received]
PHST- 2017/09/17 00:00 [revised]
PHST- 2017/09/26 00:00 [accepted]
PHST- 2017/10/31 06:00 [pubmed]
PHST- 2019/10/08 06:00 [medline]
PHST- 2017/10/31 06:00 [entrez]
PHST- 2019/04/17 00:00 [pmc-release]
AID - 10.1002/sim.7535 [doi]
PST - ppublish
SO  - Stat Med. 2018 Feb 20;37(4):627-642. doi: 10.1002/sim.7535. Epub 2017 Oct 30.