UCL Queen Square Institute of Neurology


MRI Studies on Grey Matter and Periventricular White Matter in MS Patients

MTR Gradient Image.



Dr. Declan Chard

Multiple sclerosis (MS) has long been thought of as mainly a disease of brain white matter (WM), and its most obvious clinical feature (a relapse) has been closely linked with the formation of new WM lesions. However, it has become clear that there are abnormalities in brain grey matter (GM) too, albeit difficult to see with conventional MRI scans, and that in the longer-term these are significantly associated with disability.

MS Grey Matter Lesions


As with WM, lesions form in GM too. However, in contrast to WM, where conventional MRI scans show the majority of lesions, GM lesions are much more difficult to detect. Double inversion recovery is the most used method for detecting GM lesions, and prospectively about 10% of such lesions are seen. However, using a newer type of brain scan called Phase Sensitive Inversion Recovery (PSIR), compared with previous scans we were able to more confidently detect GM lesions, and we were also able to assess their evolution over time, showing that while GM lesions may extend into WM the reverse is rarely seen (Sethi et al. 2016: https://www.ncbi.nlm.nih.gov/pubmed/26272028). This suggests that the mechanisms underlying GM lesion expansion are mainly from the outside rather than inside of the brain.

Role in the Diagnosis of MS

WM lesions can occur in people who do not have MS, and are seen as part of normal ageing. We investigated whether or not GM lesions seen using PSIR could help with the diagnosis of MS, i.e. are they found mostly in people with MS and rarely in others, and so we also looked for such lesions in people with no known neurological problems. As with WM, we found that with increasing age some GM lesions were seen, but there were significantly fewer in people who did not have MS. This suggests that PSIR may, with further development, be able to help with the diagnosis of MS. We have published the first scientific paper from this work (Sethi et al. 2012: https://www.ncbi.nlm.nih.gov/pubmed/22807559 and Sethi et al. 2013: https://www.ncbi.nlm.nih.gov/pubmed/24244381).

Long Term Clinical Relevance

MS is a very variable disease, with some people with MS developing little disability over time and others more rapidly developing impairments. We recently undertook a 30-year follow-up study of people who had a first episode suggestive of MS (a clinically isolated syndrome). In those who developed MS, and who were alive at the 30 year follow-up, about half had relapsing-remitting (RR) MS and remained fully able to walk, and half had secondary progressive (SP) MS and impaired walking. We assessed MRI scans at 30 years to look for features (such as brain atrophy, WM and GM lesions) that could distinguish people with RRMS from those who had developed SPMS. The most striking difference was in the number of GM lesions, which were seen in everyone with SPMS, but were seen in few people with RRMS (Haider et al. 2021: https://pubmed.ncbi.nlm.nih.gov/33880511/). This suggests that GM lesions may play an important role in determining a progressive MS course, and we are planning further work to assess this.

Remyelination Treatment Trials

We now have many treatments designed to reduce the formation of WM lesions and associated clinical relapses. However, once a lesion forms there are no established treatments that enhance natural repair mechanisms, in particular that promote remyelination. Working with teams in Cambridge and Edinburgh, we undertook the MRI analysis for a trial of bexarotene. The drug caused significant side effects and so is not suitable for clinical use in MS, and the main trial outcome (changes in all MS brain lesions) was negative. However, when we looked at WM and GM lesions separately we found a clear treatment effect in GM lesions, and more subtle effects in some WM lesions (Brown et al 2021: https://pubmed.ncbi.nlm.nih.gov/34418398/). This result was surprising as we did not set out to image GM lesions, but is potentially very relevant to future remyelination trials as it highlights that the potential for remyelination differs substantially between lesions, and that when looking for treatment effects this should be taken into account.

Grey Matter Magnetisation Transfer Imaging

Using an MRI technique called Magnetisation Transfer Ratio (MTR) imaging, which is sensitive to the amount of myelin in brain tissue, we can look for subtle changes in brain GM tissue structure. With MTR we found differences in the GM of people with MS compared with people with no known neurological problem, and in particular that GM MTR was more abnormal towards the outer surface of the brain (Samson et al. 2014 https://www.ncbi.nlm.nih.gov/pubmed/24552746). This gradient was more noticeable in people who had secondary progressive MS when compared with those who had progressive symptoms from the outset (primary progressive MS), and again suggests that mechanisms underlying GM pathology in MS may be influenced by processes arising close to the surface or outside the brain. This finding may be important when developing new treatments and tailoring them to individual people with MS.

Periventricular White Matter Magnetisation Transfer Imaging

Arising from our work on GM, which highlights the possibility that factors outside the brain (for example in the cerebrospinal fluid) may affect pathology within it, using MTR we have looked for evidence of similar gradients in WM. We found these in people with relapsing-remitting and secondary progressive MS and, as with GM MTR gradients, these were greater in people with secondary progressive MS (Liu et al. 2015: https://www.ncbi.nlm.nih.gov/pubmed/25823475). In more recent work, we have shown that these periventricular gradients may be found soon after the first clinical episode in people with MS, and that they predict the risk of having a further relapse (Brown et al. 2017: https://www.ncbi.nlm.nih.gov/pubmed/28043954). We have also shown that treatment with alemtuzumab, a potent MS disease modifying treatment, reduces abnormal periventricular MTR gradients. The steepness of the gradients before the treatment in part predicts the clinical response (Brown et al. 2019: https://www.ncbi.nlm.nih.gov/pubmed/31169059). We continue to pursue this line of research.