×

Dose response signal detection under model uncertainty. (English) Zbl 1419.62338

Summary: We investigate likelihood ratio contrast tests for dose response signal detection under model uncertainty, when several competing regression models are available to describe the dose response relationship. The proposed approach uses the complete structure of the regression models, but does not require knowledge of the parameters of the competing models. Standard likelihood ratio test theory is applicable in linear models as well as in nonlinear regression models with identifiable parameters. However, for many commonly used nonlinear dose response models the regression parameters are not identifiable under the null hypothesis of no dose response and standard arguments cannot be used to obtain critical values. We thus derive the asymptotic distribution of likelihood ratio contrast tests in regression models with a lack of identifiability and use this result to simulate the quantiles based on Gaussian processes. The new method is illustrated with a real data example and compared to existing procedures using theoretical investigations as well as simulations.

MSC:

62P10 Applications of statistics to biology and medical sciences; meta analysis
62F03 Parametric hypothesis testing
62E20 Asymptotic distribution theory in statistics

Software:

bnpmr
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Akacha, The impact of dropouts on the analysis of dose-finding studies with recurrent event data, Statistics in Medicine 29 pp 1635– (2010)
[2] Andrews, Optimal tests when a nuisance parameter is present only under the alternative, Econometrica 62 pp 1383– (1994) · Zbl 0815.62033 · doi:10.2307/2951753
[3] Andrews, Admissibility of the likelihood ratio test when a nuisance parameter is present only under the alternative, Annals of Statistics 23 pp 1609– (1995) · Zbl 0843.62023 · doi:10.1214/aos/1176324316
[4] Benda, Model-based approaches for time-dependent dose finding with repeated binary data, Statistics in Medicine 29 pp 1096– (2010)
[5] Biesheuvel, Many-to-one comparisons in stratified designs, Biometrical Journal 44 pp 101– (2002) · Zbl 1052.62075 · doi:10.1002/1521-4036(200201)44:1<101::AID-BIMJ101>3.0.CO;2-H
[6] Bornkamp , B. 2006 www.statistik.tu-dortmund.de/bornkamp/diplom.pdf
[7] Bornkamp, Response-adaptive dose-finding under model uncertainty, Annals of Applied Statistics 5 pp 1611– (2011) · Zbl 1223.62158 · doi:10.1214/10-AOAS445
[8] Bornkamp, Innovative approaches for designing and analyzing adaptive dose-ranging trials, Journal of Biopharmaceutical Statistics 17 pp 965– (2007) · doi:10.1080/10543400701643848
[9] Bornkamp, Bayesian nonparametric estimation of continuous monotone functions with applications to dose-response analysis, Biometrics 65 pp 198– (2009) · Zbl 1159.62023 · doi:10.1111/j.1541-0420.2008.01060.x
[10] Bornkamp , B. Pinheiro , J. Bretz , F. 2010
[11] Bretz, Dose finding ’ a challenge in statistics, Biometrical Journal 50 pp 480– (2008) · doi:10.1002/bimj.200810438
[12] Bretz, Combining multiple comparisons and modeling techniques in dose-response studies, Biometrics 61 pp 738– (2005) · Zbl 1079.62105 · doi:10.1111/j.1541-0420.2005.00344.x
[13] Bretz, On a hybrid method in dose finding studies, Methods of Information in Medicine 43 pp 457– (2004)
[14] Chernoff, On the distribution of the likelihood ratio, Annals of Mathematical Statistics 25 pp 573– (1954) · Zbl 0056.37102 · doi:10.1214/aoms/1177728725
[15] CHMP 2014 http://www.ema.europa.eu/docs/en_GB/document_library/Regulatory_and_procedural_guideline/2014/02/WC500161027.pdf
[16] Claeskens , G. Hjort , N. L. 2008
[17] Davies, Hypothesis testing when a nuisance parameter is present only under the alternative, Biometrika 64 pp 247– (1977) · Zbl 0362.62026 · doi:10.1093/biomet/64.2.247
[18] Dette, Optimal designs for the EMAX, log-linear and exponential models, Biometrika 97 pp 513– (2010) · Zbl 1233.62146 · doi:10.1093/biomet/asq020
[19] Dette, A note on nonparametric estimation of the effective dose in quantal bioassay, Journal of the American Statistical Association 100 pp 503– (2005) · Zbl 1117.62317 · doi:10.1198/016214504000001493
[20] Genz , A. Bretz , F. 2009
[21] Gutjahr, Likelihood ratio tests for a dose-response effect using multiple nonlinear regression models, Biometrics (2015)
[22] Hotelling, Tubes and spheres in n-spaces, and a class of statistical problems, American Journal of Mathematics 61 pp 440– (1939) · JFM 65.0795.02 · doi:10.2307/2371512
[23] Hothorn, Analysis of randomised dose-finding studies: Closure test modifications based on multiple contrast tests, Biometrical Journal 39 pp 467– (1997) · Zbl 0890.62080 · doi:10.1002/bimj.4710390408
[24] ICH 1994
[25] Johnston, Linaclotide improves abdominal pain and bowel habits in a Phase II study of patients with irritable bowel syndrome with constipation, Gastroenterology 139 pp 1877– (2010) · doi:10.1053/j.gastro.2010.08.041
[26] Klingenberg, Proof of concept and dose estimation with binary responses under model uncertainty, Statistics in Medicine 28 pp 274– (2009) · doi:10.1002/sim.3477
[27] Lin, Multiple contrast test for detecting monotonic dose-response relationship and FDR-adjusted confidence intervals for selected parameters in a microarray setting, Online Journal of Bioinformatics 10 pp 67– (2009)
[28] Lin , D. Shkedy , Z. Yekutieli , D. Amaratunga , D. Bijnens , L. 2012
[29] Lindsay , B. G. 1995
[30] Liu, Asymptotics for likelihood ratio tests under loss of identifiability, Annals of Statistics 31 pp 807– (2003) · Zbl 1032.62014 · doi:10.1214/aos/1056562463
[31] Miller, Adaptive dose-finding: Proof of concept with type I error control, Biometrical Journal 52 pp 577– (2010) · Zbl 1201.62131 · doi:10.1002/bimj.200900222
[32] Morgan , B. 1992
[33] Neal, Hypothesis testing and Bayesian estimation using a sigmoid E(max) model applied to sparse dose-response designs, Journal of Biopharmaceutical Statistics 16 pp 657– (2006) · doi:10.1080/10543400600860469
[34] Pinheiro, Model-based dose finding under model uncertainty using general parametric models, Statistics in Medicine 33 pp 1646– (2014) · doi:10.1002/sim.6052
[35] Pinheiro, Analysis of dose-response studies-modeling approaches, Dose Finding in Drug Development pp 146– (2006) · doi:10.1007/0-387-33706-7_10
[36] Prakasa-Rao , B. L. S. 1992
[37] Ruberg, Dose response studies. I. Some design considerations, Journal of Biopharmaceutical Statistics 5 pp 1– (1995a) · doi:10.1080/10543409508835096
[38] Ruberg, Dose response studies. II. Analysis and interpretation, JBS 5 pp 15– (1995b)
[39] Seber , G. A. F. Wild , C. J. 1989
[40] Song, On asymptotically optimal tests under loss of identifiability in semiparametric models, Annals of Statistics 37 pp 2409– (2009) · Zbl 1173.62039 · doi:10.1214/08-AOS643
[41] Stewart, Detecting dose-response with contrasts, Statistics in Medicine 19 pp 913– (2000) · doi:10.1002/(SICI)1097-0258(20000415)19:7<913::AID-SIM397>3.0.CO;2-2
[42] Tamhane, Multiple test procedures for dose finding, Biometrics 52 pp 21– (1997) · Zbl 0906.62115 · doi:10.2307/2533141
[43] Tamhane, Multiple test procedures for identifying the maximum safe dose, Journal of the American Statistical Association 96 pp 835– (2001) · Zbl 1072.62672 · doi:10.1198/016214501753208546
[44] Tanaka, An adaptive two-stage dose-response design method for establishing proof of concept, Journal of Biopharmaceutical Statistics (2012)
[45] Tukey, Testing the statistical certainty of a response to increasing doses of a drug, Biometrics 41 pp 295– (1985) · Zbl 0613.62131 · doi:10.2307/2530666
[46] Verrier, Dose-finding studies, MCP-Mod, model selection, and model averaging: Two applications in the real world, Clinical Trials 11 pp 476– (2014) · doi:10.1177/1740774514532723
[47] Wakana, A method for therapeutic dose selection in a phase II clinical trial using contrast statistics, Statistics in Medicine 26 pp 498– (2007) · doi:10.1002/sim.2552
[48] Wilks, The large-sample distribution of the likelihood ratio for testing composite hypotheses, Annals of Mathematical Statistics 9 pp 60– (1938) · Zbl 0018.32003 · doi:10.1214/aoms/1177732360
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.