Premium
MRI characteristics predict the efficacy of meloxicam treatment in patients with desmoid‐type fibromatosis
Author(s) -
Shimizu Koki,
Hamada Shunsuke,
Sakai Tomohisa,
Ito Shinji,
Urakawa Hiroshi,
Arai Eisuke,
Ikuta Kunihiro,
Koike Hiroshi,
Ishiguro Naoki,
Nishida Yoshihiro
Publication year - 2019
Publication title -
journal of medical imaging and radiation oncology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.31
H-Index - 43
eISSN - 1754-9485
pISSN - 1754-9477
DOI - 10.1111/1754-9485.12940
Subject(s) - medicine , meloxicam , fibromatosis , gastroenterology , magnetic resonance imaging , surgery , radiology
This study aimed to determine the clinical significance of MRI characteristics as a possible predictor of responsiveness to meloxicam treatment in patients with desmoid‐type fibromatosis (DF). Additionally, it analysed the correlation between CTNNB1 mutation status and signal intensity of MRI. Methods Forty‐six patients consecutively treated with meloxicam composed this study. The low‐intensity area (LIA) on T2‐weighted MRI was determined. We divided patients into two groups based on the efficacy of meloxicam: a clinical benefit group (CB group, including CR: complete response; PR: partial response; and SD: stable disease) and non‐clinical benefit group (NB group, including PD: progressive disease). Correlations of the efficacy with LIA and CTNNB1 mutation status with LIA were investigated. Results In total, 11, 17 and 18 patients showed PR, SD and PD, respectively. The mean LIA ratio before treatment was significantly higher ( P < 0.001) in the CB group than in the NB group. For predicting the efficacy, sensitivity was 68%, and specificity was 89% when setting the cut‐off value as 20% for LIA. Mean changes in the LIA ratio before and after treatment were significantly higher ( P = 0.01) in the CB group than in the NB group. Mean LIA ratio before treatment was significantly lower ( P < 0.001) in the S45F mutation group than in the other mutation group. In multivariate analysis, the LIA ratio before treatment was a significant predictor of responsiveness ( P = 0.02). Conclusions MRI characteristics were a useful predictor of the efficacy of meloxicam in DF patients. It may be possible to predict the clinical outcome more accurately when combined with other factors, such as CTNNB1 mutantion status.