Treatment optimization in multiple sclerosis: how do we apply emerging evidence?

KEYWORDS:

• Treatment optimization
• high efficacy treatment
• treatment switch
• no evidence of disease activity
• multiple sclerosis

In the past years, the introduction of highly effective treatments has considerably expanded treatment options in relapsing remitting multiple sclerosis (RRMS). Not only the number of available treatment options has increased, but evidence has emerged that there is a limited window of opportunity to intervene effectively in multiple sclerosis (MS) patients, which appears to be critical for achieving favorable long–term outcomes [1,2]. Considering the important role of focal inflammation driving both acute and delayed neurodegeneration, a zero tolerance or no evidence of disease activity (NEDA) strategy suppressing all inflammatory activity should be the therapeutic aim [3]. This new emerging therapeutic strategy treating to target to achieve NEDA is in line with treatment concepts of other autoimmune diseases as rheumatoid arthritis where treatment algorithms aim to suppress as much joint inflammation as possible and to induce long-term remission for prevention of permanent end-organ damage.

Effective and strategical interventions imply prompt treatment optimization, a switch of therapy in patients with suboptimal response or treatment failure with their current MS treatment. A suboptimal clinical response can be defined as occurring in patients receiving disease-modifying therapies (DMTs) for ≥1 year with one or more relapses or expanded disability status scale increase. However, it is not always easy but challenging to define suboptimal response respective treatment failures precisely: On treatment, the monitoring of MS disease activity is key to achieve optimal outcomes [4]. During treatment, it has to be defined whether the patient is a treatment responder or not [5]. For this treatment monitoring, clinical and especially MRI parameters as well as potentially other biomarkers can be applied:

• Clinical disease monitoring on MS treatments should include the three key elements: disease activity as manifested in relapses (reflecting inflammation), disability (reflecting neuroaxonal loss) and functionality (reflecting the degree of compensation or cerebral reserve) [1].

• Conventional MRI techniques can serve as helpful tools in assessing evolution of demyelinating lesions in time and space [6]. However, due to the limited pathological specificity they show only weak correlations with clinical measures of disability. Perhaps, conventional MRI is more useful for predicting treatment response rather than for assessing disease evolution itself as the limited relationship between short-term MRI measurements and long-term disability has been demonstrated [7]. In addition, MRI monitoring is essential in specific treatment scenarios for safety reasons (e.g. Natalizumab) [8].

• Nonconventional MRI techniques such as magnetization transfer appear to provide better and more quantitative measures of higher pathological specificity demonstrating better correlations with standard measures of clinical disability, but they have not yet been established in clinical practice [9]. MRI measurements of more diffuse brain atrophy have shown interesting results as imaging biomarkers of MS-related neuroaxonal damage at least at the group level. So regular MRI follow-up is recommended, beginning at 3–6 months after initiation of treatment, at 6–12 months after the reference scan and annually using standardized semiquantitative comparable MRI [6].

• Possibly even more sensitive markers of subclinical disease activity are emerging, for example, with monitoring of cerebrospinal fluid parameters which is not practically relevant due to the invasive nature. That is why serum neurofilament light chain concentrations could serve as an interesting biomarker in clinical practice [10]. First data have been presented but have not yet been validated so far.

In addition to clinical and MRI monitoring, the patient´s perspective should be implemented applying patient-reported outcomes (PRO), which can be used as outcome and monitoring parameters as well. Although monitoring strategies mainly focus on clinical and MRI measures of disease activity up to now, it should be noted that patient-reported progression of symptoms, adverse effects of treatment, and an inability to tolerate injections may also constitute a basis for treatment switches. So PROs become increasingly important in treatment decision-making as they are very sensitive to changes allowing moving the focus of treatment to the prevention or delay of disability. But therefore it will be necessary to validate such PROs in clinical trials and routine practice in the context of treatment optimization.

To collect all monitoring information according a prespecified clinical pathway, a computerized patient management system as the Multiple Sclerosis Documentation System can assist to collect all necessary parameters [11,12]. As an example for this innovative strategy, the Post-authorization Noninterventional German Safety Study of GilEnyA in relapsing-remitting multiple sclerosis patients (PANGAEA) 2.0 study has been started which is an updated version of the PANGAEA 1.0 study [13,14]. This noninterventional study will not only evaluate the long-term benefit of a treatment change to fingolimod but also the applicability of new concepts of data acquisition, assessment of MS disease activity, and evaluation of treatment response for the clinical routine as the modified Rio criteria [6,15]. In the PANGAEA 2.0 study, the disease activity status of patients receiving a DMT is evaluated in order to identify patients at risk of disease progression. This evaluation is based on outcome parameters for both clinical disease activity and MRI, and subclinical measures using the two-dimensional functional disability score, describing disease activity from the physician’s and the patient’s perspective. In a connected model of patient’s satisfaction, perceived effectiveness, side effects, convenience and adherence, patients’ individual needs and concerns have to be addressed [16]. Generally, treatment decisions in clinical practice are framed in terms of a hierarchy of efficacy and risk for treatments based on data from their pivotal clinical trials [17]. Treatment decision then represents the consideration of these data in the context of estimates of relative disease severity for any given patient. Patient- and neurologist-specific factors of preference and access to the treatments also play a role.

Which treatment strategies should be applied if suboptimal treatment response is observed? In clinical practice today, it is still a common approach to cycle between several first-line therapies with different doses of drug or treatments (so-called horizontal switching) in an active patient instead of trying more efficacious therapies. There are reports that horizontal switching may have positive effects [18]. So, it is still an ongoing discussion which strategy is best for the individual active patient [19,20]. In our opinion, the available evidence suggests that switching to a more efficient disease-modifying treatment is more effective than dose escalation or switching to another baseline therapy [21]. Our important arguments to move to higher efficacy MS treatment are the higher responder rate and the earlier onset of action in comparison with baseline therapies [1].

In addition, there are no current data suggesting that the early use of high-efficacy treatments presents a risk of therapeutic burnout. Rather, the clear evidence for a limited therapeutic window militates in favor of early intervention and treatment optimization and would suggest that regular monitoring during treatment with DMTs and prompt intervention in cases of suboptimal response or treatment failure are essential to prevent long-term outcomes [2].

An important question is which high-efficacy drug is the best for treatment optimization? This has to be decided individually for every patient. Especially potential side effects usually play a prominent role. So risk stratification of John Cunningham virus antibody index or comorbidities can provide important guidance [8]. Potential side effects are mostly related to the individual immunological mechanism of action of each disease-modifying treatment which focus on cell migration (natalizumab, fingolimod), cell depletion (alemtuzumab, ocrelizumab), or immune cell regulation (daclizumab) [22,23]. Specific adverse events have to be evaluated in the context of the respective risk management or risk stratification plan.

In addition, regarding treatment strategy we have to distinguish ‘impulse’ or induction-like high-efficacy therapy as alemtuzumab which provides positive long-term outcome mostly by only two depletion courses in contrast to the continuous high-efficacy therapies whose individual pharmacodynamic effects have shorter half-life time that ongoing treatment is needed [24]. Different labels in different countries have to be considered as well, for example, alemtuzumab and daclizumab are approved as first-line disease-modifying therapies only in Europe but not in the United States [25].

Data of the pivotal clinical studies do not provide enough information for individual treatment decisions between the high-efficacy treatments. One strategy is to perform subgroup analyses in randomized controlled trials [26]. To generate further scientific data for the evidence of early treatment optimization and treatment selection, real-world data are increasingly used for comparison studies to examine therapy choice and sequencing decisions [27]. Identification and mitigation of biases and careful consideration of study power are key factors for designing appropriate real-world evidence (RWE) studies [28]. Indeed, various biases exist and require careful consideration in selecting appropriate comparators, patient populations, data sources, outcomes, and statistical analyses.

In summary, implementing treat to target strategy in RRMS treatment requires detailed multidimensional monitoring and collection of different outcomes to evaluate treatment response. Treatment optimization to high-efficacy treatment should be considered relying on strategic and immunological characteristics and RWE.

Declaration of interest

Tjalf Ziemssen has received reimbursements for participation in scientific advisory boards from Bayer Healthcare, Biogen Idec, Novartis Pharma AG, Merck Serono, Teva, Genzyme, and Synthon. He has also received speaker honorarium from Bayer Healthcare, Biogen Idec, Genzyme, Merck Sharp & Dohme, GlaxoSmithKline, Novartis Pharma AG, Teva, Sanofi Aventis, and Almirall. He has also received research support from Bayer Healthcare, Biogen Idec, Genzyme, Novartis Pharma AG, Teva, and Sanofi Aventis. Katja Thomas received reimbursements for participation in scientific advisory boards from Novartis Pharma AG and Roche. Furthermore, she received speaker honorarium from Novartis, Bayer and Biogen idec. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This article was not funded.