304. Role of texture analysis (TA) as prognostic factor in patients with glioblastoma diagnosis subject to radio-chemotherapic treatment concerning post-operating tomozolomide. (December 2018)

• Record Type: Journal Article
• Authors: Biondi, M.; Vanzi, E.; De Otto, G.; Belmonte, G.; Nardone, V.; Tini, P.; Sebaste, L.; Carfagno, T.; Battaglia, G.; Rubino, G.; Pastina, P.; Pirtoli, L.; Banci Buonamici, F.
• Is Part Of: Physica medica. Volume 56(2018)Supplement 2
• Page Start: 248
• Page End: 249
• Publication Date: 2018-12
• Subjects: Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2018.04.313
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Purpose: The purpose of this retrospective study was to evaluate the prognostic-predictive role of texture analysis parameters (TA) measured on preoperative MRI and their correlation with known clinical and molecular factors in radiotherapy-treated glioblastoma patients. Methods: This study regarded patients affected to glioblastoma and treated with concomitant postoperative radiotherapy with before surgery MRI exams (T1, FLAIR and DWI). TA GLCM parameters were obtained with LIFEx software and subsequently selected according to reproducibility through the intra-class correlation coefficient (ICC). A cut-off correlated with overall survival (OS) and free disease survival (PFS) was identified using X-Tile analysis software considering parameters like MGMT methylation and EGFR expression (i.e. Table). Results: This study included 73 patients and took to the definition of three classes of risk related to the TA considering clinical outcomes. Multivariate OS analysis showed that the significantly related survival parameters were KPS and TA risk classes (p < 0.001). The AUC curve of 0.80 (DS: 0.062 95% CI: 0.684–0926). In the group with MGMT non-methylated - EGFR high expression (prognostically worse), the risk class influences statistically significant OS with a median survival rate of 4 months vs 10 months vs 12 (p: 0.028). In the MGMT group, low-expression EGFR (prognostically favourable class) patients were found, for the three classes, an OS with a median survival rate of 4, 9 and 68.8 months (Fig). Conclusion: This study shows that there is a correlation between TA GLCM parameters extracted from RM sequences by building a score based on quantifiable TA parameters that seem to improve prognostic stratification of patients significantly. The results of this work require external validation, possibly prospective, that can improve the robustness of the results obtained.