Background: In vivo biomarkers to predict progression of brain tumors are of great value in clinical practice. Therefore, the authors tested the hypothesis that changes in choline ratios by magnetic resonance (MR) spectroscopic imaging and/or relative tumor blood volume (rTBV) can differentiate clinically stable from progressive pediatric brain tumors.

Methods: MR spectroscopic imaging examinations were performed on 27 children with neuroglial brain tumors during therapy on a 1.5-Tesla MR system. Normalized rTBV values were measured in 11 of 27 patients. Each examination was rated as stable or progressive by clinical and imaging criteria.

Results: The percent change in normalized choline (Cho) was significantly greater in patients who had progressive examinations compared with patients who had stable examinations (P = 0.03). The percent change in Cho/N-acetylaspartate (Cho/NAA) was significantly higher in patients who had progressive outcomes (n = 18 patients) compared with patients who had stable outcomes (n = 32 patients; P < 0.001; sensitivity, 0.89; specificity, 0.88) and was identified as the most important prognostic indicator of tumor progression by logistic regression (likelihood ratio test, 33.4; P < 0.001). The odds of tumor progression were approximately 55 times greater for patients who showed at least a 20% change in Cho/NAA. rTBV distinguished between progressing and stable tumors (P = 0.03), and Cho/NAA and rTBV values showed interaction to predict the probability of a progressing clinical outcome.

Conclusions: The percent change in Cho/NAA by proton MR spectroscopic imaging, assisted by rTBV, was useful in predicting tumor progression in children with brain tumors.

Download full-text PDF

Source
http://dx.doi.org/10.1002/cncr.20096DOI Listing

Publication Analysis

Top Keywords

brain tumors
16
spectroscopic imaging
12
percent change
12
tumor progression
12
magnetic resonance
8
pediatric brain
8
stable progressive
8
rtbv values
8
patients
8
greater patients
8

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!