Bad cop behaviour: Pushing NEDA over the RIM!


Over the last years, it has become apparent that some MS brain lesions are more detrimental than others. One MS lesion subtype that is specifically attracting attention is the “iron rim containing lesion” or IRL. A ‘normal’ MS lesion represents inflamed brain tissue. Over time, the ‘normal’ MS lesion is likely to shrink a little bit because of the resolution of inflammation-induced swelling and the loss of myelin and axons. However, some lesions seem to be expanding rather than shrinking over time, albeit slowly. We can distinguish these bad lesion cops from less bad lesion cops (nu such thing as a good lesion) by looking at the presence of an iron rim with a special MRI sequence (susceptibility-weighted imaging). This iron rim corresponds to an edge of macrophages/microglia that are actively phagocytosing or eating myelin. As the destruction of oligodendrocytes and myelin leads to the release of their iron content, this binge eating leads to piling up iron. Consequentially, these IRLs are more destructive with a tendency to leave black-holes, which we know are associated with more loss of axons. 

There is now a new study that provides interesting longitudinal data on these IRLs in 33 pwMS. The study was also able to document the appearance of some new IRLs. In the study, there were 16 new IRLs in 6/33 pwMS during follow-up. Interestingly, 10/16 new IRLs were seen in 2/6 patients which were both very early in their disease course. When a new IRL was formed, iron accumulated diffusely in the lesion. Subsequently, the iron became organised as a rim around the lesion and over time the iron rim (but not the lesion itself) could disappear. As many IRLs were already present at study onset, the study did not allow to document whether gadolinium-enhancement and thus florid inflammation preceded the formation of these lesions. These observations imply that you do not necessarily develop more IRL’s when living longer with MS. Some of the lesions that we label as ‘normal’ MS lesions might have been IRLs in a previous life. Hence, there is a continuum between chronically active IRLs and inactive lesions. 

Unfortunately, it is not possible (yet) to visualise the iron rims on the routine brain scans we do in the context MS follow-up; visualising the iron rim is still in a research phase. Nonetheless: new data, new wanderings on the topic:  

First, IRLs clearly provide real-time information about the destructiveness of a chronic active MS lesions. When they expand locally, they might be the underlying substrate for progression and thus new or expanding disability. An IRL in a motor pathway can cause progressive weakness down one side of the body. However, it is currently unclear if these IRLs are only locally destructive or whether they also mark more widespread neurodegeneration which would potentially result into an increase of neurofilament levels (= brain debris marker) or thinning of the retina measured by OCT. In other words: Do we need to monitor and target them separately?, or do they correlate with other markers of neurodegeneration in MS? 

The key second question to solve is what triggers the formation of these iron rims: Does an IRL always appear after an inflammatory brain insult (i.e. gadolinium enhancement)? In this situation, the microglia would be responding to something in the surrounding tissue that is activating them, and they are basically doing their job by clearing this trigger. Or can they appear spontaneously in regions that have never been inflamed before? The latter is less likely based on the current data but would point to an intrinsic dysfunction of the microglia (and would be difficult to treat).

Third, a philosophical question that applies to many things in life: how can we get rid of it when it’s already there. Currently available DMT’s could potentially prevent the formation of new IRLs by preventing the appearance of new inflammatory lesions. However, if IRLs are already present at baseline, it is unlikely for DMTs to have a big impact. DMT’s namely work on B/T cells while the rim is full of macrophages. The available data on Ocrevus showed that the drug has only a very limited effect on the destructive nature of existing (iron rim) expanding lesions. If an individual has more IRLs at baseline, do we manage expectations about our ability to control their disease with a DMT?  Would it be advantageous to refer them for aHSCT or Lemtrada instead of Ocrevus? 

Altogether, IRLs are a more destructive lesion subtype that is present early in the MS disease course. When they are present, it becomes very difficult to label someone as a real NEDA or ‘no evidence of disease activity’, irrespective of other factors. This means that even if there are no entirely new lesions on MRI and no clinical relapses, there would be still a fair chance of disease progression. Unfortunately, there are still a lot of unknowns about how to “trim the iron rim”, especially DMT-wise, making it difficult to use the marker in practice. To be continued in the next decade!

Twitter: @SmetsIde

Disclaimer: Please note that the opinions expressed here are those of Ide Smets and do not necessarily reflect the position of the Barts and The London School of Medicine and Dentistry nor Barts Health NHS Trust.

Brain. 2021 Apr 12;144(3):833-847. doi: 10.1093/brain/awaa436.

Long-term evolution of multiple sclerosis iron rim lesions in 7 T MRI

Assunta Dal-Bianco 1, Günther Grabner 1 2, Claudia Kronnerwetter 3, Michael Weber 3, Barbara Kornek 1, Gregor Kasprian 3, Thomas Berger 1, Fritz Leutmezer 1, Paulus Stefan Rommer 1, Siegfried Trattnig 3, Hans Lassmann 4, Simon Hametner 1Affiliations expand

  • PMID: 33484118
  • DOI: 10.1093/brain/awaa436


Recent data suggest that multiple sclerosis white matter lesions surrounded by a rim of iron containing microglia, termed iron rim lesions, signify patients with more severe disease course and a propensity to develop progressive multiple sclerosis. So far, however, little is known regarding the dynamics of iron rim lesions over long-time follow-up. In a prospective longitudinal cohort study in 33 patients (17 females; 30 relapsing-remitting, three secondary progressive multiple sclerosis; median age 36.6 years (18.6-62.6), we characterized the evolution of iron rim lesions by MRI at 7 T with annual scanning. The longest follow-up was 7 years in a subgroup of eight patients. Median and mean observation period were 1 (0-7) and 2.9 (±2.6) years, respectively. Images were acquired using a fluid-attenuated inversion recovery sequence fused with iron-sensitive MRI phase data, termed FLAIR-SWI, as well as a magnetization prepared two rapid acquisition gradient echoes, termed MP2RAGE. Volumes and T1 relaxation times of lesions with and without iron rims were assessed by manual segmentation. The pathological substrates of periplaque signal changes outside the iron rims were corroborated by targeted histological analysis on 17 post-mortem cases (10 females; two relapsing-remitting, 13 secondary progressive and two primary progressive multiple sclerosis; median age 66 years (34-88), four of them with available post-mortem 7 T MRI data. We observed 16 nascent iron rim lesions, which mainly formed in relapsing-remitting multiple sclerosis. Iron rim lesion fraction was significantly higher in relapsing-remitting than progressive disease (17.8 versus 7.2%; P < 0.001). In secondary progressive multiple sclerosis only, iron rim lesions showed significantly different volume dynamics (P < 0.034) compared with non-rim lesions, which significantly shrank with time in both relapsing-remitting (P < 0.001) and secondary progressive multiple sclerosis (P < 0.004). The iron rims themselves gradually diminished with time (P < 0.008). Compared with relapsing-remitting multiple sclerosis, iron rim lesions in secondary progressive multiple sclerosis were significantly more destructive than non-iron rim lesions (P < 0.001), reflected by prolonged lesional T1 relaxation times and by progressively increasing changes ascribed to secondary axonal degeneration in the periplaque white matter. Our study for the first time shows that chronic active lesions in multiple sclerosis patients evolve over many years after their initial formation. The dynamics of iron rim lesions thus provide one explanation for progressive brain damage and disability accrual in patients. Their systematic recording might become useful as a tool for predicting disease progression and monitoring treatment in progressive multiple sclerosis.

About the author

Ide Smets


  • When 7T MRI was FDA approved around 2017, one of my first comments was that it would divide MS experts into those who have access to 7T and those who do NOT (at the University level, that is). Hence, those who do NOT have access but MAY have ideas to explore, are left wondering (or holding the bag) of the possibilities. How do we bridge this gap ? The have and have-nots in medicine ?

    At every University, IF the federal government(s) decides to have a 7T installed (no easy task, I know, and certainly of minimal benefit if the ‘MS faculty’ is only minimally interested in academic pursuits using such fantastic technology) then the playing field will be LEVEL.

    Cortical changes on MRI in MS as well as so-called CVS (central vein sign) and now of course the iron rim sign are all BEST visualized on 7T which is why anything and everything is NEW and publishable. And for those of us with zero access to the ‘grand opera’ of MRI technology as it currently stands today, we can only blink in awe and teach our residents of those wonderful findings on a far, far away galaxy using the power of imagination or the lack, thereof !

    • Some of them get in the CNS so that’s a good place to start and certainly may have an influence on these cells. I hope it gets investigated in more detail.

  • These observations imply that you do not necessarily develop more IRL’s when living longer with MS

    This is very interesting 🙂

    Prl are also associated with ocb

    Paramagnetic Rim Lesions on MRI are Specific for the Presence of Oligoclonal Bands in Multiple Sclerosis

    Ps they can also be view in 3 tesla mri

    They dont necessarily fully associate with SEL

    MRI Characteristics of Chronic MS Lesions by Phase Rim Detection and/or Slowly Expanding Properties

    P119 / P119 – Paramagnetic rims in treatment naive persons at the time of multiple sclerosis diagnosis

    Conclusions: Paramagnetic rims are present at the time of MS diagnosis and are characterized by lower myelin content compared to chronic lesions without a rim. This indicates that the underlying pathology is likely more aggressive, though not necessarily active. Work is ongoing in our group to enlarge the sample size and assess the relationship between paramagnetic rims and features of a more aggressive disease course at disease onset on an individual patient level.!/9245/presentation/276

    Scary and complicated stuff

    Nice post

    Thanks for this

    • “Scary and complicated stuff”…that’s neurology…

      “For years, the neurologist in Moncton, New Brunswick, has seen patients with symptoms common to Creutzfeldt-Jakob disease, a fatal brain disorder that affects one in 1 million people each year.
      But diagnostic testing for the rare neurodegenerative syndrome keeps coming back negative, more patients with similar symptoms have turned up each year, and Marrero hasn’t found another cause…….Some have hallucinations, including what Marrero said are “terrifying hallucinatory dreams” that leave them afraid to go to sleep, and tactile hallucinations in which they feel as if insects are crawling on them. One symptom, particularly devastating for loved ones, is Capgras delusion, a belief that family members have been replaced by impostors.”

    • HI Louis. Thanks for the references. Very interesting. Association with OCB is nice, and logic given both OCB and IRLs are very specific for MS. I wonder what the substrate of a SEL without an iron rim is. Would these SELs be a sort of IRL that is fading out? Or something different? TBC

  • Great post. Would increasing blood flow to brain help clear rim? Over de ade ago, Out of desperation during bad flair I gunieaed pig myself and took ginseng and ginkobolbla at doses deemed safe and beneficial to stroke victims in study trials for 6 mos. Also, with hemotolgist consult I took aggressuve doses of b12. I added vit c, e, d, a at doses recommended by neuro. Interestingly, I stayed away from Iron supplements because a significant number of MSers on open forums indicated they felt it correlated to worsening of symptoms. My rim lesion stayed enhanced greater than 6 mos, but I continued to clinically get better and lesion less apparent on MRIs for next 3 yrs.

    • Great to hear your MRI improved. Even without supplements you would have ingested a lot of ‘iron’ as dietary iron is present in red meat/vegetables/fruit/beans and thus very difficult to avoid. The fact that we are seeing an iron rim is the consequence of inflammation killing cells and cells releasing their content in the process of dying – including iron which is key for a cell’s energy production.

  • If I look on Google for:
    “deposition of complement and antibodies multiple sclerosis” several papers pop out. To me it is very probable that OCB are actively driving progression. Antibodies activate cells and complement to do their job and that is what we find in MS. So I would go for sizomus… It would be interesting to have a look to the people enrolled at least on a 3T scanner for SEL/IRL/PRL.
    “Or can they appear spontaneously in regions that have never been inflamed before?”
    Do new lesions appear in pwMS treated with Anti-CD20? If the answer is no then we need the acute lesion to have IRL. If not microglia is not dysfunctional it’s just doing its job

    • Sizomus will hopefully provide proof of concept in MS about OCB’s being pathogenic or not. The question is indeed whether removing plasma cells/OCB’s would solve everything progression-wise.. If yes, then at least no IRL should appear under treatment. If no, we need a separate strategy for the IRLs.

By Ide Smets



Recent Posts

Recent Comments