Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Modeling Heterogeneity in the Assessment of Treatment Effects on Tumor Development While Accounting for Monotone Dropout

Modeling Heterogeneity in the Assessment of Treatment Effects on Tumor Development While... Our research is motivated by a dose-finding trial study in standard chemotherapy in melanoma involving 62 patients, whose tumor size as biomarker expression and toxicity grade were measured at two cycles of treatments. The effects of different dose levels of a treatment on tumor development are the purpose of these studies. In this study, we model the clustered tumor data with dropouts by pattern mixture Tweedie mixed models where unobserved latent risks are characterized by random effects. An optimal estimation of our model has been done using the orthodox best linear unbiased predictors (BLUP) of random effects. The predicted latent risks help us classify patients into high- and low-risk groups by using cluster analysis approach. We illustrated the method with analysis of the clustered tumor development data. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bulletin of the Malaysian Mathematical Sciences Society Springer Journals

Modeling Heterogeneity in the Assessment of Treatment Effects on Tumor Development While Accounting for Monotone Dropout

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/modeling-heterogeneity-in-the-assessment-of-treatment-effects-on-tumor-rLpy6VZ3ej
Publisher
Springer Journals
Copyright
Copyright © Malaysian Mathematical Sciences Society and Penerbit Universiti Sains Malaysia 2022
ISSN
0126-6705
eISSN
2180-4206
DOI
10.1007/s40840-021-01225-5
Publisher site
See Article on Publisher Site

Abstract

Our research is motivated by a dose-finding trial study in standard chemotherapy in melanoma involving 62 patients, whose tumor size as biomarker expression and toxicity grade were measured at two cycles of treatments. The effects of different dose levels of a treatment on tumor development are the purpose of these studies. In this study, we model the clustered tumor data with dropouts by pattern mixture Tweedie mixed models where unobserved latent risks are characterized by random effects. An optimal estimation of our model has been done using the orthodox best linear unbiased predictors (BLUP) of random effects. The predicted latent risks help us classify patients into high- and low-risk groups by using cluster analysis approach. We illustrated the method with analysis of the clustered tumor development data.

Journal

Bulletin of the Malaysian Mathematical Sciences Society – Springer Journals

Published: Jan 4, 2022

Keywords: Best linear unbiased predictor; Clustered data; Missing data; Random effects models; Tweedie family; 62F10; 62J12

References

  • Series evaluation of Tweedie exponential dispersion model densities
    Dunn, PK; Smyth, GK
  • Joint modelling of a binary and a continuous outcome measured at two cycles to determine the optimal dose
    Ezzalfani, M; Burzykowski, T; Paoletti, X
  • Application of random-effects pattern-mixture models for missing data in longitudinal studies
    Hedeker, D; Gibbons, RD
  • Exponential dispersion models (with discussion)
    Jørgensen, B
  • Bayesian dose-finding in two treatment cycles based on the joint utility of efficacy and toxicity
    Lee, J; Thall, PF; Ji, Y; Muller, P
  • Pattern-mixture models for multivariate incomplete data
    Little, RJA
  • A class of pattern-mixture models for normal incomplete data
    Little, RJA
  • Modeling the drop-out mechanism in repeated measures studies
    Little, RJA
  • Random effects Cox models: a poisson modelling approach
    Ma, R; Krewski, D; Burnett, RT
  • Nested generalized linear mixed models: an orthodox best linear unbiased predictor approach
    Ma, R; Jørgensen, B
  • Ckmeans.1d.dp: Optimal k\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$k$$\end{document}-means clustering in one dimension by dynamic programming
    Wang, H; Song, M