#NeuroSpeak: DMF as a second-line agent or not

Did you know that dimethyl fumarate is not as effective when used 2nd- or 3rd-line? #NeuroSpeak #MSBlog


The Mouse Doctor did a good deed this weekend; he kindly agreed to speak about MS treatments at a meeting the allied HCPs hosted at Queen Square our former Institution. At the meeting he was asked about why dimethyl fumarate (DMF) had not being recommended by NICE as a second-line therapy for patients with highly-active or rapidly-evolving severe MS? He wasn’t sure so he dropped me an email.  The reason is that DMF is a much more effective treatment in pwMS who are naive to treatment. When DMF is used 2nd-line it has an impact on relapses, but in the post-hoc analysis it had no impact on disability progression. All the NICE cost-effective models are driven by disability progression and hence in this subgroup of patients it was not considered cost-effective.


This data was presented at ECTRIMS in 2013, but has not subsequently been published. I have uploaded the poster for you so that you can see the data for yourself. The answer to your question is in Figure 4 in relation to disability progression. Please note that the data on DMF in newly diagnosed patients, which is very good has been published. I have suggested to Biogen that they should also publish the 2nd-line data, but my request seems to have fallen on deaf ears. I wonder why?


Based on the post-hoc analysis below I don’t recommend DMF second-line unless the reason for switching treatments is due to a tolerance issue or there are specific reasons for someone requesting DMF. The latter is usually linked to fingolimod or teriflunomide being contraindicated. Whilst we are the topic of 1st-line vs. 2nd-line efficacy some of you may be remember a post I did on Terifluomide being more effective as a 2nd- or 3rd-line DMT compared to when it is used 1st-line. It the only DMT to be more effective when used later than earlier. The finding was consistent across both phase 3 trials and therefore must be linked to real biology. If you can work out why Teriflunomide is an outlier in this regard you may be able to explain something important about the biology of MS. We have some ideas, but I will keep you in suspense. We are trying to get a grant from Genzyme-Sanofi to explore our hypothesis in more detail and will keep you posted. 





BACKGROUND: Delayed-release dimethyl fumarate (DMF) demonstrated efficacy and safety in the Phase 3 DEFINE and CONFIRM trials.

OBJECTIVE: To evaluate delayed-release DMF in newly diagnosed relapsing-remitting multiple sclerosis (RRMS) patients, in a post-hoc analysis of integrated data from DEFINE and CONFIRM.

METHODS: Patients included in the analysis were diagnosed with RRMS within 1 year prior to study entry and naive to MS disease-modifying therapy.

RESULTS: The newly diagnosed population comprised 678 patients treated with placebo (n = 223) or delayed-release DMF 240 mg BID (n = 221) or TID (n = 234). At 2 years, delayed-release DMF BID and TID reduced the annualized relapse rate by 56% and 60% (both p < 0.0001), risk of relapse by 54% and 57% (both p < 0.0001), and risk of 12-week confirmed disability progression by 71% (p < 0.0001) and 47% (p = 0.0085) versus placebo. In a subset of patients (MRI cohort), delayed-release DMF BID and TID reduced the mean number of new or enlarging T2-hyperintense lesions by 80% and 81%, gadolinium-enhancing lesion activity by 92% and 92%, and mean number of new non-enhancing T1-hypointense lesions by 68% and 70% (all p < 0.0001 versus placebo). Flushing and gastrointestinal events were associated with delayed-release DMF.

CONCLUSION: Delayed-release DMF improved clinical and neuroradiological outcomes relative to placebo in newly diagnosed RRMS patients.

CoI: multiple

12 thoughts on “#NeuroSpeak: DMF as a second-line agent or not”

  1. This was meeting for physiotherapists. Over hundred gave up their saturday morning to learn. I gave them the bog=standard this is how it works and then gave them an alternative view.Queen square seemingly didn't have any neurologists that can give up thier saturday morning to educate :-(, so I stepped in. A conference in London…I seem to be drawing the short straw or am fed on scraps.

  2. I enjoy talking to people about my ideas and amazingly I was offered some dosh for doing this..I donated to charity. Better than donating half to the taxman

  3. this is exciting and shows the need for quicker diagnosis… but what what do the physiotherapists have to do with it? lol

  4. The Queen Square MSologists should be ashamed of themselves. Surely one of them could have found the time?

  5. Shouldn't this information about reduced 2nd and 3rd line effectiveness be known to everybody?Some years ago I had to find a new treatment after failing betainterferon and copaxone and TysabriAt that time the DMF trials were over, but DMF was not available yet. The neurologist tried to get DMF, but it wasn't possible to do so.Then, for lack of anything else, the neuro prescribed off-label rituximab. Means it was a lucky escape. Just a year later he would have prescribed DMF instead.

    1. Should information be known to everybody…simple answer is yes. Perhpas this is the type of info that needs to be added into the MS Society/Trust websites.

    2. Can't you guy's do basic arithmetic? 3-1 = 2If you take the whole population of MSers in the phase-3 Tecfidera programme and 'cherry-pick' the responders – MSers newly diagnosed and naive to DMTs – what you are left with in the bottom of the barrel are the non-responders or those previously exposed to other DMTs. Biogen and ProfG has published on the cherries, but expect you to use you nous and calculate what is left behind. Rotten apples?

    3. Anonymous, interesting point. Because I remember being told (neuros) that HSCT works best if you have failed other DMTs. Which is bullshit. There is no data I have come across which shows that HSCT works best for people who have failed *insert DMT of choice*. It's just a dogma that developed along the Chinese Whispers line of discussions on the issue. The funny thing is that half the neuros didn't actually realise this – they honestly thought that HSCT works best if you fail something else first.

    4. I think it is considered the nuclear option after other options have failed. It was the risk profile which made them cautious. Remember in the past the death rate was about 5 percent.It is not now.

Leave a Reply to MouseDoctorCancel reply

Discover more from Prof G's MS Blog Archive

Subscribe now to keep reading and get access to the full archive.

Continue reading