Friday, 15 March 2013

Natalizumab Safety Update - February 2013

#MSBlog: February 2013 update: 331 natalizumab-related cases of PML; 23% mortality.

"As of the 5th February 2013 there have been 331 confirmed cases of natalizumab associated PML. This is on denominator of over 112,000 treated MSers with over 260,000 years of natalizumab exposure. 76 out or the 331 MSers with PML, or 23% of the cases, have died as a result of PML."

"For accurate risk profiling please refer to slide 18. This is based on JCV exposure, duration of treatment and prior use of immunosuppressive therapies."

"If you are on Natalizumab please help out by completing this survey. Thanks."


  1. How do we read the risk stratification tool with respect to the time periods referenced? In other words, if someone falls into the 5.2/1000 (or 1 in 192) box, then what does that mean? If 5.2 out of 1000 will get the disease, does this take into account the number of doses the person has received? Obviously, someone who is JC positive and has 25 doses should be at a lower risk of developing PML than someone who receives 47 doses.
    But my question is this: 5.2 MSers (JCV+/No IS use/>24 months tx) out of 1000 will develop PML … when? Yearly? Over the 2 years from 25 - 48 infusions? Or do you "roll the dice" each time you have an infusion with a 1 in 192 chance of developing PML?
    Does my question make sense?

    1. 5.2/1000 means that there are 5.2 (i.e. 5-6 cases) cases for every thousand people treatedwhich means you have a 1 in 192 chance of developing PML if you are JC viruspositive and been on treatment for more than 2 years (that will be say 24 doses if you have have less than 2 years of dosing then the chance is about 1 in 1 thousand six hundred and seventy people. In regard to differences between 25and 47 doses i.e 2-4 years of treamtent this is the windows of treatment that biogen looked and based their risk stratefication at. After 2 years then your risk is about 1 in 200 of getting PML and this remains for the next two years. As more people develop PML it may be possible to reduce the interval say to one years vereses two years.

    2. Your question makes sense. The data is based on all subjects that have had more than 24 infusions. The problem is that it is confounded by JCV+ people coming off the drug, which falsely lowers the risk for a JCV+ve individual continuing on treatment. So the data is being confounded my the impact of JCV serological screening. The bottom line is that if you are in the high-risk category for developing PML you have to assume that your risk will increase with time, which is why a lot of MSers are choosing to stop natalizumab and move onto something else.

  2. Following on from Prof G's answer - does the JCV risk really go away once Tysabri is stopped ? I thought I had read that there had been PML cases in people who had stopped Tysabri.

    Also I presume there is scant if any knowledge of the specific risk of moving from Tysabri to either Fingolimod or other monoclonal antibody ?

    1. There is a period of asymptomatic PML, i.e. a period of time before it presents with overt symptoms. We know this from serial MRI studies. Therefore if you stop Natalizumab there is still a risk of developing PML. However, once your immune cells get back into the brain and spinal cord they should see the virus and clear it.

      You are correct when you say we have no data about the risk of PML on other therapies post-natalizumab. I agree it can't be zero because the virus must have a niche it lives in where it can undergo mutations. These mutation can't simply disappear when you stop natalizumab therefore the must increase your risk above baseline.


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