MS returns after you stop taking natalizumab (Tysabri)

Rebound post natalizumab; defining and interpreting the problem. #MSBlog #MSResearch

“Does rebound occur after stopping natalizumab? Yes and no; it all depends on how you define rebound and how you measure it.”

“Firstly what does rebound mean. Rebound is disease activity greater than what was present at baseline. Most people define baseline activity as what has happened to you in the last 1 or 2 years; if you have had 3 relapses in the last 2 years then your annualized relapse rate (ARR) is 1.5 or if you have had 3 relapses in the last 12 months then your AAR is 3.  The AAR at an individual level is very difficult to interpret; therefore, I prefer to look at disease activity in terms of MRI activity, i.e. how many new or enlarging lesions are there on your MRI in the last 2 years or how many Gd-enhancing lesions are there at present. Gd-enhancing lesions are newly active lesions; the average enhancement time for an active lesion is 3-4 weeks. If you define rebound using ARR it is difficult to find evidence for rebound simply because relapses occur infrequently and you need large numbers of MSers to get an answer; this study below is under-powered for this. If you use MRI activity to define rebound the answer is very clear; once natalizumab washes  out of the system you tend to see a large number of enhancing lesions appear 3-6 months after the last dose of natalizumab. In some of our MSers we have seen over 50 enhancing lesions.”

“What is rebound telling us about MS? The main message is that natalizumab is not a cure for MS and whatever is causing MS is still there and once you allow immune surveillance to re-start the immune system triggers a massive surge in MS disease activity. In some MSers this rebound has proved fatal and in other it has caused severe relapses. How do we prevent the rebound from occurring? I think all DMTs have the ability to take the edge off the rebound and the more effective DMTs may be able to prevent it altogether.”

“As a result of rebound we have stopped doing long wash-out when we stop natalizumab. All we do  now is an MRI and a lumbar puncture to exclude asymptomatic PML, i.e. PML that has started before you develop symptoms. If these tests are clear we then start the next drug ASAP, which at present tends to be fingolimod (it is the only other licensed highly-active treatment we have). We usually start it about 3-4 weeks after the last natalizumab infusion. As it takes up to 2 months for fingolimod to start working properly, by the time natalizumab is washed-out of the system fingolimod should be on top of the MS activity. Despite this there are still a number of MSers who have been reported, by others, to breakthrough with relapses and MRI activity.  This tells me that fingolimod is not as effective as natalizumab at reducing CNS lymphocyte trafficking, which is not necessarily a bad thing as it implies the risk of PML with fingolimod is low. The safety data  that is emerging with fingolimod supports this statement.  I would not treat rebound from natalizumab on fingolimod as a fingolimod failure. The mechanisms of action of fingolimod are complex and may take several months to start working. I would therefore wait until 6 months before rebaselining someone and I would compare future MRI scans and clinical activity to this new baseline time point. This issue of rebaseling is relevant to most DMTs; the all take weeks to months to start working properly therefore any relapses or MRI activity early on need to be excluded when assessing a treatment response.”

“Does all this make sense to you?”

Epub: Melis et al. Post-natalizumab clinical and radiological findings in a cohort of multiple sclerosis patients: 12-month follow-up. Neurol Sci. 2013 Aug.

Background: There is an urgent need to identify the best strategies to prevent the loss of natalizumab (Nz) beneficial effects after its suspension. 

Objective: The objective is to evaluate the clinical and radiological disease activity and to test the efficacy of immunomodulatory/immunosuppressive drugs (IT) after Nz suspension. 

Methods: Clinical and radiological data from 54 MSers 2 years before treatment (pre-Nz), during treatment (on-Nz) and after interruption, during 1-year follow-up (post-Nz) were retrospectively collected. Annualized relapse rate (ARR), expanded disability status scale (EDSS), presence of new T2 lesions and Gd+ (gadolinium enhancing) T1 lesions were evaluated. 

Results: Pre-Nz ARR at 1 year was 1.74 while post-Nz ARR was 0.94 (p = 0.0053). We observed that post-Nz disease activity never raised over pre-Nz levels, neither post-Nz ARR nor post-Nz EDSS. In MSers retreated with Nz after suspension, post-Nz ARR was significantly lower than pre-Nz ARR (p = 0.017), but not in MSers treated with other IT or in MSers not treated with any disease modifying drugs (DMD). The mean time of freedom from new T2 lesions and new Gd+ lesions was lower in post-Nz period compared to on-Nz (T2 lesions p = 0.0000, Gd+ lesions p = 0.0000). 

Conclusion: In conclusion, a “rebound” pattern was not identified in this cohort, though the disease activity rapidly returned after Nz, regardless of the treatment used.

CoI: multiple

Other posts on rebound activity:

01 Dec 2012
“The potential rebound in MS disease activity that is commonly seen after withdrawal of natalizumab, and now fingolimod, reminds me of the book ‘Waiting for the Barbarians’ by JM Coetzee, which I read as a teenager.
27 Jul 2012
Epub: Hakiki et al. Withdrawal of fingolimod treatment for relapsing-remitting multiple sclerosis: report of six cases. Mult Scler. 2012 Jul. The objective of this study is to report the limited experience on fingolimod suspension in 
19 Feb 2012
RESULTS: Three months after discontinuation of treatment with fingolimod, the patient experienced a severe relapse, with Expanded Disability Status Scale score progression from 2.5 to 4.5. On brain and spinal magnetic 
08 Aug 2013
This homeostatic, vital or secondary action of the organism is scientifically explained by the rebound effect of drugs, resulting in worsening of symptoms after enantiopathic treatment withdrawal. Natalizumab reduces relapses 
31 Dec 2012
“The potential rebound in MS disease activity that is commonly seen after withdrawal of natalizumab, and now fingolimod, reminds me of the book ‘Waiting for the Barbarians’ by JM Coetzee, which I read as a teenager.
03 Jul 2012
No evidence of rebound was observed in MRI scans. Furthermore, EDSS and MSFC were stable in MSers, again suggesting that 12 months of post-natalizumab-GA therapy is not associated with clinical deterioration.
30 Oct 2012
Some of us have started prescribing pulsed monthly steroids to cover this period, but this does not appear to prevent the rebound entirely. There are several trials testing different washout periods and switching options that will 
31 Oct 2012
Natalizumab dramatically reduces relapses in MSers with active MS, but it may induce progressive multifocal leukoencephalopathy (PML).(1) A rebound of MS or an immune reconstitution inflammatory syndrome (IRIS) were 
13 Mar 2013
“I think we should be focusing on the rebound activity that occurs post-natalizumab therapy. Natalizumab is a very effective therapy that switches off focal inflammation in MSers; i.e. it stops relapses and MRI activity. However 
03 Jan 2012
Both the moment of rebound disease activity and the outcome of PML are related to clearance of the drug. Specific features of this IgG4 antibody (i.e. half-antibody exchange) may result in underestimated drug levels. Here, we 
15 Jul 2012
They will publish it this year and this will be another element in the risk stratification of PML, likely. is this true?what is the best option after Tysabri to prevent rebound of disease activity? in our hospital they use 6 months on 
21 Jul 2013
NHS England is fully aware of the MS rebound that occurs when natalizumab is withdrawn and that there has been deaths as a result of it. No wonder MSers in England who are JCV+ve and at high-risk of PML, but cannot 

About the author

Prof G

Professor of Neurology, Barts & The London. MS & Preventive Neurology thinker, blogger, runner, vegetable gardener, husband, father, cook and wine & food lover.


  • If disease activity returns then that means natalizumab is not 'curing' MS, merely supressing it. This makes me wonder if Campath 1-H is also just masking the effects of MS and not really addressing factors driving progression.

    I suppose if we are still unsure of why MS happens then the disease cannot be properly treated.

    • The difference between natiluzumab and Campath 1 is that Campath destroys the cells that are causing disease whereas natiluzumab merely stops them from getting into the CNS. All the evidence so far is that treating with Campath very early does seem (so far) to stop disease progression.

    • How early is your definition of early? Will it work for every RRMSer in early disease stage or will some still progress?

    • On diagnosis. Will it work for every RRMSer, who knows but early indications seem to be promising. At the very least any progression should be greatly diminished.

  • Instead of fingolimod why not use a corticosteroid taper when discontinuing natalizumab treatment as a standard treatment? IRIS in HIV patients with rebounding immunity has shown to be lessened with steroids.

    • This is not IRIS that occurs in PML. This is MS rebound. Several groups have tried steroids and they have not been that effective in preventing rebound. The problem is there is little, or no, evidence that steroids modifies the course of MS. This is why most evidence-led neurologists don't use steroids outside the context of a clinical relapse.

    • Thanks Dr. G. To follow you said that MS rebound was defined as disease activity that is greater than that at baseline. New Gd enhancing lesions appear 3-6 mos post therapy cessation. Is it incorrect to view stopping natalizumab and restoring immune surveillance as comparable to increasing CD4+ using HAART for HIV patients? Immune reconstitution inflammation in HIV can be due to any of a number of infectious agents in the CNS. In MS the driver of new inflammation is unknown. If EBV is the driver could anti-virals be used as prophylaxis?

    • I think you are confusing IRIS with rebound. Rebound is the disease activity returning where white blood cells say against myelin are entering the brain whereas IRIS is due to the presence of virus in the brain and when nataluzimab wears off white cells against the JCV virus enter the brain to kill infected cells so with HAART you increase T cells and they get rid of virus. Baseline in this context is not when you start treatment but when you stop treatment. If the driver of the response was viral agent then indeed anti-virals may work. At present we do not know this so best to stick to immunomodulator

      which hade o off then the immune response is allowed to nte the

    • Yeah thanks…….the mechanism of IRIS is due to a known infectious agent that can be detected in CSF in the lab or indicated by abnormal CSF markers. We only know half the story in MS.

By Prof G



Recent Posts

Recent Comments