PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of neuroncolAboutAuthor GuidelinesEditorial BoardNeuro-Oncology
 
Neuro Oncol. 2017 January; 19(Suppl 1): i25.
Published online 2017 March 2. doi:  10.1093/neuonc/now292.004
PMCID: PMC5358554

OP05. Arterial spin labelling MRI of cerebral tumours in rats

Abstract

Brain perfusion imaging can contribute valuable information in diseases such as cancer and stroke. Arterial spin labelling (ASL) is a non-invasive MRI technique enabling quantification of perfusion by imaging tissue following exchanges of water with magnetically tagged blood. Although ASL has recently been standardised in the clinic1, pre-clinical usage of ASL remains inconsistent between studies. Our aim here was to identify optimal parameters for pre-clinical ASL, taking into account the anatomical and MRI system differences between humans and rodents, for reproducible and accurate perfusion imaging of metastatic brain tumours in rats. Multiphase pseudo-continuous ASL (pCASL) data were acquired on a 9.4T MRI (Agilent) with a 72mm volume transmit coil and a 4-channel surface receive array (Rapid Biomedical), using radiofrequency pulses with eight phase angles between 0° and 315°, each 45° apart2. pCASL data from eight phases were fit to a modified Fermi function (α=66, β=21) using a modified version of BASIL3 before processing with oxford_asl3. Imaging readout was a multislice single-shot spin echo EPI with FOV=32x32mm (64x64 matrix, slice thickness=1mm), TR=4 s, TE=12.4ms. To better replicate clinical ASL MRI for robust pre-clinical measurements of cerebral blood flow (CBF) in rat, key parameters relating to blood tagging were optimised across 3 strains of rats (Sprague-Dawley, Wistar, and BD-IX) to ensure broad applicability. Sufficient inversion of the blood in the tagging plane in the rat’s neck was estimated to occur at a width of 6.2mm and optimal angle of 45˚, such that it lay perpendicular to the carotid arteries and enabled more homogenous tagging of in-flowing blood. A blood tagging duration of 1.4s (train of 600μs 40° Hanning-shaped pulses, starting every 1.2ms) was found to be the most efficient tag duration, resulting in similar low variation of cerebral blood flow values as in longer tag durations while retaining a shorter TR. A post-label delay of 0.55-0.65s allowed labelled blood water transit to the tissue, prior to image acquisition. This optimised pCASL sequence was next tested in a BD-IX rat model of breast cancer brain metastasis4. Female rats were injected intra-cerebrally with 1000 ENU1564 (N-ethyl-N-nitrosurea-induced rat mammary adenocarcinoma) cells in 1µL PBS, or PBS alone as injection control. Six weeks after tumour cell injection, tumours were readily detected on anatomical images by increased signal on T2-weighted images, and found to have significantly reduced CBF (40 ± 8mL/100g/min) compared to the contralateral striatum (92 ± 12mL/100g/min, p=1.04x10-3) or to healthy brain tissue in PBS injected rats (95 ± 15mL/100g/min, p=3.35x10-5; Figure 1). This is the first instance in which ASL MRI has been optimised for pre-clinical applications. We demonstrate the ability of this optimised sequence to reliably determine changes in tumour perfusion in a rat model of brain metastasis. Pre-clinical ASL has previously been used for glioma perfusion in rats, where similar results were found5.

References

1. Aslop DC, et al. Magn Reson Med. 2014;102–116;

2. Jung Y, et al. Magn Reson Med. 2010;799–810;

3. Chappell MA, et al. IEEE Trans Signal Process. 2009;57:223–236;

4. Serres S, et al. Intern Journ Cancer. 2014;134;4:885–896;

5. Silva AC, et al. Magn Reson Med. 2000;44:169–173.

Brain perfusion imaging can contribute valuable information in diseases such as cancer and stroke. Arterial spin labelling (ASL) is a non-invasive MRI technique enabling quantification of perfusion by imaging tissue following exchanges of water with magnetically tagged blood. Although ASL has recently been standardised in the clinic1, pre-clinical usage of ASL remains inconsistent between studies. Our aim here was to identify optimal parameters for pre-clinical ASL, taking into account the anatomical and MRI system differences between humans and rodents, for reproducible and accurate perfusion imaging of metastatic brain tumours in rats. Multiphase pseudo-continuous ASL (pCASL) data were acquired on a 9.4T MRI (Agilent) with a 72mm volume transmit coil and a 4-channel surface receive array (Rapid Biomedical), using radiofrequency pulses with eight phase angles between 0° and 315°, each 45° apart2. pCASL data from eight phases were fit to a modified Fermi function (α=66, β=21) using a modified version of BASIL3 before processing with oxford_asl3. Imaging readout was a multislice single-shot spin echo EPI with FOV=32x32mm (64x64 matrix, slice thickness=1mm), TR=4 s, TE=12.4ms. To better replicate clinical ASL MRI for robust pre-clinical measurements of cerebral blood flow (CBF) in rat, key parameters relating to blood tagging were optimised across 3 strains of rats (Sprague-Dawley, Wistar, and BD-IX) to ensure broad applicability. Sufficient inversion of the blood in the tagging plane in the rat’s neck was estimated to occur at a width of 6.2mm and optimal angle of 45˚, such that it lay perpendicular to the carotid arteries and enabled more homogenous tagging of in-flowing blood. A blood tagging duration of 1.4s (train of 600μs 40° Hanning-shaped pulses, starting every 1.2ms) was found to be the most efficient tag duration, resulting in similar low variation of cerebral blood flow values as in longer tag durations while retaining a shorter TR. A post-label delay of 0.55-0.65s allowed labelled blood water transit to the tissue, prior to image acquisition. This optimised pCASL sequence was next tested in a BD-IX rat model of breast cancer brain metastasis4. Female rats were injected intra-cerebrally with 1000 ENU1564 (N-ethyl-N-nitrosurea-induced rat mammary adenocarcinoma) cells in 1µL PBS, or PBS alone as injection control. Six weeks after tumour cell injection, tumours were readily detected on anatomical images by increased signal on T2-weighted images, and found to have significantly reduced CBF (40 ± 8mL/100g/min) compared to the contralateral striatum (92 ± 12mL/100g/min, p=1.04x10-3) or to healthy brain tissue in PBS injected rats (95 ± 15mL/100g/min, p=3.35x10-5; Figure 1). This is the first instance in which ASL MRI has been optimised for pre-clinical applications. We demonstrate the ability of this optimised sequence to reliably determine changes in tumour perfusion in a rat model of brain metastasis. Pre-clinical ASL has previously been used for glioma perfusion in rats, where similar results were found5.

References

1. Aslop DC, et al. Magn Reson Med. 2014;102–116;

2. Jung Y, et al. Magn Reson Med. 2010;799–810;

3. Chappell MA, et al. IEEE Trans Signal Process. 2009;57:223–236;

4. Serres S, et al. Intern Journ Cancer. 2014;134;4:885–896;

5. Silva AC, et al. Magn Reson Med. 2000;44:169–173.


Articles from Neuro-Oncology are provided here courtesy of Society for Neuro-Oncology and Oxford University Press