The doubling of therapeutic options over the past decade for the management of relapsing-remitting multiple sclerosis (RRMS) has made treatment decision-making considerably more complex. Most patients will require or request a change in medication over the first few years of treatment, so it is important to have a plan from the outset that will optimize outcomes and minimize risk. A good plan today, as General Patton is quoted as saying, is better than a perfect plan tomorrow, not least because of the innumerable unknowns about MS and its treatment.
The overall goal of therapy is to delay the onset of irreversible neurological and cognitive deficits, notably the conversion to a secondary-progressive course, and maintain patient self-defined quality of life. Most patients initially receive an injectable disease-modifying therapy (DMT); an increasing proportion are started on an oral platform therapy (teriflunomide, dimethyl fumarate). However, it may be argued that a majority of patients will not attain an adequate level of disease control with a first-choice agent, and will not achieve an optimal outcome.
The EPIC (expression/genomics, proteomics, imaging, and clinical) study was a long-term follow-up of a contemporary cohort of DMT-treated patients (n=407) with clinically isolated syndrome (CIS; 18% of sample) and RRMS (82% of sample) (University of California, San Francisco MS-EPIC Team, et al. Ann Neurol 2016;80:499-510). Mean age at onset was 33.6 years; mean disease duration at study entry was 6 years; and baseline EDSS was low (0 to 1.5) in 52% of patients. At baseline, 1.5% were receiving a higher-efficacy therapy and most (82.1%) experienced clinical and/or MRI disease activity during the first two years of treatment. As expected, the short-term risk of worsening disease was low: 6.4% of treated RRMS patients transitioned to SPMS and 4.7% reached EDSS 6 at 10 years after disease onset (i.e. in the first fours years of treatment). However, 10 years later, 55.3% had worsened on their EDSS score, and 24.2% had transitioned to SPMS despite the use of a front-line therapy. The proportion switching to a higher-efficacy therapy was not reported.
A low rate of treatment switching was seen in an analysis of survey responders in the NARCOMS registry (n=10,591) (Salter et al. Patient Prefer Adherence 2014:8:971-979). The rate of treatment switching was 6.5% (10.9% of treated patients). Overall, 73% were on an injectable DMT at the time of switch: one-third were switched to another injectable, 41% were switched to fingolimod, and 17% received an infusion agent (the survey was in 2011, before other options were available). At the time of switch, the mean duration of MS was about 14 years and 77% had moderate to severe disability, suggesting that the therapeutic window was already partially closed.
Among those patients who are candidates for switching, few will be escalated to one of the higher-efficacy therapies. For example, a 1-year observational study looked at whether the first iteration of the Canadian treatment optimization recommendations was implemented in clinical practice in patients on an injectable DMT (Grand’Maison et al. Can J Neurol Sci 2013;40:527-535). During the treatment year, 16.8% of patients had one or more relapses, and 19.6% had evidence of progression. MRI was performed in only 21.2% of subjects; in the subset for which MRIs were obtained, 51.2% had new/enlarging T2 lesions, 9.8% had new Gd-enhancing lesions, 4.9% had new/enlarging hypointense T1 lesions, and 2.9% had brain atrophy. Despite these findings, clinicians’ level of concern was generally low, and treatment was switched in only 5.4% of patients. Clinicians were in agreement with the treatment optimization recommendations if no action was required (95% concordance), but not if treatment needed to be optimized (29.4% concordance).
This tendency to maintain a treatment strategy despite clinical-radiological evidence of worsening disease has been termed therapeutic inertia (Saposnik et al. ECTRIMS 2016; abstract P749; see also Saposnik et al. AAN 2017; abstract P3.406). In this case-based survey, therapeutic inertia was recorded in two-thirds of neurologists treating MS. The authors concluded that a reluctance to escalate therapy may be due to clinicians’ aversion to ambiguity, i.e. a preference for known risks over unknown risks. The best-known risks are treatment-related adverse events, which have been well-characterized. Perhaps the least well known is the risk of worsening physical and cognitive disability in an individual patient; this would appear to be an acceptable risk, at least in the short term. Over the longer term, the risk of a poor outcome becomes more known as disability accumulates, and it is only then that treatment-associated risks become more acceptable both to clinician and patient. Unfortunately, this approach is a model for secondary prevention – which forces DMTs into a role for which they were not developed and for which they are not well-suited.
Another factor that weighs heavily in the decision to stay the course is patient preference, which may be at odds with the optimal treatment approach. Patient preferences largely reflect wellness, lifestyle and other factors that are unrelated to disease activity or, indeed, the need to treat. A truly patient-preference approach in many cases would eschew medications in favour of dietary modification and a healthier lifestyle. Surveys suggest that patients who are earlier in their treatment course or with milder symptoms will favour side-effect avoidance (or treatment avoidance), whereas those with worse disability are more likely to opt for more effective treatments (Mansfield et al. AAN 2016; abstract P3.108). As a consequence, less effective treatments may be started when more effective therapies are needed; and higher-efficacy agents will be employed too late to make a difference.
One therapeutic strategy that has become fashionable is “treat to target”, a term borrowed from rheumatology. The three requirements for treat-to-target are: to have a target; have a treatment that can hit the target; and have a way of determining if the target has been hit. Arguably, none of these criteria applies to MS (see Part 1 of this series), leaving the usefulness of this concept in question.
A more practical approach to treatment planning is to initiate therapy with the most efficacious agent, with the view to escalating as soon as circumstances dictate or insurers allow. Risk factors for early worsening are generally known: relapse frequency/severity; lesion location; early disability; and so on (see Part 2 of this series). What is unclear is if these factors are truly prognostic, or whether there is “presentation bias”. What may be the key factor is symptom deniability: mild paresthesia is easier to dismiss than Charcot’s Triad of nystagmus, intention tremor and dysarthria, symptoms that are alarming enough to seek medical attention. Thus, it may be argued that patients diagnosed with MS have, by definition, symptoms severe enough to overcome their denial, and are therefore at risk of worsening disease.
A treatment plan is ideally developed for the long term. and weighs the short- and long-term risks of treatment (known and unknown) against the four Ms of MS that are unambiguously at risk: mobility, mind, mood and marriage. The goal is to prevent further neurodegenerative changes and brain tissue loss, which are irreversible. As such, a plan that opts for an earlier switch as soon as it is feasible is generally preferred. Once the unknown risk of progression has become the known disability, no treatment – however efficacious – is likely to regain what has been lost.
In part 5, we’ll discuss sequencing strategies and treatment options.
Dr. Daniel Selchen: A rational treatment plan for an individual with MS involves trying to balance the risks and side effects of treatment with the risks of that person’s disease. In too many instances, both clinicians and patients devote considerably more time and emotional energy to the risks of treatment than to the well-known risks of disease described in this series. The “therapeutic inertia” documented by Saposnik (2016), and implied by Grand’Maison (2013), describes failure on the part of neurologists to actually identify patients who are not doing well and, even worse, recognizing it and ignoring it. In addition to modifying our own behaviour, it is incumbent on us to educate our patients – not only about treatment risks and side effects of drugs, but also about the natural history of a potentially devastating condition.
Regulators and payers have compounded this problem by their rigid and non-evidence- based application of rules about ‘first-’ and ‘second-line’ therapies – absurd in the context of treatments where only one of soon-to-be 14 (hopefully) medications has ever even been studied in a second-line setting.
The difficulties of “treating to target” in MS have been described above. Earlier in this series, Dr. Giacomini reminded us about the unknowns involved both with sequencing of therapies and with the newer, more aggressive treatments. What we know, however, is that many patients (Gavin Giovannoni suggests 80%) will fail on a ‘first-line’ therapy in a relatively short period of time. There are many uncertainties, as described in this series by Drs. Freedman, Yeung and Giacomini. But the goal of therapy should be to treat patients with bad disease aggressively from the beginning, and to closely monitor patients with initially less aggressive disease, with the plan to early intervention if they fail. The evidence suggests that we are not doing a great job in either of these tasks. As indicated in the above article, waiting too long defeats the purpose of treatment, which is the preservation of neural tissue and good function in multiple domains.
As detailed in this series, there is a lot that we need to find out about sequencing of treatments, and that is going to take years. While we learn, we have to make the best judgments we can about maximizing both safety and efficacy for our patients. The field has seen tremendous advances in the last 10 years. We now have highly effective therapies. But we need to overcome “therapeutic inertia” to ensure that these agents are offered to patients who need them in a timely fashion. In the absence of strong evidence on the optimal approach, we will need to generate expert opinion to provide guidance to clinicians managing MS patients on the most effective and safe use and sequencing of these treatments.
NeuroSens survey on sequencing – Part 4