Extension of the SAEM algorithm to left-censored data in nonlinear mixed-effects model: Application to HIV dynamics model

The reduction of viral load is frequently used as a primary endpoint in HIV clinical trials. Nonlinear mixed-effects models are thus proposed to model this decrease of the viral load after initiation of treatment and to evaluate the intra- and inter-patient variability. However, left censoring due to quantification limits in the viral load measurement is an additional challenge in the analysis of longitudinal HIV data. An extension of the stochastic approximation expectation-maximization (SAEM) algorithm is proposed to estimate parameters of these models. This algorithm includes the simulation of the left-censored data in a right-truncated Gaussian distribution. Simulation results show that the proposed estimates are less biased than the usual naive methods of handling such data: omission of all censored data points, or imputation of half the quantification limit to the first point below the limit and omission of the following points. The viral load measurements obtained in the TRIANON-ANRS81 clinical trial are analyzed with this method and a significant difference is found between the two treatment groups of this trial.

[1]  L B Sheiner,et al.  Estimating population kinetics. , 1982, Critical reviews in biomedical engineering.

[2]  J. Hughes,et al.  Mixed Effects Models with Censored Data with Application to HIV RNA Levels , 1999, Biometrics.

[3]  V De Gruttola,et al.  Estimation of HIV dynamic parameters. , 1998, Statistics in medicine.

[4]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[5]  Alan S. Perelson,et al.  Decay characteristics of HIV-1-infected compartments during combination therapy , 1997, Nature.

[6]  A. Ding,et al.  Assessing antiviral potency of anti-HIV therapies in vivo by comparing viral decay rates in viral dynamic models. , 2001, Biostatistics.

[7]  Lang Wu,et al.  A Joint Model for Nonlinear Mixed-Effects Models With Censoring and Covariates Measured With Error, With Application to AIDS Studies , 2002 .

[8]  C. Robert.  Simulation of truncated normal variables , 2009, 0907.4010.

[9]  E. Kuhn,et al.  Coupling a stochastic approximation version of EM with an MCMC procedure , 2004 .

[10]  Stuart L. Beal,et al.  Ways to Fit a PK Model with Some Data Below the Quantification Limit , 2001, Journal of Pharmacokinetics and Pharmacodynamics.

[11]  É. Moulines,et al.  Convergence of a stochastic approximation version of the EM algorithm , 1999 .

[12]  D Commenges,et al.  Analysis of left-censored longitudinal data with application to viral load in HIV infection. , 2000, Biostatistics.

[13]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[14]  D. Rubin,et al.  Parameter expansion to accelerate EM: The PX-EM algorithm , 1998 .

[15]  Russell D. Wolfinger,et al.  Laplace's approximation for nonlinear mixed models. , 1993 .

[16]  Lang Wu.  Exact and Approximate Inferences for Nonlinear Mixed-Effects Models With Missing Covariates , 2004 .

[17]  Philippe Flandre,et al.  Nevirapine or lamivudine plus stavudine and indinavir: examples of 2-class versus 3-class regimens for the treatment of human immunodeficiency virus type 1. , 2002, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America.

[18]  Mats O. Karlsson,et al.  Impact of Omission or Replacement of Data Below the Limit of Quantification on Parameter Estimates in a Two-Compartment Model , 2002, Pharmaceutical Research.

[19]  France Mentré,et al.  The SAEM algorithm for group comparison tests in longitudinal data analysis based on non‐linear mixed‐effects model , 2007, Statistics in medicine.

[20]  A A Ding,et al.  Relationships between antiviral treatment effects and biphasic viral decay rates in modeling HIV dynamics. , 1999, Mathematical biosciences.

[21]  T. Louis.  Finding the Observed Information Matrix When Using the EM Algorithm , 1982 .

[22]  Hulin Wu,et al.  A Comparison Study of Models and Fitting Procedures for Biphasic Viral Dynamics in HIV‐1 Infected Patients Treated with Antiviral Therapies , 2000, Biometrics.

[23]  Hulin Wu,et al.  The study of long‐term HIV dynamics using semi‐parametric non‐linear mixed‐effects models , 2002, Statistics in medicine.

[24]  D. Bates,et al.  Nonlinear mixed effects models for repeated measures data. , 1990, Biometrics.

[25]  Hulin Wu,et al.  Identification of significant host factors for HIV dynamics modelled by non‐linear mixed‐effects models , 2002, Statistics in medicine.

[26]  L Wu,et al.  A multiple imputation method for missing covariates in non‐linear mixed‐effects models with application to HIV dynamics , 2001, Statistics in medicine.

[27]  Marc Lavielle,et al.  Maximum likelihood estimation in nonlinear mixed effects models , 2005, Comput. Stat. Data Anal..