Tuesday, 11 September 2018

The leader of the pack

Mult Scler Relat Disord. 2018 Aug 29;25:322-328. doi: 10.1016/j.msard.2018.08.026. [Epub ahead of print]

Efficacy and safety of monoclonal antibody therapies for relapsing remitting multiple sclerosis: A network meta-analysis.

Xu X, Chi S, Wang Q, Li C, Xu B, Zhang J, Chen X.


Several monoclonal antibodies have been licensed for relapsing remitting multiple sclerosis (RRMS). It is still unclear which treatment regimen should be recommended due to the lack of head-to-head randomized controlled trials (RCTs). This study aims to investigate the relative efficacy and safety of existing monoclonal antibody therapies in treating RRMS.


We searched PubMed, Embase, and the Cochrane Library for RCTs of monoclonal antibodies for treatment of RRMS. We performed a network meta-analysis to identify evidence comparing monoclonal antibodies with one another, interferon beta-1a (INFβ-1a), or placebo in adult patients with RRMS. The primary efficacy outcome was annualized relapse rate and the primary safety outcome was incidence rate of serious adverse events.


A total of 14 eligible studies containing 9412 patients treated with 7 regimens were analyzed. INFβ-1a was the most common comparison treatment and showed an annualized relapse rate of 45.3%. All monoclonal antibody regimens, including natalizumab, natalizumab plus INFβ-1a, alemtuzumab, daclizumab, and ocrelizumab, were associated with significant reduction in annualized relapse rate and similar risks of serious adverse events. Cluster analysis showed that natalizumab plus INFβ-1a and alemtuzumab performed best in terms of high efficacy and safety. Natalizumab and daclizumab were characterized by high efficacy but relatively high risk of serious adverse events. Ocrelizumab was differentiated by high safety but relatively poor efficacy.


This network meta-analysis provided a comprehensive summary of efficacy and safety of monoclonal antibodies for RRMS, which might provide a reference for treatment. More direct comparison studies are warranted.

Figure: Network meta-analysis of monoclonal antibodies compared with INFβ-1a for primary efficacy and safety outcomes. (a) Annualized relapse rate, (b) incidence rate of sever adverse events, (c) cluster rank plot of ranking estimation for reducing annualized relapse and serious adverse events. Mean with 95% CI represent the risk ratio and its 95% confidence interval of annualized relapse rate and serious adverse events. The dashed lines represent the different quadrants of the ranking estimation. ALE = alemtuzumab; DAC = daclizumab; NAT = natalizumab; OCR = ocrelizumab; PBO = placebo.

If I'm asking you a question, chances are I already know the answer...How you might add? My reply would be that you get answers only when you ask the correct questions. Pause for a moment and think about it.

RRMS therapeutics is a smorgasbord of the new and the old, but the question on everyone's lips is who is the leader of the pack? There is then the annoying concept of  how to sequence these treatments in order to avoid short and long-term complications. Whilst the pharmaceutical industry brings out the latest kid on the block, neurologists and regulatory authorities watch agog at the runaway horse, whilst at the same time trying to bring back some semblance of order. The regular neurologist is content to follow the pack (nothing wrong with this), but its the 'experts' who will be deciding on which questions we should all be asking. It is a race to the finish line. And, the guidance you end up following is very much dependent upon which corner of the world you currently inhabit.

Here is an interesting attempt by Xu and colleagues to make our life a bit easier by telling us which treatments are best for RRMS in terms of efficacy and safety (note daclizumab has been withdrawn owing to safety issues). The monoclonals of course seem to be the obvious choice (natalizumab, alemtuzumab, ocrelizumab, dacalizumab), all know to be superior to interferon-beta 1a - but which one is the best? The issue remains that there are no head-to-head studies, but with some statistical jiggery anything is possible (for those who're interested, its a network meta-analysis using a consistency model by frequentist approaches with interferon as the comparator). Luckily interferon hasn't changed much!

They found that natalizumab, natalizumab plus INFβ-1a, alemtuzumab, daclizumab, and ocrelizumab were associated with a significantly lowered risk of annualized relapse rate compared with INFβ-1a (RR 0.14 [95% CI 0.11–0.19] for natalizumab plus INFβ-1a, 0.31 [0.24–0.39] for alemtuzumab, 0.41 [0.26–0.64] for natalizumab, 0.45 [0.37–0.55] for daclizumab, and 0.45 [0.36–0.56] for ocrelizumab). Ranking according to relapse rates the best treatments were: natalizumab plus interferon > alemtuzumab > natalizumab > daclizumab (withdrawn) > ocrelizumab. All biologicals had a similar incidence of serious adverse events. If you combined relapse rate reduction to serious adverse events, natalizumab plus interferon beta-1a was associated with the best outcome. Whilst, in terms of a single drug alemtuzumab was the leader of the pack, and ocrelizumab at the bottom of the monoclonals (see figure above).

Without stepping on anybody's toes, this study confirms what we already know, which is that monoclonals offer the best option for a person with RRMS.


  1. Wondering how Cladribine compares with these treatments listed above. Where do you reckon it should be placed in the ranking from what you already know about its efficacy and safety? Thankyou.

    1. So the CLARITY study was placebo controlled; there was an extension study but this compared being in both cladribine and IFNB together versus placebo and IFNB. There was a study called CLASE which looked at those with suboptimal response to the injectables but I don’t know the outcome of this. I would hazard a guess that Cladrabine would be below natalizumab and Alemtuzumab.

    2. And would cladribine be below ocrelizumab?

    3. Meta-analysis of Cladribine tablets versus other DMTs:

  2. While I think that these studies are immensely important, it is a sad state of affairs that manuscripts like this pass the peer-review process and the second sentence of the results states that "INFβ-1a was the most common comparison treatment and showed an annualized relapse rate of 45.3%".

    Relapse rates are given in number of relapses / year (annualized) and not in percent, the number makes no sense and it makes me wonder how rigorous the remaining statistics could be if the first one is so blatantly wrong. I assume they mean reduction in annualized relapse rate (?), but this does not make this much better.

    1. Yes I skipped past that one and went directly to the raw relapse rate outcomes presented in the results. I don’t think they meant mean reduction either, as in the discussion they say “the annualised relapse rate was 23.4% in patients receiving monoclonal antibodies compared with 45.3% in those treated with IFNB-1a and 72.5% in those treated with placebos”.

  3. I am very confused!
    I have received 98 Tysabri infusions so far and have never seen, heard of or been offered natalizumab + interferon beta-1a!

    What is this black magic and is it prescribed at Barts?

    1. The natalizumab and IFNB-1a study was one of the combination studies. There were some fingers pointed when in the extension of this study was when the first PML case on natalizumab was encountered- and if you remember the FDA pulled the plug soon after. So unless someone wants to try this combo again we’ll never know!

  4. There is too much noise in this analysis. Based on this study, Natalizumab is the worst in terms of safety, but if you get interferon on top of that, then the combination the safest of all. Does it make any sense?

    1. This type of confusion is seen in multiple treatment comparisons using Bayesian and frequentist approaches. A drug may not be in the category of being least safe (depending on serious adverse events reported in a study) but if it is in the category being highly ranked in terms of efficacy or has 50% chance of being in the higher ranking then it would emerge as the more attractive option. So the RR for Nat+IFNB is 0.14 versus 0.41 for Nat alone. The authors did look at funnel plots to examine asymmetries between the studies for efficacy and safety outcomes and didn’t find one.


Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.