Quantitative Perfusion and Permeability in Brain Tumors Before and After Laser Treatment
Poster No:
2089
Submission Type:
Abstract Submission
Authors:
Sagada Penano1, Daniel Biro1, Peter Warnke1, Timothy Carroll1
Institutions:
1University of Chicago, Chicago, IL
First Author:
Co-Author(s):
Introduction:
Drug delivery to brain tumors is difficult due to the low permeability of the blood brain barrier (BBB) and highly variable permeability of leaky blood vessels in the tumor [1]. However, laser interstitial thermal therapy (LITT) has the capability to increase permeability in regions adjacent to the LITT ablated tumor and decrease permeability of the tumor itself, facilitating drug delivery while inhibiting leakage [2]. Quantitative values of perfusion (capillary-level blood flow in ml/min) and permeability (min-1) are necessary to determine the exact amount of low molecular weight molecules that can be delivered to the tumor site. Two promising imaging techniques to accomplish this are dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) and dynamic susceptibility magnetic resonance imaging (DSC-MRI) respectively. This study presents a clinical application of quantitative perfusion and permeability mapping in tumor sites before and after LITT.
Methods:
Five consecutive patients with brain tumors were consented and underwent a preoperative MRI scan on a Philips Achieva 3T MRI scanner. In addition to anatomical reference scans, contrast enhanced DCE and DSC acquisitions were performed for the purposes of research. All image processing and analysis was done using software written "in-house" with MATLAB 2023b. One patient had two distinct tumors that underwent treatment, resulting in 6 cases in total. Within 24 hours of the preoperative scan, the patients were treated with LITT at the tumor site. Immediately after the LITT procedure, a postoperative scan was performed.
The preoperative scans and some postoperative scans consisted of an echo planar imaging Look-Locker (EPI-LL) sequence for T1 mapping, then a Gadolinium based contrast agent (GBCA) was injected. The permeability of GBCA was imaged with gradient echo DCE-MRI (series of T1 weighted images) with variable flip angles for T1 mapping. A second injection of GBCA was performed followed by DSC-MRI (series of T2* weighted images) to measure perfusion of the GBCA. Following the DSC-MRI sequence, another EPI-LL was performed for T1 mapping post injection. A contrast enhanced anatomical T1 weighted image was also acquired.
Quantitative cerebral blood flow (qCBF) was calculated from the DSC-MRI scan by first converting signal intensity to contrast agent concentration and deconvolving the signal with the arterial input function (AIF). The AIF was chosen automatically to reduce operator error and streamline post processing of the perfusion data. However, this calculation is very sensitive to errors in the AIF. Thus, a correction factor was implemented based on an AIF-independent steady state measurement of cerebral blood volume [3]. To obtain quantitative permeability from the DCE images, the transfer coefficient Ktrans was derived from compartmental analysis using the Patlak model [4].
The preoperative scans and some postoperative scans consisted of an echo planar imaging Look-Locker (EPI-LL) sequence for T1 mapping, then a Gadolinium based contrast agent (GBCA) was injected. The permeability of GBCA was imaged with gradient echo DCE-MRI (series of T1 weighted images) with variable flip angles for T1 mapping. A second injection of GBCA was performed followed by DSC-MRI (series of T2* weighted images) to measure perfusion of the GBCA. Following the DSC-MRI sequence, another EPI-LL was performed for T1 mapping post injection. A contrast enhanced anatomical T1 weighted image was also acquired.
Quantitative cerebral blood flow (qCBF) was calculated from the DSC-MRI scan by first converting signal intensity to contrast agent concentration and deconvolving the signal with the arterial input function (AIF). The AIF was chosen automatically to reduce operator error and streamline post processing of the perfusion data. However, this calculation is very sensitive to errors in the AIF. Thus, a correction factor was implemented based on an AIF-independent steady state measurement of cerebral blood volume [3]. To obtain quantitative permeability from the DCE images, the transfer coefficient Ktrans was derived from compartmental analysis using the Patlak model [4].
Results:
A decrease in the permeability of the tumor was noted in all 6 cases after LITT with an average decrease of 0.66 min-1 (-58%). Patient 6 exhibited two tumors and at the time of this study, one was treated with LITT while the other was left untreated (Representative case Figure 1). The tumor treated with LITT showed a significant decrease in permeability of 0.80 min-1 (-44%) while the untreated tumor showed a decrease in permeability of 0.04 min-1 (-4%). Perfusion before and after LITT appeared to have no significant change.
·Untreated brain tumor (top row) shows no significant change in permeability while LITT ablated tumor (bottom row) shows significant change in permeability (-44%)
Conclusions:
A significant decrease in the permeability of brain tumors was observed and with quantitative values of perfusion and permeability. This new protocol will allow physicians to create accurate drug delivery maps to better treat patients and pave the way for MR data reproducibility.
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Anatomy and Functional Systems 1
Neuroinformatics and Data Sharing:
Databasing and Data Sharing
Novel Imaging Acquisition Methods:
Anatomical MRI 2
Keywords:
Cerebral Blood Flow
Data analysis
MRI PHYSICS
STRUCTURAL MRI
Treatment
Workflows
Other - Permeability
1|2Indicates the priority used for review
Provide references using author date format
2. Luo H, Shusta EV. (2020 Nov 12), 'Blood-Brain Barrier Modulation to Improve Glioma Drug Delivery', Pharmaceutics. 2020;12(11):1085.
3. Carroll TJ, Horowitz S, Shin W, et al. (2008), 'Quantification of cerebral perfusion using the "bookend technique": an evaluation in CNS tumors', Magn Reson Imaging.
4. Sourbron SP, Buckley DL. (2013), 'Classic models for dynamic contrast-enhanced MR'. NMR Biomed.