Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer

Nai Ming Cheng, Yu Hua Dean Fang, Li yu Lee, Joseph Tung Chieh Chang, Din Li Tsan, Shu Hang Ng, Hung Ming Wang, Chun Ta Liao, Lan Yan Yang, Ching Han Hsu, Tzu Chen Yen

Research output: Contribution to journalArticle

56 Citations (Scopus)

Abstract

Methods: We retrospectively reviewed the records of 88 patients with T3 or T4 OPSCC who had completed primary therapy. Progression-free survival (PFS) and disease-specific survival (DSS) were the main outcome measures. In an exploratory analysis, a standardized uptake value of 2.5 (SUV 2.5) was taken as the cut-off value for the detection of tumour boundaries. A fixed threshold at 42 % of the maximum SUV (SUVmax 42 %) and an adaptive threshold method were then used for validation. Regional textural features were extracted from pretreatment 18F-FDG PET/CT images using the grey-level run length encoding method and grey-level size zone matrix. The prognostic significance of PET textural features was examined using receiver operating characteristic (ROC) curves and Cox regression analysis.

Purpose: The question as to whether the regional textural features extracted from PET images predict prognosis in oropharyngeal squamous cell carcinoma (OPSCC) remains open. In this study, we investigated the prognostic impact of regional heterogeneity in patients with T3/T4 OPSCC.

Results: Zone-size nonuniformity (ZSNU) was identified as an independent predictor of PFS and DSS. Its prognostic impact was confirmed using both the SUVmax 42 % and the adaptive threshold segmentation methods. Based on (1) total lesion glycolysis, (2) uniformity (a local scale texture parameter), and (3) ZSNU, we devised a prognostic stratification system that allowed the identification of four distinct risk groups. The model combining the three prognostic parameters showed a higher predictive value than each variable alone.

Conclusion: ZSNU is an independent predictor of outcome in patients with advanced T-stage OPSCC, and may improve their prognostic stratification.

Original languageEnglish
Pages (from-to)419-428
Number of pages10
JournalEuropean Journal of Nuclear Medicine and Molecular Imaging
Volume42
Issue number3
DOIs
Publication statusPublished - 2015 Jan 1

Fingerprint

Oropharyngeal Neoplasms
Fluorodeoxyglucose F18
Squamous Cell Carcinoma
Survival
Disease-Free Survival
Glycolysis
ROC Curve
Regression Analysis
Outcome Assessment (Health Care)
Neoplasms

All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Cite this

Cheng, Nai Ming ; Fang, Yu Hua Dean ; Lee, Li yu ; Chang, Joseph Tung Chieh ; Tsan, Din Li ; Ng, Shu Hang ; Wang, Hung Ming ; Liao, Chun Ta ; Yang, Lan Yan ; Hsu, Ching Han ; Yen, Tzu Chen. / Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer. In: European Journal of Nuclear Medicine and Molecular Imaging. 2015 ; Vol. 42, No. 3. pp. 419-428.
@article{4ad8d17d2e6845b69352faa5657a30dd,
title = "Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer",
abstract = "Methods: We retrospectively reviewed the records of 88 patients with T3 or T4 OPSCC who had completed primary therapy. Progression-free survival (PFS) and disease-specific survival (DSS) were the main outcome measures. In an exploratory analysis, a standardized uptake value of 2.5 (SUV 2.5) was taken as the cut-off value for the detection of tumour boundaries. A fixed threshold at 42 {\%} of the maximum SUV (SUVmax 42 {\%}) and an adaptive threshold method were then used for validation. Regional textural features were extracted from pretreatment 18F-FDG PET/CT images using the grey-level run length encoding method and grey-level size zone matrix. The prognostic significance of PET textural features was examined using receiver operating characteristic (ROC) curves and Cox regression analysis.Purpose: The question as to whether the regional textural features extracted from PET images predict prognosis in oropharyngeal squamous cell carcinoma (OPSCC) remains open. In this study, we investigated the prognostic impact of regional heterogeneity in patients with T3/T4 OPSCC.Results: Zone-size nonuniformity (ZSNU) was identified as an independent predictor of PFS and DSS. Its prognostic impact was confirmed using both the SUVmax 42 {\%} and the adaptive threshold segmentation methods. Based on (1) total lesion glycolysis, (2) uniformity (a local scale texture parameter), and (3) ZSNU, we devised a prognostic stratification system that allowed the identification of four distinct risk groups. The model combining the three prognostic parameters showed a higher predictive value than each variable alone.Conclusion: ZSNU is an independent predictor of outcome in patients with advanced T-stage OPSCC, and may improve their prognostic stratification.",
author = "Cheng, {Nai Ming} and Fang, {Yu Hua Dean} and Lee, {Li yu} and Chang, {Joseph Tung Chieh} and Tsan, {Din Li} and Ng, {Shu Hang} and Wang, {Hung Ming} and Liao, {Chun Ta} and Yang, {Lan Yan} and Hsu, {Ching Han} and Yen, {Tzu Chen}",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/s00259-014-2933-1",
language = "English",
volume = "42",
pages = "419--428",
journal = "European Journal of Nuclear Medicine and Molecular Imaging",
issn = "1619-7070",
publisher = "Springer Verlag",
number = "3",

}

Cheng, NM, Fang, YHD, Lee, LY, Chang, JTC, Tsan, DL, Ng, SH, Wang, HM, Liao, CT, Yang, LY, Hsu, CH & Yen, TC 2015, 'Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer', European Journal of Nuclear Medicine and Molecular Imaging, vol. 42, no. 3, pp. 419-428. https://doi.org/10.1007/s00259-014-2933-1

Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer. / Cheng, Nai Ming; Fang, Yu Hua Dean; Lee, Li yu; Chang, Joseph Tung Chieh; Tsan, Din Li; Ng, Shu Hang; Wang, Hung Ming; Liao, Chun Ta; Yang, Lan Yan; Hsu, Ching Han; Yen, Tzu Chen.

In: European Journal of Nuclear Medicine and Molecular Imaging, Vol. 42, No. 3, 01.01.2015, p. 419-428.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer

AU - Cheng, Nai Ming

AU - Fang, Yu Hua Dean

AU - Lee, Li yu

AU - Chang, Joseph Tung Chieh

AU - Tsan, Din Li

AU - Ng, Shu Hang

AU - Wang, Hung Ming

AU - Liao, Chun Ta

AU - Yang, Lan Yan

AU - Hsu, Ching Han

AU - Yen, Tzu Chen

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Methods: We retrospectively reviewed the records of 88 patients with T3 or T4 OPSCC who had completed primary therapy. Progression-free survival (PFS) and disease-specific survival (DSS) were the main outcome measures. In an exploratory analysis, a standardized uptake value of 2.5 (SUV 2.5) was taken as the cut-off value for the detection of tumour boundaries. A fixed threshold at 42 % of the maximum SUV (SUVmax 42 %) and an adaptive threshold method were then used for validation. Regional textural features were extracted from pretreatment 18F-FDG PET/CT images using the grey-level run length encoding method and grey-level size zone matrix. The prognostic significance of PET textural features was examined using receiver operating characteristic (ROC) curves and Cox regression analysis.Purpose: The question as to whether the regional textural features extracted from PET images predict prognosis in oropharyngeal squamous cell carcinoma (OPSCC) remains open. In this study, we investigated the prognostic impact of regional heterogeneity in patients with T3/T4 OPSCC.Results: Zone-size nonuniformity (ZSNU) was identified as an independent predictor of PFS and DSS. Its prognostic impact was confirmed using both the SUVmax 42 % and the adaptive threshold segmentation methods. Based on (1) total lesion glycolysis, (2) uniformity (a local scale texture parameter), and (3) ZSNU, we devised a prognostic stratification system that allowed the identification of four distinct risk groups. The model combining the three prognostic parameters showed a higher predictive value than each variable alone.Conclusion: ZSNU is an independent predictor of outcome in patients with advanced T-stage OPSCC, and may improve their prognostic stratification.

AB - Methods: We retrospectively reviewed the records of 88 patients with T3 or T4 OPSCC who had completed primary therapy. Progression-free survival (PFS) and disease-specific survival (DSS) were the main outcome measures. In an exploratory analysis, a standardized uptake value of 2.5 (SUV 2.5) was taken as the cut-off value for the detection of tumour boundaries. A fixed threshold at 42 % of the maximum SUV (SUVmax 42 %) and an adaptive threshold method were then used for validation. Regional textural features were extracted from pretreatment 18F-FDG PET/CT images using the grey-level run length encoding method and grey-level size zone matrix. The prognostic significance of PET textural features was examined using receiver operating characteristic (ROC) curves and Cox regression analysis.Purpose: The question as to whether the regional textural features extracted from PET images predict prognosis in oropharyngeal squamous cell carcinoma (OPSCC) remains open. In this study, we investigated the prognostic impact of regional heterogeneity in patients with T3/T4 OPSCC.Results: Zone-size nonuniformity (ZSNU) was identified as an independent predictor of PFS and DSS. Its prognostic impact was confirmed using both the SUVmax 42 % and the adaptive threshold segmentation methods. Based on (1) total lesion glycolysis, (2) uniformity (a local scale texture parameter), and (3) ZSNU, we devised a prognostic stratification system that allowed the identification of four distinct risk groups. The model combining the three prognostic parameters showed a higher predictive value than each variable alone.Conclusion: ZSNU is an independent predictor of outcome in patients with advanced T-stage OPSCC, and may improve their prognostic stratification.

UR - http://www.scopus.com/inward/record.url?scp=84925507745&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925507745&partnerID=8YFLogxK

U2 - 10.1007/s00259-014-2933-1

DO - 10.1007/s00259-014-2933-1

M3 - Article

C2 - 25339524

AN - SCOPUS:84925507745

VL - 42

SP - 419

EP - 428

JO - European Journal of Nuclear Medicine and Molecular Imaging

JF - European Journal of Nuclear Medicine and Molecular Imaging

SN - 1619-7070

IS - 3

ER -