Unraveling biophysical interactions of radiation pneumonitis in non-small-cell lung cancer via Bayesian network analysis

Yi Luo, Issam El Naqa, Daniel L. McShan, Dipankar Ray, Ines Lohse, Martha M. Matuszak, Dawn Owen, Shruti Jolly, Theodore S. Lawrence, Feng Ming (Spring) Kong, Randall K. Ten Haken

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Background In non-small-cell lung cancer radiotherapy, radiation pneumonitis ≥ grade 2 (RP2) depends on patients’ dosimetric, clinical, biological and genomic characteristics. Methods We developed a Bayesian network (BN) approach to explore its potential for interpreting biophysical signaling pathways influencing RP2 from a heterogeneous dataset including single nucleotide polymorphisms, micro RNAs, cytokines, clinical data, and radiation treatment plans before and during the course of radiotherapy. Model building utilized 79 patients (21 with RP2) with complete data, and model testing used 50 additional patients with incomplete data. A developed large-scale Markov blanket approach selected relevant predictors. Resampling by k-fold cross-validation determined the optimal BN structure. Area under the receiver-operating characteristics curve (AUC) measured performance. Results Pre- and during-treatment BNs identified biophysical signaling pathways from the patients’ relevant variables to RP2 risk. Internal cross-validation for the pre-BN yielded an AUC = 0.82 which improved to 0.87 by incorporating during treatment changes. In the testing dataset, the pre- and during AUCs were 0.78 and 0.82, respectively. Conclusions Our developed BN approach successfully handled a high number of heterogeneous variables in a small dataset, demonstrating potential for unraveling relevant biophysical features that could enhance prediction of RP2, although the current observations would require further independent validation.

Original languageEnglish (US)
Pages (from-to)85-92
Number of pages8
JournalRadiotherapy and Oncology
Volume123
Issue number1
DOIs
StatePublished - Apr 1 2017

Fingerprint

Radiation Pneumonitis
Bayes Theorem
Non-Small Cell Lung Carcinoma
Radiation
Area Under Curve
Radiotherapy
MicroRNAs
ROC Curve
Single Nucleotide Polymorphism
Therapeutics
Cytokines
Datasets

Keywords

  • Bayesian network analysis
  • Biophysical interactions
  • Lung cancer
  • Radiation pneumonitis

ASJC Scopus subject areas

  • Hematology
  • Oncology
  • Radiology Nuclear Medicine and imaging

Cite this

Unraveling biophysical interactions of radiation pneumonitis in non-small-cell lung cancer via Bayesian network analysis. / Luo, Yi; El Naqa, Issam; McShan, Daniel L.; Ray, Dipankar; Lohse, Ines; Matuszak, Martha M.; Owen, Dawn; Jolly, Shruti; Lawrence, Theodore S.; Kong, Feng Ming (Spring); Ten Haken, Randall K.

In: Radiotherapy and Oncology, Vol. 123, No. 1, 01.04.2017, p. 85-92.

Research output: Contribution to journalArticle

Luo, Y, El Naqa, I, McShan, DL, Ray, D, Lohse, I, Matuszak, MM, Owen, D, Jolly, S, Lawrence, TS, Kong, FMS & Ten Haken, RK 2017, 'Unraveling biophysical interactions of radiation pneumonitis in non-small-cell lung cancer via Bayesian network analysis', Radiotherapy and Oncology, vol. 123, no. 1, pp. 85-92. https://doi.org/10.1016/j.radonc.2017.02.004
Luo, Yi ; El Naqa, Issam ; McShan, Daniel L. ; Ray, Dipankar ; Lohse, Ines ; Matuszak, Martha M. ; Owen, Dawn ; Jolly, Shruti ; Lawrence, Theodore S. ; Kong, Feng Ming (Spring) ; Ten Haken, Randall K. / Unraveling biophysical interactions of radiation pneumonitis in non-small-cell lung cancer via Bayesian network analysis. In: Radiotherapy and Oncology. 2017 ; Vol. 123, No. 1. pp. 85-92.
@article{ac2dc718bb4a490da9a1f0a61c65bab8,
title = "Unraveling biophysical interactions of radiation pneumonitis in non-small-cell lung cancer via Bayesian network analysis",
abstract = "Background In non-small-cell lung cancer radiotherapy, radiation pneumonitis ≥ grade 2 (RP2) depends on patients’ dosimetric, clinical, biological and genomic characteristics. Methods We developed a Bayesian network (BN) approach to explore its potential for interpreting biophysical signaling pathways influencing RP2 from a heterogeneous dataset including single nucleotide polymorphisms, micro RNAs, cytokines, clinical data, and radiation treatment plans before and during the course of radiotherapy. Model building utilized 79 patients (21 with RP2) with complete data, and model testing used 50 additional patients with incomplete data. A developed large-scale Markov blanket approach selected relevant predictors. Resampling by k-fold cross-validation determined the optimal BN structure. Area under the receiver-operating characteristics curve (AUC) measured performance. Results Pre- and during-treatment BNs identified biophysical signaling pathways from the patients’ relevant variables to RP2 risk. Internal cross-validation for the pre-BN yielded an AUC = 0.82 which improved to 0.87 by incorporating during treatment changes. In the testing dataset, the pre- and during AUCs were 0.78 and 0.82, respectively. Conclusions Our developed BN approach successfully handled a high number of heterogeneous variables in a small dataset, demonstrating potential for unraveling relevant biophysical features that could enhance prediction of RP2, although the current observations would require further independent validation.",
keywords = "Bayesian network analysis, Biophysical interactions, Lung cancer, Radiation pneumonitis",
author = "Yi Luo and {El Naqa}, Issam and McShan, {Daniel L.} and Dipankar Ray and Ines Lohse and Matuszak, {Martha M.} and Dawn Owen and Shruti Jolly and Lawrence, {Theodore S.} and Kong, {Feng Ming (Spring)} and {Ten Haken}, {Randall K.}",
year = "2017",
month = "4",
day = "1",
doi = "10.1016/j.radonc.2017.02.004",
language = "English (US)",
volume = "123",
pages = "85--92",
journal = "Radiotherapy and Oncology",
issn = "0167-8140",
publisher = "Elsevier Ireland Ltd",
number = "1",

}

TY - JOUR

T1 - Unraveling biophysical interactions of radiation pneumonitis in non-small-cell lung cancer via Bayesian network analysis

AU - Luo, Yi

AU - El Naqa, Issam

AU - McShan, Daniel L.

AU - Ray, Dipankar

AU - Lohse, Ines

AU - Matuszak, Martha M.

AU - Owen, Dawn

AU - Jolly, Shruti

AU - Lawrence, Theodore S.

AU - Kong, Feng Ming (Spring)

AU - Ten Haken, Randall K.

PY - 2017/4/1

Y1 - 2017/4/1

N2 - Background In non-small-cell lung cancer radiotherapy, radiation pneumonitis ≥ grade 2 (RP2) depends on patients’ dosimetric, clinical, biological and genomic characteristics. Methods We developed a Bayesian network (BN) approach to explore its potential for interpreting biophysical signaling pathways influencing RP2 from a heterogeneous dataset including single nucleotide polymorphisms, micro RNAs, cytokines, clinical data, and radiation treatment plans before and during the course of radiotherapy. Model building utilized 79 patients (21 with RP2) with complete data, and model testing used 50 additional patients with incomplete data. A developed large-scale Markov blanket approach selected relevant predictors. Resampling by k-fold cross-validation determined the optimal BN structure. Area under the receiver-operating characteristics curve (AUC) measured performance. Results Pre- and during-treatment BNs identified biophysical signaling pathways from the patients’ relevant variables to RP2 risk. Internal cross-validation for the pre-BN yielded an AUC = 0.82 which improved to 0.87 by incorporating during treatment changes. In the testing dataset, the pre- and during AUCs were 0.78 and 0.82, respectively. Conclusions Our developed BN approach successfully handled a high number of heterogeneous variables in a small dataset, demonstrating potential for unraveling relevant biophysical features that could enhance prediction of RP2, although the current observations would require further independent validation.

AB - Background In non-small-cell lung cancer radiotherapy, radiation pneumonitis ≥ grade 2 (RP2) depends on patients’ dosimetric, clinical, biological and genomic characteristics. Methods We developed a Bayesian network (BN) approach to explore its potential for interpreting biophysical signaling pathways influencing RP2 from a heterogeneous dataset including single nucleotide polymorphisms, micro RNAs, cytokines, clinical data, and radiation treatment plans before and during the course of radiotherapy. Model building utilized 79 patients (21 with RP2) with complete data, and model testing used 50 additional patients with incomplete data. A developed large-scale Markov blanket approach selected relevant predictors. Resampling by k-fold cross-validation determined the optimal BN structure. Area under the receiver-operating characteristics curve (AUC) measured performance. Results Pre- and during-treatment BNs identified biophysical signaling pathways from the patients’ relevant variables to RP2 risk. Internal cross-validation for the pre-BN yielded an AUC = 0.82 which improved to 0.87 by incorporating during treatment changes. In the testing dataset, the pre- and during AUCs were 0.78 and 0.82, respectively. Conclusions Our developed BN approach successfully handled a high number of heterogeneous variables in a small dataset, demonstrating potential for unraveling relevant biophysical features that could enhance prediction of RP2, although the current observations would require further independent validation.

KW - Bayesian network analysis

KW - Biophysical interactions

KW - Lung cancer

KW - Radiation pneumonitis

UR - http://www.scopus.com/inward/record.url?scp=85010761422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85010761422&partnerID=8YFLogxK

U2 - 10.1016/j.radonc.2017.02.004

DO - 10.1016/j.radonc.2017.02.004

M3 - Article

C2 - 28237401

AN - SCOPUS:85010761422

VL - 123

SP - 85

EP - 92

JO - Radiotherapy and Oncology

JF - Radiotherapy and Oncology

SN - 0167-8140

IS - 1

ER -