Treatment of epilepsy using newer antiseizure medications: A retrospective cohort study of prescription patterns in Sweden 2013–2022
Center for Health Governance, Department of Economics, University of Gothenburg, Gothenburg, Sweden
UCB, Stockholm, Sweden
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
Abstract
Objective
This study aimed to analyze the choice of newly introduced antiseizure medications (ASMs; brivaracetam, cannabidiol, cenobamate, fenfluramine, and perampanel) by age and sex, and the use of these ASMs in relation to regulatory approval.
Methods
Patients were identified with dispensation of any ASM subsequent to at least one health care contact due to epilepsy (International Classification of Diseases, 10th Revision: G40) during 2000–2022. Incidence rate ratios (IRRs) for starting a newly introduced ASM, retention rates, number of ASM treatments preceding the first dispensation of a new ASM and concurrent ASM treatments at first dispensed new ASM, and share of patients using new ASMs as adjunctive treatment were calculated.
Results
Three percent of all patients identified with epilepsy treatment during the period 2013–2022 used at least one new ASM. IRRs for ever being dispensed a new ASM differed between age groups for all ASMs except cenobamate. The corresponding IRRs associated with sex were insignificantly different from 1, except for cannabidiol; men were more likely to start using cannabidiol. One‐year retention rates were between 42% (cannabidiol, men) and 64% (brivaracetam, men). The majority (70%) of patients had used >2 ASMs prior to the first dispensation of a new ASM, and a majority (98%) of patients were treated with another ASM at first dispensation of a new ASM (≤90 days before). Similarly, 82% of the patients who started using a new ASM did so as an adjunctive treatment (ASMs dispensed ≤90 days before and ≥90 days after first new ASM).
Significance
Our observations indicate that prescription patterns of the newer ASMs comply with the currently approved indications for the respective medications for a majority of patients (prevalence of brivaracetam as adjunctive treatment is somewhat lower). Retention on the dispensed medications are in line with what would be expected for adjunctive treatment of patients with drug‐resistant epilepsies.
Untitled section
Keywords: add‐on therapy, antiseizure medications, epilepsy, monotherapy, treatment pathways
Article notes
Untitled section
Revised 2025 Jul 9; Received 2025 Mar 7; Accepted 2025 Jul 9; Issue date 2025 Dec.
Boxed Text
- In Sweden, 3% of treated epilepsy patients in 2013–2022 were dispensed either brivaracetam, cannabidiol, cenobamate, or perampanel.
- A majority (98%) of these patients were treated with another ASM at first dispensation of brivaracetam, cannabidiol, cenobamate, or perampanel.
- The 1‐year retention on these new ASMs ranged from 42% (cannabidiol) to 64% (brivaracetam).
1.INTRODUCTION
Since 1989, 20 new antiseizure medications (ASMs) have been introduced to the markets in the EU as well as in the USA. 1 Although, with the possible exception of cenobamate, in general none of these new compounds has demonstrated superior efficacy to older ASMs, most are less likely to cause pharmacokinetic interactions, and some have a more favorable adverse effect and safety profiles compared to the first generation ASMs. 1 , 2 Therefore, one might expect that the introduction of a new ASM would lead to a noticeable shift from the older to the newly introduced drugs. However, only few of the newer ASMs have established themselves as first‐line treatment in newly diagnosed epilepsy patients. 3 In addition to differences in costs, a plausible explanation for this is that the knowledge regarding adverse effects associated with a particular drug is incomplete at the time of its introduction, but improves as real‐world evidence accumulates, which suggests that the proclivity to use newly introduced ASMs increases over time if a more favorable safety and adverse effect profile is confirmed. An analysis of population‐based Swedish prescription data has recently shown that it took 24 years for lamotrigine (1989–2013) and 14 years for levetiracetam (1999–2013) to be more prescribed as initial monotherapy than any of the older ASMs. 4 Much less is known about the prescribers' uptake of the newest of the new ASMs, namely, those introduced and approved by the European Medicines Agency (EMA) after 2011 (perampanel, brivaracetam, cannabidiol, fenfluramine, and cenobamate). In contrast to lamotrigine and levetiracetam, which have broad indications including monotherapy, these newer ASMs are only licensed for adjunctive treatment and for drug‐resistant epilepsies or for less common specific epilepsy types (Table 1). 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 Hence, in the EU cannabidiol is indicated for the treatment of seizures associated with Lennox–Gastaut and Dravet syndromes and tuberous sclerosis complex, and fenfluramine for Lennox–Gastaut and Dravet syndromes. Furthermore, there are specific additional requirements for reimbursement of costs for treatment with perampanel and brivaracetam, whereas cenobamate was included in the Swedish reimbursement system on September 24, 2021 (approximately 3 months before our study end point), and cannabidiol has not been included (Table 1). To the best of our knowledge, the utilization of these newest ASMs has been explored in only a small number of published studies. 15 , 16
| Brivaracetam | Cannabidiol | Cenobamate | Perampanel | Fenfluramine | |
|---|---|---|---|---|---|
| Indication (source: European Medicines Agency) | Adjunctive treatment in the treatment of partial onset seizures with or without secondary generalization in adults, adolescents, and children from 2 years of age with epilepsy Date of market authorization: January 14, 2016 | Adjunctive treatment of seizures associated with Lennox–Gastaut or Dravet syndrome, in conjunction with clobazam, for patients 2 years and older Adjunctive therapy of seizures associated with TSC for patients 2 years and older Date of market authorization: September 19, 2019 | Adjunctive treatment of focal onset seizures with or without secondary generalization in adult patients with epilepsy who have not been adequately controlled despite a history of treatment with at least 2 antiepileptic medicinal products Date of market authorization: March 26, 2021 | Adjunctive treatment of focal onset seizures with or without secondarily generalized seizures in patients 4 years and older Adjunctive treatment of primary generalized tonic–clonic seizures in patients 7 years and older with idiopathic generalized epilepsy Date of market authorization: July 23, 2012 | Adjunctive treatment of seizures associated with Dravet syndrome and Lennox–Gastaut syndrome in patients 2 years and older Dates of market authorization: Lennox–Gastaut syndrome: February 24, 2023 Dravet syndrome: December 18, 2020 |
| Reimbursement and reimbursement criteria (Sweden) | Only reimbursed as adjunctive treatment for patients for whom levetiracetam is not suitable, and who have tried at least one adjunctive treatment but did not tolerate or benefit from it Included in the reimbursement system: February 23, 2018 | Not included in the reimbursement system | No restrictions Included in the reimbursement system: September 24, 2021 | Only reimbursed for adjunctive treatment in focal seizures for drug‐resistant patients who have tried at least 1 adjunctive treatment but did not tolerate or benefit from it Included in the reimbursement system: December 8, 2012 | Only reimbursed as treatment of Lennox–Gastaut syndrome for patients with unsatisfactory treatment response on conventional treatment Included in the reimbursement system: May 1, 2025 |
The main objective in this study was to describe and analyze the use of these five new ASMs in Sweden from the time of approval—brivaracetam, cannabidiol, cenobamate, fenfluramine, and perampanel—in the following, collectively labeled “new ASMs.” More specifically, the objective was to describe the prevalence and incidence of patients using one of these ASMs, the retention rates, the prevalence of mono‐ and combination therapies, and the ASM treatment paths preceding the first dispensed new ASM, by age and sex.
2.MATERIALS AND METHODS
The data used in this study were collected from Swedish total population individual‐level registers comprising information on ASMs and out‐ and inpatient care visits. These national registers are administered by the Swedish National Board of Health and Welfare (more detailed descriptions of the employed registers can be found on the webpage of the board), and anonymized data are provided to qualified researchers after approval by the Swedish Ethical Review Authority (see formal statement below). Epilepsy was defined as having at least one hospital visit due to epilepsy (International Classification of Diseases, 10th Revision [ICD‐10] G40 as main diagnosis in the Swedish Patient Register) and at least one subsequently dispensed ASM. 17 Thus, the study population was constructed by identifying all patients with a dispensed ASM (Swedish Prescribed Drug Register, July 2005–2022; Swedish National Board of Health and Welfare) and at least one hospital visit due to epilepsy reported in the Swedish Patient Register (2000–2022; Swedish National Board of Health and Welfare) before or on the same date as a dispensed ASM. For each patient, information about health care contacts and dispensed ASMs was collected prospectively from the date of identification. The aforementioned new ASMs are identified in the Prescribed Drug Register by the following Anatomical Therapeutic Chemical codes: N03AX23 (brivaracetam), N03AX24 (cannabidiol), N03AX25 (cenobamate), N03AX26 (fenfluramine), and N03AX22 (perampanel).
The longitudinal content of the collected register data regarding dispensed pharmaceuticals, and data on health care contacts, were used not only to distinguish between patients treated for epilepsy and other patients (as described above), but also to determine the number and share of patients who remain on a new ASM.
2.1.Data analysis
All reported numbers and other statistics have been calculated for the total population of patients treated for epilepsy as defined above in Sweden during the period 2005–2022. Epilepsy patients treated with any of the five new ASMs are a subset of this population. The following statistics were calculated: (1) the total number of epilepsy patients receiving a new ASM and the incidence rate ratios (IRRs), by new ASM, age, and sex; and (2) the 90‐day and 1‐year retention rates by the aforementioned groups (90‐day retention is defined as having at least one subsequent dispensation at least 90 days beyond the first observed dispensation and analogously for 1‐year retention; retention calculations included only dispensed ASMs before 2022). Furthermore, separately for the years 2012–2022 (2013 is the first year for which any of the new included ASMs was dispensed), the following statistics were calculated: (3) the prevalent and incident number of epilepsy patients (incidence: annual number of new patients fulfilling our epilepsy criteria) and the corresponding frequencies pertaining to new ASMs, (4) the cumulative number of epilepsy patients receiving each of the newer ASMs, and by new ASM; (5) the number of ASMs tried (dispensed) before the first dispensation; and (6) the number of ASMs at the time (≤90 days before) of the first dispensation.
The new ASMs were introduced into the market at different points in time. The IRRs—(1) above—are calculated by modifying observed person time according to date of introduction. More specifically, person time at risk is calculated as the time between date of introduction or date of first health care contact due to epilepsy, whichever came last, and the date of the first observed dispensation.
Because brivaracetam, cannabidiol, cenobamate, fenfluramine, and perampanel are approved by EMA (and the Swedish Medical Products Agency) only as adjunctive therapies, we identified the number of patients receiving a new ASM and used it as an adjunctive therapy at the time of dispensation. There is no available information in the registries employed about intention to treat and, hence, whether an observed patient used a combination therapy (a new ASM as adjunctive therapy) was inferred from the observed dispensation patterns of ASMs. Adjunctive therapy was defined as follows. A patient was assumed to initialize his or her use of a new ASM as adjunctive treatment if, and only if, other ASMs were dispensed before and after the initial dispensation of a new ASM, in both cases within 90 days. This means that a necessary, but not sufficient, condition for using a new ASM as adjunctive treatment is that the patient has treatment with another/older ASM at the time of first dispensation of a new ASM. More specifically, the adjunctive ASMs were identified by first identifying the date of dispensation of ASMs closest in time, within the 90‐day time limit, to the date of dispensation of the first new ASM.
2.2.Standard protocol approvals, registrations, and patient consents
This study has been approved by the Swedish Ethical Review Authority (full review; numbers 2022‐03377‐02 and 2020‐06211). Informed patient consent is not necessary for register‐based research using secondary data.
3.RESULTS
There was no observed dispensation of fenfluramine in the Swedish Prescribed Drug Register before 2023 and hence, fenfluramine does not appear when reporting the results. Only a small fraction of epilepsy patients on pharmacological treatment used any of the new ASMs during the studied period. Of 106 315 patients with epilepsy and on ASM treatment during the period 2013–2022, only 3145 used a new ASM (some patients have used more than one new ASM and hence, the number of patients summed over all new ASMs exceeds the number of patients having used a new ASM; see Table 2).
| Brivaracetam | Cannabidiol c | Cenobamate | Perampanel | ||||||
|---|---|---|---|---|---|---|---|---|---|
| n patients | IRR (95% CI) | n patients | IRR (95% CI) | n patients | IRR (95% CI) | n patients | IRR (95% CI) | ||
| Age, years | |||||||||
| 0–2 [<2] | 3 | 9.39 (1.93–27.50) | 1 | 6.45 (.16–37.45) | 0 | .00 (.00–.00) | 0–7 [<7] | 35 | 2.22 (1.55–3.11) |
| 2–15 | 141 | 1.20 (1.00–1.43) | 26 | .51 (.29–.89) | 3 | .82 (.17–2.40) | 7–15 | 150 | 1.45 (1.22–1.71) |
| 16–40 | 559 | .87 (.78–.98) | 26 | 1.40 (.80–2.44) | 198 | .98 (.80–1.22) | 16–40 | 967 | .98 (.89–1.07) |
| 41–65 | 400 | .87 (.78–.98) | 4 | – d | 137 | 1.05 (.84–1.31) | 41–65 | 630 | .91 (.83–1.00) |
| >65 | 264 | 1.36 (1.19–1.56) | 0 | – | 18 | .88 (.51–1.41) | >65 | 167 | .93 (.79–1.09) |
| Sex | |||||||||
| Male | 634 | 1.05 (.95–1.17) | 32 | 2.38 (1.37–4.20) | 166 | .99 (.80–1.23) | 905 | .99 (.91–1.08) | |
| Female | 733 | 25 | 190 | 1044 | |||||
| Patients starting before 2022, n (male/female) | 423 (516) | 26 (18) | 40 (42) | 800 (924) | |||||
| Retention at 90 days (1 year) e | |||||||||
| Male | 348 (270) | 82% (64%) | 14 (11) | 54% (42%) | 35 (17) | 88% (43%) | 615 (459) | 77% (57%) | |
| Female | 403 (311) | 78% (60%) | 14 (11) | 78% (61%) | 35 (22) | 83% (52%) | 655 (477) | 71% (52%) | |
Table 1 provides a summary of approved indications and reimbursement criteria (Swedish setting). Table 2 reports (1) IRRs (for using a new ASM) pertaining to age group and sex; and (2) the number of patients retaining a new ASM over 90 days and 1 year, respectively. There were no significant (95% confidence interval) differences in IRRs between age groups, except that patients older than 65 years were more likely to start using brivaracetam. Similarly, there were no differences in IRRs between men and women, except that men were more than twice as likely to start using cannabidiol. The 90‐day retention rates were estimated between 54% (cannabidiol, men) and 88% (cenobamate, men). Similarly, the 1‐year retention rates were estimated between 42% (cannabidiol, men) and 64% (brivaracetam, men). Of a total of 57 patients treated for epilepsy using cannabidiol, 32 had prior to the initiation of treatment been diagnosed with an ICD code compatible with Dravet syndrome (G404), and six patients had been diagnosed with Lennox–Gastaut syndrome (G404B).
Table 3 reports, per year 2013–2022: (1) the total number of epilepsy patients; (2) the number of incident epilepsy patients; (3) the total number of patients using brivaracetam, perampanel, cenobamate, and cannabidiol, respectively; (4) the number of patients who used any of the aforementioned new ASMs for the first time (the number of unique patients is reported for each year separately); and (5) the cumulative number of patients who have ever used brivaracetam, perampanel, cenobamate, or cannabidiol, respectively. The incidence of epilepsy was roughly constant over the observed time period (2013–2022), whereas the total number of patients increased by approximately 25%. The share of epilepsy patients treated with an ASM who were treated with a new ASM increased from approximately .6% in 2013 to approximately 3% in 2022. The annual number of patients who used a new ASM the first time more than doubled over the observed period. As a share of prevalent epilepsy patients, this corresponds to an increase from .6% in 2013 to approximately 1% in 2022. The increase in the overall number of epilepsy patients from 2013 to 2022 corresponds to an increase in prevalence from .57% in 2013 to .65% in 2022.
| 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total number of epilepsy patients | 54 953 | 56 832 | 58 454 | 60 183 | 61 817 | 63 310 | 64 798 | 65 962 | 67 229 | 68 830 |
| Total number of new (incident) epilepsy patients | 5468 | 5611 | 5444 | 5484 | 5482 | 5398 | 5420 | 5281 | 5298 | 5420 |
| Total number of epilepsy patients treated with new ASM | ||||||||||
| Brivaracetam | 0 (.00%) | 0 (.00%) | 0 (.00%) | 22 (.04%) | 22 (.04%) | 137 (.22%) | 324 (.50%) | 499 (.76%) | 704 (1.05%) | 1008 (1.46%) |
| Cannabidiol | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 7 (.01%) | 12 (.02%) | 28 (.04%) | 25 (.04%) | 30 (.04%) |
| Cenobamate | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 82 (.12%) | 349 (.51%) |
| Perampanel | 317 (.58%) | 408 (.72%) | 422 (.72%) | 473 (.79%) | 511 (.83%) | 544 (.86%) | 653 (1.01%) | 738 (1.12%) | 815 (1.21%) | 901 (1.31%) |
| Unique patients | 317 (.58%) | 408 (.72%) | 422 (.72%) | 492 (.82%) | 530 (.86%) | 665 (1.05%) | 949 (1.46%) | 1210 (1.83%) | 1526 (2.27%) | 2098 (3.05%) |
| Epilepsy patients with a first dispensation of new ASM (patients can have >1 new ASM in first year) | ||||||||||
| Brivaracetam | 0 | 0 | 0 | 14 | 4 | 87 | 178 | 223 | 289 | 377 |
| Cannabidiol | 0 | 0 | 0 | 0 | 0 | 2 | 3 | 12 | 5 | 9 |
| Cenobamate | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 21 | 107 |
| Perampanel | 317 | 194 | 168 | 156 | 141 | 144 | 193 | 196 | 175 | 207 |
| Unique individuals | 317 | 194 | 168 | 170 | 145 | 229 | 365 | 413 | 473 | 671 |
| Cumulative number of patients who ever used a new ASM | ||||||||||
| Brivaracetam | 0 (.00%) | 0 (.00%) | 0 (.00%) | 22 (.04%) | 28 (.05%) | 144 (.23%) | 359 (.55%) | 620 (.94%) | 939 (1.40%) | 1367 (1.99%) |
| Cannabidiol | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 7 (.01%) | 14 (.02%) | 35 (.05%) | 44 (.07%) | 57 (.08%) |
| Cenobamate | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 0 (.00%) | 82 (.12%) | 356 (.52%) |
| Perampanel | 317 (.58%) | 511 (.90%) | 679 (1.16%) | 835 (1.39%) | 976 (1.58%) | 1120 (1.77%) | 1321 (2.04%) | 1526 (2.31%) | 1724 (2.56%) | 1949 (2.83%) |
| Unique individuals | 317 (.58%) | 511 (.90%) | 679 (1.16%) | 849 (1.41%) | 994 (1.61%) | 1223 (1.93%) | 1588 (2.45%) | 2001 (3.03%) | 2474 (3.68%) | 3145 (4.57%) |
Figure 1 illustrates the number of ASMs used before or at the time of the first dispensation of a new ASM. The proportion with more than five prescribed ASMs prior to first dispensation of a new ASM was highest among those dispensed cannabidiol (68%) and lowest with brivaracetam (22%). The vast majority of patients had used at least one ASM previously to the first observed dispensation of a new ASM. The share of patients starting with a new ASM who had not used any other ASMs before that was calculated at 5% (brivaracetam), 0% (cannabidiol), 1% (cenobamate), and 1% (perampanel). Similarly, the majority of patients were treated with another ASM at the time of first dispensation of a new ASM (ASMs dispensed ≤90 days before first dispensation of new ASM). The share of patients initiating treatment with a new ASM not dispensed another ASM at the time (≤90 days before) of first dispensation of a new ASM varied between approximately 13% (brivaracetam) and approximately 5% (cannabidiol).
Table 4 reports, per new ASM: (1) the number of patients using an adjunctive therapy and the total number of patients and (2) the frequency of the five most common adjunctive therapies before and after the first dispensation of a new ASM. The first observation pertains to the share of patients who use a new ASM as an adjunctive treatment; patients starting with brivaracetam as the first new ASM did use it as an adjunctive treatment to a lesser extent than patients starting with any of the other three new ASMs (70% vs. 87%–91%). New ASMs were used as adjunctive treatments most frequently together with lamotrigine, levetiracetam, valproic acid, and lacosamide. Of the 957 patients who initiated brivaracetam as an adjunctive treatment, 83 patients (10%) were observed having subsequent dispensations of levetiracetam (Table 4). However, of 518 patients who remained on brivaracetam for at least 1 year, only 24 also remained on levetiracetam (4%).
| ASM before | Patients, n | ASM after | Patients, n |
|---|---|---|---|
| Brivaracetam, 957 (1367) | |||
| Lamotrigine | 292 | Lamotrigine | 314 |
| Levetiracetam | 192 | Lacosamide | 181 |
| Lacosamide | 166 | Valproic acid | 158 |
| Valproic acid | 164 | Levetiracetam | 98 b |
| Carbamazepine | 90 | Carbamazepine | 97 |
| Cannabidiol, 52 (57) | |||
| Valproic acid | 14 | Valproic acid | 13 |
| Lamotrigine | 9 | Lamotrigine | 10 |
| Levetiracetam | 8 | Levetiracetam | 7 |
| Perampanel | 6 | Topiramate | 6 |
| Topiramate | 6 | Clonazepam/perampanel c | 5/5 |
| Cenobamate, 309 (356) | |||
| Lamotrigine | 78 | Lamotrigine | 81 |
| Lacosamide | 65 | Lacosamide | 66 |
| Levetiracetam | 60 | Levetiracetam | 56 |
| Valproic acid | 59 | Valproic acid | 54 |
| Carbamazepine | 37 | Carbamazepine | 40 |
| Perampanel, 1732 (1949) | |||
| Lamotrigine | 490 | Lamotrigine | 487 |
| Levetiracetam | 488 | Levetiracetam | 470 |
| Valproic acid | 305 | Valproic acid | 309 |
| Lacosamide | 290 | Lacosamide | 265 |
| Carbamazepine | 235 | Carbamazepine | 241 |
In addition to the central result, we calculated mean observed duration of epilepsy at first dispensation of brivaracetam and perampanel (time from first dispensation of an ASM after a health care contact due to epilepsy to first dispensation of the new ASM), respectively, distinguishing between patients observed at the first health care visit due to epilepsy before and after the date of introduction, respectively. The mean duration of epilepsy at first dispensation of brivaracetam and perampanel was estimated at 13.5 and 13.6 years, respectively, for patients diagnosed prior to market introduction. The corresponding figures for patients diagnosed after market introduction were estimated at 2.6 and 2.5 years.
4.DISCUSSION
Using Swedish comprehensive patient‐level, nationwide, health care and pharmaceutical dispensation data, for the years 2013–2022, we studied the use of new ASMs (brivaracetam, cannabidiol, cenobamate, and perampanel) and noted a gradual increase in the share of epilepsy patients being dispensed a new ASM, to approximately 3% of epilepsy patients in 2022. By the end of 2022, 1367 patients had used brivaracetam, 57 cannabidiol, 356 cenobamate, and 1949 perampanel. Market approvals for all four (new) ASMs are conditional on clinical use (e.g., only as adjunctive therapy). The most restrictive conditions are those for cannabidiol, which was approved for the treatment of seizures associated with Lennox–Gastaut or Dravet syndrome or tuberous sclerosis complex, that is, rare epilepsy syndromes. 18 , 19 , 20 Perampanel has the broadest set indications, including adjunctive treatment of focal seizures as well as “primary generalized tonic–clonic seizures.” The new ASMs also differ as to age limitations and in date of market approval and reimbursement; perampanel was the first to be introduced to the European market, followed by brivaracetam, cannabidiol, and cenobamate. Except for cannabidiol, these new ASMs were included in the Swedish pharmaceutical reimbursement system in that same chronological order (cannabidiol is not included).
Previous studies provide limited relevant data for comparison. 15 , 16 A cohort study of Medicare beneficiaries that assessed ASM prescriptions during 2008–2018 16 reported that of the four ASMs in our current study, perampanel was the most prescribed, followed by brivaracetam and cannabidiol. However, no details were given on the actual numbers nor on whether these ASMs were used in accordance with the licensing conditions. A questionnaire‐based study targeting 464 patients with epilepsy from epilepsy centers in Germany analyzed prescription patterns in the year 2022. 14 Brivaracetam was prescribed to 10.6% (4.1% in monotherapy), perampanel in 7.2% (1.5% in monotherapy), cannabidiol in .6% (none in monotherapy), and fenfluramine in .1% (none in monotherapy). The higher use of brivaracetam and perampanel can be because it was based on responses from patients attending epilepsy centers rather than being population‐based as was ours.
Our data indicate that in the vast majority of cases, the new ASMs were dispensed in compliance with market approval and reimbursement requirements, that is, as adjunctive therapy and when applicable after failing other ASMs. However, there were a few exceptions, also regarding adherence to age limitations. A few patients (1%) had not been dispensed another ASM prior to the first dispensation of perampanel, and <2% of the patients were younger than 7 years. However, approximately 11% did not use it as adjunctive treatment (at the time of first observed dispensation). Correspondingly, as to brivaracetam, approximately 5% of the patients were not observed with any dispensation of an ASM prior to the first dispensation of brivaracetam, and approximately 30% did not use it as adjunctive therapy. A significant proportion (20%) of patients with a first dispensation of brivaracetam used as adjunctive therapy had prior dispensation of levetiracetam, which could be because brivaracetam may be better tolerated and possibly in some patients more effective than levetiracetam, a possibility that could also explain the finding that a large proportion of patients dispensed brivaracetam as monotherapy seem to have been switched from levetiracetam. 21 , 22 Of note, in the USA, the US Food and Drug Administration has approved brivaracetam for monotherapy. More specifically, in the current study, of the 410 patients who initiated treatment with brivaracetam but did not use it as an adjunctive treatment, 257 had used only levetiracetam as the most recent ASM. It is more surprising that some patients (10%) continue to be dispensed levetiracetam after being initiated on brivaracetam, but this apparently irrational combination has been reported in other observational studies 23 and is not in conflict with conditions for market approval. Moreover, <1% of the patients dispensed brivaracetam were younger than 2 years (the market‐approved minimum age in Europe; the corresponding minimum age in the USA is 1 month). All patients dispensed cannabidiol had tried other ASMs before, and almost 68% more than five different ASMs before being prescribed cannabidiol. However, our data suggest that there were a few (5%) who were not on treatment with any other ASM at the time of the first cannabidiol dispensation. This can be because we did not access clobazam dispensation, a treatment that is a condition for the use of cannabidiol for seizures associated with Dravet or Lennox–Gastaut syndrome. The cenobamate dispensation pattern was observed to be qualitatively similar to that observed for perampanel. A small share (approximately 3%) of patients had tried fewer than two other ASMs before, as required according to the approved indications, and relatively few (11%) received the drug as monotherapy.
The 1‐year retention rates were estimated between 42% (cannabidiol, men) and 64% (brivaracetam, men). Estimated retention rates revealed moderate differences between men and women; men were more likely than women to remain on brivaracetam and perampanel over a 1‐year period, whereas the opposite was true for cannabidiol and cenobamate. Although 1‐year retention is likely to reflect tolerability as well as efficacy, 90‐day retention is probably more an expression of short‐term tolerability. However, the comparatively high 90‐day retention on cenobamate should be interpreted in the light of the very slow uptitration of this ASM over 11 weeks. This means that patients will be exposed to the target maintenance dose of cenobamate for a very short period over the first 90 days. In a recent study, 4 5‐year retention rates for the most frequently used monotherapies were calculated and reported. Using the same data, the corresponding 1‐year retention rates (not reported in the article) were estimated in the range 62%–71%. Thus, patients using new ASMs in the current analysis were more likely to change their treatment within 1 year than patients initiating epilepsy treatment with one of the most frequently used monotherapies (the exception being patients using brivaracetam), which is to be expected given their more severe and drug‐resistant epilepsies.
4.1.Strengths and limitations
The major strength of the current study is that it is population‐based, utilizing national registers with complete coverage, thus reducing the risk of bias. However, a few caveats need to be addressed. First, we do not distinguish between dispensed pharmaceuticals and usage/treatment. This is for obvious reasons; we do not have individual information about actual use of dispensed ASMs. Second, in our analyses, we did not take doses of the dispensed ASMs into account. Third, the observed dispensation patterns were interpreted in terms of mono or combination therapy, where combination therapy was defined as having dispensation of older ASMs before and after the incident dispensation of a new ASM (within 90 days). This definition, however, is not unique in the sense of being the only plausible definition. For example, the definition applied in the aforementioned recent study—demanding a specific sequence of ASMs being dispensed—would entail radically different results; our calculations using this alternative definition result in <50% of patients using a new ASM as adjunctive treatment. Fourth, available data did not allow for detailing the type of epilepsy to the extent that it could be compared to approved indications. Neither was it possible to detail seizure control or adverse effects to attain a complete understanding of retention. A further important limitation, as mentioned above, was that we did not access the dispensation of clobazam. Finally, market authorization and reimbursement systems will inevitably impact prescription and dispensation patterns of ASMs and, thus, affect the generalizability of the current results to settings with different systems.
5.CONCLUSIONS
When new ASMs are first introduced to the market, indications are often restrictive, and it is important to understand to what extent these are adhered to. Our observations in this nationwide population‐based study indicate that prescription patterns of the newer ASMs in general comply with the currently approved indications for the respective medications. Brivaracetam is an exception, where 30% appeared to be dispensed this medication as monotherapy. Almost half of these patients used levetiracetam as monotherapy before starting brivaracetam. Retention on the dispensed medications are in line with what would be expected for adjunctive treatment of patients with drug‐resistant epilepsies.
FUNDING INFORMATION
UCB has financially supported this research. The funder had the opportunity to review and comment on the final manuscript.
CONFLICT OF INTEREST STATEMENT
T.T. reports funding from Accord, Glenmark, GSK, UCB, Eisai, Ecu Pharma, Bial, Teva, Sanofi, SF Group, GW Pharma, Zentiva, and Angelini as donations to the EURAP pregnancy registry; and speaker honoraria from Eisai, Angelini, GSK, and UCB. K.B. reports funding from UCB. P.B. is an employee and stockholder of UCB Pharma. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
ACKNOWLEDGMENTS
The authors thank Vincent Laporte, PhD, of UCB for publication management support.
Untitled section
Bolin K, Berling P, Tomson T. Treatment of epilepsy using newer antiseizure medications: A retrospective cohort study of prescription patterns in Sweden 2013–2022. Epilepsia. 2025;66:4809–4820. 10.1111/epi.18576
DATA AVAILABILITY STATEMENT
Data were based on Swedish national registers, and individual level data cannot be shared due to national regulations. Original data are available upon application to the relevant authorities.
REFERENCES
Untitled section
References
- 1. Perucca E, Brodie MJ, Kwan P, Tomson T. 30 years of second‐generation Antiseizure medications: impact and future perspectives. Lancet Neurol. 2020;19(6):544–556.
- 2. Barbieri MA, Perucca E, Spina E, Rota P, Franco V. Cenobamate: a review of its pharmacological properties, clinical efficacy and tolerability profile in the treatment of epilepsy. CNS Neurol Disord Drug Targets. 2023;22(3):394–403.
- 3. Tomson T, Zelano J, Dang YL, Perucca P. The pharmacological treatment of epilepsy in adults. Epileptic Disord. 2023;25(5):649–669.
- 4. Bolin K, Berling P, Tomson T. Trends in antiseizure medication initiation, switch, or termination in patients with newly diagnosed epilepsy: a nationwide study. Neurology. 2024;103(2):e209500.
- 5. Klein P, Schiemann J, Sperling MR, Whitesides J, Liang W, Stalvey T, et al. A randomized, double‐blind, placebo‐controlled, multicenter, parallel‐group study to evaluate the efficacy and safety of adjunctive brivaracetam in adult patients with uncontrolled partial‐onset seizures. Epilepsia. 2015;56(12):1890–1898.
- 6. Krauss GL, Klein P, Brandt C, Lee SK, Milanov I, Milovanovic M, et al. Safety and efficacy of adjunctive cenobamate (YKP3089) in patients with uncontrolled focal seizures: a multicentre, double‐blind, randomised, placebo‐controlled, dose‐response trial. Lancet Neurol. 2020;19(1):38–48.
- 7. Chung SS, French JA, Kowalski J, Krauss GL, Lee SK, Maciejowski M, et al. Randomized phase 2 study of adjunctive cenobamate in patients with uncontrolled focal seizures. Neurology. 2020;94(22):e2311–e2322.
- 8. Devinsky O, Cross JH, Laux L, Marsh E, Miller I, Nabbout R, et al. Trial of Cannabidiol for drug‐resistant seizures in the Dravet syndrome. N Engl J Med. 2017;376(21):2011–2020.
- 9. Devinsky O, Patel AD, Cross JH, Villanueva V, Wirrell EC, Privitera M, et al. Effect of Cannabidiol on drop seizures in the Lennox–Gastaut syndrome. N Engl J Med. 2018;378(20):1888–1897.
- 10. French JA, Krauss GL, Steinhoff BJ, Squillacote D, Yang H, Kumar D, et al. Evaluation of adjunctive perampanel in patients with refractory partial‐onset seizures: results of randomized global phase III study 305. Epilepsia. 2013;54(1):117–125.
- 11. French JA, Krauss GL, Biton V, Squillacote D, Yang H, Laurenza A, et al. Adjunctive perampanel for refractory partial‐onset seizures: randomized phase III study 304. Neurology. 2012;79(6):589–596.
- 12. Biton V, Berkovic SF, Abou‐Khalil B, Sperling MR, Johnson ME, Lu S. Brivaracetam as adjunctive treatment for uncontrolled partial epilepsy in adults: a phase III randomized, double‐blind, placebo‐controlled trial. Epilepsia. 2014;55(1):57–66.
- 13. Thiele EA, Marsh ED, French JA, Mazurkiewicz‐Beldzinska M, Benbadis SR, Joshi C, et al. Cannabidiol in patients with seizures associated with Lennox‐Gastaut syndrome (GWPCARE4): a randomised, double‐blind, placebo‐controlled phase 3 trial. Lancet. 2018;391(10125):1085–1096.
- 14. Krauss GL, Serratosa JM, Laurenza A, Villanueva V, Endziniene M, Hong Z, et al. Randomized phase III study 306: adjunctive perampanel for refractory partial‐onset seizures. Neurology. 2012;78(18):1408–1415.
- 15. Hochbaum M, Kienitz R, Rosenow F, Schulz J, Habermehl L, Langenbruch L, et al. Trends in antiseizure medication prescription patterns among all adults, women, and older adults with epilepsy: a German longitudinal analysis from 2008 to 2020. Epilepsy Behav. 2022;130:108666.
- 16. Terman Lin CC, Kerr WT, DeLott LB, Callaghan BC, Burke JF. Changes in the use of brand name and generic medications and total prescription cost among medicare beneficiaries with epilepsy. Neurology. 2022;99(8):e751–e761.
- 17. Mbizvo GK, Bennett KH, Schnier C, Simpson CR, Duncan SE, Chin RFM. The accuracy of using administrative healthcare data to identify epilepsy cases: a systematic review of validation studies. Epilepsia. 2020;61(7):1319–1335.
- 18. Sullivan J, Benítez A, Roth J, Andrews JS, Shah D, Butcher E, et al. A systematic literature review on the global epidemiology of Dravet syndrome and Lennox–Gastaut syndrome: prevalence, incidence, diagnosis, and mortality. Epilepsia. 2024;65(5):1240–1263.
- 19. Strzelczyk A, Zuberi SM, Striano P, Rosenow F, Schubert‐Bast S. The burden of illness in Lennox‐Gastaut syndrome: a systematic literature review. Orphanet J Rare Dis. 2023;18(1):1–21.
- 20. Strzelczyk A, Lagae L, Wilmshurst JM, Brunklaus A, Striano P, Rosenow F, et al. Dravet syndrome: a systematic literature review of the illness burden. Epilepsia Open. 2023;8(4):1256–1270.
- 21. Klein P, Bourikas D. Narrative review of brivaracetam: preclinical profile and clinical benefits in the treatment of patients with epilepsy. Adv Ther. 2024;41(7):2682–2699.
- 22. Yates SL, Fakhoury T, Liang W, Eckhardt K, Borghs S, D'Souza J. An open‐label, prospective, exploratory study of patients with epilepsy switching from levetiracetam to brivaracetam. Epilepsy Behav. 2015;52:165–168.
- 23. Siddiqui F, Soomro BA, Rehman EU, Numan A, Bano S, Salam JU, et al. A prospective, observational, multicentre study to evaluate the efficacy of brivaracetam as adjuvant therapy for epilepsy: the Bravo study. Drugs Context. 2024;13:2024‐3‐2.
Associated Data
Data Availability Statement
Data were based on Swedish national registers, and individual level data cannot be shared due to national regulations. Original data are available upon application to the relevant authorities.