Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer

Nicola Dinapoli, Brunella Barbaro, Roberto Gatta, Giuditta Chiloiro, Calogero Casà, Carlotta Masciocchi, Andrea Damiani, Luca Boldrini, Maria Antonietta Gambacorta, Michele Dezio, Gian Carlo Mattiucci, Mario Balducci, Johan van Soest, Andre Dekker, Philippe Lambin, Claudio Fiorino, Carla Sini, Francesco De Cobelli, Nadia Di Muzio, Calogero GuminaPaolo Passoni, Riccardo Manfredi, Vincenzo Valentini

Research output: Contribution to journalArticle

Abstract

Purpose: The objective of this study is finding an intensity based histogram (IBH) signature to predict pathologic complete response (pCR) probability using only pre-treatment magnetic resonance (MR) and validate it externally in order to create a workflow for the external validation of an MR IBH signature and to apply the model out of the environment where it has been tuned. The impact of pCR and the final predictors on the survival outcome were also evaluated. Methods and Materials: Three centers using different MR scanners were involved in this retrospective study. The first center recruited 162 patients for model training, and the second and third centers provided 34 plus 25 patients for external validation. Patients provided written consent. Accrual period was from May 2008 to December 2014. After surgery pathologic response was defined. T2-weighted MR scans acquired before chemoradiation therapy (CRT) were used for analysis addressed on primary lesions. Images were pre-processed using Laplacian of Gaussian (LoG) filter with multiple σ, and first order intensity histogram-based features (kurtosis, skewness, and entropy) were extracted. Features selection was performed using Mann-Whitney test. Tumor staging (cT, cN) was added to build a logistic regression model and predict pCR. Model performance was evaluated with internal and external validation using area under the curve (AUC) of the receiver operator characteristic (ROC) and calibration with Hosmer-Lemeshow test. The linear cross-correlation matrix (Pearson's coefficient) and the variance inflation factor (VIF) were used to check the correlation and the co-linearity among the final predictors. The amount of the information added through the radiomics features was estimated by using the DeLong's test, and the impact of pCR and the final predictors on survival outcomes were evaluated through the Kaplan-Meier curves by using the log-rank test and the multivariate Cox model. Results: Candidate-to-analysis features were skewness (σ = 0.485, P value =.01) and entropy (σ = 0.344, P value <.05). Logistic regression analysis showed as significant covariates cT (P value <.01), skewness-σ = 0.485 (P value =.01), and entropy-σ = 0.344 (P value <.05). Model AUCs were 0.73 (internal) and 0.75 (external). Conclusions: This MR-based, vendor-independent model can be helpful for predicting pCR probability in locally advanced rectal cancer (LARC) patients only using pre-treatment imaging.

Original languageEnglish
Pages (from-to)765-774
Number of pages10
JournalInternational Journal of Radiation Oncology Biology Physics
Volume102
Issue number4
DOIs
Publication statusPublished - Nov 15 2018

Fingerprint

Chemoradiotherapy
Rectal Neoplasms
histograms
magnetic resonance
Magnetic Resonance Spectroscopy
cancer
Entropy
skewness
Logistic Models
Area Under Curve
logistics
entropy
pretreatment
regression analysis
Survival
Workflow
Neoplasm Staging
Economic Inflation
predictions
signatures

ASJC Scopus subject areas

  • Radiation
  • Oncology
  • Radiology Nuclear Medicine and imaging
  • Cancer Research

Cite this

Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer. / Dinapoli, Nicola; Barbaro, Brunella; Gatta, Roberto; Chiloiro, Giuditta; Casà, Calogero; Masciocchi, Carlotta; Damiani, Andrea; Boldrini, Luca; Gambacorta, Maria Antonietta; Dezio, Michele; Mattiucci, Gian Carlo; Balducci, Mario; van Soest, Johan; Dekker, Andre; Lambin, Philippe; Fiorino, Claudio; Sini, Carla; De Cobelli, Francesco; Di Muzio, Nadia; Gumina, Calogero; Passoni, Paolo; Manfredi, Riccardo; Valentini, Vincenzo.

In: International Journal of Radiation Oncology Biology Physics, Vol. 102, No. 4, 15.11.2018, p. 765-774.

Research output: Contribution to journalArticle

Dinapoli, N, Barbaro, B, Gatta, R, Chiloiro, G, Casà, C, Masciocchi, C, Damiani, A, Boldrini, L, Gambacorta, MA, Dezio, M, Mattiucci, GC, Balducci, M, van Soest, J, Dekker, A, Lambin, P, Fiorino, C, Sini, C, De Cobelli, F, Di Muzio, N, Gumina, C, Passoni, P, Manfredi, R & Valentini, V 2018, 'Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer', International Journal of Radiation Oncology Biology Physics, vol. 102, no. 4, pp. 765-774. https://doi.org/10.1016/j.ijrobp.2018.04.065
Dinapoli, Nicola ; Barbaro, Brunella ; Gatta, Roberto ; Chiloiro, Giuditta ; Casà, Calogero ; Masciocchi, Carlotta ; Damiani, Andrea ; Boldrini, Luca ; Gambacorta, Maria Antonietta ; Dezio, Michele ; Mattiucci, Gian Carlo ; Balducci, Mario ; van Soest, Johan ; Dekker, Andre ; Lambin, Philippe ; Fiorino, Claudio ; Sini, Carla ; De Cobelli, Francesco ; Di Muzio, Nadia ; Gumina, Calogero ; Passoni, Paolo ; Manfredi, Riccardo ; Valentini, Vincenzo. / Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer. In: International Journal of Radiation Oncology Biology Physics. 2018 ; Vol. 102, No. 4. pp. 765-774.
@article{7238880eb5894c089411d2f849ae0b39,
title = "Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer",
abstract = "Purpose: The objective of this study is finding an intensity based histogram (IBH) signature to predict pathologic complete response (pCR) probability using only pre-treatment magnetic resonance (MR) and validate it externally in order to create a workflow for the external validation of an MR IBH signature and to apply the model out of the environment where it has been tuned. The impact of pCR and the final predictors on the survival outcome were also evaluated. Methods and Materials: Three centers using different MR scanners were involved in this retrospective study. The first center recruited 162 patients for model training, and the second and third centers provided 34 plus 25 patients for external validation. Patients provided written consent. Accrual period was from May 2008 to December 2014. After surgery pathologic response was defined. T2-weighted MR scans acquired before chemoradiation therapy (CRT) were used for analysis addressed on primary lesions. Images were pre-processed using Laplacian of Gaussian (LoG) filter with multiple σ, and first order intensity histogram-based features (kurtosis, skewness, and entropy) were extracted. Features selection was performed using Mann-Whitney test. Tumor staging (cT, cN) was added to build a logistic regression model and predict pCR. Model performance was evaluated with internal and external validation using area under the curve (AUC) of the receiver operator characteristic (ROC) and calibration with Hosmer-Lemeshow test. The linear cross-correlation matrix (Pearson's coefficient) and the variance inflation factor (VIF) were used to check the correlation and the co-linearity among the final predictors. The amount of the information added through the radiomics features was estimated by using the DeLong's test, and the impact of pCR and the final predictors on survival outcomes were evaluated through the Kaplan-Meier curves by using the log-rank test and the multivariate Cox model. Results: Candidate-to-analysis features were skewness (σ = 0.485, P value =.01) and entropy (σ = 0.344, P value <.05). Logistic regression analysis showed as significant covariates cT (P value <.01), skewness-σ = 0.485 (P value =.01), and entropy-σ = 0.344 (P value <.05). Model AUCs were 0.73 (internal) and 0.75 (external). Conclusions: This MR-based, vendor-independent model can be helpful for predicting pCR probability in locally advanced rectal cancer (LARC) patients only using pre-treatment imaging.",
author = "Nicola Dinapoli and Brunella Barbaro and Roberto Gatta and Giuditta Chiloiro and Calogero Cas{\`a} and Carlotta Masciocchi and Andrea Damiani and Luca Boldrini and Gambacorta, {Maria Antonietta} and Michele Dezio and Mattiucci, {Gian Carlo} and Mario Balducci and {van Soest}, Johan and Andre Dekker and Philippe Lambin and Claudio Fiorino and Carla Sini and {De Cobelli}, Francesco and {Di Muzio}, Nadia and Calogero Gumina and Paolo Passoni and Riccardo Manfredi and Vincenzo Valentini",
year = "2018",
month = "11",
day = "15",
doi = "10.1016/j.ijrobp.2018.04.065",
language = "English",
volume = "102",
pages = "765--774",
journal = "International Journal of Radiation Oncology Biology Physics",
issn = "0360-3016",
publisher = "Elsevier Inc.",
number = "4",

}

TY - JOUR

T1 - Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer

AU - Dinapoli, Nicola

AU - Barbaro, Brunella

AU - Gatta, Roberto

AU - Chiloiro, Giuditta

AU - Casà, Calogero

AU - Masciocchi, Carlotta

AU - Damiani, Andrea

AU - Boldrini, Luca

AU - Gambacorta, Maria Antonietta

AU - Dezio, Michele

AU - Mattiucci, Gian Carlo

AU - Balducci, Mario

AU - van Soest, Johan

AU - Dekker, Andre

AU - Lambin, Philippe

AU - Fiorino, Claudio

AU - Sini, Carla

AU - De Cobelli, Francesco

AU - Di Muzio, Nadia

AU - Gumina, Calogero

AU - Passoni, Paolo

AU - Manfredi, Riccardo

AU - Valentini, Vincenzo

PY - 2018/11/15

Y1 - 2018/11/15

N2 - Purpose: The objective of this study is finding an intensity based histogram (IBH) signature to predict pathologic complete response (pCR) probability using only pre-treatment magnetic resonance (MR) and validate it externally in order to create a workflow for the external validation of an MR IBH signature and to apply the model out of the environment where it has been tuned. The impact of pCR and the final predictors on the survival outcome were also evaluated. Methods and Materials: Three centers using different MR scanners were involved in this retrospective study. The first center recruited 162 patients for model training, and the second and third centers provided 34 plus 25 patients for external validation. Patients provided written consent. Accrual period was from May 2008 to December 2014. After surgery pathologic response was defined. T2-weighted MR scans acquired before chemoradiation therapy (CRT) were used for analysis addressed on primary lesions. Images were pre-processed using Laplacian of Gaussian (LoG) filter with multiple σ, and first order intensity histogram-based features (kurtosis, skewness, and entropy) were extracted. Features selection was performed using Mann-Whitney test. Tumor staging (cT, cN) was added to build a logistic regression model and predict pCR. Model performance was evaluated with internal and external validation using area under the curve (AUC) of the receiver operator characteristic (ROC) and calibration with Hosmer-Lemeshow test. The linear cross-correlation matrix (Pearson's coefficient) and the variance inflation factor (VIF) were used to check the correlation and the co-linearity among the final predictors. The amount of the information added through the radiomics features was estimated by using the DeLong's test, and the impact of pCR and the final predictors on survival outcomes were evaluated through the Kaplan-Meier curves by using the log-rank test and the multivariate Cox model. Results: Candidate-to-analysis features were skewness (σ = 0.485, P value =.01) and entropy (σ = 0.344, P value <.05). Logistic regression analysis showed as significant covariates cT (P value <.01), skewness-σ = 0.485 (P value =.01), and entropy-σ = 0.344 (P value <.05). Model AUCs were 0.73 (internal) and 0.75 (external). Conclusions: This MR-based, vendor-independent model can be helpful for predicting pCR probability in locally advanced rectal cancer (LARC) patients only using pre-treatment imaging.

AB - Purpose: The objective of this study is finding an intensity based histogram (IBH) signature to predict pathologic complete response (pCR) probability using only pre-treatment magnetic resonance (MR) and validate it externally in order to create a workflow for the external validation of an MR IBH signature and to apply the model out of the environment where it has been tuned. The impact of pCR and the final predictors on the survival outcome were also evaluated. Methods and Materials: Three centers using different MR scanners were involved in this retrospective study. The first center recruited 162 patients for model training, and the second and third centers provided 34 plus 25 patients for external validation. Patients provided written consent. Accrual period was from May 2008 to December 2014. After surgery pathologic response was defined. T2-weighted MR scans acquired before chemoradiation therapy (CRT) were used for analysis addressed on primary lesions. Images were pre-processed using Laplacian of Gaussian (LoG) filter with multiple σ, and first order intensity histogram-based features (kurtosis, skewness, and entropy) were extracted. Features selection was performed using Mann-Whitney test. Tumor staging (cT, cN) was added to build a logistic regression model and predict pCR. Model performance was evaluated with internal and external validation using area under the curve (AUC) of the receiver operator characteristic (ROC) and calibration with Hosmer-Lemeshow test. The linear cross-correlation matrix (Pearson's coefficient) and the variance inflation factor (VIF) were used to check the correlation and the co-linearity among the final predictors. The amount of the information added through the radiomics features was estimated by using the DeLong's test, and the impact of pCR and the final predictors on survival outcomes were evaluated through the Kaplan-Meier curves by using the log-rank test and the multivariate Cox model. Results: Candidate-to-analysis features were skewness (σ = 0.485, P value =.01) and entropy (σ = 0.344, P value <.05). Logistic regression analysis showed as significant covariates cT (P value <.01), skewness-σ = 0.485 (P value =.01), and entropy-σ = 0.344 (P value <.05). Model AUCs were 0.73 (internal) and 0.75 (external). Conclusions: This MR-based, vendor-independent model can be helpful for predicting pCR probability in locally advanced rectal cancer (LARC) patients only using pre-treatment imaging.

UR - http://www.scopus.com/inward/record.url?scp=85048154979&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048154979&partnerID=8YFLogxK

U2 - 10.1016/j.ijrobp.2018.04.065

DO - 10.1016/j.ijrobp.2018.04.065

M3 - Article

C2 - 29891200

AN - SCOPUS:85048154979

VL - 102

SP - 765

EP - 774

JO - International Journal of Radiation Oncology Biology Physics

JF - International Journal of Radiation Oncology Biology Physics

SN - 0360-3016

IS - 4

ER -