Applying an Automated Radiation Therapy Volume Analysis Pipeline to Determine the Utility of Spectroscopic MRI-guided Adaptive Radiation Therapy for Glioblastoma 公开
Su Hyun Kim (Fall 2023)
Published
Abstract
Due to the infiltrative nature of glioblastoma, tumor characterization is often inadequate with clinical imaging alone, resulting in suboptimal radiation therapy (RT) targeting. Precise targeting by whole-brain spectroscopic MRI (sMRI), which maps tumors using metabolites including choline (Cho) and N-acetylaspartate (NAA), can quantify early treatment-induced molecular changes that other modalities cannot measure. We have developed a pipeline to determine the degree of sMRI changes during early stages of RT and its association with patient outcomes. Such information can provide insight into the potential utility of adaptive RT (ART).
Data was obtained from a study (NCT03137888) where glioblastoma patients received high-dose RT guided by the pre-RT Cho/NAA twice normal (Cho/NAA ≥ 2x) volume, and received spectroscopic MRI scans pre- and mid-RT. After co-registering pre- and mid-RT Cho/NAA ≥2 volumes, spatial overlap statistics between the volumes for each patient (n=27) were calculated to quantify metabolic activity changes after two weeks of RT. Overall and progression-free survival (OS/PFS) Kaplan-Meier curves were generated, stratified by the median statistic begin plotted, and Log-rank tests were used to quantify the relationships between imaging metrics and OS/PFS.
Between the pre- and mid-RT Cho/NAA ≥ 2x volumes of all 27 patients, the median total overlap was 0.796 (range: 0 – 0.964), median Jaccard coefficient was 0.224 (0 – 0.723), median Dice coefficient was 0.365 (0 – 0.839), median volume similarity was -0.142 (-1.54 – 1.77), and median Hausdorff distance was 20.8 mm (8.54 – 92.2 mm).
Patients with lower Jaccard/Dice coefficients had longer PFS (p=0.045 for both), and patients with lower Jaccard/Dice coefficients had higher OS trending towards significance (p=0.060 for both). Cho/NAA ≥ 2x volumes changed significantly during early RT, putting healthy tissue at risk of irradiation, and warranting further study into using adaptive RT planning.
Table of Contents
1. Introduction
2. Materials and Methods
2.1 Data acquisition and processing
2.2 RT Planning
2.4 Metabolite map coverage
2.5 Overlap statistics
2.6 Visualization
2.7 Statistical analysis
2.8 Combined biomarkers
3. Results
3.1. Spatial overlap
3.2 Survival analysis: Kaplan Meier Cruves
3.3 Survival analysis: Cox Proportional Hazard Regression
4. Discussion
4.1 Significance of overlap statistics
4.2 Limitations
5. Conclusions and Future Direction
About this Honors Thesis
| School | |
|---|---|
| Department | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| 关键词 | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Applying an Automated Radiation Therapy Volume Analysis Pipeline to Determine the Utility of Spectroscopic MRI-guided Adaptive Radiation Therapy for Glioblastoma () | 2023-12-06 11:44:53 -0500 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|