Monday, November 10, 2014

Questions to ask?

The following is a list of questions you may want to ask your neurologist when a decision is being made about starting a disease-modifying therapy (DMT) for treating multiple sclerosis (MS).


  1. What is multiple sclerosis?
  2. Are you sure that you have MS?
  3. What type of MS do you have? 
  4. What prognostic group do you fall into? 
  5. What is the risk of you not being treated with a DMT? 
  6. Do you have active MS? 
  7. Am I eligible for treatment with a DMT? 
  8. Do you understand the difference between the treatment strategies of maintenance-and-escalation and induction therapy?
  9. Do you understand the concept of treat-2-target of no evident disease activity (NEDA)?

Do you have active MS?

In my mind MS disease activity refers to evidence of focal inflammatory activity, i.e. new lesions on MRI or relapses. In contrast the gradual progression, or worsening, of disability that occurs in people with SPMS and PPMS, may or may not occur in the presence of inflammatory activity. Most neurologists require evidence of disease activity in the last 12 months, with some of us accepting a 24 month window if there is no serial MRI support. The corollary of this is if you have had no relapses in the last 24 months and serial MRI studies, covering this period, show no new lesions then your MS is defined as being inactive. This does not mean your MS is necessarily stable, I have already said you could have worsening disability as part of the progressive phase of the disease. If you have so called inactive MS you need to be monitored as inactive MS may reactivate and you could be eligible for treatment.

In the early 2000’s we used to define disease activity using clinical criteria only; i.e. you needed to have at least 2 documented relapses in the last 2 years to be eligible for DMT.  This meant you needed to be seen by a neurologist so that he/she could examine you to confirm abnormalities on examination compatible with a relapse. The problem with this is that many MSers who didn’t have rapid access to a neurologist would recover from their attack before being assessed so that many of their relapses could not be documented. This was very frustrating for MSers wanting to start a DMT as many neurologists were not prepared to accept historical attacks. I remember seeing many MSers for a second opinion over this issue; a kind of arbitrator to decide who was right the assessing neurologist who would not accept undocumented historical relapses over the patient who knew they had had a relapse. I tended to give patients the benefit of the doubt, particularly if they had MRI evidence to support recent disease activity. How could we deny MSers access to a DMT because they couldn’t be seen in timely way to have their relapse documented in the clinical notes?

In 2009 our criteria incorporated MRI into the definition to allow us to treat so called high-risk patients with CIS (clinically-isolated syndromes compatible with demyelination). These criteria required CISers to have 9, or more, T2 lesions on MRI or at least one Gd-enhancing lesions. These MRI criteria were based on the McDonald diagnostic criteria at the time. I personally have never agreed with them. What makes someone with 2 lesions on MRI different from someone with 20 lesions on MRI? It could simply be that the person with 20 lesions has had asymptomatic MS longer than the person with 2 lesions on MRI. If we adopt the principle of treating MS early to prevent damage, why would we want to wait for the CISers to acquire more lesions and hence damage? I take the position that a CISer presenting with a low lesion load is lucky that they presented early with their disease and hence have a greater opportunity to benefit from early treatment. The CISer with a high lesion load unfortunately is the unlucky one. This dilemma is really only a UK problem; outside of the UK almost all neurologists offer CISers DMTs if they have two or more lesions on MRI.

Now in 2014 NICE (The National Institute for Health and Care Excellence) have given us permission to use Alemtuzumab in adults with active relapsing MS defined clinically or on MRI (see below). This could mean defining activity based on MRI only. I doubt NICE, or the EMA, realise the implications of this ruling. I suspect the OR refers to re-treatment criteria and not the treatment criteria to initiate treatment. It would hard to justify, in the current climate, starting Alemtuzumab in MSers without any recent clinical activity, i.e. relapses. In fact, some of my colleagues are refusing to offer alemtuzumab at all; they have taken the position that the therapy is too risky to justify it use. I have debated their position in the past and think it is unacceptable.

Rapidly-evolving severe MS

For those of you who have active MS there is a further two subdivisions you need to know about. The first is so called rapidly-evolving severe MS (RES); defined as two disabling attacks in a 12 month period with MRI evidence of activity during this period. This subgroup is eligible for natalizumab and fingolimod.

Highly-active MS

The other subgroup is so called highly active MS; these are MSers with unchanged or increased relapse rate or ongoing severe relapses compared with the previous year despite treatment with beta interferon, or another so called first-line DMT. The latter definition is part of the NICE final appraisal determination for fingolimod. Please note that this exposes a quirk in the NHS system in that fingolimod can only be prescribed as a second-line therapy. In contrast natalizumab can be used as a first-line treatment in patients who fulfil the definition for having rapidly-evolving severe MS and alemtuzumab for MSers who have active MS defined clinically or on MRI.
What these definitions don’t address is what we do with SPMSers and PPMSer with worsening disability and no recent relapses (last 2 years) who have new focal and/or Gd-enhancing lesions on their MRI. At present we don’t really have prescribing guidelines that cover these patients. Why not? It is simply because there has been no positive phase 3 studies in this group of MSers to guide our treatments. If you adopt the scientific principle that inflammation is bad for MS you would want to a treatment to tackle this inflammation.  This is why we have used rituximab in the past in a handful or young PPMSers with Gd-enhancing lesions; this is based on the positive subgroup analysis of the phase 2 rituximab in PPMS trial. However, in the UK NHS England have stopped individual funding requests (IFRs) applications to cover this indication; rituximab is simply too expensive for the NHS to be used off license. For this reason we are now offer these patient the option of being treated with subcutaneous cladribine. In the secondary progressive population we still have mitoxantrone on the table.

The following is a diagram that attempts to explain the classification of active MS based on subsets.


Are you eligible for a DMT?


Eligibility criteria for DMTs vary across the world and are dependent on licensing rules, national guidelines, payers reimbursement policy and your local MS Team’s prescribing rules. In some parts of the world, for example the US, MS DMTs are simply given a license for treating relapsing MS and it is then left up to the neurologist and the patient to negotiate a treatment. Despite this liberal licensing rule the payers, or insurance companies, set prescribing policy that is usually based on price. The following are NICE's technology appraisals that dictate how we use DMTs in the UK. These appraisals are based on cost-effective models that are designed to ensure the NHS purchases and used DMTs in a cost-effective way.

Thursday, April 17, 2014

What is the risk of you not being treated with a DMT?

Dirty fact: there are very few prognostic factors that have been validated that can predict a benign MS course

"The main argument against taking a population, rather than an individual, approach to treating all MSers with highly-active treatment early on is that we would over treating MSers destined to have benign MS. In other words exposing a large number of people, who are destined to have benign MS, to ‘unacceptable risks’. The unacceptable risks is the big issue that drives this debate and is why regulatory authorities, particularly in Europe and recently in the US, have put their foot down and have rejected drugs or severely curtailed their use by licensing them as second or third-line agents.

These unacceptable risks of highly-effective treatments need to be weighed against the risks of not treating MS. The following infographic, which I developed for ECTRIMS last year, tries to capture the consequences of untreated MS. What it does not capture very well is the ‘time is brain’ concept. In other words once damage has occurred to the brain and spinal cord we can’t go backwards and repair that damage; at least not yet. What can happen early on in the course of MS is that the undamaged parts of the nervous system can compensate for the damaged areas and allow functional recovery. This ability for compensatory recovery is time and age dependent; hence the term ‘window of opportunity’.




I try not to use the term ‘benign MS’. Benign MS is a retrospective diagnosis that can only be made many years into the disease and even then it is a hard call to make. For example, our current definition of benign MS is someone who has the disease for 15 years and has no disability, i.e. an EDSS of 3.0 or less. An EDSS of 3.0 implies the presence of neurological impairments (subtle deficits detected on neurological examination) and not overt disability. In other words someone with an EDSS of 3.0 would look normal to the untrained eye; the person in question would have to be examined by a neurologist to detect the neurological dysfunction or interrogated using detailed questionnaires to detect subtle problems. What you have to remember is that you have to pass through EDSS 3.0 to get to EDSS 4.0 and beyond. In natural history studies the majority of MSers who get to EDSS 4.0 become secondary progressive.


When you interrogate MSers with benign disease as defined above they have hidden symptoms. Approximately half of MSers with so called benign MS have cognitive impairment, depression or fatigue. Is this really benign MS? The problem with the field has is that we view MS through EDSS spectacles and we don’t take into account the hidden symptoms that can be very disabling.


In populations of MSers followed in hospital clinics about 30% fulfil the 15 year definition of having benign MS. However, if you follow this population for a further 10 years the figure drops to 15%, and at 30 years only 1 in 20 or 5% of MSers have benign disease. A recent long term follow-up of benign MSers in Gothenburg, Sweden, showed that 50% of MSers with benign MS at 40 years did not have benign MS at 50 years after the onset of the disease. The bottom line is benign MS is a moving target and depends on how you define it.


Hospital studies underestimate the true prevalence of benign MS, because MSers without problems tend to drop-out of long-term follow-up. In community studies the prevalence of benign MS is in the order of 45% at 15 years. This more optimistic community figure of having benign MS does not help you if you have regular follow-up at a hospital. The latter implies that you have more active MS or MS-related problems requiring regular follow-up by a neurological team.


A treatment strategy that I support is to convert every MSers clinical course to that of someone with benign MS, which is the aim of adopting the early effective treatment paradigm of treating-2-target of NEDA (no evidence of disease activity).


Can you identify MSers who are more likely to end-up with benign MS? The short answer is no. We do try and pigeonhole MSers into prognostic groups using the following favourable prognostic factors, but more often than not we get it wrong.

  1. Younger age at onset
  2. Female sex
  3. Initial presentation being isolated sensory symptoms or optic neuritis
  4. Complete recovery from first attack
  5. Long interval between the first and second relapse (>2 years)
  6. No disability after 5 years
  7. Normal MRI or low lesion load at baseline
  8. Negative CSF (no oligoclonal IgG bands)
What this list does not address is good or bad luck. What causes a focal MS lesion to present with symptoms is whether or not it is located in an eloquent site, for example the optic nerve (visual loss), brain stem (double-vision or unsteadiness of gait) or the spinal cord (sensory loss or limb weakness). Lesions that occur in ineloquent sites don’t cause overt symptoms because they typically affect cognition and ability of the nervous system to compensate for these lesions means you simply cope with the lesions. However, we know that all lesions cause damage to a greater or lesser extent - this is based on pathological and imaging studies in MSers. If you accumulate enough of these so called asymptomatic lesions you reduce your brain’s reserve capacity and are more likely to run into trouble in the future. Therefore someone who’s 20th lesion causes their first clinical attack will have acquired much more damage than someone who’s 2nd lesion caused their first clinical attack; the former has simply had the disease longer. In my opinion the person whose second lesion causes a clinical attack is lucky in that they presented before the next 18 lesions could cause irreversible damage. Hence the MSers with 2 lesions is more likely to do well that the person with 20 lesions. What I am trying to say is that having a low lesion load should be viewed as good news as it allows you a greater window of opportunity to prevent damage. In contrast the MSer who presents with a high lesion load has a narrower window as they have lost reserve capacity.

Similarly, being young means you will have the disease slightly longer before you become disabled compared to an older person. However, the younger person will still reach disability milestones at a younger age than MSers presenting at an older age. Because of this age of onset is not really a prognostic factor.


I prefer to focus on defining whether or not your disease is active. This can be done clinically, i.e. have you had a recent relapse in the last 12 or 24 months, or has your lesion load on MRI increased over the same period. An easier metric is the presence of gadolinium-enhancing lesions on a single scan; this is the MRI equivalent of a relapse. Enhancement tells us that the lesion is actively inflamed and that the inflammation is relatively new, usually with the last 2-4 weeks. Having active disease indicates that you are more likely to respond to DMTs. MSers with inactive disease are less likely to respond, but need to monitored closely so that if their disease flares-up they can be treated.


I don’t differentiate between types of lesions; i.e. whether or not the lesion causes severe or mild attacks, or no attacks at all (asymptomatic lesions). What causes a lesions to cause symptoms is it location and size; a tiny strategic lesion is a specific pathway can cause symptoms whereas a large lesion a silent area may go unnoticed. What matters to me, and hopefully to you, is whether or not you have MS lesions coming and going, either new or enlarging old lesions. The aim of our treatment is to prevent this from happening to protect your brain and spinal cord from further damage so that we can delay or prevent the onset of progressive MS.


The other treatment aim that is rarely discussed is ‘brain protection’; to prevent brain damage so that you have enough reserve capacity to allow you to age normally. The concept of reserve brain capacity has emerged from the Alzheimer’s and dementia fields; increased brain reserve protects you from the ravages of ageing. Why shouldn’t MSers expect to age normally?


With regard to NEDA; at the moment we define it using clinical and standard MRI metrics. These include relapses, disease progression, new and enlarging T2 (bright lesions) and new enhancing lesions. I suspect the current NEDA metric is only recording the tip of the MS disease activity iceberg and we will need to adapt it as new technologies become available. For example, gray matter lesions that are not visible on standard MRI scans, brain atrophy and spinal fluid neurofilament levels. Some of us also support including a patient-related outcome measure or PROM and cognitive testing in the definition. All of these are not ready for prime time as they need to be validated and tested in prospective studies. What is important is that we are clearly moving to individualised treatment with close monitoring to define outcomes and treatment targets. This is something rheumatologists did decades ago in rheumatoid arthritis. Treat-2-target is something we have stolen from them. Hopefully, we will be as successful as rheumatologists when it comes to treating MS. The big difference between them and us is that if their treatments fail they can always ask their orthopaedic colleagues to replace the joint in question. Unfortunately, as neurologists we don’t have this luxury; we simply can’t replace your brain and spinal cord nor can we fix them at present. That is why having the option of offering early highly-effective treatment with the hope of preventing damage is so important. Prevention is really our only strategy at present. You can ignore this window of opportunity at your own peril. Please don’t forget time is brain."


CoI: multiple

Wednesday, February 19, 2014

What prognostic group do you fall into?

A dirty fact: "You can't predict the prognosis of MS early in the course of the disease."

What prognostic group do you fall into?

An accurate prognosis in MS is a difficult call. It depends on several factors and it will never be accurate. In other words it is an actuarial science. In the same way as actuaries working in the insurance industry we need have to calculate a prognosis and give you a rough idea what to expect. An actuarial calculation is an average with a wide range of possibilities. This is the reason why I try and keep it simple and classify MSers into three prognostic categories; poor, indeterminate or good. Poor in this context simply means if you leave MS to its own devices and let it run it natural course the average person in this category will do badly. To be honest with you given sufficient time the majority of MSers will do badly, this is why I am an active treater. Why take the risk of untreated MS if you don’t have to?

The following is a list of factors that have been linked to poorer prognosis. I suggest you add up how many you have and if you have four or more factors you are less likely to have benign MS.

  1. Older age of onset (greater than 40 years) 
  2. Male sex 
  3. “Multifocal“ onset (more than one site in the nervous system involved) 
  4. Efferent system affected (motor, cerebellar, bladder) 
  5. Partial or no recovery from initial relapses 
  6. High relapse rate in the first 2 years (more than 2 relapses) 
  7. Disability after 5 years (EDSS > 3.0) 
  8. Abnormal MRI with large lesion load (more than 9 lesions on MRI) 
  9. Posterior fossa lesions (lesions in the back of the brain) 
  10. Brain atrophy (shrinkage of the brain) 
  11. CSF positive for OCBs (oligoclonal IgG bands) 
  12. Low vitamin D levels 
  13. Raised neurofilament levels in your spinal fluid (this test is not part of routine care unless you live in Sweden) 
  14. Smoker (smokers with MS do worse than non-smokers) 
  15. Co-morbidities (e.g. diabetes, hypertension and raised cholesterol) 
  16. Genomic factors (e.g. ApoE4; this is risk factor for Alzheimer's disease and some other degenerative brain diseases; not everyone in the field accepts this as a poor prognostic factor) 
Humans have an interesting psychology in that the exception is the rule. Gamblers don’t enter a casino to lose; they always believe they are going to win. When a person with lung cancer starts chemotherapy they believe they going to one of the 10% who are cured. When some with MS is diagnosed they believe they going to be in the 30% with benign disease. The current dogma is that 30% of untreated MSers will have benign disease. This definition of benign MS is based on having no, or little disability at 15 years; i.e. an EDSS of 3.0 or less (no visible disability). The problem with this is when you interrogate benign MSers you find that more than 50% of them have hidden symptoms of depression, anxiety and cognitive impairment. Can we really justify this definition of benign MS? What is more when you follow benign MSers past 15 years only 15% remain as benign at 25 years and only 5% after 30 years. If you get to 40 years follow-up with benign MS half will become disabled over the next 10 years. Time is the killer. Some of you will state that these figures are out of date and there are newer and better figures, which show MS is a more benign disease. You may be right. In population based studies the proportion of subjects with benign MS is greater than those in hospital or clinic-based studies; for example in the Ohlmstead, Mayo Clinic, population about 45% have benign disease at 15 years. The reason for this is that MSers with benign disease often drop-out of hospital follow-up and show up in population based studies and the wider use of DMTs is beginning to change its course.

I don’t think it is worth arguing over the exact figures; the message is that most MSers will not turn out have benign MS. Please note I say turn-out. We simply cannot make an accurate call on this early in the course of the disease. What we should be focusing on is how can we maximise your chances of having benign disease? Treating MSers with DMTs is one way of doing this and making sure that MSers adopt a healthy lifestyle is other strategy that can be done in parallel. The following figure illustrates what we are trying to do with DMTs. We are simply trying to move you to the left into a more favourable prognostic group. In other words we are trying to make sure you have benign MS.


MS classified by disease activity prior to DMTs

The aim of wide adoption of DMTs; to shift as many MSers into a more benign prognostic category. 

Some would argue that this approach is defeatist and that we should be trying to cure you of having MS. That is easier said than done. To find a cure for MS we have to find the cause. Part of finding the cause is showing a treatment cures you of the disease. At present most in the field believe that MS is an autoimmune disease; in other words your immune system goes awry and instead of doing what it was designed to do, i.e. fight infections and cancers, it turns on itself and attacks the myelin and nerve processes in the central nervous system. The only way to cure people of autoimmune diseases is to destroy your immune system and replace it with a new immune system that hopefully does not have any autoimmune cells. Therefore the only DMTs or treatment strategies that can potrntially cure you of MS at present are the ones that reboot your immune system. These are part of the so called induction therapies and include alemtuzumab, cladribine, mitoxantrone, bone marrow transplantation and possibly selective B cell depleting agents (anti-CD20 and anti-CD19). To prove someone is cured of MS you have wait a long time to see if the disease comes back or not. This has similarities to cancer cures; how long do you have to wait to say you are cured from a cancer? The oncologists have stopped using the term cure and simply state that there is no evidence of detectable disease or NEDD. For certain cancers if you have NEDD at 5-years after completing your treatment your likelihood of a cure is very high say 80%; in other words 20% will relapse after 5 years. In MS we don’t have this type of data yet as we have not had induction therapies for very long. What we have done is stolen the terminology from oncology and we have started to talk about no evident disease activity or NEDA. If you have NEDA after induction therapies how long do we need to wait say you are cured of MS? I am not sure, only time will tell. However, I have proposed a working definition of a cure as being NEDA at 15 years. How good are we at inducing a state of NEDA? Not good at all. One of the reasons for this is that we are not measuring NEDA very well. At present we include relapses, disease progression and MRI activity in the definition. Most studies have been including the baseline MRI as the comparator, this is a mistake. Most DMTs take weeks to months to start working hence any activity that takes place during the first few weeks or months of starting should not be counted. Therefore I am trying to promote the need to rebaseline MSers before starting the NEDA clock. The rebaselining will depend be DMT specific; the longer it takes for a DMT to start working the longer you need to wait to rebaseline. For the interferons, mitoxantrone, natalizumab, fingolimod, teriflunomide and BG-12 3-6 months would be appropriate. For GA rebaseling should be at 9 months and for alemtuzumab 18 to 24 months. Why so long? Alemtuzumab is an induction therapy and there is little to be gained by assessing whether or not there is evident disease activity until after you have had the full treatment course; i.e. two annual cycles of treatment. Therefore the current NEDA figures that have been published are not very helpful.

Time answers many questions:
  1. Do I have benign MS?
  2. What is my longterm prognosis?
  3. If post-induction therapy, I am rendered with no evident disease activity for 20 years, have I been cured of MS?
  4. If post-induction therapy, I am rendered with no evident disease activity, will I come back with SPMS? 
What we can't do is turn back time to make a different decision in the beginning.