Is it time to kill off whole brain atrophy, and neurofilament light, as a primary outcome measures in progressive MS trials

So the good news is that there is more evidence that ibudilast offers benefit in advanced MS. Let’s strive to get more good news. I think that we can be making it hard. Some people may think my rant is surprising and perhaps this is not the best forum, but it asks a question. The answer may be no.

Effects of ibudilast on MRI measures in the phase 2 SPRINT-MS study.Naismith RT, Bermel RA, Coffey CS, Goodman AD, Fedler J, Kearney M, Klawiter EC, Nakamura K, Narayanan S, Goebel C, Yankey J, Klingner E, Fox RJ; SPRINT-MS investigators.Neurology. 2020 Dec 2:10.1212/WNL.0000000000011314.

We all want to see treatments for progresive/advanced MS and there have been searchces for elusive agents to control progression. We have played our part in hunting, but we have repeatedly, and consistently, said that it doesn’t matter what agent you pick, if the trial design is not responsive to detect effective agents. People have got bored of me saying this.

However, if we had heeded this saying, we would have had agents to treat relapsing MS years before they appeared, we would have had agents to control elements of progressive MS years ago, for example if hand function had been the important outcome measure and not lower limb function. I guess hindsight is a beautiful thing

About ten to twenty years ago I went to an MS meeting in Boston to discuss progressive MS and outcome measures were on the agenda. Brain atrophy (Shrinkage) was discussed and it was apparaent that there are many problems with using it, as the brain can shrink or enlarge for reasons unrelated to nerve loss, which it aims to measure. It was said then that other outcomes such as Grey matter loss was a much better outcome, yet it has not found favour and brain atrophy continues to be used in trials over and over again and sadly progressive MS trials continue to fail. Likewise, spinal cord atrophy is another imaging biomarker, claimed to be an important outcome measure, yet histology shows that you can loose massive amount of nerves but the size of the cord does not shrink that much, so the outocme misses alot.

So let’s look at the current study with Ibudilast. This was tried in relapsing MS and it failed to impact on relapses and lesion formation but it did have an effect on brain atrophy, suggesting it may be active in progressive MS.

So the US Sprinted ahead of the UK to try ibudilast and did a trial in progressive MS. If whole brain atrophy was the primary outcome. The trial would have failed. P=0.08 (They may say it is a trend, but a trend is failure!!!). Had neurofilament light (a nerve protein released during damage), as a biomarker for neurodegeneration, been used as the major outcome. The trial would have failed also! This because ibudilast did not inhibit new T2 lesions in progressive and relapsing MS. Neurofilament becomes more evident when you have active lesions.

So as I say, if the trial design is wrong, you throw away your best candidates. I tried hard to get candidates back on the agenda, but the percieved failure in past trials, and sadly scientific egos, means it is a death knell for such agents. I guess if a big swinger were to say these things, maybe others would listen.

Now, I will also say if you are not prepared to fail you can never succeed, but one has to learn from mistakes and some times you simply have to do the difficult stuff to succeed. It is becoming increasingly clear to me that the neurofilament light signal is driven by the neurodegeneration from active inflammatory lesions, which can be inhibited by current potent disease modifying treatments. Block these lesions and the neurofilament light levels drop. This was evident with our PROXIMUS trial.

Whole brain atrophy is used because it is easier to measure, but use grey matter atrophy and you have yourself a positive result in the ibudilast trial.

“Gray matter atrophy was reduced by 35% for those on ibudilast vs. placebo (p=0.038)”.

Retinal thinning is another of the imager’s favourite but it will only be a useful marker when the lesions project into the visual pathways so whilst changes were seen in the right direction, a trial based on this imaging outcome would fail too. However, I believe we have a better chance of suceess if we combine agents.

We maybe we have a signature for a more successful progressive study. You select people with sufficient neural reserve intact to detect change, you use a potent disease modifying treatment that will block T1 and T2 lesion formation and block neurofilament light and block relapses and you add on a neuroprotective like ibudilast, but there are others that are more accessible, which does not inhibit the former but inhibits grey matter volume loss and you select a clinical outcome like hand function. A combination trial with a potent immunomodulator (in the Sprint trial about 20% or people were on CRAB drugs…sorry in my mind they are not good enough) and a neurorprotective and you can see effects. Hopefully. The PROXIMUS trial has parallels so maybe the positive elements weren’t a fluke, if only ProfK could add an extra arm to ChariotMS…… 🙁