The Neuropsychiatric Safety Profile of Lasmiditan: A Comparative Disproportionality Analysis with Triptans

Abstract

Migraine constitutes the world’s second-leading cause of disability. Triptans, as serotonin 5-HT_1B/1D receptor agonists, remain the first-line treatment, despite discouraged use in individuals at high cardiovascular risk. Lasmiditan, a selective lipophilic 5-HT_1F agonist without vasoconstrictive effects, is an emerging option. We aimed to investigate the safety profile of lasmiditan in the WHO pharmacovigilance database (VigiBase^®) using a comparative disproportionality analysis with triptans. VigiBase^® was queried for all reports involving lasmiditan and triptans. Disproportionality analyses relied on the calculation of the information component (IC), for which 95% confidence interval (CI) lower bound positivity was required for signal detection. We obtained 826 reports involving lasmiditan. Overall, 10 adverse drug reaction classes were disproportionately reported with triptans, while only neurological (IC 1.6; 95% CI 1.5–1.7) and psychiatric (IC 1.5; 95% CI 1.3–1.7) disorders were disproportionately reported with lasmiditan. Sedation, serotonin syndrome, euphoric mood, and autoscopy had the strongest signals. When compared with triptans, 19 out of 22 neuropsychiatric signals persisted. The results of our analysis provide a more precise semiology of the neuropsychiatric effects of lasmiditan, with symptoms such as autoscopy and panic attacks. The cardiovascular adverse drug reaction risk with triptans was confirmed. In contrast, caution is warranted with lasmiditan use in patients with neurological or psychiatric comorbidities or serotonin syndrome risk. Our study was hindered by pharmacovigilance flaws, and further studies should help in validating these results. Our findings suggest that lasmiditan is a safe alternative for migraine treatment, especially when the neuropsychiatric risk is outweighed by the cardiovascular burden.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13311-023-01404-1.

Introduction

Migraine is a primary headache disorder that directly affects more than one billion people worldwide and, according to the Global Burden of Diseases 2019, remains the second most common cause of disability worldwide [1]. Migraine manifests itself as recurrent headache attacks with a variety of accompanying symptoms [2]. The pain of migraine attack is widely accepted to be the result of trigeminovascular system activation, which is characterized by the stimulation of nociceptive neurons that innervate the dura mater and the release of proinflammatory and vasoactive neuropeptides, such as calcitonin gene-related peptide (CGRP) [3–5]. Triptans, which were first used in the 1990s, remain the first-line acute migraine treatment [6]. Triptans are serotonin agonists that act primarily on 5-HT_1B/1D receptors, leading to vasoconstriction and thus alleviating pain, which is related to vasodilation. In addition, triptans inhibit the release of neuropeptides by trigeminal nerves and nociceptive neurotransmission [7]. Even if triptans constitute a major clinical advance in the treatment of migraine, their use is associated with inadequate efficacy and/or tolerability in approximately 30–40% of affected people [8]. In addition, because of their vasoconstrictive effect through vascular -HT_1B/1D receptor activation, triptan use is limited in elderly individuals and those at high cerebrovascular and cardiovascular risk [9].

These unmet needs led to the development of new acute migraine treatments [10]. Among these emerging treatments, lasmiditan is a highly selective lipophilic 5-HT_1F agonist [11]. This first-in-class ‘ditan’ compound, devoid of significant 5-HT_1B/1D activity, can inhibit the release of neuropeptides from trigeminal afferent nerves [5] without vasoconstrictive effects [12]. The high lipophilicity of lasmiditan allows it to cross the blood‒brain barrier, as demonstrated by preclinical in vitro and in vivo data [11]. This central nervous system (CNS) distribution indicates that lasmiditan may alleviate migraine attack through CNS mechanisms, possibly in combination with peripheral nervous system effects. Accordingly, the most common adverse drug reactions (ADRs) related to lasmiditan gathered from pivotal studies were CNS-mediated ADRs [13]. This safety profile, in addition to a significant impairment of driving ability demonstrated in two driving simulation studies [14], has led the Food and Drug Administration (FDA) [15, 16] and the European Medicines Agency (EMA) [17, 18] to approve lasmiditan with special warnings, that is, that ‘patients should be cautioned not to drive or engage in other activities requiring increased attention for at least eight hours after each dose of lasmiditan, even if they feel well enough to do so’ and that ‘patients who cannot follow this warning should not take lasmiditan’.

As lasmiditan is becoming a valuable option for the acute treatment of migraine, our aim was to evaluate how its safety profile is reflected in a real-world setting by analyzing the WHO pharmacovigilance database (VigiBase^®, Uppsala Monitoring Centre, Sweden). Specifically, we aimed to identify the specific neurological and psychiatric ADRs involving lasmiditan, supplementing our approach with a comparative disproportionality analysis with triptans.

Materials and Methods

Data Source

The Uppsala Monitoring Centre (UMC) is mandated by the WHO to oversee drug safety [19], collect evidence on ADRs, and eventually identify safety signals [20]. In this context, since 1967, the WHO safety database VigiBase^® has collected Individual Case Safety Reports issued by more than 172 national pharmacovigilance network members as well as pharmaceutical companies. VigiBase^® ensures the preservation of the anonymity of both patients and notifiers [21]. According to French clinical research law, a review from an ethics committee is not required for such observational studies. As all data from VigiBase^® were deidentified, patient informed consent was not necessary.

Each Individual Case Safety Report contains the sociodemographic characteristics of the patient (age, sex), administrative information (country, reporter qualification), suspected drug (indication, start and cessation dates, dose), other drug(s) administered (suspect of interacting or concomitant), and characteristics of the ADR(s) (reaction(s), seriousness, onset, outcome). In pharmacovigilance, an ADR is considered serious if hospitalization or its prolongation was required; if it caused a congenital malformation, resulted in persistent or significant disability or incapacity; if it was life-threatening; if it resulted in death; or if significant medical intervention was required to prevent one of these outcomes [22, 23].

Experimental Design and Statistical Analyses

Query

First, we queried VigiBase^® for all reports involving lasmiditan and all triptans recorded in the database (almotriptan, donitriptan, eletriptan, frovatriptan, naratriptan, rizatriptan, sumatriptan, zolmitriptan) between 14 November 1967 (first reports in VigiBase^®) and 28 November 2022.

According to the Medical Dictionary for Regulatory Activities (MedDRA, version 25.1 [24]), a Preferred Term (PT) is a distinct descriptor of a symptom, sign, disease diagnosis, therapeutic indication, investigation, surgical or medical procedure, and characteristic of the social or family history of the patient. At a higher level, PTs tend to be grouped into System Organ Classes (SOCs), which are groupings by etiology, manifestation site, or purpose [24].

In this analysis, queried reports for lasmiditan and triptans (as a whole) were classified using their SOCs and sorted by PTs if necessary (SOCs ‘Nervous system disorders’ and ‘Psychiatric disorders’, most reported PTs).

Statistical Analyses

Qualitative variables were described using proportions. Statistical analyses were performed with GraphPad Prism version 8.0.2.

Disproportionality Analysis

In pharmacovigilance, disproportionality analyses are aimed at detecting and assessing signals related to drug safety [25]. The superiority of the proportion of reports with a specific ADR and a given drug (cases) over the proportion of reports with the same ADR and other drugs (noncases) allows one to suggest an association between this drug and this ADR. The Information Component (IC) is a tool validated by UMC [21, 26], providing a comparison between observed and expected values for a drug-ADR combination to check for a potential association. Favoring the reduction in the risk of false-positive signals, especially in the case of a low expected ADR frequency in the database, this tool allows more specific detection of potential signals than other measures, such as the reporting odds ratio [27]. The positivity of the IC reflects that the number of observed reports is higher than expected. In this context, the IC025, which is the lower boundary of the 95% confidence interval (CI) of the IC, is required by UMC to statistically confirm the detection of a signal in VigiBase^® [21, 26].

In this disproportionality analysis, potential drug–ADR associations were selected using IC025, calculated with Microsoft^® Excel^® 2019 Version 2210.

First, a disproportionality analysis was performed for each SOC involving lasmiditan and each SOC involving triptans to define global safety profiles. Second, another disproportionality analysis was performed among psychiatric disorders and nervous system disorders SOCs for lasmiditan to assess the potential signal involved with any PT concerned. This analysis allowed us to detect whether a given psychiatric or neurological PT was differentially reported with lasmiditan when compared to all other combinations of ADRs and active ingredients in VigiBase^®.

Comparative Disproportionality Analysis

Previous findings support the existence of an additional risk of psychiatric comorbidities in patients with migraine [28–30]. Furthermore, there is growing evidence to suggest a bidirectional relationship between migraine and numerous psychiatric disorders [31–33]. In addition, migraine (even without aura) may present with a wide range of neurological symptoms [34, 35]. To mitigate these potential confounding factors and to increase the specificity of our findings, we sought to determine whether lasmiditan was still disproportionately reported with the PTs detected beforehand when compared to triptans but also to eletriptan alone. Indeed, triptans, as 5-HT1_B/D agonists, are also associated with psychiatric and neurological symptoms [36, 37]. Among triptans, eletriptan, due to its high lipophilicity, appears to readily cross the blood‒brain barrier [38, 39]. This sensitivity analysis relied on the IC and its 95% CI.

Results

Characteristics of the Reports

As of 28 November 2022, VigiBase^® included 826 reports on lasmiditan and 47,433 reports on triptans, with a predominance of female sex (lasmiditan: 77.2%; triptans: 77.9%) and the United States of America as the main origin (lasmiditan: 92.6%; triptans: 64.3%). Details regarding the characteristics of the reports are displayed in Table 1.

Table 1.

Characteristics of the patients reports

Characteristics	Number of reports (%)
	Lasmiditan	Triptans
Total	826 (100)	47,433 (100)
Sex
Female	595 (72.2)	36,973 (77.9)
Male	99 (12.0)	7,033 (14.8)
Unknown	131 (15.9)	3,427 (7.2)
Age
0–27 days		47 (0.1)
28 days—23 months		36 (0.1)
2–11 years		109 (0.2)
12–17 years		677 (1.4)
18–44 years	119 (14.4)	14,198 (29.9)
45–64 years	91 (11.0)	12,677 (26.7)
65–74 years	16 (1.9)	1,532 (3.2)
≥ 75 years	5 (0.6)	371 (0.8)
Unknown	595 (72.0)	17,786 (37.5)
Country
Region of the Americas	765 (92.6)	31,834 (67.1)
Eastern Mediterranean Region	60 (7.3)	157 (0.3)
European Region	1 (0.1)	11,561 (24.3)
South‒East Asia and Western Pacific Region		3,881 (8.1)
Reporter qualification
Healthcare professional	595 (72.2)	18,393 (38.8)
Physician	324 (39.2)	11,276 (23.8)
Pharmacist	5 (0.6)	4,444 (9.4)
Other healthcare professional	266 (32.4)	2,673 (5.6)
Others	268 (32.4)	20,644 (43.4)
Lawyer		45 (0.1)
Consumer	268 (32.4)	20,599 (43.4)
Unknown		9,095 (19,2)

Lasmiditan

As shown in Table 2, the SOCs ‘Nervous system disorders’, ‘General disorders and administration site conditions’, and ‘Psychiatric disorders’ accounted for 494 (59.8%), 321 (38.9%) and 176 (21.3%) of the reports, respectively. Overall, the main reported PTs were dizziness (221, 26.8%), feeling abnormal (107, 13.0%), and somnolence (96, 11.6%). When available (502 reports, 60.8%), more than three-quarters of the patients (399, 79.5%) were taking the middle range dose of 100 mg of lasmiditan, while 55 (10.9%) were taking 50 mg and 15 (2.9%) were taking 200 mg. One case of prescribed overdosage that caused headache and somnolence was reported.

Table 2.

Distribution of adverse drug reactions SOCs ascribed to lasmiditan

SOC	Number of reports (%)	IC [95% CI]
Nervous system disorders	494 (59.8)	1.6 [1.5; 1.7]
General disorders and administration site conditions	321 (38.9)	0.2 [0.0; 0.31]
Psychiatric disorders	176 (21.3)	1.5 [1.3; 1.7]
Gastrointestinal disorders	83 (10.0)	-0.8 [-1.1; -0.5]
Musculoskeletal and connective tissue disorders	40 (4.8)	-1.0 [-1.5; -0.6]
Injury, poisoning and procedural complications	28 (3.4)	-1.7 [-2.3; -1.2]
Investigations	27 (3.3)	-1.6 [-2.2; -1.1]
Eye disorders	22 (2.7)	-0.9 [-1.5; -0.3]
Cardiac disorders	20 (2.4)	-0.5 [-1.2; 0.1]
Skin and subcutaneous tissue disorders	17 (2.1)	-3.0 [-3.8; -2.4]
Respiratory, thoracic and mediastinal disorders	13 (1.6)	-2.4 [-3.3; -4.0]
Ear and labyrinth disorders	11 (1.3)	0.0 [-1.0; -0.7]
Vascular disorders	7 (0.8)	-2.5 [-3.8; -1.6]
Infections and infestations	6 (0.7)	-3.4 [-4.7; -2.4]
Immune system disorders	5 (0.6)	-2.1 [-3.7; -1.1]
Renal and urinary disorders	2 (0.2)	-3.1 [-5.7; -1.8]
Reproductive system and breast disorders	2 (0.2)	-3.0 [-5.6; -1.7]
Social circumstances	2 (0.2)	-1.3 [-3.9; -0.0]
Blood and lymphatic system disorders	1 (0.1)	-4.5 [-8.3; -2.8]
Endocrine disorders	1 (0.1)	-1.3 [-5.1; 0.34]
Metabolism and nutrition disorders	1 (0.1)	-4.4 [-8.2; -2.7]
Neoplasms—benign, malignant and unspecified (including cysts and polyps)	1 (0.1)	-4.0 [-7.8; -2.4]

Aside from lasmiditan, coreported suspected or interacting drugs were highlighted in 49 reports (5.9%). The most frequently suspected or interacting active ingredients were other antimigraine agents, such as the CGRP-targeting drugs galcanezumab (29, 3.5%), rimegepant (2, 0.2%), and ubrogepant (2, 0.2%).

In addition, 63 (7.6%) cases were deemed serious, including 44 medically important conditions (5.3%), 10 deaths (1.2%), 9 caused/prolonged hospitalizations (1.1%), and 5 disabilities/incapacities (0.6%). When reports were available (294, 35.6%), 258 (87.8%) patients recovered or were recovering, and 27 (9.1%) were not recovering. The detailed distributions of PTs belonging to psychiatric disorders and nervous system disorders for lasmiditan reports are available in Tables S1 and S2, respectively.

Triptans

In decreasing order of importance, the most represented SOCs were ‘General disorders and administration site conditions’ (22,244, 46.9%), ‘Nervous system disorders’ (14,628, 30.8%), and ‘Injury, poisoning and procedural complications’ (8,428, 17.8%) (Table 3). The most frequently reported terms (PTs) were drug ineffective (6,825, 14.4%), product dose omission issue (3,096, 6.5%), and nausea (3,032, 6.4%). Sumatriptan was the most frequently reported triptan (32,220; 67.9%), followed by eletriptan (3,499, 7.4%), zolmitriptan (3,391, 7.1%), rizatriptan (2,818; 5.9%), naratriptan (939; 2.0%), frovatriptan (528, 1.1%), and almotriptan (502; 1.1%). In total, 117 cases of overdosage were retrieved, including 24 (20.5%) intentional overdosages, 12 (10.3%) accidental overdosages, and 3 (2.6%) prescribed overdosages.

Table 3.

Distribution of adverse drug reactions SOCs ascribed to triptans

SOC	Number of reports (%)	IC [95% CI]
General disorders and administration site conditions	22,244 (46.9)	0.4 [0.4; 0.5]
Nervous system disorders	14,628 (30.8)	0.7 [0.6; 0.7]
Injury, poisoning and procedural complications	8,428 (17.8)	0.6 [0.6; 0.7]
Product issues	8,228 (17.3)	3.0 [3.0; 3.0]
Gastrointestinal disorders	6,976 (14.7)	-0.3 [-0.3; -0.2]
Respiratory, thoracic and mediastinal disorders	3,996 (8.4)	-0.0 [-0.1; 0.1]
Psychiatric disorders	3,617 (7.6)	0.1 [0.0; 0.1]
Cardiac disorders	3,490 (7.4)	1.1 [1.1; 1.2]
Musculoskeletal and connective tissue disorders	3,375 (7.1)	-0.5 [-0.5; -0.4]
Skin and subcutaneous tissue disorders	3,267 (6.9)	-1.3 [-1.4; -1.3]
Vascular disorders	2,682 (5.7)	0.2 [0.1; 0.2]
Investigations	1,882 (4.0)	-1.3 [-1.4; -1.3]
Immune system disorders	1,698 (3.6)	0.3 [0.3; 0.4]
Eye disorders	1,426 (3.0)	-0.7 [-0.8; -0.6]
Infections and infestations	952 (2.0)	-2.0 [-2.1; -1.9]
Ear and labyrinth disorders	692 (1.5)	0.1 [0.0; 0.2]
Renal and urinary disorders	582 (1.2)	-1.1 [-1.2; -1.0]
Metabolism and nutrition disorders	447 (0.9)	-2.0 [-2.1; -1.8]
Pregnancy, puerperium and perinatal conditions	377 (0.8)	0.6 [0.4; 0.7]
Social circumstances	362 (0.8)	0.1 [-0.0; 0.3]
Reproductive system and breast disorders	355 (0.7)	-1.7 [-1.8; -1.5]
Surgical and medical procedures	331 (0.7)	-1.3 [-1.5; -1.1]
Blood and lymphatic system disorders	260 (0.5)	-2.9 [-3.0; -2.7]
Neoplasms—benign, malignant and unspecified (including cysts and polyps)	236 (0.5)	-2.6 [-2.7; -2.4]
Hepatobiliary disorders	207 (0.4)	-1.8 [-2.0; -1.6]
Congenital, familial and genetic disorders	198 (0.4)	0.5 [0.3; 0.7]
Endocrine disorders	101 (0.2)	-0.9 [-1.2; 0.6]

Suspect or interacting drugs were coreported in 18,135 cases (38.2%). With 515 (1.1%), 388 (0.8%), and 316 (0.7%) reports, respectively, topiramate, ibuprofen and gabapentin were the leading coreported drugs.

When reports were available (34,039, 71.8%), 9,176 (27.0%) reports were considered serious, with 6,191 medically important conditions (13.1%), 2,919 caused/prolonged hospitalizations (6.2%), 722 deaths (1.5%), 594 life-threatening reactions (1.3%), 538 disabilities/incapacities (1.1%), and 124 congenital anomalies/birth defects (0.3%). Death causes were led by suicide (105, 14.5%), followed by cardiac disorders (78, 10.8%). When reports were available (19,194, 40.5%), 14,774 (77.0%) patients recovered or were recovering, 3,434 (17.9%) were not recovering, and 427 (2.2%) recovered with sequelae.

Disproportionality Analysis

System Organ Classes (SOCs)

The SOCs ‘Nervous system disorders’ (IC 1.6; 95% CI 1.5–1.7) and ‘Psychiatric disorders’ (IC 1.5; 95% CI 1.3–1.7) were reported disproportionately with lasmiditan (Table 2). In contrast, ten SOCs were reported disproportionately with triptans as a whole, among which ‘Product issues’ (IC 3.0; 95% CI 3.0–3.0) and ‘Cardiac disorders’ (IC 1.1; 95% CI 1.1–1.2) led in terms of signal strength, as shown in Table 3. The SOC ‘Nervous system disorders’ was also subject to disproportionate reporting with triptans, but to a lesser extent (IC 0.7; 95% CI 0.6–0.7). The IC025 of the SOC ‘Psychiatric disorders’ did not reach statistical significance (IC 0.1; 95% CI 0.0–0.1].

Preferred Terms (PTs)

We aimed to shed some light on the potential neurological and psychiatric signals regarding lasmiditan. Ten psychiatric PTs were disproportionately reported with lasmiditan, among which euphoric mood (IC 5.1; 95% CI 4.5–5.6), autoscopy (i.e., the experience of perceiving one's body and environment as if disembodied from a distanced perspective (IC 4.1%, 95% CI 2.9–4.9)) and hallucination (IC 3.9%, 95% CI 3.4–4.3) showed the strongest signals (Table 4). Sedation (IC 5.8; 95% CI 5.5–6.1), serotonin syndrome (IC 5.0; 95% CI 4.4–5.5), and somnolence (IC 3.4; 95% CI 3.0–3.6) were the strongest signals of the 12 neurological PTs reported disproportionately with lasmiditan (Table 4). Then, we sought to identify potential disproportionate reporting of these twenty-two PTs with triptans as a whole to assess the specificity of our neurological and psychiatric signals involving lasmiditan (Table 5).

Table 4.

Psychiatric and neurological PTs disproportionately reported with lasmiditan

Psychiatric Preferred Terms
Preferred Term	Number of reports (%)	IC [95% CI]
Euphoric mood	27 (15.3)	5.1 [4.5; 5.6]
Hallucination	42 (23.9)	3.9 [3.4; 4.3]
Autoscopy	8 (4.5)	4.1 [2.9; 4.9]
Panic attack	12 (6.8)	3.2 [2.3; 3.9]
Abnormal dreams	11 (6.3)	3.1 [2.1; 3.8]
Visual hallucination	8 (4.5)	3.1 [1.9; 3.9]
Disorientation	8 (4.5)	2.4 [1.2; 3.2]
Nightmare	6 (3.4)	2.0 [0.6; 2.9]
Insomnia	19 (10.8)	1.0 [0.3; 1.6]
Anxiety	17 (9.7)	1.0 [0.3; 1.7]

Neurological Preferred Terms
Preferred Term	Number of reports (%)	IC [95% CI]
Sedation	78 (15.8)	5.8 [5.5; -6.1]
Serotonin syndrome	25 (5.1)	5.0 [4.4; 5.5]
Somnolence	96 (19.4)	3.4 [3.0; 3.6]
Migraine	32 (6.5)	3.1 [2.5; 3.6]
Dizziness	221 (44.7)	2.7 [2.5; 2.9]
Hypersomnia	12 (2.4)	3.3 [2.3; 4.0]
Paraesthesia	52 (10.5)	2.4 [2.0; 2.8]
Hypoesthesia	24 (4.8)	1.7 [1.0; 2.2]
Balance disorder	10 (2.0)	1.9 [0.9; 2.7]
Restless leg syndrome	4 (0.8)	2.2 [0.5; 3.3]
Dysarthria	5 (1.0)	1.8 [0.3; 2.8]
Tremor	16 (3.2)	1.0 [0.2; 1.6]

Table 5.

Disproportionality analysis of psychiatric and neurological PTs reported with triptans (among those disproportionately reported with lasmiditan)

Psychiatric Preferred Terms
Preferred Term	Number of reports (%)	IC [95% CI]
Euphoric mood	86 (2.3)	2.2 [1.9; 2.5]
Euphoric mood	86 (2.3)	2.2 [1.9; 2.5]
Hallucination	135 (3.7)	-0.0 [-0.3; 0.2]
Autoscopy	2 (0.1)	1.0 [-1.6; 2.4]
Panic attack	78 (2.2)	0.6 [0.3; 0.9]
Abnormal dreams	37 (1.0)	-0.4 [-0.9; 0.0]
Visual hallucination	34 (0.9)	0.3 [-0.2; 0.7]
Disorientation	105 (2.9)	0.7 [0.4; 1.0]
Nightmare	67 (1.8)	-0.4 [0.0; 0.3]
Insomnia	249 (6.9)	-1.0 [-1.3; -0.9]
Anxiety	569 (15.7)	0.3 [0.2; 0.4]

Neurological Preferred Terms
Preferred Term	Number of reports (%)	IC [95% CI]
Sedation	149 (1.0)	1.5 [1.3; 1.7]
Serotonin syndrome	227 (1.6)	3.6 [3.7; 3.9]
Somnolence	866 (5.9)	0.8 [0.7; 0.9]
Migraine	3,011 (20.6)	4.0 [3.9; 4.0]
Dizziness	2,059 (14.1)	0.1 [0.0; 0.2]
Hypersomnia	21 (0.1)	-1.1 [-1.8; -0.6]
Paraesthesia	1,877 (12.8)	1.8 [1.8; 1.9]
Hypoesthesia	668 (4.6)	0.6 [0.7; 0.8]
Balance disorder	96 (0.7)	-0.5 [-0.8; -0.2]
Restless leg syndrome	19 (0.1)	-0.5 [-1.2; 0.1]
Dysarthria	120 (0.8)	1.0 [0.7; 1.3]
Tremor	471 (3.2)	[-0.1; 0.2]

Four out of ten psychiatric PTs were also disproportionately reported with triptans, whereas seven out of twelve neurological PTs were subject to disproportionate reporting (Table 5).

To mitigate migraine-related confounders, we performed a comparative disproportionality analysis for lasmiditan and these neurological and psychiatric PTs using, as comparators, triptans as a group and eletriptan alone. Anxiety was the sole nondisproportionately reported psychiatric PT (IC 0.7; 95% CI -0.0; 1.3). Likewise, regarding the twelve neurological PTs, only migraine (IC -0.7; 95% CI -1.2; -0.2) and dysarthria (IC 1.1; 95% CI -0.5; 2.1) were nondisproportionate. When lasmiditan was compared with eletriptan, fewer psychiatric and neurological signals persisted, and if so, less importantly.

The comparative ICs for lasmiditan and each neurological and psychiatric PT are shown in Table 6.

Table 6.

Comparative disproportionality analysis of psychiatric and neurological PTs reported with lasmiditan, using triptans and eletriptan as comparators

Psychiatric Preferred Terms
Preferred Term	IC [95% CI]
	Triptans	Eletriptan
Euphoric mood	3.5 [2.9; 4.0]	1.9 [1.3; 2.4]
Hallucination	3.6 [3.1; 4.0]	2.0 [1.5; 2.4]
Autoscopy	3.7 [2.5; 4.5]	2.1 [0.9; 2.9]
Panic attack	2.6 [1.7; 3.3]	1.3 [0.4; 2.0]
Abnormal dreams	3.1 [2.1; 3.9]	1.8 [0.8; 2.5]
Visual hallucination	2.8 [1.6; 3.6]	1.0 [-0.2; 1.8]
Disorientation	1.8 [0.6; 2.6]	0.9 [-0.3; 1.7]
Nightmare	1.9 [0.5; 2.8]	0.9 [-0.5; 1.8]
Insomnia	1.9 [1.2; 2.5]	0.6 [-0.1; 1.2]
Anxiety	0.7 [-0.0; 1.3]	0.3 [-0.5; 0.9]

Neurological Preferred Terms
Preferred Term	IC [95% CI]
	Triptans	Eletriptan
Sedation	4.2 [3.8; 4.5]	2.3 [1.9; 2.6]
Serotonin syndrome	2.4 [1.8; 2.9]	1.3 [0.7; 1.9]
Somnolence	2.5 [2.2; 2.8]	1.2 [0.9; 1.5]
Migraine	-0.7 [-1.2; -0.2]	-1.7 [-2.2; -1.2]
Dizziness	2.5 [2.3; 2.7]	1.6 [1.4; 1.8]
Hypersomnia	3.6 [2.6; 4.3]	1.7 [0.8; 2.4]
Paraesthesia	0.7 [0.2; 1.0]	1.0 [0.6; 1.4]
Hypoesthesia	1.0 [0.4; 1.5]	0.6 [-0.1; 1.1]
Balance disorder	2.2 [1.2; 2.9]	1.0 [-0.0; 1.8]
Restless leg syndrome	2.3 [0.6; 3.4]	1.6 [-0.1; 2.7]
Dysarthria	1.1 [-0.5; 2.1]	0.6 [-1.0; 1.6]
Tremor	0.9 [0.1; 1.5]	0.8 [0.0; 1.5]

Discussion

Our analysis of the WHO pharmacovigilance database delved into the neuropsychiatric safety profile of lasmiditan, and the results provide a more precise semiology, with symptoms such as autoscopy and panic attacks. The results of relative frequency analysis, disproportionality analysis, and comparative disproportionality analysis all were in agreement and differentiated the ‘real-life’ ADR spectrum of lasmiditan from that of triptans. The safety profile of triptans showed a broad spectrum of affected systems, with a predominance of cardiovascular ADRs and a lower neuropsychiatric burden. In contrast, lasmiditan, considered almost devoid of cardiovascular risks, exhibited a variety of neuropsychiatric ADRs. The disproportionate reporting of neuropsychiatric ADRs with lasmiditan stood out even when lasmiditan was compared to triptans, as seen for the serotonin syndrome signal, which was stronger with lasmiditan.

Lasmiditan

Regarding lasmiditan, the elective disproportionality signals for ‘Nervous system disorders’ and ‘Psychiatric disorders’ confirmed its known predominance of CNS effects [11, 40–42]. At a dose of 100 mg (the most frequent dose in the database), lasmiditan is considered to exhibit a low abuse potential [43]. This, together with the fact that it has recently been granted marketing authorization, warranting a certain degree of caution by prescribers and patients, may explain the single overdosage report that was retrieved in our study. Drug abuse for recreational use is also often fueled by the reputation of a drug having a potential psychoactive effect, which, at the moment, is lacking for lasmiditan [44].

In line with data from randomized clinical trials (RCTs), dizziness, paraesthesia, balance disorder, and somnolence were frequently ascribed to lasmiditan [45]. Potential activation of 5-HT1_F receptors, located in the cerebellum and the lateral vestibular nucleus, might trigger such manifestations [46]. The occurrence of these ADRs may be increased by the CNS tropism of lasmiditan.

Likewise, 5-HT1_F receptor-specific binding sites have been detected in the frontal, parietal, temporal, and occipital cortex of the human brain [47]. As a highly selective 5-HT1_F agonist [48], lasmiditan may have substantial effects on these areas involved in multiple psychiatric symptoms, such as hallucinations and anxiety [49, 50]. However, the drug label of lasmiditan classifies vertigo, lethargy, anxiety, abnormal dreams, euphoric mood, and hallucinations as ‘less common adverse reactions’ [16]. While considered relatively infrequent in RCTs [13, 43, 51], psychiatric ADRs accounted for more than one-fifth of lasmiditan-related reports in our study. The strengths of the signals should prompt caution and therefore justify further assessment of the prevalence of these ADRs in a real-world setting. Furthermore, the semiological peculiarities of hallucinatory and anxious manifestations were detailed in our study through the identification of specific signals that involve visual hallucinations [16], autoscopy [52], and panic attacks, which persisted when compared with triptans.

Triptans

Regarding triptans, numerous SOCs were reported disproportionately, among which ‘Product issues’ were predominant, consistent with the complexity of pharmaceutical forms for this class (autoinjectors, jet injectors, transdermal patches [53]). As expected, the following greatest signals corresponded to cardiac and vascular disorders [9, 54–56]. Although exhibiting numerous benign CNS-related ADRs, as reported on drug labels [37], some of them (such as dizziness or tremor) did not reach statistical significance for disproportionality. Likewise, the SOC ‘Psychiatric disorders’ as a whole did not show a disproportionality signal. In that respect, the inconsistent ability of triptans to cross the blood‒brain barrier may reflect their predominantly peripheral safety profile [11, 57].

Comparative Analysis

Out of twenty-two neurological and psychiatric PTs disproportionality reported with lasmiditan, eleven were also disproportionately reported with triptans. Only anxiety, migraine, and dysarthria were not disproportionately reported for lasmiditan when triptans were used as a comparator. Cases of dysarthria were possibly confounded by the fact that migraine with brainstem aura [35, 58] may display such neurological symptoms. Similarly, the reporting of migraine as an ADR might result from confusion with the indication or inefficiency of the drug. An additional hypothesis is medication overuse headache, as it may arise from the overuse of any acute migraine medications [51]. Last, anxiety might be a comorbidity, a trigger, a risk factor, or a nonresponse factor for migraine [28–31, 33]. However, depression also has a bidirectional relationship with migraine [32, 33], so antidepressants could be coprescribed. Although both lasmiditan and triptans are subject to a ‘Warning and Precautions’ section dedicated to serotonin syndrome on their drug labels [16, 37], the serotonin syndrome signal was more prominent for lasmiditan than for all triptans but also for eletriptan alone in our comparative analysis. Indeed, while sumatriptan (the most reported triptan in VigiBase^®) does not display a strong ability to cross the blood‒brain barrier [59], eletriptan is considered to exhibit the greatest lipophilicity and therefore the strongest blood‒brain barrier crossing ability among triptans [39]. However, the important P-glycoprotein affinity of eletriptan may limit its central bioavailability [38]. We hypothesize that the more a migraine treatment drug crosses the blood‒brain barrier, the more disproportionate the burden of neurological and psychiatric adverse events.

Strengths and Limitations

While other studies assessed the overall safety profiles of lasmiditan and triptans [36, 43, 45, 58, 60], our exploratory analysis focused, respectively and comparatively, on their neurological and psychiatric burden. Our study findings highlighted the neuropsychiatric safety profile of lasmiditan and confirmed the cardiovascular profile of triptans.

Our disproportionality analysis of the WHO safety database benefits from the strengths of the analysis of postmarketing surveillance data, reflecting lasmiditan and triptan safety in a ‘real-world’ setting. Hence, we had the opportunity to analyze a large dataset of pharmacovigilance reports beyond the close monitoring that characterizes RCTs.

Nevertheless, the inherent flaws of spontaneous reporting systems and pharmacovigilance approaches may hinder our results, such as the lack of follow-up, underreporting, and incomplete data [61]. Coding heterogeneity might be a prevalent limitation regarding psychiatric ADRs, the understanding of which is always altered by subjectivity, therefore raising the need for a qualitative assessment of these symptoms. The scarcity of clinical data contained in the reports did not allow us to differentiate new-onset from preexisting psychiatric or neurological symptoms, which are closely related to migraine [28, 29, 31, 33], although our reliance on reasonably similar populations (between lasmiditan and triptans) might limit the impact of this bias. Healthcare professionals accounted for more than three-quarters of reports involving lasmiditan but only less than half of the reports regarding triptans, which could have biased reporting but also may have reduced the risk of coding errors in lasmiditan reports. However, the Weber effect, which implies a rise in the reporting of ADRs for a given drug during its first years of marketing [62, 63], might have induced overreporting, artificially increasing the strength of lasmiditan-related signals. Thus, the possibility that some reports were wrongly attributed to an iatrogenic trigger cannot be ruled out, particularly for symptoms that belong to migraine syndrome. Indeed, as a therapeutic option for migraine in patients with resistance or a contraindication to triptans [64], the choice of lasmiditan might have led to a substantial population bias (with an increased risk of neuropsychiatric disorders due to comorbidities). Therefore, head-to-head comparisons between drug classes should not be used as a basis for extrapolation. Last, pharmacoepidemiological studies are exploratory, aiming to raise awareness about potential drug safety signals. In that respect, no definite conclusion regarding causality can be drawn from our findings.

In this study, relying on a comprehensive analysis of the WHO pharmacovigilance database, we highlight neurological and psychiatric safety signals associated with lasmiditan. Further studies should assess their burden and deepen the qualitative analysis of these manifestations. Our findings are supported by a comparison with the safety profile of triptans, whose disproportionality analysis underlined the need for caution regarding the risk of cardiovascular events. The lack of signal regarding cardiovascular ADRs with lasmiditan is in agreement with its CNS tropism. However, caution is warranted when using lasmiditan in patients with neurological or psychiatric comorbidities or at risk of serotonin syndrome. In contrast, our study findings strengthen the arguments in favor of the legitimacy of lasmiditan as a safer alternative for the treatment of migraine than triptans in terms of cardiovascular risk, especially in patients for whom the neuropsychiatric risk is outweighed by the cardiovascular burden. Population-based studies in real-life settings should help reevaluate these outcomes. Indeed, the pharmacokinetic and pharmacodynamic profiles of these migraine treatments may underlie their respective safety profiles, paving the way to the tailored management of patients with migraine.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors' Contributions

DM, AOG, EKVO, AD, ML-M, and MDD wrote the manuscript. DM, AOG, ML-M, and MDD designed the research. DM, AOG, and EKVO performed the research. DM, AOG, and AD analyzed the data. All authors read and approved the final manuscript.

Availability of Data and Materials

The data that support the findings of this study are available from Uppsala Monitoring Center (UMC), but restrictions apply to the availability of these data, which were used under license for the current study and are not publicly available. Access to VigiBase^® is available without fees to Dr Fanny Rocher. Data are, however, available from the authors upon reasonable request and with the permission of UMC.

Declarations

Ethics Approval and Consent to Participate

Ethics committee approval was not required for this observational study because the analysis was carried out on anonymized data from a pharmacovigilance database.

Conflicts of Interest

ML-M reports personal fees for advisory boards, speaker panels or investigation studies from Allergan, Amgen, Astellas, ATI, BMS, Boehringer, Boston Scientific, CoLucid, Convergence, GlaxoSmithKline, Grunenthal, IPSEN, Lundbeck, Lilly, Medtronic, Menarini, MSD, Novartis, Pfizer, Reckitt Benckiser, Saint-Jude, Sanofi-Aventis, Teva, UCB, UPSA, and Zambon. All other authors declare no competing financial interest.