A Bayesian Approach for the Cox Proportional Hazards Model with Covariates Subject to Detection Limit
DOI:
https://doi.org/10.6000/1929-6029.2014.03.01.5Keywords:
Bayesian, Biomarker, Detection limit, Lung Injury, Proportional hazards modelsAbstract
The research on biomarkers has been limited in its effectiveness because biomarker levels can only be measured within the thresholds of assays and laboratory instruments, a challenge referred to as a detection limit (DL) problem. In this paper, we propose a Bayesian approach to the Cox proportional hazards model with explanatory variables subject to lower, upper, or interval DLs. We demonstrate that by formulating the time-to-event outcome using the Poisson density with counting process notation, implementing the proposed approach in the OpenBUGS and JAGS is straightforward. We have conducted extensive simulations to compare the proposed Bayesian approach to the other four commonly used methods and to evaluate its robustness with respect to the distribution assumption of the biomarkers. The proposed Bayesian approach and other methods were applied to an acute lung injury study, in which a panel of cytokine biomarkers was studied for the biomarkers’ association with ventilation-free survival.
References
Baker M. In biomarkers we trust? Nature biotechnology. 2005; 23(3): 297-304. http://dx.doi.org/10.1038/nbt0305-297 DOI: https://doi.org/10.1038/nbt0305-297
Ray P, et al. Statistical evaluation of a biomarker. Anesthesiology 2010; 112(4): 1023-40. http://dx.doi.org/10.1097/ALN.0b013e3181d47604 DOI: https://doi.org/10.1097/ALN.0b013e3181d47604
Morikawa T, et al. Association of CTNNB1 (β-catenin) alterations, body mass index, and physical activity with survival in patients with colorectal cancer. JAMA: the journal of the American Medical Association 2011; 305(16): 1685-94. http://dx.doi.org/10.1001/jama.2011.513 DOI: https://doi.org/10.1001/jama.2011.513
García-Bilbao A, et al. Identification of a biomarker panel for colorectal cancer diagnosis. BMC Cancer 2012; 12(1): 43. DOI: https://doi.org/10.1186/1471-2407-12-43
Braunwald E. Biomarkers in heart failure. New Engl J Med 2008; 358(20): 2148-59. http://dx.doi.org/10.1056/NEJMra0800239 DOI: https://doi.org/10.1056/NEJMra0800239
Patel DD. Prognostic significance of immunohistochemically localized biomarkers in stage II and stage III breast cancer: a multivariate analysis. Ann Surg Oncol 2000; 7(4): 305-11. http://dx.doi.org/10.1007/s10434-000-0305-5 DOI: https://doi.org/10.1007/s10434-000-0305-5
McShane LM, et al. Reporting recommendations for tumor marker prognostic studies (REMARK). J Natl Cancer Instit 2005; 97(16): 1180-84. http://dx.doi.org/10.1093/jnci/dji237 DOI: https://doi.org/10.1093/jnci/dji237
Hall P, Going J. Predicting the future: a critical appraisal of cancer prognosis studies. Histopathology 1999 35(6): 489-94. http://dx.doi.org/10.1046/j.1365-2559.1999.00862.x DOI: https://doi.org/10.1046/j.1365-2559.1999.00862.x
Cochran WG. Errors of measurement in statistics. Technometrics 1968; 10(4): 637-66. http://dx.doi.org/10.2307/1267450 DOI: https://doi.org/10.2307/1267450
Carroll RJ, et al. Measurement error in nonlinear models: a modern perspective. 2010: CRC press.
Schisterman EF, et al. The limitations due to exposure detection limits for regression models. Am J Epidemiol 2006; 163(4): 374-83. http://dx.doi.org/10.1093/aje/kwj039 DOI: https://doi.org/10.1093/aje/kwj039
Helsel DR. Nondetects and data analysis. Statistics for censored environmental data. 2005: Wiley-Interscience.
Cole SR, et al. Estimating the odds ratio when exposure has a limit of detection. Int J Epidemiol 2009; 38(6): 1674-80. http://dx.doi.org/10.1093/ije/dyp269 DOI: https://doi.org/10.1093/ije/dyp269
Nie L, et al. Linear regression with an independent variable subject to a detection limit. Epidemiology (Cambridge, Mass.) 2010; 21(Suppl 4): S17. http://dx.doi.org/10.1097/EDE.0b013e3181ce97d8 DOI: https://doi.org/10.1097/EDE.0b013e3181ce97d8
Henschel V, et al. A semiparametric Bayesian proportional hazards model for interval censored data with frailty effects. BMC Med Res Methodol 2009; 9(1): 9. DOI: https://doi.org/10.1186/1471-2288-9-9
Wang HJ, Feng X. Multiple Imputation for M-Regression With Censored Covariates. J Am Statist Assoc 2012; 107(497): 194-204. http://dx.doi.org/10.1080/01621459.2011.643198 DOI: https://doi.org/10.1080/01621459.2011.643198
Wu H, et al. A Bayesian approach for generalized linear models with explanatory biomarker measurement variables subject to detection limit: an application to acute lung injury. J Appl Statist 2012; 39(8): 1733-47. http://dx.doi.org/10.1080/02664763.2012.681362 DOI: https://doi.org/10.1080/02664763.2012.681362
Dagne GA, Huang Y. Bayesian semiparametric mixture Tobit models with left censoring, skewness, and covariate measurement errors. Statist Med 2013. DOI: https://doi.org/10.1002/sim.5799
Fellahi J-L, et al. Simultaneous Measurement of Cardiac Troponin I, B-type Natriuretic Peptide, and C-reactive Protein for the Prediction of Long-term Cardiac Outcome after Cardiac Surgery. Anesthesiology 2009; 111(2): 250-57. http://dx.doi.org/10.1097/ALN.0b013e3181a1f720 DOI: https://doi.org/10.1097/ALN.0b013e3181a1f720
Ware LB, et al. Prognostic and pathogenetic value of combining clinical and biochemical indices in patients with acute lung injury. CHEST J 2010; 137(2): 288-96. http://dx.doi.org/10.1378/chest.09-1484 DOI: https://doi.org/10.1378/chest.09-1484
Cox DR. Regression models and life-tables. Journal of the Royal Statistical Society. Series B (Methodological) 1972; 187-220. DOI: https://doi.org/10.1111/j.2517-6161.1972.tb00899.x
Ibrahim JG, Lipsitz SR, Chen MH. Missing covariates in generalized linear models when the missing data mechanism is non‐ignorable. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 1999; 61(1): 173-90. http://dx.doi.org/10.1111/1467-9868.00170 DOI: https://doi.org/10.1111/1467-9868.00170
Chen Q, Ibrahim JG. Semiparametric models for missing covariate and response data in regression models. Biometrics 2006; 62(1): 177-84. http://dx.doi.org/10.1111/j.1541-0420.2005.00438.x DOI: https://doi.org/10.1111/j.1541-0420.2005.00438.x
Clayton DG. A Monte Carlo method for Bayesian inference in frailty models. Biometrics 1991; 467-485. http://dx.doi.org/10.2307/2532139 DOI: https://doi.org/10.2307/2532139
Andersen PK, Gill RD. Cox's regression model for counting processes: a large sample study. Ann Statist 1982; 1100-1120. http://dx.doi.org/10.1214/aos/1176345976 DOI: https://doi.org/10.1214/aos/1176345976
Kalbfleisch JD. Non-parametric Bayesian analysis of survival time data. Journal of the Royal Statistical Society. Series B (Methodological) 1978; 214-221. DOI: https://doi.org/10.1111/j.2517-6161.1978.tb01666.x
Koch AL. The logarithm in biology 1. Mechanisms generating the log-normal distribution exactly. J Theoret Biol 1966; 12(2): 276-90. http://dx.doi.org/10.1016/0022-5193(66)90119-6 DOI: https://doi.org/10.1016/0022-5193(66)90119-6
Limpert E, Stahel WA, Abbt M. Log-normal distributions across the sciences: keys and clues. BioScience 2001; 51(5): 341-52. http://dx.doi.org/10.1641/0006-3568(2001)051%5B0341:LNDATS%5D2.0.CO;2 DOI: https://doi.org/10.1641/0006-3568(2001)051[0341:LNDATS]2.0.CO;2
Ibrahim JG, Chen MH, Sinha D. Bayesian survival analysis. 2005: Wiley Online Library.
Rubin DB. Inference and missing data. Biometrika 1976; 63(3): 581-92. http://dx.doi.org/10.1093/biomet/63.3.581 DOI: https://doi.org/10.1093/biomet/63.3.581
Lubin JH, et al. Epidemiologic evaluation of measurement data in the presence of detection limits. Environ Health Perspect 2004; 112(17): 1691. DOI: https://doi.org/10.1289/ehp.7199
De Groot J, et al. Multiple imputation to correct for partial verification bias revisited. Statist Med 2008; 27(28): 5880-89. http://dx.doi.org/10.1002/sim.3410 DOI: https://doi.org/10.1002/sim.3410
Harrell FE. Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis. 2001: Springer. DOI: https://doi.org/10.1007/978-1-4757-3462-1
Little R, Hyonggin A. Robust likelihood-based analysis of multivariate data with missing values 2003.
Siddique J, Belin TR. Multiple imputation using an iterative hot‐deck with distance‐based donor selection. Statist Med 2008; 27(1): 83-102. http://dx.doi.org/10.1002/sim.3001 DOI: https://doi.org/10.1002/sim.3001
Hopke PK, Liu C, Rubin DB. Multiple Imputation for Multivariate Data with Missing and Below‐Threshold Measurements: Time‐Series Concentrations of Pollutants in the Arctic. Biometrics 2001; 57(1): 22-33. http://dx.doi.org/10.1111/j.0006-341X.2001.00022.x DOI: https://doi.org/10.1111/j.0006-341X.2001.00022.x
Uh H-W, et al. Evaluation of regression methods when immunological measurements are constrained by detection limits. BMC Immunol 2008; 9(1): 59. http://dx.doi.org/10.1186/1471-2172-9-59 DOI: https://doi.org/10.1186/1471-2172-9-59
Rubin DB. Multiple imputation for nonresponse in surveys. 2009; Vol. 307. Wiley. com.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2014 Qingxia Chen, Huiyun Wu, Lorraine B. Ware, Tatsuki Koyama
![Creative Commons License](http://i.creativecommons.org/l/by/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Policy for Journals/Articles with Open Access
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are permitted and encouraged to post links to their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
Policy for Journals / Manuscript with Paid Access
Authors who publish with this journal agree to the following terms:
- Publisher retain copyright .
- Authors are permitted and encouraged to post links to their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work .