In this observational study we describe the frequency, reasons and determinants of switching or suspension of DMDs, as well as the patterns of DMD prescription in our cohort. Our results show that most patients who started treatment had a DMD switch at some point, being disease activity the most common reason for change and that the persistence on treatment was similar for the different DMDs, regardless of their efficacy or mode of administration. In addition, the insurer status of the patients was found to be the main determinant of DMD switching. Most treatment switches involved interferons fingolimod and treatment suspension.
Our findings are in line with similar previous studies, which show that most patients treated will eventually change their therapy[10–13]. However, the treatment persistence in different studies is varied, with median time to treatment switch ranging from 25 months[12] to 50 months[10, 11]. In our study, treatment persistence was longer, with a median time of 60 months. The differences in these results are still to be addressed, but the structure of the healthcare system and the availability of different DMDs in different countries might be determinant. Other factors that influence the persistence on DMDs, such as comorbidities[13, 14] might contribute to the differences in the results. However, our study did not assess the presence of comorbidities, which is a limitation that needs to be acknowledged.
We find it interesting that the treatment persistence was very similar across the different DMDs in terms of their classification based on efficacy and mode of administration. This leads to thinking that the decision to switch or stop therapy might be driven by factors different from those, such as cost, and continuous availability of the medications involved.
The most common reason for treatment change was disease activity. This is in line with previous findings[15, 16], although in these studies a minority of patients treated with high-efficacy DMDs were assessed.
In our study, age at MS onset and insurer status were the only variables associated with the risk of switching therapies. The former has several interpretations. It is possible that patients with younger age at onset had a more active disease than patients with onset at an older age, which might increase the likelihood of treatment switching due to lack of efficacy. Similar findings were reported by previous studies, which described that subjects who were younger when starting a DMD, were more likely to discontinue therapy[10, 11, 13]. On the other hand, patients with MS onset at older age are more likely to have progressive MS, for which the therapeutic options are limited, leading to a decreased likelihood of treatment switches. However, these results should be interpreted carefully. It is important that more than half of the patients received interferons as the first treatment, and thus the overall risk of treatment changes might be driven by the behaviour of treatment switches from this group of medications.
Patients with younger age at onset and enough disease duration to date were probably treated during times in which no alternatives to interferons were available; on the other hand, young patients with recent MS onset might not have had enough follow-up time to need a treatment switch. The influence of more recent use of DMDs with higher efficacy (particularly of anti-CD20 antibodies) in the behaviour of treatment switching remains to be studied.
We did not find an association of disability with treatment persistence as has been previously reported[13, 16], although it must be borne in mind that we had access to scarce EDSS determinations and could not assess the time relationships between these and the treatment switches.
The analysis of treatment patterns revealed that most switches occurred around the interferons, fingolimod and treatment suspension. This is likely explained due to the long time the interferons and fingolimod were the only available therapeutic options in our country, but also the high frequency of tolerability issues associated with the use of interferons.
The most relevant finding in our opinion is the influence of the insurer in the risk of treatment changes. Although the coverage for the different DMDs in our country is mandated to be equal regardless of the individual insurers, several aspects of the care of MS might differ across them. These include, for example, the access to continuous and specialised consultations, regular imaging monitoring and continuous DMD provision, among others, which might have influenced MS activity, and thus the need for treatment changes. However, the insurer status of the patients might be an expression of unobserved variables such as socioeconomic status and educational attainment, which are known to be related to access barriers[17], and might have influenced the treatment persistence as well.
We acknowledge several limitations of our study, the most important of which is its retrospective design, which introduces a high risk of recall bias. This is particularly important due to the fact that most of the patients were already treated and several had treatment switches before being cared for at our institution. There is also a high risk of selection bias, patients with more aggressive MS or with more administrative difficulties in the past might have been drawn to seek care at our reference centre.
In conclusion, our results suggest that a majority of patients will have treatment changes, most likely caused by disease activity, but also influenced by coverage issues. Most treatment changes have occurred involving interferons, likely due to being the first DMDs available. Further study is needed to assess persistency and its determinants in different healthcare settings.