Opening the black box of biomarker measurement error.

Epidemiology (Cambridge, Mass.) (Impact Factor: 6.18). 07/2010; 21 Suppl 4:S1-3. DOI: 10.1097/EDE.0b013e3181dda514
Source: PubMed
  • [Show abstract] [Hide abstract]
    ABSTRACT: Modeling survival data with a set of covariates usually assumes that the values of the covariates are fully observed. However, in a variety of applications, some values of a covariate may be left-censored due to inadequate instrument sensitivity to quantify the biospecimen. When data are left-censored, the true values are missing but are known to be smaller than the detection limit. The most commonly used ad-hoc method to deal with nondetect values is to substitute the nondetect values by the detection limit. Such ad-hoc analysis of survival data with an explanatory variable subject to left-censoring may provide biased and inefficient estimators of hazard ratios and survivor functions. We consider a parametric proportional hazards model to analyze time-to-event data. We propose a likelihood method for the estimation and inference of model parameters. In this likelihood approach, instead of replacing the nondetect values by the detection limit, we adopt a numerical integration technique to evaluate the observed data likelihood in the presence of a left-censored covariate. Monte Carlo simulations were used to demonstrate various properties of the proposed regression estimators including the consistency and efficiency. The simulation study shows that the proposed likelihood approach provides approximately unbiased estimators of the model parameters. The proposed method also provides estimators that are more efficient than those obtained under the ad-hoc method. Also, unlike the ad-hoc estimators, the coverage probabilities of the proposed estimators are at their nominal level. Analysis of a large cohort study, genetic and inflammatory marker of sepsis study, shows discernibly different results based on the proposed method. Naive use of detection limit in a parametric survival model may provide biased and inefficient estimators of hazard ratios and survivor functions. The proposed likelihood approach provides approximately unbiased and efficient estimators of hazard ratios and survivor functions.
    01/2012; Suppl 3(2). DOI:10.4172/2155-6180.S3-002
  • [Show abstract] [Hide abstract]
    ABSTRACT: MALDI-TOF profiling of low molecular weight (LMW) peptides (peptidome) usage is limited due to the lack of reproducibility from the confounding inferences of sample preparation, data acquisition and processing. We applied MALDI-TOF analysis to profile urine peptidome with the aims to: 1) compare centrifugal ultrafiltration and dialysis pre-treatments, 2) determine whether using signal LOD (sLOD), together with data normalization, may reduce MALDI-TOF variability. We also investigated the influence of peaks detection on reproducibility. Dialysis allowed to obtain better MALDI-TOF spectra than ultrafiltration. Within the 1000 to 4000 m/z range, we identified 120 and 129 peaks in intra- and inter-assay studies respectively. To estimate the sLOD, serial dilution of pooled urines up to 1/256 were analysed in triplicate. Six data normalization strategies were investigated - the mean, median, internal standard, relative intensity, TIC and linear rescaling normalization. Normalization methods alone performed poorly in reducing features variability while when combined to sLOD adjustment showed an overall reduction in features CVs. Applying a feedback signal processing approach, after median normalization and sLOD adjustment, CVs were reduced from 103% to 26% and 113% to 25% for the intra- and inter-assay respectively, and spectra became more comparable in terms of data dispersion. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
    Proteomics 02/2015; 15(9). DOI:10.1002/pmic.201400253 · 3.97 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We consider generalized linear regression analysis with left-censored covariate due to the lower limit of detection. Complete case analysis by eliminating observations with values below limit of detection yields valid estimates for regression coefficients, but loses efficiency; substitution methods are biased; maximum likelihood method relies on parametric models for the unobservable tail probability distribution of such covariate, thus may suffer from model misspecification. To obtain robust and more efficient results, we propose a semiparametric likelihood-based approach for the estimation of regression parameters using an accelerated failure time model for the covariate subject to limit of detection. A two-stage estimation procedure is considered, where the conditional distribution of the covariate with limit of detection given other variables is estimated prior to maximizing the likelihood function. The proposed method outperforms the complete case analysis and the substitution methods as well in simulation studies. Technical conditions for desirable asymptotic properties are provided.


Available from