32P - Heterogeneity analysis of DW MRI as a biomarker for prediction of overall survival and 6 month PFS in GBM patients

Date 17 December 2016
Event ESMO Asia 2016 Congress
Session Poster lunch
Topics Central Nervous System Malignancies
Staging Procedures (clinical staging)
Presenter Manijeh Beigi
Citation Annals of Oncology (2016) 27 (suppl_9): ix9-ix18. 10.1093/annonc/mdw574
Authors M. Beigi1, A. Fathi Kazerooni1, M. Safari2, A. Ameri3, B. Moini4, M. Shojaee Moghdam5, H. Salighehrad2
  • 1Medical Engeenering And Medical Physics, Tehran University of Medical Sciences, 1417613151 - Tehran/IR
  • 2Medical Engeenering And Medical Physics, Tehran University of Medical Sciences, Tehran/IR
  • 3Radiation Oncology, Shahid Beheshti University of Medical Sciences, Tehran/IR
  • 4Radiation Oncology, Emam Hosein Hospital, Tehran/IR
  • 5Imaging, Day Hospital, Tehran/IR



GBM is the primary malignant brain tumors with median patient survival of 12-15 months. Intratumoral heterogeneity causes to variable behavior of tumor and response to treatment. Anatomical MRI are not capable of providing information about tumor matrix. Diffusion-weighted (DW) MRI provide physiological information by quantifying the water diffusivity within tumor. The aim of this study was to evaluate the use of texture analysis of DWI for prediction of overall and 6 month progression free survival of patients.


11 patients with GBM who were referred to Jorjani Center (2012 -2014), were selected. All patients gave their written informed consent. Routin and DW MRI were performed before treatment. Apparent diffusion coefficient (ADC) maps were generated from DW images. T2 FLAIR images were registered to ADC map. ROI was placed on tumor region in T2FLAIR images. The ROI’s masks were transferred to in-house programs in MATLAB. The heterogeneity of the GBM were explored by first-order histogram of ADC included: histogram standard deviation, smoothness, third moment, uniformity, entropy, kurtosis, Percentile25, Percentile75 and Percentile95.


Positive correlation was seen between Percentile25, Percentile75 and Percentile95 with overall survival that were statistically significant (p 


Heterogeneity metrics of DWI were explored to obtain the most accurate predictive biomarker for theraputic outcome. The positive correlation of above mentioned parameters implies that tumors with higher mentioned metrics will respond better to therapy and longer overall survival. These results demonstrate the feasibility of heterogeneity analysis of DW MRI for pre-treatment prediction of outcome undergoing radiation therapy and will be validated in twenty-eight population to determine which metrics can be adopted as relevant biomarkers for outcome prediction of GBM.

Clinical trial indentification

Legal entity responsible for the study

Tehran University of Medical science


Tehran University of Medical science


All authors have declared no conflicts of interest.