Public Assessment Report

Scientific discussion

Dimenhydrinat “Trimb” 20 mg lozenges ()

DK/H/2835/001/DC

Date: 4 December 2018

This module reflects the scientific discussion for the approval of Dimenhydrinat ”Trimb”. The procedure was finalised on 20 September 2018. For information on changes after this date please refer to the module ‘Update’.

I. INTRODUCTION

Based on the review of the quality, safety and efficacy data, the Member States have granted a marketing authorisation for Dimenhydrinat ”Trimb” 20 mg lozenges, from Trimb Healthcare AB.

The product is indicated for prevention of in adults and children over 6 years of age. A comprehensive description of the indications and posology is given in the SmPC.

Motion sickness is a well-known and syndrome in otherwise healthy people. The physical signs of motion sickness occur in both humans and animals during travel by sea, automobile or airplane and in space. Furthermore, some other special situations, such as simulators, the cinema and video games, have been described as causing pseudomotion sickness. Children between 2 and 12 years old are most susceptible to motion sickness, and women are more frequently affected than men. Predisposing factors include pregnancy and migraines and possibly a side difference in the mass of otoconia in the vestibular organs. Therapy is directed towards decreasing conflicting sensory input, accelerating the process of adaptation and controlling nausea and vomiting. To control these vegetative symptoms, and are the most effective .

Motion sickness is caused by certain types of motion and is induced during passive locomotion in vehicles, generated by unfamiliar body accelerations, to which the person has not adapted, or by an intersensory conflict between vestibular and visual stimuli. Motion sickness indiscriminately affects air, sea, road and space travelers. All individuals (humans and animals) possessing an intact vestibular apparatus can get motion sickness given the right quality and quantity of provocative stimulation, although there are wide and consistent individual differences in the degree of susceptibility. The cardinal signs of motion sickness are nausea, vomiting, pallor and cold sweating.

Histamine H1 receptors are involved in the development of motion sickness signs and symptoms, including emesis. Upon provocative motion stimuli, a neural mismatch signal activates the neuron system in the hypothalamus, and the histaminergic descending impulse stimulates H1 receptors in the brainstem’s emetic center. The histaminergic input to the emetic center through H1 receptors is independent of dopamine D2 receptors in the chemoreceptor trigger zone in the area postrema and serotonin 5-HT3 receptors in the visceral afferent, which are also involved in the emetic reflex.

Dimenhydrinate is used most commonly as an because it is primarily a type 1 (H1) antagonist, but it also possesses an antimuscarinic effect. The other component of dimenhydrinate, 8- chlorotheophylline, is a xanthine derivative. By blocking adenosine receptors, xanthine derivatives likely produce excitation and psychostimulation.

The marketing authorisation has been granted pursuant to Article 10a (bibliographic application) of Directive 2001/83/EC. This type of application does not require submission of the results of preclinical tests or clinical trials if the Applicant can demonstrate that the active substance of the medicinal product has been in well-established medicinal use within the EU for at least 10 years, with recognised efficacy and an acceptable level of safety.

Dimenhydrinate has been extensively used in humans for several decades and can be considered an active substance with well-established used for the intended indication.

II. QUALITY ASPECTS

II.1 Introduction

PAR Scientific discussion 2/21

Dimenhydrinat ”Trimb” lozenges are presented as anise flavoured lozenges containing 20 mg of dimenhydrinate. The lozenge is clear to yellowish, round and with a diameter of 19±1 mm.

The finished product is to be marketed in PVC-PVDC/Aluminium blisters in pack sizes of 24.

The lozenges contain: Isomalt (E953); saccharin sodium (E954); masking (propylene glycol (E1520), triethyl citrate (E1505)); anise flavour (glyceryl triacetate (E1518) and mouth watering (arabic gum (E414), citric acid (E330) and sodium citrate (E331)).

The RMS has been assured that acceptable standards of GMP (see Directive 2003/94/EC) are in place for this product type at all sites responsible for the manufacturing of the active substance as well as for the manufacturing and assembly of this product prior to granting its national authorisation.

II.2 Substance

The manufacturer of the active substance, dimenhydrinate, has obtained a Certificate of Suitability.

The drug substance is controlled in line with the respective Ph.Eur monograph, as attested by the CEP covering the manufacturing process of the active substance.

The control tests and specification for drug substance are adequately drawn up by the Applicant and in line with the CEP and additional requirements on solvents.

The CEP does not cover stability studies as no re-test period is reported. The Applicant has provided container closure system and stability data and the proposed re-test period and storage conditions are considered acceptable.

II.3 Medicinal Product

The development of the product has been described. The choice of excipients is justified and their functions explained.

The manufacturing process is described in sufficient details and the process and its associated in process controls are considered acceptable.

The drug product is a lozenge and the product specification covers appropriate parameters for this dosage form. The shelf life and release specification for related substance are in general considered acceptable. A follow-up measure is needed since an impurity is formed above the ICH Q3B qualification threshold and should be identified and qualified. The Applicant has initiated these studies, and will submit when complete.

Validations of the analytical methods have been presented, and the Applicant provided demonstration of the stability indicating capability of the methods. Batch analyses have been performed on 3 batches. The batch analysis results show that the finished products meet the specifications proposed.

The conditions used in the stability studies are according to the ICH stability guideline. A shelf life of 24 months with no special storage condition is acceptable. The analytical methods are shown to be suitable for stability measurements and control.

PAR Scientific discussion 3/21

III. NON-CLINICAL ASPECTS

III.1

The pharmacological effects of dimenhydrinate and were described in several older publications, whereas the publications of abuse potential were of a more recent date. An overview of pharmacological studies referred in this assessment report with calculated (whenever possible) human equivalent doses is found in the table below.

Overview of active doses of dimenhydrinate and diphenhydramine in animal models of anti-emesis and abuse potential and their estimated human equivalent dose. Species Endpoint Active dose Human Compound Reference equivalent dose* Anaesthetised Diminished 1.5 mg/kg i.v. No Diphenhydramine Jaju and Cat excitability of the conversion Wang, vestibular factor 1971 complex. available Anaesthetised Diminished 2.5-5.0 mg/kg No Dimenhydrinate Jaju and Cat excitability of the i.v. conversion Wang, vestibular factor 1971 complex. available Dog Delay of onset, 50 mg/day 28 mg/kg Dimenhydrinate Gralla et reduction of p.o. al, 1979 number and (on average duration of emetic 11 episodes after mg/kg/day) irradiation of abdominal area Dog Delay of onset, 25 mg/day 14 mg/kg Diphenhydramine Gralla et reduction of p.o. al, 1979 number and (on average duration of emetic 11 episodes after mg/kg/day) irradiation of abdominal area Dog Reduced second 25 mg/day 14 mg/kg Diphenhydramine Gralla et stage of irradiation p.o. al, 1979 –induced emesis (on average 11 mg/kg/day) Rat CPP (abuse 60 mg/kg i.p. 9.6 mg/kg Dimenhydrinate Halpert et potential) al, 2003 Rat CPP (abuse 37.8 mg/kg 6 mg/kg Diphenhydramine Halpert et potential) i.p. al, 2003 Rat Locomotor 25 and 40 4 and 6.5 Dimenhydrinate Halpert et stimulation mg/kg i.p. mg/kg al, 2003 Rat Locomotor 27-37.8 4.4 – 6 Diphenhydramine Halpert et stimulation mg/kg i.p. mg/kg al, 2003 Mice CPP (abuse 30 mg/kg i.p. 2.4 mg/kg Dimenhydrinate Nguyen et potential) al, 2010 Mice CPP when co- 3 mg/kg i.p. 0.24 Dimenhydrinate Nguyen et administered with mg/kg al, 2010 low dose (7.5 mg/kg), which PAR Scientific discussion 4/21

in itself did not induce CPP *Human equivalent dose is calculated using conversion factors in Table 1 of FDA: Guidance for Industry. Estimating the Maximum Safe Starting Dose on initial Clinical Trials for Therapeutics in Adult Healthy Volunteeers, 2005. CPP: Conditioned Place Preference

The anti-motion sickness effects of diphenhydramine and dimenhydrinate was investigated in an anesthetised cat model. The neuronal activity was enhanced by physical stimulation of the labyrinth via angular (turntable) or linear (swing) acceleration, or by electrical stimulation of one of the vestibular nerve branches with a bipolar electrode. Both diphenhydramine (1.5 mg/kg i.v.) and dimenhydrinate (2.5-5.0 mg/kg i.v.) suppressed the spontaneous as well as the enhanced vestibular neuronal firing. These findings suggest that the anti-motion sickness property of diphenhydramine and dimenhydrinate may be due to diminished excitability of the vestibular complex, Jaju and Wang, 1971.

In a study in rats, dimenhydrinate was orally administered at a dose of 100 mg after rotatory stimulation. Immediately after vestibular stimulation, the animals were killed and the brain stem fixed in freshly prepared Carnoy’s fluid for 1.5 h. The brain stem of rats from the normal group were also fixed under similar conditions. Vestibular stimulation caused no significant alteration of Deiter’s cells in rats treated with dimenhydrinate. The vestibular stimulation caused a loss of basophilia in Deiter’s nucleus cells and showed that treatment with dimenhydrinate protects the cells from this effect. The results of this study suggested that the drug may act directly on the nucleus cells or indirectly by preventing stimuli from reaching the nucleus, Sikdar, 1966.

Dimenhydrinate and diphenhydramine were tested in a model of irradiation-induced emesis in dogs. Treatment with dimenhydrinate or diphenhydramine decreased both total and mean number of episodes, delayed onset and shortened the duration of first-stage emesis. Diphenhydramine also appeared to reduce second-stage emesis, Gralla et al, 1979.

In a cat model of induced emesis, dimenhydrinate showed superiority to diphenhydramine suggesting that 8-chlorotheophylline may also have a synergistic effect, Mitchell, 1950.

The antiemetic effect of dimenhydrinate and diphenhydramine was demonstrated in rat, cat and dog models of motion sickness and emesis. Furthermore, the was investigated on a cellular and electrophysiologic basis in the vestibular system. This is considered adequate.

Halpert et al, 2003 demonstrated abuse potential of high doses of dimenhydrinate and diphenhydramine in rats with diphenhydramine being most potent. Nguyen et al, 2010 demonstrated abuse potential of dimenhydrinate in mice. Especially the effect of co-administration of a low dose dimenhydrinate with cocaine is remarkable (see Table 2). The studies evaluating the pharmacological effects in rodents were not designed for dose response evaluation (Sikdar, 1966). However, the overall picture is that the dose ranges of potential abuse potential of dimenhydrinate in rodents is overlapping with the doses inducing antiemetic effects in cats and dogs. DMH was marketed as monotherapy for the first time in Denmark (June 2015) as a 50 mg sublingual tablet and can only be obtained by prescription, which limits the potential of abuse compared to a classification as an OTC drug. The Applicant’s proposal that 20 mg lozenges would present a lower risk of abuse than 50 mg tablets is supported.

The Applicant chose not to refer to the publication authored by Wang et al, 1998, where antihistamines was shown to have QT-prolongation effects. Diphenhydramine displayed intermediate potencies with respect to QT prolongation (relative EC50 values, 11-13 microM, however, this is accepted, as cardiac effects of diphenhydramine in humans is described in clinical overview 2.5.3.1.4.

The lack of any non-clinical discussion on the pharmacodynamic interaction of is acceptable, as the drug interaction profile has been discussed in the clinical overview 2.5.3.2.3.

PAR Scientific discussion 5/21

III.2

The Applicant has not performed any kinetic studies, and has therefore only presented literature references for methods for determining diphenhydramine in serum or plasma of relevant animal species. Several studies have described the pharmacokinetics of diphenhydramine in laboratory animals including the rat and guinea pig (Glazko & Dill, 1949), rabbit (Walters et al, 1993), monkey (Drach et al, 1970), dog (Wang et al, 2007) and sheep (Yoo et al, 1986 and 1990, Wong et al, 2000). Pharmacokinetic studies on 8-chlorotheophylline are scarce, only one study is presented (Walters et al, 1993).

Bioanalytical methods Various bioanalytical methods have been used for the quantification of diphenhydramine. Only one study shows pharmacokinetic data of 8-chlorotheophylline. Methods of extraction seem continuously to be liquid-liquid extraction and this technique was also used with the recent LC-MS method published in 2007. Separation methods ranged from counter current extraction to gas- and then liquid chromatography in line with development of bioanalytical technologies in general.

Pharmacokinetic studies employed bioanalytical methods, which were state of the art at the time of conduct. This is acceptable. Not all publications stated the exact bioanalytical method used. This is also acceptable as the drug has been used clinically and extensive knowledge from the clinical setting has been obtained - including relatively recent published pharmacokinetic studies in humans (Clinical Overview 2.5.3.2.).

Absorption The pharmacokinetics of diphenhydramine was studied in both pregnant and non-pregnant ewes. Non- linear kinetics with regard to clearance and volume of distribution was observed in non-pregnant ewes, however within the dose range of 25 to 100 mg, there was no significant difference for T1/2 or Vd. T½ was determined to be in the range of 34 to 68 minutes pregnant and non-pregnant ewes and 46 min in fetus. The average clearance (5 L/h/kg) remained relatively unchanged regardless of the dose (Yoo et al, 1990). Pharmacokinetics of oral (capsule with and ginger) administration of diphenhydramine was also briefly evaluated in dog, where time for maximal plasma concentration was 3.2 hour and t½ was 4.4 hours.

Propylene glycol administered i.m. prior to exposure to dimenhydrinate is apparently inhibiting clearance of both diphenhydramine and 8-chlorotheophyline in rabbits. Propylene glycol is an inhibitor of CYP2E1 (Thomsen et al, 1995). 2E1 is most likely not responsible for the metabolism of diphenhydramine in humans. On the other hand, 8-chloro-theophylline is probably a substrate of CYP2E1, but may also be metabolised by CYP1A2 in humans.

Distribution The distribution of diphenhydramine was investigated by Glazko & Dill in 1949 in rat and guinea pig using techniques of both chemical and radioactive quantification. The concentration in tissue followed this decreasing order lung>spleen>kidney>brain>liver>muscle in both rat and guinea pig after subcutaneous administration. Tissue concentrations were higher in guinea pig as compared to rat. Discrepancies in tissue content measured by the chemical or radioactivity was observed for blood and to minor degree heart in guinea pig. This could be an indication of presence of metabolites in blood not detected by the colorimetric method used for quantification in the chemical method.

Diphenhydramine degradation was evaluated in tissue homogenates. The showed the greatest activity. Rat lung showed more activity than the corresponding tissue in the rabbit or guinea pig, and some activity is also evident in rat (Glazko & Dill, 1949).

Protein binding was determined by equilibrium analysis of plasma samples from in vivo studies (ex vivo) in non-pregnant ewes (Yoo et al,) and lambs (Wong et al, 2000). Protein binding is reported as the ratio between concentration in plasma and buffer after equilibrium analysis. The mean diphenhydramine plasma unbound fraction for lambs, two weeks old (0.15±0.10), lambs, two months old (0.15±0.06) and adult sheep (0.12 ±0.07) were not significantly different. However, diphenhydramine unbound fraction PAR Scientific discussion 6/21 for both groups of postnatal lamb and adult sheep were significantly lower than values previously observed for fetal lamb (0.30±0.09) (Wong et al., 2000). Percentage unbound can be calculated from the table below and is 28% for animal No. 1. Therefore, diphenhydramine is considered not to be extensively bound to plasma proteins.

Metabolism Diphenhydramine undergoes oxidative metabolism via N-demethylation to the secondary nordiphen- hydramine, desmethyldiphenhydramine) and subsequently to the primary (dinordiphenhydramine) amine, both excreted in the urine. Dinordiphenhydramine is further oxidized to diphenylmethoxyacetic acid, which appears to become conjugated with glycine (formation of diphenylmethoxyacetic acid- glycine amide) or glutamine prior to . Another reported metabolic step of diphenhydramine is the biotransformation to diphenhydramine- N-oxide. The metabolite diphenylmethoxyacetic acid was shown to be slowly cleared in monkey and even slower in dog where no decline in plasma concentration was observed after 24 hours.

Excretion Clearance and volume of distribution of diphenhydramine was varying in lambs and sheep of different age, however the resulting terminal half-life was faster in fetus and lambs as compared to adult sheep.

The excretion mechanisms of diphenhydramine is not entirely known. However studies in rats and sheep of varying age provides some information. Glazko & Dill’s study from 1949 shows that approximately one third of the dose is excreted through the urine within 71 hours in the rat. The remainder of the dose was not accounted for. The study by Drach et al, 1970 showed that the rat was not employing the diphenylmethoxyacetic acid route of elimination. However, this was shown to be a major route in monkey and dog. Furthermore, Wong et al, 2000 showed that diphenylmethoxyacetic acid metabolites was slowly cleared in lambs two weeks of age compared to lambs two months of age. Diphenhydramine metabolism in humans is similar to the dog and the metabolite DPMA has much longer half-life than parent compound (>16 hours for DPMA versus about 9 hours for diphenhydramine). A clinical study in children revealed no concern with regard to changes in clearance between age groups from 2 to 17 years.

III.3 Toxicology

8-chlorotheophylline Molar similarity of LD50 of dimenhydrinate and diphenhydramine is striking, see Table below.

PAR Scientific discussion 7/21

This indicates that 8-chlorotheophylline does not contribute to the toxicity of diphenhydrinate when administered in combination, at least as a single dose. LD50 is higher for 8-chlorotheophylline than for theophylline and lower than , two other methylxanthines. No repeat-dose toxicity data could be retrieved for 8-chlorotheophylline from public domain, but Applicant considers repeat-dose toxicity data on caffeine and theophylline to be of some relevance due to the resemblance in molecular structure and pharmacology. NOAELs from repeat-dose toxicity studies on caffeine provides adequate safety margins whereas safety margins to human exposure of 8-chlorotheophylline using theophylline is limited. Data presented by McColl et al, 1955 indicates that theophylline is slightly more potent than 8- chlorotheophylline.

As for genotoxicity, 8-chlorotheophylline did not seem to increase genotoxic potential of diphenhydramine alone and 8-chlorotheophylline is not listed as carcinogen in the US (2009).

Regarding reproduction and developmental toxicity, an overview of data from caffeine and theophylline was submitted. These data provided only limited safety margins to foetal/maternal toxicity or teratogenicity. In any case, the proposed medicinal SmPC informs that dimenhydrinate should only be used during pregnancy when benefits outweigh the possible risks. Overall, limited data on 8- chlorotheophylline could be retrieved from public domain.

Diphenhydramine

Single dose toxicity Single dose toxicity was evaluated in a range of common laboratory species including juvenile rat (4 days old) and pregnant mice after oral, subcutaneous, intraperitoneal and intravenous routes of administration. The oral and subcutaneous routes show much higher LD50s compared to especially the intravenous route. The range of LD50s for the intravenous route is 10 to 46 mg/kg and for the oral route 167 to 856 mg/kg. LD50 in mice was 31 mg/kg given via IV administration, whereas the lowest pharmacologically active dose in mice via the IV route was 3 mg/kg, hence a safety margin is provided.

Repeated dose toxicity Repeat dose toxicity studies of diphenhydramine in rats and mice were published by U.S. National Institute of Health in 1989 (NTP, 1989). Animals were dosed orally via the diet. Several signs of toxicity was evident including dose-related decrease in body weight gain in both the 14-day and 13-weeks studies (see tables of rat data below). Food intake decreased in a dose-related manner in the 14-days study and increased in the 13-weeks study. Clinical signs such as hyperactivity and sensitivity to sound were observed in the 14-day study and to a lesser degree in the 13-weeks study. Histopathological lesions were not found in the 14-days study. In the 13-weeks study cytoplasmic vacuolization of the liver, characteristic of fat accumulation, were observed in male and female rats receiving 313-2500 ppm. The severity of this change increased with increased dose. No compound-related histopathologic effects were observed in mice. Since no toxicokinetic data is provided for these repeat-dose studies, safety margin cannot be calculated. NOAEL for the 13-weeks study was 625 ppm for male rat and 313 ppm for female rat.

PAR Scientific discussion 8/21

Toxicokinetics Toxicokinetic documentation of exposure in repeat dose toxicity studies was sparse. Only the two doses (313 and 625 ppm) used in the two-year carcinogenicity study were evaluated for plasma exposure in rats after 30 days of dosing via feed. This analysis showed that most samples contained diphenhydramine in concentrations below limit of quantification (1 ng/ml).

Albert et al, 1975 published a study on the pharmacokinetics of diphenhydramine in man where peak plasma concentration levels were approximately 50 ng/ml after an oral dose of 50 mg diphenhydramine. Assuming dose-linear pharmacokinetics, this corresponds to plasma concentrations of approximately 40 ng/ml for a single dose 40 mg dimenhydrinate (maximum single dose stated in SPC, Posology). Hence, PAR Scientific discussion 9/21 clinical relevant exposure could not be documented for the two dose levels used in the two-year study in rat.

Genotoxicity Diphenhydramine hydrochloride has been tested for mutagenicity in a variety of bacterial and animal systems, and with one exception (chromosomal abberations), the results have been uniformly negative (NTP. TR 355; 1989).

The mutagenicity of diphenhydramine was evaluated using the Salmonella typhimurium test. In this study, eight drugs that are amines or amides and which interact with nitrous acid to form potentially carcinogenic and mutagenic N-nitroso derivatives were tested for mutagenicity. None of the compounds was mutagenic alone, with or without liver S9 activation. After reaction with nitrite in acetic acid solution, diphenhydramine gave mutagenic products with or without activation, but only to strain TA98 and not strain TA1535, TA1538 or TA100 (Andrews et al., 1984). The combined data for diphenhydramine are equivocal, hence the genotoxic potential of diphenhydramine is considered to be low.

Carcinogenicity The carcinogenicity of diphenhydramine was evaluated in rats and mice at doses of 313 and 625 ppm in feed without evidence of clinical relevant exposure, hence this study is not considered eligible for carcinogenic assesment. Furthermore, after two years, tumour incidence was similar to control groups (NTP, 1989). Another group (Lijinsky, 1984) dosed diphenhydramine at 2000 ppm in feed with or without sodium nitrite and found that diphenhydramine alone did not increase incidence of tumours. However, co-administration with sodium nitrite induced liver neoplasms. Dosing diphenhydramine with compounds capable of nitrosation of diphenhydramine in the acidic environment of the stomach may increase the carcinogenic risk. This was confirmed by the in vitro genotoxicity study by Andrews et al., 1984. However, since dimenhydrinate is not supposed to be dosed continuously, but only infrequently, the carcinogenic potential of diphenhydramine is considered to be low.

Reproductive and developmental toxicity Dimenhydrinate has previously been used as an antiemetic for nausea and vomiting in pregnancy. Therefore a range of studies of the reproductive and developmental effects of diphenhydramine and dimenhydrinate has been published, also recently (Moraes, 2004, Desdicioğlu et al, 2011 and Fazliogullari et al, 2012). See table below for an overview of evaluated end-points and findings.

Overview of studies of dimenhydrinate and diphenhydramine reproductive and developmental toxicity Study type/ Species; Route & Dosing period Findings Parameters tested Reference Number dose where no effects Female/ were observed group Male fertility Not tested NA NA NA NA Female Rabbit GD 8-16 None No significant fertility N = 24 Dimen- difference from McColl, 1967 hydrinate control group in % 100 mortality, Mean mg/kg/day litter size, Mean p.o. litter weight, resorption sites per litter and pregnancy rate Embryo-foetal Rabbit GD 8-16 None No significant development N=24 Dimen- difference from McColl, 1967 hydrinate control group with 100 regard to gross mg/kg/day defects, skeletal p.o. defects, ribs or

PAR Scientific discussion 10/21

cardiovascular defects Embryo-foetal Rat Dimen- Explanted on Dose-dependent development embryos hydrinate GD 9.5, growth retardation in No difference in Fazliogullari, in vitro 2.5-20 exposed for 48 crown-rump length nuclear DNA 2012 N=10 µg/mL hours in and yolk sac fragmentation culture diameter at from control medium concentrations equal to and higher than 5 µg/mL Somite number was decreased at all concentrations including 2.5 µg/mL, see figure and picture in IV.5.2. below.

Peri & Rat Diphen- GD 0-21 Slightly accelerated Gonadal hormone postnatal N=6 hydramine pinna unfolding, eye levels and weight Chiavegatto, 20 opening and delay of gain were similar 1997 mg/kg/day testes descend and to control. s.c. vaginal opening. Slightly accelerated righting reflex and geotaxis development, see table below. Peri & Rat Diphen- GD 16-21 Delayed testis Offspring physical postnatal N=12 hydramine descent, altered parameters, female Moraes, 2004 20 patterns of sexual open-field mg/kg/day behavior in male measures, sexual s.c. offspring, (see table behavior and below). Increased striatal striatal DA, neurochemical decreased striatal measurements DOPAC and reduced similar to control DOPAC/DA, HVA/DA and 5HIAA/5-HT ratios also in male offspring. Peri & Rat Dimen- GD 1-7 to Increase in mean postnatal N=10 hydrinate imitate the morphometric Desdicioğlu, 115 first trimester parameters during 2011 mg/kg/day of pregnancy the postnatal period i.m. in humans in the DMH group was less than the control group. Total increase in maternal weight gain in pregnant rats were less than in control group (27 g vs. 50 g)

The effects of of 100 mg/kg/day dimenhydrinate on GD 8-16 on female reproduction and embryofetal development in rabbit was evaluated in 1967 by McColl et al. No significant effects different from control were observed. However, in several rat studies (s.c. or i.m.), effects were observed PAR Scientific discussion 11/21 on post natal development in pups from dimenhydrinate or diphenhydramine treated dams both after treatment in the early, late or whole time of pregnancy. Effects observed were decreased growth, delayed onset of puberty, disturbances in sexual behaviour, in striatal dopamine system and decreased maternal body weight gain during pregnancy. All the in vivo studies were conducted in one dose only and without exposure detmination. Therefore quantitative risk assesment is difficult and can only rely on allometric scaling principles.

One study of embryofetal development was conducted ex vivo. Rat embryos were cultured in vitro from gestational day 9.5 to 11.5, which, as stated by the authors, is the critical period of organogenesis in the rat, equivalent to 3–6 weeks after fertilization in human embryos. In this study, the NOAEL was 2.5 µg/mL dimenhydrinate for all end points except somite number, see also section IV.5.2. This concentration is 50 times higher than clinical relevant plasma concentrations of diphenhydramine (40 ng/ml).

It should be mentioned here, that the indication to be pursued for Dimenhydrinat ”Trimb” (20 mg dimenhydrinate in lozenges), is motion sicknes and not nausea and vomiting during the first trimester of pregnancy.

Fertility and early embryonic development Evaluation of male fertility after exposure to dimenhydrinate or diphenhydramine was not assessed. Female fertility was only tested in rabbit after p.o. administration and seem not be impacted by exposure to dimenhydrinate (McColl et al, 1967).

Rabbit may not be susceptible to effects of oral administration of dimenhydrinate as no effects were observed on embryofetal development either. Fertility was not tested in rat. Male fertility may be impacted, since male rat pups exposed to dimenhydrinate prenatally showed delayed puberty and altered patterns of sexual behaviour.

Embryo- foetal development Rat offspring seem to be susceptible to diphenhydramine and dimenhydrinate during pregnancy. Especially, the study by Desdicioğlu et al, 2011, where pregnant rats were dosed with dimenhydrinate 115 mg/kg/day i.m. during GD 1-7, significant effects were observed on both dams and offspring. In this study, the increase in mean morphometric parameters during the postnatal period in the dimenhydrinate group was less than the control group and total increase in maternal weight gain in pregnant rats were less than in control group (27 g vs. 50 g).

One study of embryofetal development was conducted ex vivo. Rat embryos were explanted on day 9.5 and exposed to dimenhydrinate for 48 hours prior to morphological examination. Statistically significant dose-dependent growth retardation in total morphological score, crown-rump length and yolk sac diameter were observed at concentrations equal to and higher than 5 µg/mL (Fazliogullari et al, 2012). Somite number was also decreased at 2.5 µg/mL.

Morphological studies of rat foetuses from dams treated with dimenhydrinate indicate that the compound induces embryonic growth retardation. Safety margin cannot be assessed, as only one dose level was used in the in vivo study, hence no NOAEL is available. The ex vivo study showed NOAEL of 2.5 µg/mL of most morphological parameters except somite number for which no NOAEL was established.

Prenatal and postnatal development, including maternal function Chiavegatto et al. (1997) analyzed the influence of prenatal exposure to diphenhydramine of rats on maternal behavior and milk production of dams; physical and reflexologic development of offspring; and long-term effects on open field behaviors and gonadal hormone levels in offspring. Female pregnant rats were injected subcutaneously, daily, with 20 mg/kg/day diphenhydramine or saline from embryonic day (E) 0 to 21. Neither maternal behavior nor milk production was affected by diphenhydramine treatment. Treated offspring showed a slightly accelerated pinna unfolding, eye opening, and a delay of testes descent and vaginal opening, see table below. Both righting reflex and negative geotaxis development were accelerated, but prenatal exposure to diphenhydramine did not modify offspring PAR Scientific discussion 12/21 locomotor activity. The findings suggested that prenatal diphenhydramine exposure influences physical and reflex development of rat pups (Chiavegatto et al., 1997).

Chiavagettos study in rats, dosed a lower dose (s.c. 20 mg/kg/day diphenhydramine) than other studies, showed only slight effects on end-points. Nevertheless, some were significant, such as delay in vaginal opening and testes descent. Rat offspring from diphenhydramine-dosed dams also seemed to be faster in pinna unfolding and eye opening and to develop righting and startle reflex, as compared to control offspring.

Studies in which the offspring (juvenile animals) are further evaluated The effects of 20 mg/kg diphenhydramine administration to pregnant rats was assessed on gestation days 16–21, a critical period for sexual differentiation and CNS maturation (Moraes et al., 2004). Diphenhydramine treatment decreased maternal body weight gain during the treatment period. Offspring physical parameters were not altered in the treated group, and no significant treatment-related changes were found in female openfield measures, sexual behavior or in striatal neurochemical measurements. However, delayed testis descent an altered patterns of sexual behavior, (see table below) occurred in male offspring accompanied by increased striatal DA, decreased striatal DOPAC as well as reduced DOPAC/DA, HVA/DA and 5-HIAA/5-HT ratios. Taken together, these data suggest that exposure to diphenhydramine during the fetal period of rat development altered postnatal CNS maturation and sexual development of male offspring via changes in striatal bioamine systems (Moraes et al., 2004).

The study of Moraes et al, 2004 indicate male offspring to be particular susceptible to effects of diphenhydramine, where significant effects on e.g. sexual behaviour was shown in adult offspring from dams dosed with diphenhydramine 20 mg/kg/day during GD 16-21. The human equivalent dose of 20 mg/kg/day s.c. in rat is 20 * 0.16a = 3.2 mg/kg. The human maximum recommended daily dose at any one time is 140 mg p.o. corresponding to 140 mg/70 kg = 2 mg/kg/day. There is a discrepancy between the between rat and human, nevertheless, no safety margin can be provided at the level of dosing and no exposure of diphenhydramine in dams was reported in the study. a: FDA. Guidance for Industry. Estimating the Maximum Safe Starting Dose on initial Clinical Trials for Therapeutics in Adult Healthy Volunteeers.

Toxicological studies of excipients The excipients of dimenhydrinate lozenges are isomalt, saccharin sodium, masking, anise flavor, mouth watering and purified water. These are commonly used in medicinal products and described in pharmacopoeias and textbooks concerning inactive substances used in pharmaceutical industry.

Toxicity studies were performed with the excipients isomalt and saccharin sodium included in the Applicant’s formulation. Toxic effects in animals were rare and when reported, corresponded to a dose range considerably higher than that employed in the formulation. Accordingly, the amount of each substance needed to reach the LD50 was much higher than that found in the formulation. A review of publicly available safety data on the major constituents in the mixtures of flavouring agents “masking”, anise flavor” and “mouthwatering” was provided. No data gave rise to concerns when comparing to daily intake of Dimenhydrinat ”Trimb” lozenges.

III.4 Ecotoxicity/environmental risk assessment (ERA)

Applicant claims that, although no environmental impact is expected from the introduction of the Applicant’s product; due to the fact that dimenhydrinate is a substance with well-established use, no increase in the consumption is expected, and therefore, an Environmental Risk Assessment (ERA) is not deemed necessary.

Berninger et al, 2011 states the diphenhydramine, has been specifically identified in several major environmental compartments (water, sediment, tissue). In streams receiving significant discharges of treated municipal effluent, diphenhydramine has been detected in the water at concentrations ranging from 0.01 to 0.10 mg/L. In the sediment, diphenhydramine concentrations were PAR Scientific discussion 13/21 much higher (20–50mg/kg), two and three orders of magnitude higher than associated water concentrations. Perhaps most important, diphenhydramine has been found in the tissues of fish. Ramirez et al. found diphenhydramine in the muscle tissue of fish living downstream of a North Texas municipal effluent outflow at a mean concentration of approximately 1mg/kg. These concentrations found in US are already at or higher than the action limit of 0.01 µg/mL as stated in Guideline on the environmental risk assessment of medicinal products for human use (EMEA, 2006). Berninger et al, 2011 reports a NOEC of 0.8 µg/L for reproduction in daphnia magna for diphenhydramine.

According to Directive 2001/83/EC, Applicants are required to submit an ERA also for applications under Art. 10a-well established use/bibliographic medicinal products.

The Applicant has not provided an environmental risk assessment report. The Applicant has committed to perform Phase 1 of environmental risk assessment on dimenhydrinate for submission 2 months post approval date. This should include screening for PBT and PECsurfacewater. If the environmental risk assessment enter Phase 2, then the Applicant should be committed to submit a Phase 2 environmental risk assessment including all relevant studies according to the ERA guideline* within 24 months from approval date. Applicant should submit via appropriate variation procedure. *Guideline on the environmental risk assessment of medicinal products for human use. EMEA/CHMP/SWP/4447/00 corr 2.

IV. CLINICAL ASPECTS

IV.1 Introduction

Dimenhydrinate and its active substance diphenhydramine are sedating antagonist and have been used since the 1950'ies to alleviate nausea, vomiting, and associated with motion sickness. The substance has been in well-established used during the last almost 70 years, and this is sufficiently justified in the Applicant’s clinical overview. The anti-nauseant and antiemetic effect is well established and this has also been sufficiently justified in the clinical overview presented by the Applicant.

IV.2 Pharmacokinetics

The Applicant has provided information regarding the pharmacokinetic profile of dimenhydrinate and its active component diphenhydramine. The majority of studies are based on the formulations: tablets, oral suspension, and IV. A single study has been reported using chewing gum as formulation in seven volunteers, to describe the plasma kinetics of a low dose of dimenhydrinate. Overall, it is agreed that the pharmacokinetic properties of diphenhydramine are well-established in several oral formulations, but importantly, there is no pharmacokinetic study of the formulation which the Applicant seeks approval for.

To account for this the Applicant presented a study on the pharmacokinetics of the proposed product during a clinical trial performed in 2017. The Study G14-12 was a randomised, open-label, single-dose, two-period crossover study to assess the comparative bioavailability of two formulations of 20 mg of dimenhydrinate administered to healthy volunteers under fasting conditions. Dimenhydrinate was formulated as a chewing-gum and a lozenge. Importantly, the Applicant confirms that the formulation of the lozenge is similar to the applied-for product. The individual PK profiles of diphenhydramine for the 36 healthy volunteers included in the study show that for the lozenge, diphenhydramine can be measured in plasma 15 minutes after starting to suck the lozenge, which indicates a fast raising in the concentrations. Measurable concentrations were observed in all the volunteers 24 hours after dosing, with the majority presenting measurable concentrations 48 hours after dosing. The mean Cmax was 19.2 PAR Scientific discussion 14/21 ng/mL and was reached approximately 2 hours after dose intake. The elimination half-life observed during the study was approximately 10 hours. The primary objection of the study was to compare the pharmacokinetic profiles of the lozenge with the chewing-gum formulation. For dimenhydrinate, Cmax was notable higher when administered as a lozenge as compared to the chewing-gum formulation (18.41 ng/mL vs. 7.61 ng/mL). The ratio (CI90%) was 242.13 (224.13-261.58). Similar pattern was observed for the AUC0-t: 63.40 vs. 172.02 ng*h/mL with a ratio of 271.32 (CI90%: 253.28-290.63). Thus both Cmax and AUC0-t was substantially higher for the lozenge formulation and not at all bioequivalent with the chewing-gum formulation. The Applicant has taken this into account, when deciding the recommended dose (see below). On the contrary to the values for dimenhydrinate, bioequivalence between the two formulations was found for the active metabolite 8-chlorotheophylline; the ratios (CI90%) for Cmax and AUC0-t being 114.43 (107.34-121.99) and 104.29 (88.75-122.55), respectively.

Due to the findings for ‘non-bioequivalence’, the Applicant was requested to present further justification for extrapolation of the pharmacokinetic values for chewing-gum formulation to the present lozenge formulation taking into account that there are significant differences for Cmax and AUCt-0. In their response, the Applicant justified the extrapolation of the pharmacokinetic values for chewing-gum formulation to the present lozenge formulation by presenting box-plots of values presented in the literature for Cmax and AUC for the dose normalised to the to-be-marketed 10.8 mg lozenge formulation. The values for Cmax er similar and of no concern. The values for AUC appear to be slightly higher for the test-product (lozenge formulation), though the values are overlapping. The literature values includes 14 references, mostly pharmacokinetic studies, where healthy volunteers have received higher than the doses recommended in the present posology (often up to 50 mg of dimenhydrinate). Overall, and most importantly the efficacy and safety of the product in equi-potent doses administered orally as tablets or chewing gum has been well established and therefore, the issue was not be pursued further.

The formulation (dose-normalised to diphenhydramine base) is compared with a wide range of oral formulations (solutions, capsules, tablets and chewing gums; and the Applicant argues that the product does not deviate from the literature data in terms of overall exposure (AUC), Cmax and also Tmax. The Applicant argues that lozenges, as well chewing gum formulations, are expected to have a larger residence time in the oral cavity, when compared with other solid oral formulations, like capsules and tablets. Therefore, the lozenge formulation is expected to have a higher bioavailability in comparison to regular oral formulations because a higher fraction of the drug is expected to be absorbed through the oral mucosa, thereby avoiding the first pass metabolism. The Applicant therefore proposes a product with a dose at the lower end of the conventional oral doses.

According to the Applicant, if a subject is dosed with 2 lozenges dimenhydrinate concentrations reach the level of the average conventional dose range of dimenhydrinate (50 mg) solid oral doses, or even higher doses. Predicted plasma concentrations after dimenhydrinate 40 mg administration (two lozenges), corresponding to 21.6 mg of diphenhydramine; simulations (n=1080) with inter-subject variability and ruv, the Applicant argues that the proposed formulation completely blends with the PK of other diphenhydramine salts (hydrochloride and theoclate salts), not only in terms of overall exposure, but also maximum concentration and time to reach the maximum concentration. Therefore, the formulation can be reliably bridged to the literature, and extrapolation of efficacy and safety can be done.

Dimenhydrinate is unstable in gastrointestinal pH and thus rapidly releases its diphenhydramine moiety after oral administration. It is well-established that diphenhydramine is highly protein bound, with less than 2% of unbound fraction, independent of the concentration (within the range 113 to 1130 ng/ml). The Applicant argues that Pgp-mediated efflux appears not to be a limiting factor to the CNS penetration.

Diphenhydramine exhibits a large first-pass effect with about 50% metabolism occurring before the drug reaches the general circulation following oral administration. The drug is metabolised by demethylation to DMDP followed by demethylation and oxidative deamination. Another reported metabolic step of diphenhydramine is the biotransformation to diphenhydramine-N-oxide.

PAR Scientific discussion 15/21

Total AUC for DMDP after IV diphenhydramine was significantly correlated with clearance of the parent compound. Furthermore, total AUC for the metabolite was greater following oral as opposed to intravenous dosage. The apparent disappearance half-life of DMDP was similar to that of the parent compound, suggesting that its apparent rate of disappearance is limited by its rate of formation (“flip- flop” effect).

Regarding the elimination of diphenhydramine, it has been shown that diphenhydramine undergoes extensive hepatic metabolisation primarily via CYP2D6. Only 1-3% is eliminated unchanged in the urine. Urinary excretion of the total diphenhydramine metabolites represented about 64% of the dose in single dose studies, and 49% after multiple doses. The T1/2 is estimated to range between 3-9 hours. All relevant pharmacokinetic information is included in section 5.2 of the SmPC.

The Applicant provides information that no dose adjustment is needed among elderly, or children, but dose adjustment of the total daily dose is advisable among patients with hepatic impairment due to the extensive hepatic metabolisation. This is in accordance with the observations of Meredith et al. (1984) and considering the impact of the first pass metabolism after oral administration, a reduction in 50% of the clearance capacity is considered as the worst-case scenario. This worst-case scenario was selected for the initial simulations (simulations performed in a population of 1080 subjects). The overall exposure is expected to increase from 171 ng*h/mL to about 288 ng*h/mL while the peak concentration will raise from 19.1 to approximately 21.7 ng/mL. The increase in the Cmax can be considered as non-significant and lacking clinical relevance. The increase in the overall exposure after one single dose is equally non- relevant, but the accumulation is more likely to occur after multiple doses.

For adults, a total daily dose of 4 lozenges (2 lozenges within a 10 minutes interval between each one, repeated after 6 hours) is by the Applicant considered to be an adequate dosing regimen for the maximal proposed period of 7 days. The Applicant states that the self-titration concept of the formulation is very helpful to avoid the undesirable sedation. This is due to the fact that the lowest effective dose is advised. Consequently, a subject experiencing sedation will reduce the intake in the next dose. This will be further discussed in the clinical (efficacy-safety) part of the overview. The Applicant argues that pharmacokinetic drug-drug interactions are a risk when treating patients with diphenhydramine due to its properties as a substrate of CYP2D6 as well as being a potent inhibitor of CYP2D6.

With regard to the clinical pharmacology relevant for the safety of the product, the toxic value of diphenhydramine is reported to be above 1000 ng/mL. The pharmacological actions leads to drowsiness, sedation, and mental performance impairment, which correlates to the plasma concentration. The Applicant reports that a effect has been shown when administering a single oral dose of 50 mg. Moreover, the Applicant argues that a single 25 mg dose of diphenhydramine (corresponding to a double-dose of the formulation, which the Applicant applies for) is indistinguishable from placebo in terms of sedative, performance impairing or memory, or impairing properties. The evidence of this statement is from a single study, which was a double-blind randomised placebo-controlled cross-over study of 37 young and elderly males and females.

The most predominant lack of information is that there are no published data to support a bridging from existing formulations to the proposed lozenge formulation. This was initially considered a major limitation of the application – both in terms of pharmacokinetic properties, as well as for efficacy and safety perspectives. Therefore, the Applicant studied the pharmacokinetics of the proposed product during a clinical trial performed in 2017. Evaluation of this study is discussed above.

IV.3 Pharmacodynamics

Diphenhydramine is considered a first-generation, reversible antihistamine included in the class of drugs that also presents certain anti-muscarinic activity. Diphenhydramine penetrates the blood– brain barrier and interferes with neurotransmission by histamine at central H1 receptors. The H1-receptors present in all of the major parts of brain. Results indicate that dimenhydrinate exerts

PAR Scientific discussion 16/21 its effect in motion sickness by reducing the vestibular and visual influx and by partly inhibiting the integrative functioning of the vestibular nuclei.

The motion sickness drugs seem to assist the cerebellum by diminishing impulses from various orientation reflexes in order to preserve the functional capacity of the central nervous system. Diphenhydramine also has affinity for muscarinic and adrenergic receptors. Therefore, central antagonism of may also play a role in preventing motion sickness. The effects of diphenhydramine on the H1 histamine receptors in the brain can potentially lead to adverse CNS symptoms such as drowsiness, sedation, , fatigue, and headache. The Applicant states, that sedation is diphenhydramine is dose dependent.

The pharmacodynamic effects of diphenhydramine as anti-nauseant and antiemetic are well-established, as well as the safety pharmacology. The Applicant presents data showing, that the QTc interval is significantly prolonged among 126 patients with diphenhydramine overdose, compared to a control group. Furthermore, the heart rate was significantly increased. No patient experienced torsade de pointes ventricular .

IV.4 Clinical efficacy

Dimenhydrinate and its active substance diphenhydramine are sedating histamine H1 receptor antagonist and have been used since the 1950'ies to alleviate nausea, vomiting, and dizziness associated with motion sickness. The substance has been well-established used during the last almost 70 years, and this is sufficiently justified in the Applicant’s clinical overview. The anti-nauseant and antiemetic effect is well established and this has been sufficiently justified in the clinical overview presented by the Applicant.

The Applicant has sought approval for dimenhydrinate 20 mg in a formulation not previously marketed, namely a hard-candy lozenge formulation. No studies have been performed using this formulation; the majority of previous studies include the substance in oral formulation or tablets. Only one randomised trial has been conducted using a chewing gum formulation showing a significantly reduction in of sodium and sweat after caloric stimulation. The main efficacy outcome was a dimenhydrinate induced significant reduction in secretion of sodium in sweat after caloric stimulation. Sweating is one of the common symptoms of motion sickness and as such a relevant pharmacodynamics surrogate endpoint. It is more questionable as a clinical endpoint. The Applicant has conducted three in vitro studies showing similar dissolution of the chewing gum formulation and the hard-candy lozenge formulation. Furthermore, the Applicant has performed a pharmacokinetic study of the proposed product during a clinical trial in 2017 (please see pharmacokinetic section above).

Two systematic reviews are presented to substantiate the efficacy of dimenhydrinate in preventing motion sickness. The first systematic review was from 1966 in which the authors found 15 studies encompassing around 5,000 subjects. The authors made up an effectiveness measure based on the percentage of patients who reported motion sickness in the placebo group and in the active substance group. The authors found 72% percentage “effectiveness”. However, this is prone to bias. It is not reported whether the studies were blinded; the primary outcome was not reported other than “motion sickness”, which makes impossible to assess potential differences among studies included. The Applicants argues that although the report of Wood et al. (1966) did not presented the results with the statistical analysis employed in the current meta-analysis, the work was capable to gather more than 5000 subjects on the efficacy of dimenhydrinate and quantified the effect of this drug on the prevention of motion sickness. This argument is acknowledged.

A Cochrane review from 2011 on the effects of scopolamine in preventing and treating motion sickness is also mentioned in the clinical overview. This study also included dimenhydrinate indirectly, by combining the studies in which both drugs were compared. The Applicant reports that the results confirmed the efficacy of scopolamine against placebo and the drug was considered being as effective as dimenhydrinate. However, this approach to indirectly assess the effectiveness of a treatment is not

PAR Scientific discussion 17/21 without problems. The most predominant is the non-systematic approach. Only selected studies with both scopolamine and dimenhydrinate is mentioned. This could potentially induce selection bias. The reported evidence of efficacy in the applied therapeutic indication is primarily from old randomised trials. The summary of the most relevant publications mentioned by the Applicant presents 8 studies conducted between 1949 and 1956. The route of administration was oral and rectally applied, and the doses are reported to be between 50-100 mg dimenhydrinate. This is substantially stronger than the applied dose.

Based on the PK modelling and the literature, the Applicant concluded that with one lozenge, the lowest efficacy threshold of 14.5 ng/mL is expected to be overcome for 84-90% of the subjects (84% for the adolescents aged 16 years and 90% for children aged 6 and 8 years). With this dose, <1% of all patients (regardless of age) will reach the safety threshold of 50 ng/mL. With two lozenges, >99% of all patients (regardless of age) with reach the 14.5ng/mL efficacy threshold but 19% of the adolescents (16 years) and 28-31% of the children (age 8 and 6 years) will have concentrations exceeding the safety threshold of 50 ng/mL. Thus, it appears that when treated with two lozenges additional approximately 10% of the children age 6 and 8 years old will reach the motion sickness efficacy threshold of 14,5 ng/mL (from approximately 90% to almost 100%) but also additional approximately 30% will reach the safety threshold of 50 ng/mL (from approximately 0-1% to approximately 30%). This combined with a doubling (or almost doubling) of the AUC in children aged 6 and 8 years as compared to adults (344 ng/mL and 310 ng/mL vs. 171 ng/mL for 6 years, 8 years and adults, respectively) needed further discussion of the benefit-risk assessment for children aged <16 years old. Therefore, as a major objection, the Applicant was requested to further justify the better benefit-risk-balance of recommending 1-2 lozenges (20-40 mg) (rather than 1 lozenge [20 mg]) per dose to the younger age groups. The Applicant was also requested to discuss if the posology should recommend 1 lozenge (20 mg) per dose for children <16 years. In their response, the Applicant acknowledged the doubtful benefit-risk assessment for the children and agreed to amend the posology for children and adolescents to 20 mg of dimenhydrinate (1 lozenge) in each dose.

IV.5 Clinical safety

The summary of safety demonstrates the safety profile of dimenhydrinate, by discussing clinical trials, systematic reviews, and case reports. Nevertheless, the Applicant states that side effects of anti-motion sickness drugs are very difficult to appraise critically, because many of the more subjective symptoms of motion sickness are indistinguishable from the common complaints that follow with medication. The Applicant presents data from systematic reviews and meta-analysis of diphenhydramine safety. The adverse effects include effects, CNS depression, dizziness, hypersensitivity reactions, and impaired psychomotor functions. Serious adverse reactions including tachycardia, palpitation, ECG changes , hypotension, and hypertension are rare and mostly reported in association with overdose.

Eight clinical, randomised, placebo-controlled trials of dimenhydrinate vs. placebo are presented in the clinical overview. The time-period was spanning from 1949 to 2002, and was primarily conducted among soldiers. The earlier studies were not blinded, thus potentially biased.

A major limitation in the presentation is the lack of safety data on the proposed lozenge formulation. To account for this, the Applicant has provided literature on diphenhydramine dosing and the association to adverse events. A single study on the use of dimenhydrinate in a chewing gum is presented. This was a study conducted on 24 healthy subjects, where a lower incidence of sedating effects was observed compared to tablet administration. The proposed mechanism of this is due to the divided dose principle of the chewing gum. Further, a population pharmacokinetic (pop-PK) study has been conducted using information from a recent study on diphenhydramine in children and adolesents showing that sedative properties (defined as a plasma concentration above 100 ng/mL) cannot be reached with one lozenge, and the consumption of 2 might lead to 0.1% of the subjects to exceed 100 ng/mL. The Applicant presents data, which show that 0.1% of the subjects will experience a Cmax above 100 ng/mL. This

PAR Scientific discussion 18/21 concentration is associated with somnolence. Upon request, the Applicant has presented a discussion of whether drowsiness is included in this definition and concludes that these terms are normally used interchangeable. The Applicant presents data from other diphenhydramine drugs and states that the frequency reported for the present study is similar to the frequency reported for these other drugs, and the frequencies mentioned in the SmPC have been adjusted.

Three studies on the safety in comparison to other anti-motion have been included in the clinical overview. The first study showed an increased incidence of drowsiness when treated with dimenhydrinate compared to transdermal scopolamine. Another study showed that the number of adverse effects under the effect of or dimenhydrinate was similar to that reported with their respective placebos. The third study found, among 20 subjects included in a 4x4 cross-over study, a total of 12 adverse events, where three was accounted for by dimenhydrinate. The four adverse events included muscular tension, burning sensation, sensation of burning and swelling. Other SmPCs of diphenhydramine products are able to state the frequencies of undesirable effects and the Applicant was requested also to state the frequencies of undesirable effects. In the updated SmPC, the Applicant has presented frequencies, which is endorsed.

It is well-established, that dimenhydrinate is substance of abuse due to its euphoric and hallucinogenic properties. Both acute abuse and chronic abuse have been reported. The Applicant argues that the formulation may presents an advantage compared to other oral formulations. This statement is supported. The Applicant also argues that the physical limitation (size of the lozenge) will prevent the abuse for recreational purposes, because of the large number of lozenges that should be ingested. Furthermore, the low dosage compared to other oral formulations would decrease the interest of abusers. The active substance is more easily absorbed from the mouth avoiding first pass metabolism with the present formulation. The lozenges, while sucked, are expected to have an approximate residence time of 10 minutes inside the mouth, which allows absorption through the oral mucosa. With conventional oral formulations (e.g. solutions, capsules, tablets) this time is reduced and absorption through the mucosa is not expected.

Several scenarios are proposed by the Applicant. If the lozenge is chewed and small pieces are swallowed, the behaviour should be more like a conventional oral formulation with higher first-pass metabolism. The Applicant simulated a 20% reduction in the total bioavailability, and conclude that lower concentrations would be reached and no safety concerns are expected. Another scenario is that the lozenge is chewed and the small pieces are completely sucked until disappearance in the mouth. Higher proportion of the lozenge may be absorbed through this route. A 20% increase is simulated, representing the worst-case scenario based on observations from a study from 1990 on systemic availability of diphenhydramine after sublingual and oral dimenhydrinate.

The Applicant performed a pop-PK simulation with information from the findings by Gelotte et al. 2017 showing none of the expected scenarios seem to be associated with safety concerns in terms of toxic effects.

In regards to abuse, the Applicant argues that a drug abuser is characterized by engaging in a compulsive behaviour, triggered by the reward pathway. The abuser seeks achieving the reward as quickly as possible. Therefore, a lozenge formulation is proposed by the Applicant to be counteractive as higher concentrations may be obtained by chewing and sucking completely the small pieces, but this procedure requires a longer time. Similarly, an overdose would require the use of several lozenges, which would need a longer time in comparison to another type of formulation that just require a direct ingestion. Initially, the Applicant did not discuss the potential risks and consequences of abuse in the younger population, which was therefore requested in the day 70 LoQ. The Applicant responded that abuse by the younger population is unlikely due to several factors, some of them listed above. Further, the Applicant presented data from the EudraVigilance database (2018) for reports of abuse of dimenhydrinate. These data support that abuse of demenhydrinate (among the younger population) is not constituting a major problem, and the issue will not be pursued. Of note, the prescription status, is a national issue and will be decided after approval of the application. Also, this decision (regarding prescription status) may limit the exposure to children/adolescents.

PAR Scientific discussion 19/21

The symptoms of overdose follows those of anticholinergic overdose, which the Applicant sufficiently has discussed in the clinical overview. Information regarding use of active charcoal (in the hours after ingestion) and recommendations regarding monitoring of vital (and laboratory) parameters has been included in section 4.9 of the SmPC.

IV.6 Risk Management Plan

The MAH has submitted a risk management plan, in accordance with the requirements of Directive 2001/83/EC as amended, describing the pharmacovigilance activities and interventions designed to identify, characterise, prevent or minimise risks relating to Dimenhydrinat ”Trimb”.

Safety specification Summary of safety concerns Important identified risks  Hypersensitivity  Aggravation of narrow-angle and prostatic hypertrophia  Administration in case of bronchial asthma, COPD, emphysema and chronic bronchitis  Elderly with susceptibility to the central and peripheral anticholinergic effects of dimenhydrinate  Risk of accumulation in patients with kidney or liver failure  Combined administration with ototoxic antibiotics  Increased anticholinergic effects in co-administration with tricyclic , MAO-I, neuroleptic and other drugs with anticholinergic effects  Increased drowsiness when co-administrated with  Increased CNS adverse events and sedative effects when co- administrated with other CNS Important potential risks  QTc prolongation  Use in patients phaechromocytoma Missing information  Children under 2 years of age  Use during pregnancy and lactation

Pharmacovigilance Plan Routine pharmacovigilance is suggested and no additional pharmacovigilance activities are proposed by the Applicant, which is endorsed.

Risk minimisation measures Routine risk minimisation is suggested and no additional risk minimisation activities are proposed by the Applicant, which is endorsed.

V. USER CONSULTATION

The package leaflet has been evaluated via a user consultation study in accordance with the requirements of Articles 59(3) and 61(1) of Directive 2001/83/EC. The language used for the purpose of user testing the PIL was English. The results show that the package leaflet meets the criteria for readability as set out in the Guideline on the readability of the label and package leaflet of medicinal products for human use.

The test consisted of: a pilot test with 2 participants, followed by two rounds with 10 participants each. The questions covered the following areas sufficiently: traceability, comprehensibility and applicability. PAR Scientific discussion 20/21

VI. OVERALL CONCLUSION, BENEFIT/RISK ASSESSMENT AND RECOMMENDATION

Dimenhydrinat ”Trimb” 20 mg lozenges has a proven chemical-pharmaceutical quality and an established favourable efficacy and safety profile. The MAH has demonstrated that the active substance, dimenhydrinate, has been in well-established medicinal use with recognised efficacy and an acceptable level of safety within the Community for at least 10 years in the specific therapeutic use.

The MAH presented a risk management plan summarising the safety concerns. There are no additional pharmacovigilance or risk minimisation measures.

Agreement between Member States was reached during a written procedure. There was no discussion in the CMD(h). The Concerned Member States, on the basis of the data submitted, considered that a marketing authorisation for Dimenhydrinat ”Trimb” could be granted. The decentralised procedure was finalised on 20 September 2018. Dimenhydrinat ”Trimb” was authorised in Denmark on 30 November 2018.

According to the List of Union reference dates and frequency of submission of periodic safety update reports (PSURs), no routine PSURs are required for this product.

The date for the first renewal will be: 19 September 2023.

The following post-approval commitments have been made during the procedure:

1. The Applicant should submit identification studies for impurity RRT 1.05 and a toxicological / qualification justification for the proposed limit of NMT 0.4% via appropriate variation procedure. Depending on the results of these studies, the shelf life specification and the shelf life / storage condition of the drug product should be amended if applicable. Unless requested otherwise by the Applicant, 6 months from end of procedure, is considered sufficient time.

2. Applicant have committed to perform Phase 1 of environmental risk assessment on dimenhydrinate for submission 2 months post approval date. This should include screening for PBT and PECsurfacewater. If the environmental risk assessment enter Phase 2, then Applicant should be committed to submit a Phase 2 environmental risk assessment including all relevant studies according to the ERA guideline* within 24 months from approval date. Applicant will submit via appropriate variation procedure. *Guideline on the environmental risk assessment of medicinal products for human use. EMEA/CHMP/SWP/4447/00 corr 2.

PAR Scientific discussion 21/21