Sunday, 28 August 2016

Does the early bird get the worm in CIS? The 11 year follow-up to the BENEFIT study

The paper: Kappos, et al. "The 11-year long-term follow-up study from the randomized BENEFIT CIS trial." Neurology (2016): 10-1212.

Background: Multiple Sclerosis (MS) often starts as an individual episode of neurological dysfunction. At this early stage, termed a clinically isolated syndrome (CIS), it is often difficult to predict who will go on to develop definite MS, and who will have no further episodes. Previous trials have shown that early treatment with disease-modifying therapies like interferon β-1b can delay the conversion from CIS to clinically-definite MS. However, it is not known whether early treatment reduces disability and disease progression in the long-term.

Aim: To assess the effect of early vs. delayed treatment on long-term outcomes at 11 years after onset of CIS.

Method: The initial BENEFIT trial was a randomised, double-blind, placebo-controlled trial in which participants received either early (n=292) or delayed (n=176) treatment with interferon β-1b after onset of CIS. The early treatment group began treatment within 60 days of onset. The delayed treatment group received placebo until either 2 years after onset of CIS or onset of clinically-definite MS.

After the initial 2 year trial, patients were invited to participate in a 3-year open-label extension. In the 3 year extension period, patients could choose to have any disease-modifying drug/s, or to have no drugs at all. Patients and investigators were not blinded to treatments in this stage.

The 11 year follow-up study reassessed patients who were part of the original trial 11 years after onset of CIS. 278 of the original 468 patients were assessed. This was a cross-sectional, observational study that compared several outcome measures in the early vs delayed treatment groups. The outcomes measured were: risk of conversion to clinically-definite MS, time to conversion, relapse rate, conversion to secondary progressive MS, disability scores, neuropsychometric scores, MRI appearance, and health resource utilisation.

Results: Patients in the early treatment group had a lower risk of conversion to clinically definite MS (66.6% vs 75.0%), a longer lag time until the conversion to clinically definite MS, a lower average annualised relapse rate over the whole study period (0.21 vs 0.26), and appeared to have a lower risk of conversion to secondary progressive MS (4.5% vs 8.3%). Average EDSS scores were similar in the early and delayed treatment groups.

The early treatment group performed better on the Paced Auditory Serial Addition Task-3 (PASAT-3), in which patients have to add up a series of numbers presented to them via audio at 3-second intervals. There was little difference between early and delayed treatment groups’ MRI appearances or health resource utilisation.

Conclusion: Early treatment (within 6 months) of CIS with interferon β-1b leads to clinical benefit over delayed treatment, and this benefit is sustained 11 years after the onset of CIS.

Interpretation: This impressive study provides a fascinating insight into the long-term consequences of different treatment strategies in early MS/CIS. What these results appear to show is that an aggressive, ‘see and treat’, approach in CIS is better than a ‘watch and wait’ strategy. We knew already from previous incarnations of BENEFIT that early interferon β-1b treatment delays the onset of and reduces the risk of conversion to clinically-definite MS. This 11 year follow-up allows us to work out whether these benefits actually translate into better long-term quality of life for patients.

It is therefore interesting that EDSS scores – a commonly-used measure of disability in MS – were almost identical in the early and delayed treatment groups. This implies that while early treatment may stave off the onset of definite MS, it does not improve long-term disability. There are several possible explanations for this finding. Firstly, the EDSS scores are very low and stable in the study population, and so the study might not be sufficiently powered to detect small differences between the groups. Even still, a small difference in EDSS scores, which might be missed if the study were underpowered, would not necessarily be clinically significant. A biological explanation for the lack of obvious benefit in terms of EDSS score is that the processes involved in disability progression in MS, such as neuronal and synaptic loss, might be quite separate to the processes responsible for relapses.

Another interesting observation is that whilst the overall annualised relapse rate (ARR) was lower in the early treatment group, the ARR was actually slightly lower in the delayed treatment group in the final year of follow-up. This might imply that early treatment produces a long-term but not indefinite reduction in relapse rate, which levels off by year 11. Intuitively, this makes sense, as the participants spent 9 years on whatever treatment they and their doctors opted for, and only 2 years in the stringent, double-blind study period.

Given that participants had received 9 years of treatment since the original study period, it is remarkable that there were sustained benefits of early treatment in terms of overall ARR, risk of conversion to clinically-definite MS, time to conversion, and risk of secondary progressive MS. This could be interpreted as evidence that there is a crucial window of opportunity when patients have their first episode suggestive of MS, and that initiating treatment at this stage produces long-lasting benefits.

On balance, there seems to be little harm in starting CIS treatment as early as possible. However, BENEFIT-11 provides no evidence that this approach leads to long-term differences in disability, the outcome which probably matters most for quality of life. It does provide some evidence that early treatment influences the course of the disease, but this is very difficult to disentangle from effects of other treatment which patients may have tried in the intervening follow-up period.


  1. Paper from Dec 2015 suggests Vit D3 daily makes interferon B-1b work even better (Lower rates of MS activity observed on MRI). Results for brain atrophy and clinical progression were more equivocal.

  2. Agreed Dr. Ben,

    I read this some days back and was surprised by it. At CMSC 16' I believe it was Avonex. As I recall stated the efficacy of interferon in long term study. In the news report I'd read I thought there must be errors.

    In some ways might make complete sense. Many diseases when slapped hard right at onset seem with many folks to really cause the disease and activity to be considerably less long term.

    The question would be why towards MS given interferons thought mechanism of action.

  3. It's imperative that neuros give patients the option of treatment at he CIS stage.
    How many lesions does one need to have to be diagnosed with CIS? What is the definition of CIS in terms of lesions? Is there one? When I was diagnosed with CIS I'm pretty sure I had had a couple of episodes, not just one. However the definition is one episode is CIS. I'm confused.

    1. This is an important question - CIS is now defined as a first episode of neurological symptoms which gets at least partially better. The new definition excludes people who have evidence on MRI of old lesions. This has changed since the BENEFIT trial started - they included people who we would now diagnose as having MS based on symptoms + MRI changes suggestive of old lesions.

    2. Considering how many people more than likely have a CIS incident that goes undiagnosed the criteria while making sense may not fit in the real world.

      Its a shame there is no big data gathering towards these aspects of the disease onset.

      It makes complete sense towards research so patients can be filtered towards perhaps early onset trials to both learn and try hit the disease hard at onset.

  4. This is a quite confusing result. Perhaps its far to early for the EDSS scores to be a useful metric, perhaps we need another 10+ years to see how the disability progresses between the two groups. I would assume that perhaps after 20 years it much more distinct between disease progression. But if not and all the groups end up at the same EDSS point after the same period of time, then what is the benefit of early intervention apart from to reduce relapses?

    1. Very true - i'd add that it's especially hard to detect an EDSS benefit in this study population, as the participants all had a very mild disease course compared to participants in other big natural history studies. Also worth noting that EDSS is a good metric of disability, but it's still just a scoring system and so it can't fully encapsulate everything that leads to a good quality of life! This is why I like that this study and others use outcomes like employment alongside EDSS - it gives a much better all-round picture


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