Final Report Summary - PTMETMRI (Probing the tissue microenvironment of tumours by Magnetic Resonance Imaging)
Magnetic Resonance Spectroscopy (MRS) thermometry techniques have been used for monitoring hypothermic treatment temperatures for hypoxic ischaemic brain injury and could be used as cancer tumour environment probes. The absolute temperature measure needs to be accurate to ensure the measure has clinic value. MRS thermometry is a measure of chemical shift between the water position in the spectra and a temperature independent reference. The water position, proton resonant Frequency (PRF), is dependent on temperature but it is also affected by non-temperature factors, such as ionic concentration and fast chemical exchange. If these factors have a large contribution to the MRS thermometry measure then the accuracy of the absolute temperature comes into question and apparent temperature measure can only be obtained. However the non-temperature factor may provide further information about the tissue environment which could be especially useful in cancer tumours. The research aim for PTMETMRI was to investigate the non-temperature factors for increasing the accuracy of absolute temperature measures and utilising the non-temperature factors as tumour microenvironment probes, which has not been investigated before. Therefore the project was split into two areas; a) using the MRS thermometry measure as a tumour diagnosis, prognosis and treatment response marker and b) investigating the underlying mechanisms affecting the MRS thermometry measure. The aim was to contribute novel and unique research to the field but also to investigate if prior work using MRS thermometry would be affected by this projects finding.
The clinical part of the project involved investigating the apparent temperature measure for distinguishing two types of common childhood brain tumours as an additional and complimentary measure to MRI methods utilised in the clinic. MRS is used clinically as a measure of metabolic activity in the brain, which provides multiple metabolite concentration measures. This can be combined as a spectral profile for differentiating tumour types but has reduced efficiency for understanding how the tumour is behaving. The MRS thermometry measure is gained through a simple analysis of the MRS spectra (chemical shift) and is therefore a complimentary additional measure. One of the main focuses of the project is how the measure can add to the understanding of tumour function. In achieving this thoroughly the project also investigated other MRI measures, the T2, metabolite T2s and diffusion weighted imaging (DWI) for their complimentary nature in aiding clinical management of cancer patients and improving outcomes.
The investigation into the non-temperature based effects was performed using temperature controlled phantoms, translatable to the International Temperature Standards 90. The equipment was provided by the Temperature Group at the National Physics Laboratory to ensure the temperature was extremely accurate. The phantom temperature could be maintained to less than 0.1oC during experiments. There have been literature studies suggesting that two main factors may affect the MRS thermometry measure; ionic concentration and chemical exchange. However these studies were either not translatable to the ITS 90 standards or did not thoroughly investigate aspects of the non-temperature effects. The initial work in this project found that ionic concentration and protein content, a model for chemical exchange, affected the MRS thermometry measure. Specifically, an increase in ionic concentration to 100 mM changed the apparent temperature (while the actual temperature remained constant) by 0.75°C and an increase of 1% protein content changed the measure by -0.2oC. Ionic concentrations have been show to vary in healthy brain regions by ~10mM and between healthy and tumour tissue by ~50mM. Protein content has been shown to change up to ~3% in healthy regions of the brain but protein content in childhood brain tumour types has not been specifically investigated. While the protein content difference is small between brain regions the water content variability is larger, for example there is ~10% water content difference between white and grey matter, increasing the spatial variability of the chemical exchange effect on the temperature measure. The gel experiments found that the increase in agar concentration changed the MRS thermometry measure, however chemical exchange was found not to be the main driver as it is a different type of exchange to the protein content experiment. This result was important as it reduces a factor in this complex system that will affect the measure in an in vivo system. The gel phantom experiments also concluded that using scanners with different magnetic field strengths does not significantly affect the measures. Therefore the measure may be robust across different scanner field strengths. These factors may increase the error of the absolute temperature in healthy brain monitoring and most certainly will affect brain tumour measures.
A high grade tumour type (Medulloblastoma) and low grade tumour type (low grade glioma) cohort was chosen to investigate the MRS thermometry measure. Results showed a significant difference between the two tumour types with the higher grade tumour type being ~1.4°C lower in temperature compared to the low grade gliomas, assuming temperature was the only contributor to the measure. This was the opposite result to expectations; high grade tumours have greater metabolic activity, which may be expected to drive the temperature higher. Increased thermal transportation may play a part in explaining the unexpected result as medulloblastomas will have higher blood flow and regulate the tumour temperature better but Dr Babourina-Brooks theorised that non-temperature based factors were significantly contributing to the results as well. Comparisons between the DWI results showed that the increased cellular density/metabolic rate did correlate with the MRS thermometry results when all the tumour data was combined, however this was not significant within each tumour type. This suggests the cellular density/metabolic rate may not be the only factor affecting the MRS thermometry measure. Comparisons with metabolite and water T2 values also showed that the water T2 was significantly different between tumour types. This can be explained by the contribution of the types of water signal within the tumour. The water signal can be split into bulk and chemically bound water. In gliomas more MRI signal comes from the bulk water and chemically bound water has less of contribution to the overall signal due to their micro-cystic nature. This relates to the MRS thermometry measure as more bound water means more chemical exchange, which was shown by our protein content phantom results. Increased chemical exchange was shown to decrease apparent temperature and may be a main driver for the difference observed between these tumour groups. Ionic concentration may be significantly different in tumours and may be affecting the measure also; however the ionic effect is in the same direction as the temperature. Therefore for this study it would mean the low grade glioma’s had a higher concentration of ions, which has not been investigated to the author’s knowledge. Both of these factors will change with tumour environment and can serve as a probe for tumour function.
The research involved in PTMETMRI has shown that the MRS thermometry can be useful in distinguishing tumour types and by understanding the complex non-temperature based effects on the measure it is possible to probe the tumour even further. It was found that the MRS thermometry measure is complimentary to the standard MRI values gained through T2 weighted, MRS and DWI at no extra cost. Additional information gained through imaging can aid in improving the clinical management of patients with brain tumours and lead to better outcomes and exploring novel techniques in improving this.
Furthermore, Dr Babourina-Brooks’ transfer of knowledge activities inspired a PhD project to investigate more advanced models of diffusion for body and brain tumours, which he is involved in. He consulted with clinicians from different hospitals on diffusion protocols and changed the way the clinic utilised the technique. Dr Babourina-Brooks provided presentations and consultation to the host institutions researchers on diffusion-MRI related matters. His integration and enthusiasm for the work have earned him an NIHR research fellow position within the same host institution.