A flexible multivariate random effects proportional odds model with application to adverse effects during radiation therapy

Nicole Augustin, Sung Won Kim, Annemarie Uhlig, Christina Hanser, Michael Henke, Martin Schumacher

Research output: Contribution to journalArticle

Abstract

Radiation therapy in patients with head and neck cancer has a toxic effect onmucosa, the soft tissue in and around the mouth. Hence mucositis is a serious common side effect and is a condition characterized by pain and inflammation ofthe surface of the mucosa. Although the mucosa recovers during breaks of and following the radiotherapy course the recovery will depend on the type of tissue involved and on its location. We present a novel flexible multivariate random effects proportional odds model which takes account of the longitudinal course of oral mucositis at different mouth sites and of the radiation dosage(in terms of cumulative dose). The model is an extension of the {\emproportional odds model} which is used for ordinal response variables. Our model includes the ordinal multivariate response of the mucositis score by location, random intercepts for individuals and includes a non-linear function of cumulative radiation dose. The model allows to test whether sensitivitydiffers by mouth sites after having adjusted for site specific cumulative radiation dose. The model also allows to check whether and how the (non-linear) effect of site specific dose differs by site. We fit the model to longitudinal patient data from a prospective observation and find that after adjusting for cumulative dose, upper, lower lips and mouth floor are associated with the lowest mucositis scores and hard and soft palate are associated with the highest mucositis scores. This implies the possibility that tissues at different mouth locations differ in their sensitivity to the toxic effect of radiation. We also find that cumulative dose followed by mouth site are the strongest predictors of mucositis, and the effects of age and gender are negligible.
LanguageEnglish
Article number201600142
Pages1339–1351
JournalBiometrical Journal
Volume59
Issue number6
Early online date10 Jul 2017
DOIs
StatusPublished - 15 Nov 2017

Fingerprint

Proportional Odds Model
Radiation Therapy
Random Effects Model
Dose
Radiation
Model
Multivariate Response
Inflammation
Odds
Radiotherapy
Soft Tissue
Intercept
Pain
Nonlinear Effects
Nonlinear Function
Therapy
Random effects
Predictors
Lowest
Cancer

Keywords

  • generalised additive mixed model, proportional odds model, cumulative threshold model, radiation therapy, multivariate repeated measures, ordered categorical response, biologically effective dose (BED), mucositis

Cite this

A flexible multivariate random effects proportional odds model with application to adverse effects during radiation therapy. / Augustin, Nicole; Kim, Sung Won; Uhlig, Annemarie; Hanser, Christina; Henke, Michael; Schumacher, Martin.

In: Biometrical Journal, Vol. 59, No. 6, 201600142, 15.11.2017, p. 1339–1351.

Research output: Contribution to journalArticle

Augustin, Nicole ; Kim, Sung Won ; Uhlig, Annemarie ; Hanser, Christina ; Henke, Michael ; Schumacher, Martin. / A flexible multivariate random effects proportional odds model with application to adverse effects during radiation therapy. In: Biometrical Journal. 2017 ; Vol. 59, No. 6. pp. 1339–1351.
@article{afb35c2aa5fb4922a8f5939d50c0e3f8,
title = "A flexible multivariate random effects proportional odds model with application to adverse effects during radiation therapy",
abstract = "Radiation therapy in patients with head and neck cancer has a toxic effect onmucosa, the soft tissue in and around the mouth. Hence mucositis is a serious common side effect and is a condition characterized by pain and inflammation ofthe surface of the mucosa. Although the mucosa recovers during breaks of and following the radiotherapy course the recovery will depend on the type of tissue involved and on its location. We present a novel flexible multivariate random effects proportional odds model which takes account of the longitudinal course of oral mucositis at different mouth sites and of the radiation dosage(in terms of cumulative dose). The model is an extension of the {\emproportional odds model} which is used for ordinal response variables. Our model includes the ordinal multivariate response of the mucositis score by location, random intercepts for individuals and includes a non-linear function of cumulative radiation dose. The model allows to test whether sensitivitydiffers by mouth sites after having adjusted for site specific cumulative radiation dose. The model also allows to check whether and how the (non-linear) effect of site specific dose differs by site. We fit the model to longitudinal patient data from a prospective observation and find that after adjusting for cumulative dose, upper, lower lips and mouth floor are associated with the lowest mucositis scores and hard and soft palate are associated with the highest mucositis scores. This implies the possibility that tissues at different mouth locations differ in their sensitivity to the toxic effect of radiation. We also find that cumulative dose followed by mouth site are the strongest predictors of mucositis, and the effects of age and gender are negligible.",
keywords = "generalised additive mixed model, proportional odds model, cumulative threshold model, radiation therapy, multivariate repeated measures, ordered categorical response, biologically effective dose (BED), mucositis",
author = "Nicole Augustin and Kim, {Sung Won} and Annemarie Uhlig and Christina Hanser and Michael Henke and Martin Schumacher",
year = "2017",
month = "11",
day = "15",
doi = "10.1002/bimj.201600142",
language = "English",
volume = "59",
pages = "1339–1351",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",
number = "6",

}

TY - JOUR

T1 - A flexible multivariate random effects proportional odds model with application to adverse effects during radiation therapy

AU - Augustin, Nicole

AU - Kim, Sung Won

AU - Uhlig, Annemarie

AU - Hanser, Christina

AU - Henke, Michael

AU - Schumacher, Martin

PY - 2017/11/15

Y1 - 2017/11/15

N2 - Radiation therapy in patients with head and neck cancer has a toxic effect onmucosa, the soft tissue in and around the mouth. Hence mucositis is a serious common side effect and is a condition characterized by pain and inflammation ofthe surface of the mucosa. Although the mucosa recovers during breaks of and following the radiotherapy course the recovery will depend on the type of tissue involved and on its location. We present a novel flexible multivariate random effects proportional odds model which takes account of the longitudinal course of oral mucositis at different mouth sites and of the radiation dosage(in terms of cumulative dose). The model is an extension of the {\emproportional odds model} which is used for ordinal response variables. Our model includes the ordinal multivariate response of the mucositis score by location, random intercepts for individuals and includes a non-linear function of cumulative radiation dose. The model allows to test whether sensitivitydiffers by mouth sites after having adjusted for site specific cumulative radiation dose. The model also allows to check whether and how the (non-linear) effect of site specific dose differs by site. We fit the model to longitudinal patient data from a prospective observation and find that after adjusting for cumulative dose, upper, lower lips and mouth floor are associated with the lowest mucositis scores and hard and soft palate are associated with the highest mucositis scores. This implies the possibility that tissues at different mouth locations differ in their sensitivity to the toxic effect of radiation. We also find that cumulative dose followed by mouth site are the strongest predictors of mucositis, and the effects of age and gender are negligible.

AB - Radiation therapy in patients with head and neck cancer has a toxic effect onmucosa, the soft tissue in and around the mouth. Hence mucositis is a serious common side effect and is a condition characterized by pain and inflammation ofthe surface of the mucosa. Although the mucosa recovers during breaks of and following the radiotherapy course the recovery will depend on the type of tissue involved and on its location. We present a novel flexible multivariate random effects proportional odds model which takes account of the longitudinal course of oral mucositis at different mouth sites and of the radiation dosage(in terms of cumulative dose). The model is an extension of the {\emproportional odds model} which is used for ordinal response variables. Our model includes the ordinal multivariate response of the mucositis score by location, random intercepts for individuals and includes a non-linear function of cumulative radiation dose. The model allows to test whether sensitivitydiffers by mouth sites after having adjusted for site specific cumulative radiation dose. The model also allows to check whether and how the (non-linear) effect of site specific dose differs by site. We fit the model to longitudinal patient data from a prospective observation and find that after adjusting for cumulative dose, upper, lower lips and mouth floor are associated with the lowest mucositis scores and hard and soft palate are associated with the highest mucositis scores. This implies the possibility that tissues at different mouth locations differ in their sensitivity to the toxic effect of radiation. We also find that cumulative dose followed by mouth site are the strongest predictors of mucositis, and the effects of age and gender are negligible.

KW - generalised additive mixed model, proportional odds model, cumulative threshold model, radiation therapy, multivariate repeated measures, ordered categorical response, biologically effective dose (BED), mucositis

UR - https://doi.org/10.1002/bimj.201600142

U2 - 10.1002/bimj.201600142

DO - 10.1002/bimj.201600142

M3 - Article

VL - 59

SP - 1339

EP - 1351

JO - Biometrical Journal

T2 - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

IS - 6

M1 - 201600142

ER -