Supporting Persistence in the First Year After a Therapy Change

Key Takeaways:

• In United States claims data, persistence after a first therapy switch fell from 91 percent at 3 months to 55 percent at 12 months, with a third of the loss between months 3 and 6.
• Each oral class has a predictable early adverse-effect window. Most of what a patient will feel in the first months is already in the label.
• Nearly 1 in 3 respondents with relapsing forms in the 2026 Multiple Sclerosis In America survey had used a disease-modifying therapy in the past but were not on one at the time of the survey.

A therapy change only works if the patient stays on it. Most of the loss happens earlier than clinicians tend to expect, and early enough that the visit where the change is made is the place to address it.

Where early discontinuation happens

In a claims analysis of 4,121 United States patients switching disease-modifying therapy (DMT) for the first time, 54.6 percent were still on that therapy at 12 months. Non-persistence was counted as a gap of 60 days or more in coverage, or another switch. The shape of the drop matters more than the 12-month figure: 91 percent still on the new DMT at 3 months, 75 percent at 6 months, 55 percent at 12.¹ About a third of patients who stopped by 12 months did so between months 3 and 6, before most patients would have had a follow-up visit.

Route of administration mattered too. Patients who switched to an oral agent had 1.95 times (95 percent confidence interval 1.66 to 2.31) the adjusted odds of still being on therapy at 12 months compared with those who switched to an injectable, and 0.82 (0.68 to 0.98) times the odds of a relapse afterward. Switching itself cut relapses by 32.4 percent between the pre-switch and post-switch periods.¹ That is the benefit a patient may be giving up by stopping their new therapy early.

What the tolerability data shows

A single-center cohort covering 213 treatment cycles on oral DMTs reported overall continuation of 72.8 percent, with no significant difference between the 3 agents studied. What did differ was reporting. Patients on dimethyl fumarate reported adverse events far more often than those on fingolimod or teriflunomide – 49 percent against 28 and 27.3 percent – and the investigators note that the extra reporting did not affect continuation on that agent, suggesting some of these adverse events with dimethyl fumarate may have been mild or transient.²

One caveat with adverse event reporting is under-reporting. Patients may not report, or may deny outright, adverse events that are mild, tolerable, or transient such that they resolve by the time of the next follow-up visit. Also, patients should be told in advance which adverse events are most common and how long they usually last, which may make those events, especially if mild, easier to tolerate and less likely to be reported.² In this study, they were not able to assess these potential confounding factors. Nonetheless, this data highlights the importance of talking to patients about the potential common side effects of therapies before the switch and monitoring for these side effects after the switch.

Early adverse effects after a switch

For diroximel fumarate, similar to other fumarates, gastrointestinal events cluster early, mostly in month 1, and 4 percent of patients stopped because of them against less than 1 percent on placebo. Flushing occurred in 40 percent, started soon after the first dose, and led 3 percent to stop.³ Some fumarate labels specify a 1-week starting dose before escalation, and allow a temporary dose reduction with a return to full dose within 4 weeks.^3,4

For teriflunomide, the mean white blood cell count falls during the first 6 weeks.5 Alanine aminotransferase elevation – the most common reason for discontinuation in teriflunomide trials – shows up mostly within the first year, and the label requires monthly liver-enzyme monitoring for the first 6 months.⁵ For fingolimod, macular edema occurs mainly in the first 3 to 4 months and blood pressure elevation is first detected at about 1 month, on top of the first-dose observation requirement.⁶ For cladribine tablets, the lymphocyte nadir comes 2 to 3 months after each treatment course.⁷

The adverse effects mentioned above are not a full safety profile. They are some of the ones that tend to appear in the first weeks to months after starting particular agents, when patients who have just switched may stop therapy.

When to check in

If persistence falls fastest between months 3 and 6, a first follow-up at 6 months arrives after the decision has been made. A check-in inside that window, around 3-4 months after a therapy switch, is the one that can change the outcome, and it takes 3 questions: did the expected effects show up, did they settle on schedule, and is the patient still taking the drug.

In the 2026 In America survey, 31.5 percent of respondents with relapsing forms had used a DMT in the past but were not using one at the time of the survey, against 7.5 percent who had never used one. One respondent, 25 years after diagnosis and no longer on therapy, put it plainly: "I am very sensitive to medication side effects and can no longer tolerate many meds."⁸