“Only your eyes can show the suffering” dixit Fausto Coppi


Once upon a time, there were cyclists as Fausto Coppi. He was the kind of guy that attacked 5 km after the start of the race (Milan San-Remo 1946), and then dropped the rest of the lead group with 14 minutes on the ‘Passo del Turchino’ or ‘who-wins-on-the-Turchino-wins-the-race’. He not only won the Tour de France twice, he did it with a ‘martian’ dominance. In 1949, there was a 14’ difference with his lifelong rival Gino Bartali and in 1952 there was an 28’ difference. No Swiss watch needed. May I remind you that Pogacar only beat Roglic with 59’’ in the TDF of 2020? Once Fausto broke away from the peloton, you never saw him again. This happened five times in both the Giro di Lombardia and Giro d’Italia. It was this legend that claimed “Only your eyes can show the suffering,” and we could not agree more (although calves and eyes would probably be more accurate).

In a new study, Lambe et al. showed that in pwMS the thickness of the ganglion cell layer which is located at the back of your eye corresponded with disability worsening 10 years down the line. If the ganglion cell layer measured less than 70 micron, there was a 4-fold increased risk of a meaningful rise on the EDSS disability scale. The ganglion cell layer is a very specific layer of the retina that contains the naked, unmyelinated cell bodies of neurons. The branches of these neurons, also ‘axons’, form the optic nerve. If the optic nerve becomes inflamed and demyelinated, neurons die and this leads to retrograde thinning of the ganglion cell layer. Importantly, this also happens in pwMS that did not have obvious eye problems which is one of the hallmarks of smouldering MS. Therefore, the thickness of the ganglion cell layer is probably a reflection of your brain reserve, and thus mirrors how much neurons you are left with after MS inflammation attacks your brain. This layer can be measured using ‘OCT’ or ‘optical coherence tomography’’ in which an infrared laser beam is directed at your retina. The reflection of this beam allows to model the 3D-shape of the retina. OCT is essentially very similar to an echography.

These are very important findings as the MS community is literally craving to have a reliable, easy-to-implement biomarker of brain volume loss or ‘end organ damage’. Of note, OCT is more promising than the other techniques that aspire to quantify brain volume loss. 

Brain volume measurements on MRI have been extensively researched, but have the following shortcomings:

  • Brain volume is a composite measure. It does not only incorporate axons/neuronal bodies but also a lot of glial cells, myelin/oligodendrocytes and water.
  • If pwMS start an MS treatment, the brain volume initially disproportionately reduces because the inflammatory oedematous lesions shrink and thus not because there is ongoing damage to the brain. To avoid this confounder, repeated measurements of brain volume over time are needed. 

Neurofilament light levels are released when neurons die and can be measured in blood and the cerebrospinal fluid. However, there are the following shortcomings:

  • Cerebrospinal fluid levels are more reliable to quantify but require a lumbar puncture. Hence, they are not suited for repeat measurements. 
  • Blood neurofilament levels are easier to access but levels are so low there is a big overlap with healthy individuals. This means it is very difficult to establish reference ranges that would allow to distinguish healthy ageing from pathological ageing in MS. 

The next line of research should focus on how OCT measurements could be implemented in clinic practice. Imagine we would have a measurement of ganglion cell layer thickness when you first attend our clinic and when we decide on a treatment strategy. If your ganglion cell layer is thinner than average (for example lower than 70 micron), how should we act on this – especially if the person with MS at that point is relatively unaffected? Should this prompt a more aggressive second or even third line treatment strategy (e.g. stem cell transplantation)? Personally, I think the answer is yes. Low levels of ganglion cell layer thickness flag there is not much brain reserve left to deal with the damage caused by potential future relapses. In other words, there is a narrow window for disease modification and it needs to be seized asap. This is the trap and beauty of human biology. As we have so much reserve and parallel circuits in the brain, we only present with clinically meaningful symptoms when already a significant amount of our neurons has died. Moreover, the OCT measurements may also be used to properly counsel patient. If your levels are low already at the outset of your disease, it would be wise to manage expectations personally and professionally. 

When Fausto Coppi rode his first race at age 15, he won first prize: 20 lire and a salami sandwich. This article is the first race of OCT as a clinically useful marker of brain damage in pwMS. We need more data, preferably from clinical trials, to understand how ganglion cell layer thickness evolves over time, which cut-offs can be used, how it reflects for example spinal cord damage, what would be a meaningful reduction in ganglion cell layer thickness and to what extent treatments can modify it. Nonetheless, a glimpse at the back of your eye can lift the veil on a suffering MS brain.

Twitter: @SmetsIde

. 2021 Mar 2;10.1212/WNL.0000000000011788.doi: 10.1212/WNL.0000000000011788. Online ahead of print.

Association of Spectral-Domain OCT With Long-term Disability Worsening in Multiple Sclerosis

Jeffrey Lambe 1, Kathryn C Fitzgerald 1, Olwen C Murphy 1, Angeliki G Filippatou 1, Elias S Sotirchos 1, Grigorios Kalaitzidis 1, Elena Vasileiou 1, Nicole Pellegrini 1, Esther Ogbuokiri 1, Brandon Toliver 1, Nicholas J Luciano 1, Simidele Davis 1, Nicholas Fioravante 1, Ohemaa Kwakyi 1, Hunter Risher 1, Ciprian M Crainiceanu 2, Jerry L Prince 3, Scott D Newsome 1, Ellen M Mowry 1, Shiv Saidha 1, Peter A Calabresi 4Affiliations expand

  • PMID: 33653904
  • DOI: 10.1212/WNL.0000000000011788


Objective: To evaluate whether a retinal spectral-domain optical coherence tomography (SD-OCT) assessment at baseline is associated with long-term disability worsening in people with multiple sclerosis (PwMS), we performed SD-OCT and Expanded Disability Status Scale (EDSS) assessments among 132 PwMS at baseline and at a median of 10 years later.

Methods: In this prospective, longitudinal study, participants underwent SD-OCT, EDSS, and visual acuity (VA) assessments at baseline and at follow-up. Statistical analyses were performed using generalized linear regression models, adjusted for age, sex, race, MS subtype, and baseline disability. We defined clinically meaningful EDSS worsening as an increase of ≥2.0 if baseline EDSS score was <6.0, or an increase of ≥1.0 if baseline EDSS score was ≥6.0.

Results: 132 PwMS (mean age: 43 years; n=106 patients with relapsing remitting MS) were included in analyses. Median duration of follow-up was 10.4 years. In multivariable models excluding eyes with prior optic neuritis, relative to patients with an average baseline ganglion cell+inner plexiform layer (GCIPL) thickness ≥70µm (the mean GCIPL thickness of all eyes at baseline), an average baseline GCIPL thickness <70µm was associated with a 4-fold increased odds of meaningful EDSS worsening (adjusted odds ratio: 3.97; 95% CI: 1.24-12.70; p=0.02), and an almost 3-fold increased odds of low-contrast VA worsening (adjusted odds ratio: 2.93; 95% CI: 1.40-6.13; p=0.04).

Conclusions: Lower baseline GCIPL thickness on SD-OCT is independently associated with long-term disability worsening in MS. Accordingly, SD-OCT at a single time-point may help to guide therapeutic decision making among individual PwMS.

Classification of evidence: This study provides Class I evidence that lower baseline GCIPL thickness on SD-OCT is independently associated with long-term disability worsening in MS.

About the author

Ide Smets


  • Ide,

    Nice post. I still cycle 30 mins every day (now on a hybrid bike not a racing bike). Your post sent me back to the early 80s and the disbelief of my father at the amount I had spent on a Campagolo super record chainset!

    “The ganglion cell layer is a very specific layer of the retina that contains the naked, unmyelinated cell bodies of neurons. The branches of these neurons, also ‘axons’, form the optic nerve. If the optic nerve becomes inflamed and demyelinated, neurons die and this leads to retrograde thinning of the ganglion cell layer.” Does the fact that the ganglion cell layer is unmyelinated suggest that MS is not really a “disease where the immune system attack myelin” as the various MS societies always claim on their websites?

    • Campagnolo’s are definitely the best. I still have an old retro bike with campagnolo gears in the cellar!

      Good question. MS is definitely a disease that attacks myelin and leads to demyelination but subsequently demyelination leads to dying of axons/neurons. It’s the latter that will ultimately determine the level of disability and not necessarily the extent of demyelination. So although the demyelination is the trigger, the damage that matters in the end is the one to neurons.

      This is the big advantage of OCT. It focuses only on neuronal density and is not confounded by other brain components such as myelin and glial cells.

      • Ide, is it possible to assign a timeframe to the dying of nuerons / axons associated with an MS lesion…..for instance, 5 years after an ‘attack’ if you havent lost any function etc is it safe to say that the death of those nuerons / axons has been avoided, or at least enough of them have survived to maintain function?

        • I wish I could do this but unfortunately there is no fixed term after which a lesion becomes ‘benign’. Especially, as recently it has become clear that some of the lesions in the brain are ‘SEL’ or ‘Slow Expanding Lesions’. This means that over the course of years they gradually expand in volume driving disability accumulation.

          Moreover, the reduction in neuronal density is not always associated with the number of MS lesions = focal inflammation but rather the consequence of diffuse widespread inflammation covering the entire brain. This widespread inflammation can unfortunately not be visualised on MRI but is also a big driver of neuronal death – independent of MS lesions.

  • Can anyone with MS even get OCT on the NHS? I’m seeing an ophthalmologist in 3 weeks. Is it even worth asking, given that it’s only to see if my cataract removal went OK? Have either of you ever tried a Rohlof hub. Extremely good but ferociously expensive.

    • Hi Kit! It is available through the NHS and performed by the ophthalmologists but not for “prognostic” purposes like described in this study. Currently, it is used to help out with the differential diagnosis when people would present with an atypical optic neuritis. So if we want to use OCT in pwMS it would have to be in the context of a research study.

  • I am extremely fortunate to have gotten OCT annually in the US and totally agree and wish eventually it will become SoC.

    • How is the information coming from an OCT measurement used in the US? What happens with the results?

      • Ide – While some U.S. based University Hospitals perform annual OCT scans, there is not much done with the results. Most Do not believe in OCT as a marker of disease progression. OCT in the U.S. is mainly still in clinical trials: https://my.clevelandclinic.org/health/diagnostics/17293-optical-coherence-tomography

        In most cases, OCT scans are used to monitor adverse effects of certain DMTs.

        When I was part of the TREAT-MS trial, John Hopkins was collecting OCT data.

        • I think they were right in the early days of OCT but there is more and more evidence that OCT is indeed prognostic, especially if you use the ganglion cell layer thickness rather than the older retinal nerve layer thickness. Overall, it’s correct to include OCT in research trials because it’s not ready yet to be implemented in clinical practice.

  • Correct me if I’m wrong, but doesn’t Specsavers and John Lewis offer OCT screening as part of your annual sight test?

    • Yes, I think this is correct but I’m not sure whether their software would be suited to dissect and accurately quantify this ganglion cell layer thickness. This measure is quite novel and not so commonly used in other ocular pathologies such as glaucoma or diabetic retinopathy.

  • I.m in US and also get OCT with my annual exam since about 2010. It’s not covered by insurance and costs btwn $35-$50. The optomolgist pulls image up on screen right after and gives brief review of how healthy retina looks and blood vessels. I ask for comparison from previous yr, and Dr. Does split screen and says yep looks good. No info in chart other than it was done, and prior years at mercy of dr storage system. I will look into if thickness value can be charged. No pun intended.

    • Interesting, although I think ‘qualitative’ comparison by the naked eye between results from different years might not be that informative in terms of MS prognosis.

    • Hi Rogier, interesting study. This study shows that eye movements abnormalities ‘square wave jerks’ are more common among pwMS. This is not odd at all as many pwMS present with symptoms such as double vision or nystagmus which is a reflection inflammation at brain stem level. However, this is a descriptive study and cannot be used for prognosis such as OCT. Eye movement abnormalities do not reflect widespread neuronal damage in the central nervous system but are rather a reflection of local pathology at brain stem level.

  • My doctors use inputs from several sources – how you’re feeling, mri, neuropsych eval, and oct to measure progression. If any one measure shows change year to year then further discussion and reevaluation of treatment is at least considered. If all are unchanged it’s considered positive. I think it’s useful and hoping more people have access to all of these and that more metrics can be added too to better predict progression and prognosis and inform treatment decisions

    • Yes, indeed, very similar study design and many more outcomes such as cognition, QoL and MRI seem to be incorporated! Will lead to interesting results.

By Ide Smets



Recent Posts

Recent Comments