Comparative sensitivity of laboratory animals used for preclinical convulsion risk assessment to drug-induced convulsion

Abstract

Drug-induced convulsion is a serious concern in drug development, such that the convulsion liability of drug candidates must be evaluated in preclinical safety studies. However, information on the differences among species regarding their sensitivity to convulsions induced by convulsant drugs in humans remains limited. Here, we selected 11 test articles from several pharmacological classes and compared the sensitivities of three types of laboratory animal to convulsion. All 11 test articles were examined in mice via intraperitoneal injection and in rats via intravenous bolus; and 6 of the 11 test articles, selected mainly based on availabilities of data on drug plasma concentrations in humans at convulsion, were examined in non-human primates (NHPs) via intravenous infusion. Plasma concentrations of the test articles shortly after convulsion onset or 5 min after administration were measured. All 11 articles tested in mice, 10 of 11 articles tested in rats, and all 6 articles tested in NHPs induced convulsion with premonitory signs. Although there was a general tendency that rats and NHPs exhibited convulsions at lower plasma drug concentrations than did mice, the plasma concentrations at convulsion onset were generally comparable, within 3-fold differences, across the animal species. We conclude that the mice, rats, and NHPs examined in the present study generally showed similar sensitivities to convulsion induced by the test articles. Thus, each of these laboratory animals can be used for the assessment of convulsion risk in the early stages of drug development, depending on throughput, cost, and test article-specific requirements.

INTRODUCTION

Drug-induced convulsion is a serious concern in drug development because it can potentially lead to life-threatening adverse events in humans. Database-derived information has shown that among 390 marketed drugs reported to have induced convulsion in mice, rats, dogs, or non-human primates (NHPs) during nonclinical studies, convulsion was also reported to have occurred in humans in clinical trials of 65 of the drugs (17%) (Abt et al., 2019). Ideally, drug candidates with no risk or a wide safety margin for convulsion should be selected at the preclinical stage; however, translation of convulsion risk from laboratory animals to humans is a huge challenge because differences in the sensitivity to drug-induced convulsions among species are not fully understood (Clark and Steger-Hartmann, 2018; Abt et al., 2019).

To obtain drug candidates with low convulsion liability, which is particularly important for candidates intended for development as central nervous system-active drugs because they are intended to penetrate the blood-brain barrier, there are several in vitro and ex vivo systems available for compound screening at the early drug discovery stage, such as the convulsion panel assay (convulsion-specified binding assay to voltage-gated or ligand-gated ion channels, G-protein-coupled receptors, enzymes, or transporters; Easter et al., 2009; Roberts et al., 2021), microelectrode array in human induced pluripotent stem cell-derived neurons (Bradley et al., 2018; Kreir et al., 2018; Odawara et al., 2018; Tukker and Westerink, 2021), and hippocampal tissue slices (Accardi et al., 2018). At lead optimization or later, in vivo animal studies become more important because they provide data on drug exposure levels (peak plasma concentration [C_max]) that evoke convulsions, which can aid in setting exposure caps for clinical trials, premonitory signs of convulsions that should be monitored in clinical trials, contributions of drug metabolites produced in vivo to onset of convulsion, and tolerance to the onset of convulsion after repeated administrations. However, for emerging new drug modalities (e.g., RNA therapeutics, antibody-drug conjugates, and gene therapies), the utility of current in vitro or ex vivo screening assays has not been fully validated to detect convulsion risk, meaning in vivo animal studies remain an important means of evaluating drug-induced convulsion at the preclinical stage for these modalities.

A recent survey of 11 pharmaceutical companies from around the world examining inter-species differences in sensitivity to drug-induced convulsion found that dogs were considered the most sensitive animal species, and other species such as mice, rats, and NHPs showed overall similar sensitivity profiles to one another and less sensitivity compared with dogs (DaSilva et al., 2020). However, beagle dogs are known to be excessively sensitive to idiopathic epilepsy (Edmonds et al., 1979; Hoskins, 2000), and therefore the use of beagle dogs may mislead convulsion risk assessment. For mice, rats, and NHPs, several studies have investigated inter-species differences regarding sensitivity to convulsion liability (Easter et al., 2009; Bassett et al., 2014; Accardi et al., 2018); however, there are many drugs known to induce convulsions in humans that remain to be investigated for possible inter-species differences of sensitivity to convulsion in those animal species.

Here, we investigated the inter-species differences among three common laboratory animal species regarding their sensitivity to drug-induced convulsion using 11 test articles for mice and rats, 9 of which are reported to induce convulsions in humans. Six of the 9 test articles (convulsants in humans) were also tested for NHPs, which were selected mainly based on availabilities of data on drug plasma concentrations in humans at convulsion. The test articles were administered to the animals, and plasma concentrations at the onset of convulsion were compared across the animal species. When clinical case reports were available, plasma concentrations at the onset of convulsions in animals were compared with those during convulsion in humans.

MATERIALS AND METHODS

Ethical statement

The present study was approval by the Laboratory Animal Care and Use Committee of Eisai Co., Ltd. (Tokyo, Japan) prior to study initiation and was performed in compliance with the Laboratory Animal Policy of Eisai Co., Ltd. The facilities where the present study was conducted (Tsukuba Research Laboratories at Eisai Co., Ltd.) are accredited by the Japan Health Sciences Foundation.

Animals

DBA2 mice (male, 4 weeks old, n = 4 per dose level), Wistar Han rats (male, 9 to 11 weeks old, n = 2 or 4 per test article [dose escalation regimen]), and cynomolgus monkeys (male and female, 3 to 5 years old, n = 2 per test article [dose escalation regimen], except for Compound A which was n = 1) were used. Animal numbers were determined to be the minimum needed to detect drug-induced convulsions based on previous experience in the facility and preliminary studies (unpublished data). DBA2 mice are a genetic model of audiogenic seizures (Löscher, 2011) and our previous studies have shown that the incidence of audiogenic seizure decreases with age. Therefore, to avoid possible false-negative results, we used young mice (4-week-old) in the present study, which also did not show convulsions after negative control treatment with saline, amoxicillin (up to 1000 mg/kg), or acetaminophen (up to 600 mg/kg). Only male mice and rats were used because males are used more frequently than females in pharmacological studies and the concentrations of some neurotransmitters (e.g., gamma-aminobutyric acid [GABA] and glutamate) in various brain regions or GABA_A receptor subunit expression in the hippocampus are known to change with estrous cycle in female mice (Maguire et al., 2005) or rats (Frankfurt et al., 1984). For NHPs, however, both males and females were used due to the recent constrained supply of NHPs, i.e. limited number of males.

Preparation of test articles

The following 9 test articles, each a reported convulsant in humans, were chosen from a variety of pharmacological classes: 4-aminopyridine (4-AP; Fujifilm Wako, Japan), bupropion (Combi-Blocks, USA), diphenhydramine (Fujifilm Wako), pentylenetetrazole (PTZ; Tokyo Chemical Industry, Japan), strychnine (Fujifilm Wako), theophylline (Fujifilm Wako) (or aminophylline [theophylline/ethylenediamine 2:1 chemical mixture] for rats, Sigma-Aldrich), tiagabine (Tokyo Chemical Industry), tramadol (Sigma-Aldrich), and an Eisai proprietary compound (hereafter “Compound A”). In addition, linopirdine (Sigma-Aldrich, USA) and pilocarpine (Fujifilm Wako), which are known to induce convulsions in laboratory animals such as rodents and marmoset (Otto et al., 2004; Tetz et al., 2006; Pontes et al., 2016), were chosen as reference convulsants for mice and rat studies; these 2 compounds have not been reported to induce convulsion in humans possibly because overdose might not happen in linopirdine due to the termination of the drug development (Pieniaszek et al., 1995; Rockwood et al., 1997) and overdose may be rare with pilocarpine due to its small patient population (xerostomia) (Hendrickson et al., 2004). The mechanisms of action of the selected compounds are shown in Table 1.

Table 1. Mechanisms of action of the test articles.

The following vehicles were used to prepare solutions of each dosing formulation (see Supplementary Table 1 for details): saline for intraperitoneal (IP) administration to mice (adding hydrochloric acid [HCl] and sodium hydroxide [NaOH] for linopirdine and theophylline, respectively), saline or 5% glucose solution for intravenous (IV) administration to rats (adding 2-hydroxypropyl-β-cyclodextrin solution, HCl, and dimethyl sulfoxide & HCl, for Compound A, linopirdine, and strychnine, respectively), and saline or 5% glucose solution for IV infusion to NHPs. In the present study, no vehicle control group was set because we had no experiences of convulsion with these vehicles in the previous toxicity studies in mice, rats, and NHPs.

Experimental design

The dose levels used are described in Table 2. For the mice and rats, preliminary dose-range finding studies were conducted prior to the present study to understand the appropriate doses levels where convulsions occurred (data not shown). For the NHPs, the dose levels were determined with reference to the results from the mouse and rat studies.

Table 2. Doses used and premonitory signs of convulsion observed for each test article.

Mouse

The test articles were administered by IP injection at low, middle, and high dose levels (4 dose levels for linopirdine: additional high dose was set due to no convulsions in 3 dose levels), and then clinical signs were observed under freely moving conditions for up to 1 hr post-dose (n = 4 per dose level). Once convulsions were noted, blood was collected from the abdominal vena cava under isoflurane anesthesia.

Rat

The test articles were administered to rats according to the standard protocol for cardiovascular screening in rats used at Eisai Co., Ltd., which allowed us to assess the potential to combine screenings for cardiovascular and convulsion events. Prior to the administration day, catheters were placed in the animals: one in the jugular vein for drug administration and another in the femoral artery for blood collection and measurement of blood pressure. Electrodes were also implanted subcutaneously at the upper-right chest (negative pole) and lower-left chest (positive pole) for electrocardiogram recording (lead II). The catheters and electrodes were connected to a free-moving cannulation system (Sugiyama-gen Co., Ltd., Japan) by a tether. All signals were acquired by a Ponemah physiology software platform (ver. 4.90-SP2; Data Sciences International Inc., USA). The test articles were administered to the animals (n = 2 per group) by IV bolus following a dose escalation regimen of low and high doses with a 20-min pause between administrations. When the low dose unexpectedly induced convulsions, another cohort with lower dose levels was added (2 cohorts [total 4 rats] were used for 4-AP [3 and 10 mg/kg and 1 and 3 mg/kg], bupropion [15 and 45 mg/kg and 10 and 30 mg/kg], and PTZ [25 and 75 mg/kg and 5 and 15 mg/kg]). The bolus was administered for 1 min at 1 mL/kg/min, except for theophylline and pilocarpine (theophylline was at 2 mL/kg/min to achieve a higher C_max, and pilocarpine was at 0.5 mL/kg/min for 2 min to avoid excessive pharmacological effects in the peripheral system, such as bradycardia). The clinical signs under freely moving conditions were observed and blood was collected via the catheter in the femoral artery at 5 min post-dose. Clinical observation continued until 30-min post-dose (but sometimes continued up to 240-min post-dose, associated with capturing any prolonged/delayed effects of the test articles on electrocardiogram). An anticonvulsant drug, diazepam (Cercine Injection; Teva Takeda Yakuhin Ltd., Japan), was administered by IV injection to suppress severe convulsions, such as those involving opisthotonus, jumping, and/or ananastasia.

NHP

Due to the limited number of animals, 5 of the test articles reported to cause convulsions in humans were selected mainly based on availabilities of data on drug plasma concentrations in humans at convulsion and administered to NHPs: 4-AP, bupropion, theophylline, tiagabine, and tramadol. The test articles were administered by IV infusion (n = 2 [1 male and 1 female]) via the saphenous or brachial vein up to the onset of convulsion or a maximum of 1 hr. The animals were restrained with a monkey chair (well acclimated to the NHPs) for all of the drug administrations, except for that of tramadol for which the NHPs were freely moving in their home-cage with a tethered infusion pump, not used for 4-AP, bupropion, theophylline, and tiagabine due to a mechanical trouble. To avoid rapid onset of a moribund condition due to convulsion, the 1-hr IV infusion was performed by administration of a low-concentration formulation for the first 30 min and then a high-concentration formulation for the second 30 min. Infusion rates and target blood concentrations at the end of the 1-hr IV infusion were determined according to available in-house data in animals and publications in humans (the mouse and rat studies were conducted prior to the NHP study, and the mouse and rat data were used as references for 4-AP, theophylline, tiagabine, and tramadol and the Drug Label information on Forfivo XL^® was referred for bupropion). Clinical signs during the IV infusion were observed and blood was collected shortly after the onset of convulsion, or at completion of the 1-hr IV infusion if no convulsion occurred, without anesthesia. Diazepam was administered intramuscularly to suppress the convulsions.

Compound A was administered to an NHP by IV infusion in a previous study under comparable conditions to those of the present study (n = 1 [female], unpublished data); therefore, this result was incorporated into the present study.

Plasma pretreatment to measure drug concentration in vivo

Blood samples collected from mice, rats, and NHPs were centrifuged to obtain plasma. For pilocarpine, approximately 1% (v/v) of 500 mmol/L sodium ethylenediaminetetraacetic acid solution was added to the blood samples just after collection to prevent hydrolysis of the pilocarpine. An aliquot of the in vivo plasma samples or calibration standard samples was mixed with methanol/acetonitrile (3/7, v/v) containing propranolol as an internal standard, followed by centrifugation. The supernatant was filtered and then injected into a liquid chromatography tandem mass spectroscopy system for analysis.

Determination of plasma protein binding and unbound plasma concentration

The plasma protein binding of the test articles (except for pilocarpine, see below) in mice, rats, NHPs, and humans was determined by an equilibrium dialysis method using a Rapid Equilibrium Dialysis device (Thermo Fisher Scientific, USA) (Supplementary Table 2). Pooled human plasma (heparin sodium) was purchased from Biopredic International (France). Blank plasma (200 μL) spiked with a test article at 1 μg/mL (100 μg/mL for theophylline considering high plasma concentration) was dialyzed against phosphate buffered saline (PBS) (350 μL). The Rapid Equilibrium Dialysis device was then sealed with a gas-permeable membrane and shaken at 100 rpm for approximately 16 hr at 37°C under 5% CO₂. After incubation, plasma and PBS samples were collected from the donor and receiver chambers, respectively, and matrix-matched samples were prepared using blank samples. The samples were deproteinized by methanol/acetonitrile (3/7, v/v) containing the internal standard and analyzed by liquid chromatography tandem mass spectroscopy. The fraction unbound in plasma (f_u) was calculated by dividing the peak area ratio of the analyte to the internal standard in PBS samples by that in plasma samples. The f_u of pilocarpine was not determined in this study but was assumed to be unity based on available information (United States Food and Drug Administration [U.S. FDA], 2010). The unbound plasma concentration was calculated from the product of f_u and total plasma concentration.

Liquid chromatography tandem mass spectroscopy analysis

For bupropion, Compound A, diphenhydramine, linopirdine, PTZ, strychnine, theophylline, tiagabine, and tramadol, a Shimadzu 20A high-performance liquid chromatography system (Shimadzu, Japan) coupled with an API 3200 or API 4000 mass spectrometer (AB Sciex, USA) was used. Distilled water containing 0.02% (v/v) formic acid and acetonitrile containing 0.02% (v/v) formic acid were used as the mobile phases. The test articles were separated with a linear gradient using an L-Column ODS column (5 μm, 2.1 × 150 mm; Chemicals Evaluation and Research Institute, Japan). The multiple reaction monitoring transitions in positive electrospray ionization mode were m/z 240.1 to 184.3 for bupropion, 392.1 to 93.1 (or 392.1 to 299.8) for linopirdine, 264.0 to 58.1 for tramadol, 139.0 to 96.2 for PTZ, 181.1 to 124.0 for theophylline, 256.2 to 167.1 for diphenhydramine, 376.2 to 247.2 for tiagabine, 335.1 to 184.1 for strychnine, and 260.1 to 183.0 for the internal standard propranolol.

For 4-AP and pilocarpine, an Acquity UPLC system (Waters, USA) coupled with a Xevo TQ-XS mass spectrometer (Waters) was used. As the mobile phases, distilled water containing 0.02% (v/v) formic acid and distilled water/acetonitrile/1 mol/L ammonium acetate (17/80/3, v/v/v) were used for 4-AP, and distilled water/acetonitrile/1 mol/L ammonium acetate (89/10/1, v/v/v) and distilled water/acetonitrile/1 mol/L ammonium acetate (9/90/1, v/v/v) were used for pilocarpine. The test articles were separated with a linear gradient using a Scherzo SW-C18 column (3 μm, 50 × 2 mm; Imtact, Japan) for 4-AP and a Kinetex XB-C18 column (2.6 μm, 2.1 × 100 mm; Phenomenex, USA) for pilocarpine. The multiple reaction monitoring transitions in positive electrospray ionization mode were: m/z 95.43 to 77.81 for 4-AP, 209.02 to 94.83 for pilocarpine, and 260.11 to 115.84 for the internal standard propranolol.

RESULTS

The dose levels at which convulsions were induced in the animals are summarized in Table 2. “Premonitory signs” was defined here as the signs that were observed at the highest non-convulsant dose level or prior to the onset of a convulsion.

Mouse

All of the 11 test articles (including 2 reference convulsants in rodents) induced convulsions. The convulsion occurred within 15 min post-dose in 3 of 4 mice administered high-dose linopirdine or strychnine, 2 of 4 mice administered middle-dose tiagabine or diphenhydramine, and all 4 mice administered a high dose of the other test articles. Table 2 shows the premonitory signs of convulsion that were observed.

Rat

Convulsion was induced in 1 of 2 rats administered diphenhydramine, 3 of 4 rats administered bupropion or PTZ, and 2 of 2 rats administered the other test articles, except for tiagabine. Convulsion onset occurred within 5 min post-dose for most of the test articles, but at 10–105 min and 15–29 min post-dose for pilocarpine and theophylline. Table 2 shows the premonitory signs observed, including electrocardiogram changes. For Compound A, although QRS prolongation was observed due to an excessive on-target effect (i.e., sodium channel blockade), no clinical signs were noted prior to the onset of convulsion unlike what was observed for the other test articles. High-dose tiagabine failed to induce convulsion in 1 rat and resulted in death without convulsion of the other animal within 5-min post-dose.

NHP

Due to limited animal availability, only 6 test articles (4-AP, bupropion, Compound A, theophylline, tiagabine, and tramadol) were evaluated in NHPs. All 6 test articles induced convulsion during or shortly after completion of IV infusion in all animals, except for theophylline, in which convulsion was noted in only 1 (female) of 2 animals the following morning (approximately 24 hr after completion of IV infusion; the reason for the delayed convulsion is unknown). The total dose infused before convulsion occurred (or at the completion of IV infusion for theophylline) was calculated and are listed in Table 2. Table 2 also shows the premonitory signs that were observed for Compound A and tramadol with the animals under freely moving conditions. For the other test articles, no apparent clinical abnormalities were observed prior to convulsion, although this is partially because the animals were restrained in a monkey chair for those administrations (some premonitory signs reported in NHPs, such as tremor, twitching, retching/vomiting, salivation, ptosis, chewing, head shaking, excessive licking, and repeated yawning [Authier et al., 2019], are detectable even when restrained in a monkey chair, but were not present).

Species comparison of sensitivity to convulsion based on drug exposure

Figure 1 shows the plasma concentration of each test article shortly after convulsion onset (mice and NHPs [for theophylline, at the completion of IV infusion in NHPs]) or at 5-min post-dose (rats). The use of unbound versus total drug plasma concentration remains debatable when comparing sensitivity between species (DaSilva et al., 2020); however, Fig. 1 shows only unbound plasma concentrations because only unbound drug in the plasma can distribute to the brain parenchyma through the blood-brain barrier in line with the “free drug hypothesis” (Chen et al., 2020; Loryan et al., 2022). In the figure, the absence of convulsion is indicated by “-”. The human plasma concentrations at convulsion onset or during convulsion shown in the figure were estimated based on data taken from the published literature (see Table 3), or from in-house data for Compound A, except for those for linopirdine and pilocarpine, which have not been reported to induce convulsion in humans, or PTZ, for which drug concentrations during convulsion have not been reported.

Fig. 1

Unbound plasma concentrations of the test articles shortly after convulsion onset (in mouse and non-human primate [NHP]) or at 5-min post-dose (in rat). For theophylline in NHPs (n = 2; one NHP did not show convulsion and the other did the following morning), the values were at the completion of the 1-hr intravenous infusion. Symbols, including triangle, square, and circle, indicate the plasma concentrations of individual animals. Plus and minus signs indicate animals in which convulsion was observed or not observed, respectively. The clinical category shows human plasma concentration ranges at convulsion onset estimated based on available reference data. The dotted lines indicate published or estimated therapeutic blood concentrations (unbound) in humans based on available reference data. The horizontal lines indicate the means. 4-AP: Aminopyridine; PTZ: Pentylenetetrazole.

Table 3. Estimated human plasma concentrations during convulsion.

The unbound plasma concentrations of the test articles at convulsion onset were generally similar across the animal species, although there were some noteworthy trends. Mice showed the highest mean plasma concentrations in all three laboratory animals for 9 out of the 11 test articles (4-AP, bupropion, Compound A, diphenhydramine, linopirdine, PTZ, theophylline, tiagabine, and tramadol), with rats showing a higher concentration than mice for strychnine and equal plasma concentrations for mice and rats with pilocarpine. Mice showed the lowest mean exposure for 1 of 11 test articles (strychnine), rats for 3 of 11 test articles (diphenhydramine, linopirdine, PTZ), and NHPs for 3 of 6 test articles (Compound A, theophylline, and tiagabine). Comparing the mean plasma concentrations in mice with those in rats or NHPs, less than 3-fold (or more than 1/3-fold) differences was found for almost all test articles, except for a ≥ 6-fold difference for Compound A and a 5-fold difference for tramadol.

DISCUSSION

The current literature contains a number of publications concerning drug-induced convulsions in animals (Czuczwar et al., 1982; Akahane et al., 1993; Kita et al., 1999; Sansig et al., 2001; Luszczki et al., 2005; Easter et al., 2009; Authier et al., 2009, 2016, 2019; Bassett et al., 2014; Fujimoto et al., 2015); however, there is still a lack of information regarding convulsions in animals induced by convulsant drugs in humans. Here, we administered a total of 11 test articles from a variety of pharmacological classes, 9 of which are reported to induce convulsions in humans and 2 of which are known as convulsant in rodents, to mice, rats, and NHPs and compared the sensitivity of these laboratory animals to the convulsion liability of these drugs.

Although the dosing route of all of the test articles in humans is (or was reported) oral, dosing route of IP or IV was adopted in the present study because convulsion evaluations in the preclinical stages of drug development are usually conducted by IP or IV administration due to limited small amounts of a test article. The mice received the test articles via IP injection due to a technical difficulty to conduct IV administration to 4-week-old mice. The rats were catheterized and electrode-implanted and received the test articles via IV bolus. Such surgically modified rats are already used for preclinical screening of cardiovascular effects such as blood pressure changes and electrocardiogram abnormalities; therefore, this approach could allow for a simultaneous approach to cardiovascular and convulsion screening in rat. The NHPs received the test articles via a 1-hr IV infusion designed to allow precise determination of the concentration of the test article in the blood at the onset of convulsion. The results showed that 11 out of 11 test articles (100%) induced convulsion in mice, 10 out of 11 (91%) induced convulsion in rats, and 6 out of 6 (100%) induced convulsion in NHPs. These results are consistent with the results of a survey conducted by DaSilva et al. (2020) in which 11 pharmaceutical companies were asked for convulsion-related data for their proprietary compounds. In a non-exposure-based comparison of convulsion outcome across animal species, DaSilva et al. reported there to be no statistically significant inter-species difference in sensitivity to convulsion outcome between mouse and rat (n = 16 compounds) or between rat and NHP (n = 27 compounds), implying there to be no clear sensitivity ranking across these 3 species. In mice, Easter et al. (2009) assessed the convulsion liability of 15 convulsant test articles (4-AP, aminophylline, bupropion, cefazolin, chlorpromazine, enoxacin, isoniazid, maprotiline, meperidine, PTZ, pergolide, pilocarpine, reserpine, SNC-80, and strychnine) in PTZ convulsion threshold test, resulting in a predictability for convulsant liability in mice of 93% (increased sensitivity to PTZ-evoked convulsion was detected for 14 out of 15 test articles, not with the antibiotic cefazolin). Together, these previous findings and our present results indicate that mice can be used to detect convulsion liability with high sensitivity by either simple clinical observation after IP administration or evaluation by PTZ convulsion threshold test.

One interesting finding was that a rat treated with tiagabine died without first experiencing a convulsion; the death was possibly due to severe respiratory depression, as has been reported in human overdose cases (Spiller et al., 2005; Bauer and Cooper-Mahkorn, 2008) and no other fatal changes such as severe hypotension or life-threatening arrhythmia in the rat. The present authors have also previously observed other test articles (Eisai proprietary compounds), convulsants when oral administration to animals, to cause deaths without convulsion in rats after IV bolus administration, and those deaths are possibly attributed to excessive drug effects on vital signs and/or respiratory suppression. Hence, the convulsion liability of drug candidates would be carefully considered if death is discovered to occur after IV bolus administration in rats.

In the present study, unbound plasma concentrations at convulsion onset were presented and used for inter-species comparison (Fig. 1) because only unbound drug in the plasma can cross the blood-brain barrier and distribute to the brain parenchyma assuming the “free drug hypothesis” (Chen et al., 2020; Loryan et al., 2022). In addition, unbound plasma concentrations at convulsion onset, available in humans in some cases, will be informative for future comparisons with drug concentrations in the brain or cerebrospinal fluid at convulsion onset or the results of in vitro convulsion assessments such as microelectrode array using human induced pluripotent stem cell-derived neurons. In the present study, although mice tended to show high plasma concentrations compared with those in rats or NHPs, exposure at convulsion onset was generally comparable across the species, especially for rats and NHPs (Fig. 1). According to the recent survey by DaSilva et al. (2020), an exposure-based analysis of 44 compounds revealed that, other than dogs, rats and NHPs were the most sensitive species (comparable likelihood of drug-induced convulsions), suggesting that our exposure results are consistent with those from the survey of DaSilva et al. Figure 1 and Table 3 also show estimated ranges of human plasma concentrations during convulsion taken from the published literature (Ball et al., 1979; Bever et al., 1994; Fil et al., 2015; Stork and Hoffman, 1994; Davidson, 1989; Abdelmalek et al., 2014; Abdi et al., 2014; Jang et al., 2010; McKeown et al., 2011; Pragst et al., 2006; Pieniaszek et al., 1995; U.S. FDA Drug Label for Salagen^®; Edmunds et al., 1986; Hernandez et al., 1998; Marques et al., 2000; Wood et al., 2002; Bahls et al., 1991; Covelli et al., 1985; Fulton et al., 2005; Kazzi et al., 2006; Ostrovskiy et al., 2002; Taghaddosinejad et al., 2011). The plasma concentrations at convulsion onset in the laboratory animals were comparable or higher than the ranges observed in humans, although it should be noted that there is a lack of detailed information on the incidence of convulsions in humans and of information on the presence or absence of other proconvulsant factors in the patients at the time of convulsion onset. In addition, Fig. 1 also shows the published or estimated therapeutic level for each test article, except for linopirdine, PTZ, and strychnine, which are not marketed medicines (Supplementary Table 3). Convulsions in animals were noted at plasma concentrations generally 10- to 60-fold higher than the therapeutic level in NHPs, 20- to 100-fold higher in rats, and 20- to 170-fold higher in mice. The test articles used in the present study were clinically developed by 2003, except for strychnine used as pesticide, before the recent decision by U.S. FDA to limit human exposure to 1/10th the C_max at the NOEL for convulsion in preclinical animal studies (Bassett et al., 2014; Authier et al., 2016; Abt et al., 2019; DaSilva et al., 2020). Thus, when conducting animal model-based convulsion screenings for new drug candidates in the future, the margin between the plasma concentration at convulsion onset versus the therapeutic level should be wider than the values observed in the present study, although the best scenario remains to select a non-convulsant compound if that is an option.

Regarding nonclinical species differences, rats are reported to be more sensitive to cannabinoid-induced convulsion than are dogs, which was thought to be due to differences between the species in cannabinoid signaling (Whalley et al., 2019). In the present study, mice showed less sensitivity to convulsion than did rats or NHPs for Compound A and tramadol, and rats were less sensitive to strychnine-induced convulsion than were mice. To our knowledge, there are no differences in the biosignaling related to the mechanisms of action of the present test articles (e.g., differences of receptor expressions/distributions of sodium, serotonin, noradrenaline, or glycine in the brain) among mice, rats, and NHPs, so the reasons for these differences remain to be explored further.

In the present study, several premonitory signs of convulsions were observed in the animals (Table 2) but no common premonitory indicator of convulsion was identified across the 11 test articles, which is consistent with the findings of previous studies (Gauvin et al., 2018; DaSilva et al., 2020). Since there are only limited reports of specific premonitory signs of convulsion in humans, we collated the adverse signs and symptoms reported after overdose, or close to or at the convulsive dose, of the test articles in humans and considered those to be the premonitory signs of convulsion in humans (Table 4). Comparing the premonitory signs in animals and humans, there was 1 sign or more observed in both animals and humans for each test article (underlined in Table 4), suggesting that some premonitory signs of convulsion observed in preclinical studies might also be observed prior to convulsion in humans. Some drugs, such as bupropion, diphenhydramine, theophylline, and tramadol, are reported to show cardiovascular abnormalities at high doses in humans (Table 4). Cardiac changes are potential biomarkers that may provide an extra-cerebral indicator of epileptic seizure onset in some situations (Eggleston et al., 2014; Gauvin et al., 2018). Indeed, the catheterized and electrode-implanted rats used in the present study did show cardiovascular effects as premonitory signs of convulsion in Compound A, pilocarpine, and theophylline (Table 2).

Table 4. Adverse signs and symptoms reported at high doses (close to or at the convulsive dose level and/or after overdose) of each test article in humans.

Currently, there are no safety biomarkers authorized for use by a regulatory agency for the predicting of drug-induced convulsion in humans or animals (Gauvin et al., 2018; Walker et al., 2018; Abt et al., 2019; Schomaker et al., 2019). In clinical trials with proconvulsant drug candidates, in addition to monitoring of possible premonitory signs of convulsion, including cardiovascular effects, noted during preclinical studies, other options that could be incorporated for safer human trials include screening to exclude healthy volunteers or patients with spontaneous electroencephalographic epileptiform abnormalities and regular monitoring for electroencephalographic abnormalities (Abt et al., 2019), as well as neurologic examinations such as for saccade eye movement to detect brain abnormalities (Sakuma et al., 2008). For prediction of drug-induced convulsion in humans and animals, some progress has been made recently with the development of an electroencephalographic-based continual reassessment method in humans (Abt et al., 2019), a frequency-intensity distribution histogram-based approach using deep-learning in rats (Matsuda et al., 2021), and a heart rate variability-based approach using machine-learning in NHPs (Nagata et al., 2021). These new analytical methods are expected to facilitate the avoidance of convulsions in humans.

The present study had some limitations: drug concentration in the brain or cerebrospinal fluid was not measured; the occurrence of drug-induced convulsions in female NHPs might be influenced by menstrual cycle (estrogen and progesterone are presumed to have proconvulsant and anticonvulsant effects, respectively, in women with epilepsy [Scharfman and MacLusky, 2006]); only 6 of the 11 test articles could be examined in NHPs; only 2 NHPs were used per test article (except for Compound A [n = 1]), despite NHPs sometimes showing large individual variations in nonclinical studies; and the timing of blood collection was not exactly the same between mice, rats, and NHPs due to limitations of the study conditions. However, diphenhydramine, PTZ, and pilocarpine, not examined in the present study, are reported to induce convulsions in NHPs (Evans and Johanson, 1989; Bassett et al., 2014; Nagata et al., 2021; Perez-Mendes et al., 2011), and our convulsion sensitivity results across the laboratory animals are consistent with those from the survey conducted by DaSilva et al. (2020). Therefore, we conclude that the present approach was appropriate for evaluating the sensitivity to drug-induced convulsion across these laboratory animals.

To provide information on which animal species is the most suitable for assessing potential convulsion risk in humans, the advantages and disadvantages of using mice, rats, or NHPs at the lead identification/optimization stage of drug development are summarized in Table 5 based on the results from the present study. Given that these 3 species generally showed similar sensitivity to convulsion and taking into account the animal welfare aspect of conducting research in NHPs and possibly the continued constrained supply of NHPs (Brown and Wange, 2023), using mice or rats should be considered the most ethical approach for in vivo convulsion screening although NHPs generally showed the lowest drug plasma concentration at convulsion onset. Thus, a stepwise cascade of using mouse and/or rat and then NHP would be considered reasonable.

Table 5. Advantages and disadvantages of each laboratory animal for convulsion screening at the lead identification/optimization stage of drug development based on the findings from the present study.

In conclusion, among mice, rats, and NHPs we compared their sensitivity to convulsion induced by 11 test articles from a variety of pharmacological classes, including 9 test articles with known convulsant liability in humans and 2 reference convulsants in rodents. Although there was a general tendency that rats and NHPs exhibited convulsions at lower plasma drug concentrations than did mice, the plasma concentrations at convulsion onset were generally comparable, within 3-fold differences, across the animal species. These results suggest that these animals can be used for convulsion screening in the preclinical stages of drug development, provided that the advantages and disadvantages associated with the use of each animal are appropriately considered.

ACKNOWLEDGMENTS

The authors gratefully thank Mr. Toshio Imade, Katsuya Enjoji, Yuki Tanaka (Drug Safety & Animal Care Technology Unit, Tsukuba Division, Sunplanet Co., Ltd.), and Kazumasa Aoyama (Pharmacological Evaluation Unit, Tsukuba Division, Sunplanet Co., Ltd.) for their excellent technical assistance.

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Abdelmalek,  D.,  Schwarz,  E.S.,  Sampson,  C.,  Halcomb,  S.E.,  McCammon,  C.,  Arroyo-Plasencia,  A.,  Stenger,  A.,  Krehbiel,  N. and  Mullins,  M.E. (2014): Life-threatening diphenhydramine toxicity presenting with seizures and a wide complex tachycardia improved with intravenous fat emulsion. Am. J. Ther., 21, 542-544.
•  Abdi,  A.,  Rose,  E. and  Levine,  M. (2014): Diphenhydramine overdose with intraventricular conduction delay treated with hypertonic sodium bicarbonate and i.v. lipid emulsion. West. J. Emerg. Med., 15, 855-858.
•  Abt,  M.,  Dinklo,  T.,  Rothfuss,  A.,  Husar,  E.,  Dannecker,  R.,  Kallivroussis,  K.,  Peck,  R.,  Doessegger,  L. and  Wandel,  C. (2019): A Framework Proposal to Follow-Up on Preclinical Convulsive Signals of a New Molecular Entity in First-in-Human Studies Using Electroencephalographic Monitoring. Clin. Pharmacol. Ther., 106, 968-980.
•  Accardi,  M.V.,  Huang,  H. and  Authier,  S. (2018): Seizure liability assessments using the hippocampal tissue slice: comparison of non-clinical species. J. Pharmacol. Toxicol. Methods, 93, 59-68.
•  Akahane,  K.,  Kato,  M. and  Takayama,  S. (1993): Involvement of inhibitory and excitatory neurotransmitters in levofloxacin- and ciprofloxacin-induced convulsions in mice. Antimicrob. Agents Chemother., 37, 1764-1770.
•  Authier,  S.,  Arezzo,  J.,  Pouliot,  M.,  Accardi,  M.V.,  Boulay,  E.,  Troncy,  E.,  Dubuc Mageau,  M.,  Tan,  W.,  Sanfacon,  A.,  Mignault Goulet,  S. and  Paquette,  D. (2019): EEG: characteristics of drug-induced seizures in rats, dogs and non-human primates. J. Pharmacol. Toxicol. Methods, 97, 52-58.
•  Authier,  S.,  Delatte,  M.S.,  Kallman,  M.J.,  Stevens,  J. and  Markgraf,  C. (2016): EEG in non-clinical drug safety assessments: current and emerging considerations. J. Pharmacol. Toxicol. Methods, 81, 274-285.
•  Authier,  S.,  Paquette,  D.,  Gauvin,  D.,  Sammut,  V.,  Fournier,  S.,  Chaurand,  F. and  Troncy,  E. (2009): Video-electroencephalography in conscious non human primate using radiotelemetry and computerized analysis: refinement of a safety pharmacology model. J. Pharmacol. Toxicol. Methods, 60, 88-93.
•  Bahls,  F.H.,  Ma,  K.K. and  Bird,  T.D. (1991): Theophylline-associated seizures with “therapeutic” or low toxic serum concentrations: risk factors for serious outcome in adults. Neurology, 41, 1309-1312.
•  Ball,  A.P.,  Hopkinson,  R.B.,  Farrell,  I.D.,  Hutchison,  J.G.,  Paul,  R.,  Watson,  R.D.,  Page,  A.J.,  Parker,  R.G.,  Edwards,  C.W.,  Snow,  M.,  Scott,  D.K.,  Leone-Ganado,  A.,  Hastings,  A.,  Ghosh,  A.C. and  Gilbert,  R.J. (1979): Human botulism caused by Clostridium botulinum type E: the Birmingham outbreak. Q. J. Med., 48, 473-491.
•  Bassett,  L.,  Troncy,  E.,  Pouliot,  M.,  Paquette,  D.,  Ascah,  A. and  Authier,  S. (2014): Telemetry video-electroencephalography (EEG) in rats, dogs and non-human primates: methods in follow-up safety pharmacology seizure liability assessments. J. Pharmacol. Toxicol. Methods, 70, 230-240.
•  Bauer,  J. and  Cooper-Mahkorn,  D. (2008): Tiagabine: efficacy and safety in partial seizures - current status. Neuropsychiatr. Dis. Treat., 4, 731-736.
•  Bever,  C.T. Jr.,  Young,  D.,  Anderson,  P.A.,  Krumholz,  A.,  Conway,  K., et al. (1994): The effects of 4-aminopyridine in multiple sclerosis patients: results of a randomized, placebo-controlled, double-blind, concentration-controlled, crossover trial. Neurology, 44, 1054-1059.
•  Boostani,  R. and  Derakhshan,  S. (2012): Tramadol induced seizure: A 3-year study. Caspian J. Intern. Med., 3, 484-487.
•  Bradley,  J.A.,  Luithardt,  H.H.,  Metea,  M.R. and  Strock,  C.J. (2018): In Vitro Screening for Seizure Liability Using Microelectrode Array Technology. Toxicol. Sci., 163, 240-253.
•  Brown,  K.M. and  Crouch,  B.I. (2017): Bupropion overdose: significant toxicity in pediatrics. Clin. Pediatr. Emerg. Med., 18, 212-217.
•  Brown,  P.C. and  Wange,  R.L. (2023): Considerations regarding the use of nonhuman primates in assessing safety endpoints for pharmaceuticals. Regul. Toxicol. Pharmacol., 143, 105449.
•  Bruccoleri,  R.E. and  Burns,  M.M. (2016): A Literature Review of the Use of Sodium Bicarbonate for the Treatment of QRS Widening. J. Med. Toxicol., 12, 121-129.
•  Cantrell,  F.L.,  Ritter,  M. and  Himes,  E. (2004): Intentional overdose with tiagabine: an unusual clinical presentation. J. Emerg. Med., 27, 271-272.
•  Chen,  C.,  Zhou,  H.,  Guan,  C.,  Zhang,  H.,  Li,  Y.,  Jiang,  X.,  Dong,  Z.,  Tao,  Y.,  Du,  J.,  Wang,  S.,  Zhang,  T.,  Du,  N.,  Guo,  J.,  Wu,  Y.,  Song,  Z.,  Luan,  H.,  Wang,  Y.,  Du,  H.,  Zhang,  S.,  Li,  C.,  Chang,  H. and  Wang,  T. (2020): Applicability of free drug hypothesis to drugs with good membrane permeability that are not efflux transporter substrates: A microdialysis study in rats. Pharmacol. Res. Perspect., 8, e00575.
•  Clark,  M. and  Steger-Hartmann,  T. (2018): A big data approach to the concordance of the toxicity of pharmaceuticals in animals and humans. Regul. Toxicol. Pharmacol., 96, 94-105.
•  Covelli,  H.D.,  Knodel,  A.R. and  Heppner,  B.T. (1985): Predisposing factors to apparent theophylline-induced seizures. Ann. Allergy, 54, 411-415.
•  Czuczwar,  S.J.,  Turski,  L. and  Kleinrok,  Z. (1982): Effects of combined treatment with diphenylhydantoin and different benzodiazepines on pentylenetetrazol- and bicuculline-induced seizures in mice. Neuropharmacology, 21, 563-567.
•  DaSilva,  J.K.,  Breidenbach,  L.,  Deats,  T.,  Li,  D.,  Treinen,  K.,  Dinklo,  T.,  Kervyn,  S.,  Teuns,  G.,  Traebert,  M. and  Hempel,  K. (2020): Nonclinical species sensitivity to convulsions: an IQ DruSafe consortium working group initiative. J. Pharmacol. Toxicol. Methods, 103, 106683.
•  Davidson,  J. (1989): Seizures and bupropion: a review. J. Clin. Psychiatry, 50, 256-261.
•  Easter,  A.,  Bell,  M.E.,  Damewood,  J.R. Jr.,  Redfern,  W.S.,  Valentin,  J.P.,  Winter,  M.J.,  Fonck,  C. and  Bialecki,  R.A. (2009): Approaches to seizure risk assessment in preclinical drug discovery. Drug Discov. Today, 14, 876-884.
•  Edmonds,  H.L. Jr.,  Hegreberg,  G.A.,  vanGelder,  N.M.,  Sylvester,  D.M.,  Clemmons,  R.M. and  Chatburn,  C.G. (1979): Spontaneous convulsions in beagle dogs. Fed. Proc., 38, 2424-2428.
•  Edmunds,  M.,  Sheehan,  T.M. and  Van’t Hoff,  W. (1986): Strychnine poisoning: clinical and toxicological observations on a non-fatal case. J. Toxicol. Clin. Toxicol., 24, 245-255.
•  Eggleston,  K.S.,  Olin,  B.D. and  Fisher,  R.S. (2014): Ictal tachycardia: the head-heart connection. Seizure, 23, 496-505.
•  Evans,  S.M. and  Johanson,  C.E. (1989): Discriminative stimulus properties of histamine H1-antagonists in animals trained to discriminate d-amphetamine or pentobarbital. J. Pharmacol. Exp. Ther., 250, 779-787.
•  Fil,  L.J.,  Sud,  P. and  Sattler,  S. (2015): A massive overdose of dalfampridine. West. J. Emerg. Med., 16, 1177-1179.
•  Frankfurt,  M.,  Fuchs,  E. and  Wuttke,  W. (1984): Sex differences in gamma-aminobutyric acid and glutamate concentrations in discrete rat brain nuclei. Neurosci. Lett., 50, 245-250.
•  Fujimoto,  Y.,  Funao,  T.,  Suehiro,  K.,  Takahashi,  R.,  Mori,  T. and  Nishikawa,  K. (2015): Brain serotonin content regulates the manifestation of tramadol-induced seizures in rats: disparity between tramadol-induced seizure and serotonin syndrome. Anesthesiology, 122, 178-189.
•  Fulton,  J.A.,  Hoffman,  R.S. and  Nelson,  L.S. (2005): Tiagabine overdose: a case of status epilepticus in a non-epileptic patient. Clin. Toxicol. (Phila.), 43, 869-871.
•  Gauvin,  D.V.,  Zimmermann,  Z.J.,  Yoder,  J.,  Harter,  M.,  Holdsworth,  D.,  Kilgus,  Q.,  May,  J.,  Dalton,  J. and  Baird,  T.J. (2018): A predictive index of biomarkers for ictogenesis from tier I safety pharmacology testing that may warrant tier II EEG studies. J. Pharmacol. Toxicol. Methods, 94, 50-63.
•  Goodman,  A.D.,  Cohen,  J.A.,  Cross,  A.,  Vollmer,  T.,  Rizzo,  M.,  Cohen,  R.,  Marinucci,  L. and  Blight,  A.R. (2007): Fampridine-SR in multiple sclerosis: a randomized, double-blind, placebo-controlled, dose-ranging study. Mult. Scler., 13, 357-368.
•  Hendrickson,  R.G.,  Morocco,  A.P. and  Greenberg,  M.I. (2004): Pilocarpine toxicity and the treatment of xerostomia. J. Emerg. Med., 26, 429-432.
•  Hernandez,  A.F.,  Pomares,  J.,  Schiaffino,  S.,  Pla,  A. and  Villanueva,  E. (1998): Acute chemical pancreatitis associated with nonfatal strychnine poisoning. J. Toxicol. Clin. Toxicol., 36, 67-71.
• Hoskins, J.D. (2000). Congenital defects of the dog. In: S. J. Ettinger, & E. C. Feldman (Eds.), Textbook of veterinary internal medicine (5th ed.), pp.1983-1996, W.B. Saunders Company, Philadelphia.
•  Jang,  D.H.,  Manini,  A.F.,  Trueger,  N.S.,  Duque,  D.,  Nestor,  N.B.,  Nelson,  L.S. and  Hoffman,  R.S. (2010): Status epilepticus and wide-complex tachycardia secondary to diphenhydramine overdose. Clin. Toxicol. (Phila.), 48, 945-948.
•  Kazzi,  Z.N.,  Jones,  C.C. and  Morgan,  B.W. (2006): Seizures in a pediatric patient with a tiagabine overdose. J. Med. Toxicol., 2, 160-162.
•  Kita,  H.,  Matsuo,  H.,  Takanaga,  H.,  Kawakami,  J.,  Yamamoto,  K.,  Iga,  T.,  Naito,  M.,  Tsuruo,  T.,  Asanuma,  A.,  Yanagisawa,  K. and  Sawada,  Y. (1999): In vivo and in vitro toxicodynamic analyses of new quinolone-and nonsteroidal anti-inflammatory drug-induced effects on the central nervous system. Antimicrob. Agents Chemother., 43, 1091-1097.
•  Kreir,  M.,  Van Deuren,  B.,  Versweyveld,  S.,  De Bondt,  A.,  Van den Wyngaert,  I.,  Van der Linde,  H.,  Lu,  H.R.,  Teuns,  G. and  Gallacher,  D.J. (2018): Do in vitro assays in rat primary neurons predict drug-induced seizure liability in humans? Toxicol. Appl. Pharmacol., 346, 45-57.
•  Loryan,  I.,  Hammarlund-Udenaes,  M. and  Syvänen,  S. (2022): Brain Distribution of Drugs: Pharmacokinetic Consideration. Handb. Exp. Pharmacol., 273, 121-150. Springer Nature, Switzerland.
•  Löscher,  W. (2011): Critical review of current animal models of seizures and epilepsy used in the discovery and development of new antiepileptic drugs. Seizure, 20, 359-368.
•  Luszczki,  J.J.,  Wojcik-Cwikla,  J.,  Andres,  M.M. and  Czuczwar,  S.J. (2005): Pharmacological and behavioral characteristics of interactions between vigabatrin and conventional antiepileptic drugs in pentylenetetrazole-induced seizures in mice: an isobolographic analysis. Neuropsychopharmacology, 30, 958-973.
•  Maguire,  J.L.,  Stell,  B.M.,  Rafizadeh,  M. and  Mody,  I. (2005): Ovarian cycle-linked changes in GABA_A receptors mediating tonic inhibition alter seizure susceptibility and anxiety. Nat. Neurosci., 8, 797-804.
•  Makarovsky,  I.,  Markel,  G.,  Hoffman,  A.,  Schein,  O.,  Brosh-Nissimov,  T.,  Tashma,  Z.,  Dushnitsky,  T. and  Eisenkraft,  A. (2008): Strychnine--a killer from the past. Isr. Med. Assoc. J., 10, 142-145.
•  Marques,  E.P.,  Gil,  F.,  Proenca,  P.,  Monsanto,  P.,  Oliveira,  M.F.,  Castanheira,  A. and  Vieira,  D.N. (2000): Analytical method for the determination of strychnine in tissues by gas chromatography/mass spectrometry: two case reports. Forensic Sci. Int., 110, 145-152.
•  Matsuda,  N.,  Kinoshita,  K.,  Okamura,  A.,  Shirakawa,  T. and  Suzuki,  I. (2021): Histograms of Frequency-Intensity Distribution Deep Learning to Predict the Seizure Liability of Drugs in Electroencephalography. Toxicol. Sci., 182, 229-242.
•  McKeown,  N.J.,  West,  P.L.,  Hendrickson,  R.G. and  Horowitz,  B.Z. (2011): Survival after diphenhydramine ingestion with hemodialysis in a toddler. J. Med. Toxicol., 7, 147-150.
•  Nagata,  S.,  Fujiwara,  K.,  Kuga,  K. and  Ozaki,  H. (2021): Prediction of GABA receptor antagonist-induced convulsion in cynomolgus monkeys by combining machine learning and heart rate variability analysis. J. Pharmacol. Toxicol. Methods, 112, 107127.
•  Odawara,  A.,  Matsuda,  N.,  Ishibashi,  Y.,  Yokoi,  R. and  Suzuki,  I. (2018): Toxicological evaluation of convulsant and anticonvulsant drugs in human induced pluripotent stem cell-derived cortical neuronal networks using an MEA system. Sci. Rep., 8, 10416.
•  Ostrovskiy,  D.,  Spanaki,  M.V. and  Morris,  G.L. 3rd. (2002): Tiagabine overdose can induce convulsive status epilepticus. Epilepsia, 43, 773-774.
•  Otto,  J.F.,  Yang,  Y.,  Frankel,  W.N.,  Wilcox,  K.S. and  White,  H.S. (2004): Mice carrying the szt1 mutation exhibit increased seizure susceptibility and altered sensitivity to compounds acting at the m-channel. Epilepsia, 45, 1009-1016.
•  Perez-Mendes,  P.,  Blanco,  M.M.,  Calcagnotto,  M.E.,  Cinini,  S.M.,  Bachiega,  J.,  Papoti,  D.,  Covolan,  L.,  Tannus,  A. and  Mello,  L.E. (2011): Modeling epileptogenesis and temporal lobe epilepsy in a non-human primate. Epilepsy Res., 96, 45-57.
•  Pieniaszek,  H.J. Jr.,  Fiske,  W.D.,  Saxton,  T.D.,  Kim,  Y.S.,  Garner,  D.M.,  Xilinas,  M. and  Martz,  R. (1995): Single-dose pharmacokinetics, safety, and tolerance of linopirdine (DuP 996) in healthy young adults and elderly volunteers. J. Clin. Pharmacol., 35, 22-30.
•  Pontes,  J.C.,  Lima,  T.Z.,  Queiroz,  C.M.,  Cinini,  S.M.,  Blanco,  M.M. and  Mello,  L.E. (2016): Seizures triggered by pentylenetetrazol in marmosets made chronically epileptic with pilocarpine show greater refractoriness to treatment. Epilepsy Res., 126, 16-25.
•  Pragst,  F.,  Herre,  S. and  Bakdash,  A. (2006): Poisonings with diphenhydramine--a survey of 68 clinical and 55 death cases. Forensic Sci. Int., 161, 189-197.
•  Radovanovic,  D.,  Meier,  P.J.,  Guirguis,  M.,  Lorent,  J.P. and  Kupferschmidt,  H. (2000): Dose-dependent toxicity of diphenhydramine overdose. Hum. Exp. Toxicol., 19, 489-495.
•  Roberts,  R.,  Authier,  S.,  Mellon,  R.D.,  Morton,  M.,  Suzuki,  I.,  Tjalkens,  R.B.,  Valentin,  J.P. and  Pierson,  J.B. (2021): Can we panelize seizure? Toxicol. Sci., 179, 3-13.
•  Rockwood,  K.,  Beattie,  B.L.,  Eastwood,  M.R.,  Feldman,  H.,  Mohr,  E.,  Pryse-Phillips,  W. and  Gauthier,  S. (1997): A randomized, controlled trial of linopirdine in the treatment of Alzheimer’s disease. Can. J. Neurol. Sci., 24, 140-145.
•  Ryan,  N.M. and  Isbister,  G.K. (2015): Tramadol overdose causes seizures and respiratory depression but serotonin toxicity appears unlikely. Clin. Toxicol. (Phila.), 53, 545-550.
•  Sakuma,  D.,  Uchida,  N.,  Iwase,  M.,  Saito,  Y.,  Mochii,  M.,  Sakurai,  K.,  Yasuda,  K. and  Yasuhara,  H. (2008): Clinical Pharmacological Study to Evaluate the Drug Interaction between Histamine H1 Receptor Antagonist Ebastine and Verapamil, P-gp Inhibitor at the Blood Brain Barrier: Objective Evaluation Using Saccadic Eye Movement Analysis. Jpn. J. Clin. Pharmacol. Ther., 39, 71-78.
•  Sansig,  G.,  Bushell,  T.J.,  Clarke,  V.R.,  Rozov,  A.,  Burnashev,  N.,  Portet,  C.,  Gasparini,  F.,  Schmutz,  M.,  Klebs,  K.,  Shigemoto,  R.,  Flor,  P.J.,  Kuhn,  R.,  Knoepfel,  T.,  Schroeder,  M.,  Hampson,  D.R.,  Collett,  V.J.,  Zhang,  C.,  Duvoisin,  R.M.,  Collingridge,  G.L. and  van Der Putten,  H. (2001): Increased seizure susceptibility in mice lacking metabotropic glutamate receptor 7. J. Neurosci., 21, 8734-8745.
•  Scharfman,  H.E. and  MacLusky,  N.J. (2006): The influence of gonadal hormones on neuronal excitability, seizures, and epilepsy in the female. Epilepsia, 47, 1423-1440.
•  Schomaker,  S.,  Ramaiah,  S.,  Khan,  N. and  Burkhardt,  J. (2019): Safety biomarker applications in drug development. J. Toxicol. Sci., 44, 225-235.
•  Shannon,  M. (1999): Life-threatening events after theophylline overdose: a 10-year prospective analysis. Arch. Intern. Med., 159, 989-994.
•  Spiller,  H.A.,  Winter,  M.L.,  Ryan,  M.,  Krenzelok,  E.P.,  Anderson,  D.L.,  Thompson,  M. and  Kumar,  S. (2005): Retrospective evaluation of tiagabine overdose. Clin. Toxicol. (Phila.), 43, 855-859.
•  Stork,  C.M. and  Hoffman,  R.S. (1994): Characterization of 4-aminopyridine in overdose. J. Toxicol. Clin. Toxicol., 32, 583-587.
•  Taghaddosinejad,  F.,  Mehrpour,  O.,  Afshari,  R.,  Seghatoleslami,  A.,  Abdollahi,  M. and  Dart,  R.C. (2011): Factors related to seizure in tramadol poisoning and its blood concentration. J. Med. Toxicol., 7, 183-188.
•  Tetz,  L.M.,  Rezk,  P.E.,  Ratcliffe,  R.H.,  Gordon,  R.K.,  Steele,  K.E. and  Nambiar,  M.P. (2006): Development of a rat pilocarpine model of seizure/status epilepticus that mimics chemical warfare nerve agent exposure. Toxicol. Ind. Health, 22, 255-266.
•  Tukker,  A.M. and  Westerink,  R.H.S. (2021): Novel test strategies for in vitro seizure liability assessment. Expert Opin. Drug Metab. Toxicol., 17, 923-936.
• United States Food and Drug Administration. (U.S. FDA). Clinical Pharmacology Review for Pilocarpine hydrochloride. (2010). https://www.accessdata.fda.gov/drugsatfda_docs/nda/2010/200890s000ClinPharmR.pdf
• United States Food and Drug Administration. (U.S. FDA). Drug Label for Forfivo XL (Bupropion hydrochloride extended-release tablets). https://www.accessdata.fda.gov/drugsatfda_docs/nda/2011/022497Orig1s000Lbl.pdf
• United States Food and Drug Administration. (U.S. FDA). Drug Label for Salagen (pilocarpine hydrochloride). https://www.accessdata.fda.gov/spl/data/4e407b8d-c8ad-4907-9246-45685ee55052/4e407b8d-c8ad-4907-9246-45685ee55052.xml
•  Walker,  A.L.,  Imam,  S.Z. and  Roberts,  R.A. (2018): Drug discovery and development: Biomarkers of neurotoxicity and neurodegeneration. Exp. Biol. Med. (Maywood), 243, 1037-1045.
•  Whalley,  B.J.,  Lin,  H.,  Bell,  L.,  Hill,  T.,  Patel,  A.,  Gray,  R.A.,  Elizabeth Roberts,  C.,  Devinsky,  O.,  Bazelot,  M.,  Williams,  C.M. and  Stephens,  G.J. (2019): Species-specific susceptibility to cannabis-induced convulsions. Br. J. Pharmacol., 176, 1506-1523.
•  Wood,  D.,  Webster,  E.,  Martinez,  D.,  Dargan,  P. and  Jones,  A. (2002): Case report: Survival after deliberate strychnine self-poisoning, with toxicokinetic data. Crit. Care, 6, 456-459.