Spline-based models for predictiveness curves and surfaces.

A biomarker is defined to be a biological characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention. The use of biomarkers in cancer has been advocated for a variety of purposes, which include use as surrogate endpoints, early detection of disease, proxies for environmental exposure and risk prediction. We deal with the latter issue in this paper.Several authors have proposed use of the predictiveness curve for assessing the capacity of a biomarker for risk prediction. For most situations, it is reasonable to assume monotonicity of the biomarker effects on disease risk. In this article, we propose the use of flexible modelling of the predictiveness curve and its bivariate analogue, the predictiveness surface, through the use of spline algorithms that incorporate the appropriate monotonicity constraints. Estimation proceeds through use of a two-step algorithm that represents the "smooth, then monotonize" approach. Subsampling procedures are used for inference. The methods are illustrated to data from a melanoma study.

[1]  Susan A. Murphy,et al.  Semiparametric likelihood ratio inference , 1997 .

[2]  Ziding Feng,et al.  Evaluating the Predictiveness of a Continuous Marker , 2007, Biometrics.

[3]  R. Dykstra,et al.  An Algorithm for Isotonic Regression for Two or More Independent Variables , 1982 .

[4]  F. T. Wright,et al.  Likelihood ratio tests involving a bivariate trend in two-factor designs: the level probabilities , 1994 .

[5]  Pranab Kumar Sen,et al.  An appraisal of some aspects of statistical inference under inequality constraints , 2002 .

[6]  Michael G Hudgens,et al.  Evaluating Candidate Principal Surrogate Endpoints , 2008, Biometrics.

[7]  Lutz Dümbgen,et al.  Least squares and shrinkage estimation under bimonotonicity constraints , 2010, Stat. Comput..

[8]  Allan R. Sampson,et al.  Structure Algorithms for Partially Ordered Isotonic Regression , 1994 .

[9]  F. T. Wright,et al.  Order restricted statistical inference , 1988 .

[10]  UsingSmoothing SplinesbyXihong Liny,et al.  Inference in Generalized Additive Mixed Models , 1999 .

[11]  E. Mammen,et al.  A General Projection Framework for Constrained Smoothing , 2001 .

[12]  Steven Stern,et al.  Feasible Nonparametric Estimation of Multiargument Monotone Functions , 1994 .

[13]  R. Pyke,et al.  Logistic disease incidence models and case-control studies , 1979 .

[14]  D. Ghosh,et al.  Semiparametric binary regression under monotonicity constraints , 2004 .

[15]  M. Wand,et al.  Semiparametric Regression: Parametric Regression , 2003 .

[16]  Jon A. Wellner,et al.  Likelihood Ratio Tests for Monotone Functions , 2001 .

[17]  B. Silverman,et al.  Nonparametric regression and generalized linear models , 1994 .

[18]  D. Ghosh.  Incorporating monotonicity into the evaluation of a biomarker. , 2007, Biostatistics.

[19]  Douglas W. Nychka,et al.  FUNFITS: data analysis and statistical tools for estimating functions , 2008 .

[20]  William F. Eddy,et al.  An Algorithm for Isotonic Regression on Ordered Rectangular Grids , 1996 .

[21]  J. Wellner,et al.  Information Bounds and Nonparametric Maximum Likelihood Estimation , 1992 .

[22]  Terry Speed,et al.  [That BLUP is a Good Thing: The Estimation of Random Effects]: Comment , 1991 .

[23]  Jeremy MG Taylor,et al.  Mitotic Rate and Younger Age Are Predictors of Sentinel Lymph Node Positivity: Lessons Learned From the Generation of a Probabilistic Model , 2004, Annals of Surgical Oncology.