Absolute bioavailability and safety of a novel rivastigmine nasal spray in healthy elderly individuals

Abstract

Aims

To test the feasibility of a novel rivastigmine nasal spray as prospective treatment for dementia.

Methods

A single dose, crossover absolute bioavailability and safety study was conducted with rivastigmine intravenous solution (1 mg) and nasal spray (3.126 mg) in eight healthy elderly individuals, aged 58–75 years.

Results

Absolute bioavailability (F) of the nasal spray was significant at 0.62 (0.15) for F > 0 (P < 0.001, n = 8). The systemic dose absorbed was 2.0 (0.6) mg, time to maximum plasma concentration was 1.1 (0.5) h and maximum plasma concentration was 6.9 (2.0) ng ml⁻¹. The NAP226–90 to rivastigmine AUC_0–∞ ratio was 0.78 (0.19). The single dose safety was good with two of five mild adverse events related to the nasal spray. Nasal and throat irritation were perceived as mild and transient, and both had resolved at 20 min post‐nasal dose. An estimated dose of two or three sprays twice‐daily with nasal spray would deliver comparable rivastigmine exposure and efficacy as a 6–9.7 mg day^–1 oral dose and a 10 cm² transdermal patch, respectively.

Conclusions

The rivastigmine nasal spray had superior absolute bioavailability compared to historical values for oral capsule and transdermal patch determined by other researchers. It had rapid onset of action, low NAP226–90 to rivastigmine exposure ratio and a favourable safety and tolerability profile. The ability to achieve adjustable, individual, twice‐daily dosing during waking hours has good potential to minimise undesirable cholinergic burden and sleep disturbances whilst delivering an effective dose for the treatment of dementia associated with Alzheimer's and Parkinson's disease.

What is Already Known about this Subject

• Rivastigmine is a moderately effective treatment for mild to severe dementia associated with Alzheimer's disease when delivered by oral capsule or transdermal patch

• A range of in vitro and in vivo animal studies of nasal absorption exist for cholinesterase inhibitors

What this Study Adds

• First human study of nasal absorption of rivastigmine, a cholinesterase inhibitor

• Superior rivastigmine bioavailability with a novel nasal spray compared to historical values for capsule and patch

• Good potential to improve efficacy and tolerability by providing an effective rivastigmine dose during waking hours whilst reducing the problems of undesirable cholinergic burden, sleep disturbances and local irritation

Tables of Links

TARGETS
Enzymes 2
Acetylcholinesterase
Butyrylcholinesterase http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=257

LIGANDS
Rivastigmine http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=3726	http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=388

These Tables lists key protein targets and ligands in this article that are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 1, and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 2.

Introduction

Nasal delivery of physostigmine, an acetylcholinesterase (AChE) inhibitor, in rats was probably the first reported study showing the potential of the nasal route for the treatment of Alzheimer's disease 3. Rivastigmine is a potent, reversible inhibitor of AChE in the brain 4, 5. When it is delivered as an oral capsule or transdermal patch it is moderately effective for the treatment of mild to severe dementia of the Alzheimer's type and treatment of mild to moderate dementia associated with Parkinson's disease 6.

The original aim for rivastigmine was to provide a “rapid, sustained dose‐dependent inhibition of central AChE, attended by a favorable tolerability profile...” 4. The first part was achieved by the oral capsule, but its clinical utility has been limited by a high incidence of nausea, vomiting, diarrhoea (NVD) and asthenia 6. The transdermal patch successfully reduces NVD but its slow onset of action and high cholinergic burden due to 24 h rivastigmine delivery means NVD remains, albeit to a lesser extent and the therapy discontinuation rate for the transdermal patch is not superior to the oral capsule 6. This is probably due to the patch additionally causing skin irritation and sleep disturbances (due to 24 h drug delivery) 7, 8.

The pharmacological action of rivastigmine as a cholinesterase inhibitor is often described as a “pseudoirreversible” enzyme inhibitor, because it produces an effect that persists much longer than the drug is present in plasma 9. Like acetylcholine, rivastigmine undergoes hydrolysis by AChE. For acetylcholine, the acetyl moiety quickly dissociates from AChE (within microseconds), allowing rapid regeneration of the enzyme. For rivastigmine, the carbamyl moiety of rivastigmine remains bound for about 8.5 h in humans despite the drug having a short half‐life of 1–2 h 4, 9.

The metabolic pathway for the elimination of rivastigmine is integrally related to this enzyme carbamylation as it is the principal step required for cholinesterase inhibition 4. The conversion of rivastigmine to its main phenol metabolite (NAP 226–90) follows Michaelis–Menten kinetics, but for all practical purposes has linear pharmacokinetics up to a 6 mg day^–1 oral dose and at higher doses exhibits nonlinear pharmacokinetics 4, 10. Rivastigmine exposure is likely to be primarily responsible for the clinical effect because its metabolites have very little activity against AChE, being at least 10‐fold less potent than rivastigmine 9.

To determine the absolute bioavailability of a novel rivastigmine nasal spray (WIPO PCT Patent Appl. WO 2016/049 700 A1) a single‐dose, sequential, crossover, pharmacokinetic and safety study was conducted with intravenous and nasal rivastigmine in healthy elderly individuals.

Methods

Eight healthy elderly Caucasian volunteers, male and female aged between 55 and 85 years were enrolled into an open label, sequential, crossover study. The study was conducted following IRB approval (The Alfred Ethics Committee No. 538/14), and in compliance with GCP standards. Following written informed consent and screening evaluations, eligible participants were administered 1 mg of intravenous rivastigmine as a constant infusion over 30 min on Day 1 of the trial. Following a 2‐day washout, a single dose of rivastigmine nasal spray (3.126 mg) was administered (one spray in each nostril).

Participants with a history of symptomatic allergic rhinitis, currently active asthma or chronic obstructive pulmonary disease (excluding childhood asthma) were excluded. The complete inclusion and exclusion criteria are contained in Supporting Information (Appendix S1).

Clinical

The nasal spray formulation consisted of rivastigmine tartrate EP 2.5% w/v, citric acid monohydrate EP 0.055% w/v, l‐carveol Food Grade 0.05% v/v, ethanol 10% v/v, benzyl alcohol USP 0.65% v/v and polyvinyl pyrrolidone USP 1.0% w/v dissolved in water for irrigation. This formulation was filled into 20 ml high‐density polyethylene bottles and sealed with 100 μl VP7 metered‐dose pump valves with nasal spray actuators and caps (Aptar, Le Vaudreuil, France) and had a final pH of 3.6.

One 100 μl spray (2.5 mg rivastigmine tartrate equivalent to 1.563 mg rivastigmine free base) was administered into each nostril by study staff for a total single‐dose of 3.126 mg rivastigmine. The shot weight delivered to each participant was determined by weighing the nasal spray device before and after each dosing to each participant. The amount of rivastigmine delivered to each participant was determined from the shot weight, density and concentration of the nasal spray formulation.

The intravenous (IV) solution consisted of rivastigmine tartrate EP 0.032% w/v and citric acid monohydrate EP 0.022% w/v dissolved in distilled water (i.e. 1 mg rivastigmine per 5 ml). This was sterilised by filtration then aseptically filled into sealed Type I borosilicate glass vials. For IV administration, 6 ml of the IV solution was diluted to 30 ml with 5% glucose for injection and 25 ml (1 mg rivastigmine) was administered as a constant infusion over 30 min using a volumetrically controlled syringe driver. The infusion line was primed prior to start of the infusion.

Blood samples were taken from the nondominant arm of each participant for: (1) IV treatment at predose, 10, 20, 30, 45, 60, 75, 90 min, 2, 3, 4, 6, 8, 12 and 24 h; and (2) nasal treatment at predose, 5, 10, 15, 30, 60, 90 min, 2, 3, 4, 6, 8, 10, 12 and 24 h. Blood was collected into K₂ EDTA tubes, and immediately transferred to a prechilled polypropylene centrifuge tube, containing sodium fluoride to inhibit any ex vivo enzymatic breakdown of rivastigmine to NAP226–90. The whole blood sample was centrifuged at 1900 g, 3–5°C, for 10 min, and the harvested plasma transferred to a polypropylene cryotube and stored frozen at –20°C pending analysis.

A health screen pre‐ and poststudy was completed for all participants. Visual nasal mucosal examination occurred at screening, check‐in, Day 4 (pre nasal dose) and Day 5 (24 h post nasal dose). Adverse events and vital signs (blood pressure, heart rate, respiratory rate, oral body temperature) were monitored from Day 1 until Day 5 (24 h post nasal dose) and at follow‐up visit (Day 9 ± 1). For the elicitation of potential adverse events relating to nasal mucosal irritation a series of relevant tolerability questions (i.e. stinging, itching, burning sensations in nose or throat, rhinalgia and lacrimation) were sought from the participants with a five‐point Likert‐scale for the responses using a perceived nasal irritation questionnaire. This assessment was completed at 20 and 75 min following the nasally administered dose. All questions for the perceived nasal irritation questionnaire and the tests used for the healthy screen are contained in Supporting Information (Appendix S1).

Bioanalytical assay

Analysis of rivastigmine and NAP 226–90 was performed by the bioanalytical division of Anapharm Europe (Barcelona, Spain). Rivastigmine tartrate salt, NAP 226–90 and deuterated internal standards (rivastigmine‐d4 tartrate salt and NAP 226–90‐d6) supplied by Toronto Research Chemicals (Toronto, Canada) were extracted from an aliquot of human EDTA plasma using a liquid–liquid extraction procedure with ethyl acetate and then injected into a liquid chromatograph equipped with a tandem mass spectrometry detector. Separations were performed on a reversed‐phase column (Zorbax SB‐ C18, 4.6 × 50 mm, 5 μm, from Agilent Technologies). Mobile phase A was ammonium acetate 10 mmol l⁻¹ at pH 5 prepared in water and mobile phase B was ammonium acetate 10 mmol l⁻¹ prepared in methanol. The chromatographic separation was gradiently performed at room temperature at a flow‐rate from 1.00 to 1.10 ml min⁻¹.

The calibration range used for this assay was 0.05–20 ng ml⁻¹ for rivastigmine (as the free base) and 0.05–10 ng ml⁻¹ for NAP226–90. The assay passed linearity for rivastigmine (r > 0.997) and NAP226–90 (r > 0.997) over each of the calibration ranges tested. Accuracy and precision at the lowest limit of quantitation (0.05 ng ml⁻¹) for rivastigmine were 5.65% and ±2.72%, respectively. Accuracy and precision at the lowest limit of quantitation (0.05 ng ml⁻¹) for NAP226–90 were 5.79% and ±2.72%, respectively. Precision and accuracy for all the remaining concentrations in each calibration range were also within their acceptance limits.

Noncompartmental pharmacokinetic analysis

Area under the curve (AUC) measurement was taken from the blood plasma concentration–time profile. AUC to the last measured concentration (AUC_0–l), AUC over a time interval (AUC_0–6 h) and AUC zero to infinity (AUC_0–∞) were calculated by the linear trapezoidal rule using Sigmaplot (version 12.5; Systat Software, Inc. San Jose, CA, USA). Absolute bioavailability (F) was calculated as AUC_0–∞ nasal/nasal dose divided by AUC_0–∞ IV/IV dose. Fluctuation index (FI) was calculated as (C _max – C _min) divided by C _avg0–l. C _avg0–l was calculated as AUC_0–l divided by the time for the last measured concentration. Metabolite to parent ratio was calculated by dividing AUC_0–∞ of NAP‐226‐90 divided by AUC_0–∞ of rivastigmine. The maximum plasma concentration (C _max) and time to C _max were determined by visual inspection of the data. Terminal elimination half‐life (t _1/2) was defined as 0.693 / λ, where λ is the terminal elimination rate constant (calculated from the slope of the regression line of the terminal phase of the natural logarithm of concentration vs. time). Average plasma concentration (C _avg) was calculated from the AUC over a time interval (AUC_{0–6 h}) divided by the prospective dosage interval (6 h). Systemic clearance (CL = IV dose / AUC_0–∞, IV) and volume of distribution (V _z = IV dose / λ ● AUC_0–∞, IV) were calculated for the IV dose.

Statistical analysis

To meet the primary hypothesis that absolute bioavailability (F) is >0, a one‐sample Student t test using a mean ± standard deviation (SD) F = 0.5 ± 0.25 and n = 4, provides a Power of 0.908 when performed as a one‐tailed test with alpha = 0.050 using Sigmaplot analysis (version 12.5; Systat Software, Inc. San Jose, CA, USA). This conservative (mean and variation) estimate for F was based on the existing in vitro diffusion data for the rivastigmine nasal spray formulation and consideration of the extent and variability of nasal absorption in humans for other drugs. Statistical significance was determined by a one‐sample Student t test, or a nonparametric one‐sample Wilcoxon signed‐rank test, if for example the raw F data are not normally distributed. Pharmacokinetic parameters were tabulated as mean values with their SD of the mean shown as percentage coefficient of variation (CV%).

Results

Four female and four male volunteers were recruited into the clinical study. They had a mean ± SD age of 64.5 ± 6.4 years (range 58–75), weight of 74.9 ± 9.7 kg (range 52.2–82.8), height of 1.69 ± 0.09 m (range 1.59–1.79) and body mass index of 26.1 ± 2.8 kg m⁻² (range 20.6–29.5). The mean ± SD dose of rivastigmine administered by the nasal spray device during the study was 3.124 ± 0.260 mg. This dose was concordant with the target dose of 3.126 mg rivastigmine.

Pharmacokinetics

To test the hypothesis that the absolute bioavailability (F) for the rivastigmine nasal spray was greater than zero the mean ± SD percentage F of 62 ± 15% (P < 0.001, n = 8) was determined using noncompartmental pharmacokinetic analysis. The pharmacokinetic parameters for IV infusion and nasal spray are shown in Tables 1 and 2, respectively. All individual AUC_0–l values were >91% of their corresponding AUC_0–∞ for both rivastigmine and NAP226–90. The mean (SD) plasma concentration profiles for rivastigmine and NAP226–90 are shown for IV and nasal administration in Figures 1 and 2, respectively. Figure S1 shows the mean (SD) plasma concentration profiles as semilogarithmic plots to emphasise the elimination phase for rivastigmine and NAP 226–90 after IV and nasal dosing with rivastigmine and is contained in Supporting Information (Appendix S1).

Table 1.

Noncompartmental pharmacokinetic parameters of rivastigmine and NAP226–90 after administration of 1 mg rivastigmine as a 30 min constant intravenous (IV) infusion, n = 8

Rivastigmine	C max, IV (ng ml−1)	AUC0–∞ IV (ng h ml−1)	t 1/2, IV (h)	CL IV (l h−1)	V z IV (l)
Mean	8.5	11.6	1.3	86.6	163.8
Min–max	5.8–11.2	8.9–13.9	1.2–1.4	71.8–112.2	131.5–210.8
CV (%)	22	12	8	14	16
NAP226–90	Cmax, IV (ng ml−1)	AUC0–∞ IV (ng h ml−1)	t1/2, IV (h)	N:R, IV
Mean	1.2	7.2	2.8	0.63
Min–max	1.0–1.6	6.1–8.8	2.5–3.2	0.46–0.79
CV (%)	17	14	11	18
95% CI				0.53, 0.72

C _max, peak plasma concentration; AUC_0–∞, area under the plasma concentration–time curve from time zero to infinity; CL, clearance following IV administration; V_z, apparent volume of distribution following IV administration; t _1/2, terminal half‐life; N:R, NAP226–90 AUC_0–∞ to rivastigmine AUC_0–∞ ratio; CV, coefficient of variation; 95% CI, 95% confidence interval (lower, upper)

Table 2.

Noncompartmental pharmacokinetic parameters of rivastigmine and NAP226–90 after administration of 3.126 mg rivastigmine nasal spray (one spray each nostril), n = 8

Rivastigmine	C max, NS(ng ml−1)	t max, NS(h)	AUC0–∞ NS(ng h ml−1)	t 1/2, NS (h)	F(%)	FI, NS
Mean	6.9	1.1	22.6	2.6a	62	3.7a
Min–max	4.6–9.8	0.5–2	13.5–32.7	2.0–13.3	45–85	2.8–10.0
CV (%)	29	46	28		25
95 % CI				2.2, 5.0	49, 75	3.1, 4.5
NAP226–90	Cmax, NS(ng ml−1)	tmax, NS(h)	AUC0–∞ NS(ng h ml−1)	t1/2, NS (h)	N:R, NS	FI, NS
Mean	2.5	2.1	16.9	3.9	0.78	2.8
Min–max	2.0–3.4	1.0–3.0	13.8–20.5	2.9–4.6	0.66–1.15	2.0–3.6
CV (%)	24	39	14	21	25	21
95% CI					0.62, 0.95

C _max, peak plasma concentration; t _max, time of C _max; AUC_0–∞, area under the plasma concentration–time curve from time zero to infinity; t _1/2, terminal half‐life; F, bioavailability; FI, Fluctuation Index; N:R, NAP226–90 AUC_0–∞ to rivastigmine AUC_0–∞ ratio; NS, nasal; CV, coefficient of variation; 95% CI, 95% confidence interval (lower, upper)

^aMedian

Figure 1.

Mean (± standard deviation) plasma concentration–time profile of rivastigmine after administration of 1 mg rivastigmine as a 30 min constant intravenous infusion and 3.126 mg rivastigmine nasal spray, n = 8

Figure 2.

Mean (± standard deviation) plasma concentration‐time profile of NAP226–90 after administration of 1 mg rivastigmine as a 30 min constant intravenous infusion and 3.126 mg rivastigmine nasal spray, n = 8

In order to compare the cholinergic burden of the nasal spray with the transdermal patch, we also calculated the geometric mean AUC_0–∞, which was 21.9 and 16.9 ng h ml⁻¹ for rivastigmine and NAP226–90, respectively. To estimate a prospective dosing interval for the nasal spray we calculated the mean proportion (86%) of the total rivastigmine AUC_0–∞ absorbed at 6 h and the corresponding mean ± SD C _{avg0–6 h}, which was 3.2 ± 1.0 ng ml⁻¹ (range 2.0–4.6).

Safety

Adverse events were monitored throughout the study and local tolerance and irritation following nasal delivery was queried 20 and 75 min postdose. Two of the five adverse events reported were plausibly related to rivastigmine nasal spray (one individual had mild nasal congestion and another had a mild, red, itchy stomach rash; both recovered within 12 h without treatment). The remaining three adverse events where all mild (cough, drowsiness and rash) and were not related to the study drug.

Three out of eight volunteers reported mild nasal irritation, and four out of eight volunteers reported mild throat irritation shortly after nasal rivastigmine dosing, which subsequently resolved 20 min postdose. No further localised symptoms were reported at the 75 min postdose time point. The breakdowns for the initial 20 min post dose irritation scores are shown in Table 3.

Table 3.

Local tolerability question responses for 20 min period after nasal spray dose, n = 8

	Response, % of total
Question	Mildly	Not at all
Nasal
Stinging your nose	12.5	87.5
Itching your nose	25	75
Burning your nose	0	100
Causing a runny nose	25	75
Throat
Stinging your throat	25	75
Itching your throat	0	100
Burning your throat	25	75
Eyes
Causing watery eyes	0	100

Discussion

The rivastigmine nasal spray met the primary study endpoint with a mean absolute bioavailability (F) of 62%, which was significant (P < 0.001, F > 0). The variability in nasal absorption and metered‐dosing was about 16%. This was estimated by subtracting the nondose adjusted CV (28%) for nasal rivastigmine AUC_0–∞ from the CV (12%) for IV. The nasal spray had significantly (P < 0.001) higher absolute bioavailability (62 ± 15%, n = 8) compared to historical values determined by other researchers for the oral capsule (36 ± 13%, n = 12) and transdermal patch (45 ± 10%, n = 30; measured from postusage drug residuals) in healthy elderly volunteers 9, 11, 12, 13. In this particular case, comparison to historical F values is acceptable due to the similar variances observed and the reliable comparators (IV infusion or patch drug content) used as the 100% bioavailability reference.

The mean clearance (86.6 l h⁻¹) and volume of distribution (163.8 l) for the 1 mg IV dose in this study were within the ranges previously observed for Alzheimer's disease patients administered a 2 mg IV dose (CL 21.6–82.8 l h⁻¹ and V _z 53.2–227 l) 10. Whilst animal studies (rats) report sex differences in the pharmacokinetics of rivastigmine 14 we found no sex differences in rivastigmine CL and V _z after IV dosing in humans from our study, which is consistent with population pharmacokinetic data reported to the Food and Drug Administration by other researchers for the oral capsule and transdermal patch, which shows no sex effect 11, 13.

Rivastigmine was rapidly absorbed across the nasal mucosa with a mean ± SD time to maximum plasma concentration of 1.1 ± 0.5 h and a mean C _max of 6.9 ± 2.0 ng ml⁻¹. The median t _1/2 of the rivastigmine nasal spray was 2.6 h, about twice as long as that measured for the IV dose. The extent of rivastigmine exposure was clinically significant and the C _avg over the first 6 h after administration could be expected to have an appreciable inhibitory effect on central AChE levels in Alzheimer's disease patients 4, 5.

The NAP226–90 to rivastigmine ratio for AUC_0–∞ were 0.78 ± 0.19 and 0.63 ± 0.11, for the nasal and IV treatments, respectively. This ratio was comparable to that previously measured for the transdermal patch (0.67), but four‐fold lower than the oral capsule (3.49) 12. A high degree of first‐pass metabolism in the liver and gut explains the much higher peripheral exposure to the metabolite (NAP226–90) after oral dosing 9. This study confirmed nasal absorption of rivastigmine across the nasal mucosa due to the high F and low NAP226–90 to rivastigmine AUC ratio.

The single dose safety for the nasal spray was good with two mild adverse events, which resolved within 12 h without treatment, and any nasal or throat irritation was perceived as mild and transient; however, further studies are required to confirm the safety profile of this route of administration. Notably, no NVD was observed despite the systemic dose being equivalent to a single 5.6 mg oral dose (i.e. 2.0 mg / F _oral 0.355). This was not entirely unexpected. The Food and Drug Administration previously noted that in its New Drug Application review for rivastigmine oral capsule that gastrointestinal adverse events (NVD, weight loss and anorexia) were significantly correlated with NAP226–90 exposure (both C _max and AUC) using logistic regression analysis of Phase 3 clinical study data (n = 625). No such correlation was found for rivastigmine exposure 11, 15.

Previous researchers have suggested that NVD might be due to high rivastigmine FI 8, 16. In contrast, we observed no NVD in our study, despite the median rivastigmine FI being 3.7 and 4.2 after nasal and oral dosing 16, respectively. Given that a single 3 mg oral dose caused NVD in a significant proportion of individuals (33%) 12, it is unlikely that our sample size (n = 8) was a factor in failing to observe any corresponding NVD adverse events. By taking into account the good safety profile of 2 mg IV rivastigmine 10 and the findings of this study, we would suggest that rivastigmine FI is unlikely to be the primary cause of NVD in an individual.

The metered‐dose nasal spray has inherent capability to provide improved individual dosage adjustment within, below and above an effective dosing range. This may be beneficial because the intrinsic CL of rivastigmine varies up to four‐fold in Alzheimer's disease patients 10. In our study we saw a practical example of this because mean IV CL was 38% higher than that observed in Alzheimer's disease patients 10. This was similar to the 29% difference previously observed by other researchers 11.

In practical terms, a nasal dose of two sprays (i.e. one 100 μL spray in each nostril) delivered a mean systemic rivastigmine dose of 2 mg. For rivastigmine a few hours of the correct pharmacokinetic exposure in an Alzheimer's disease patient can provide up to half a day of central AChE inhibition 4, 5. An estimated dose of two or three sprays twice‐daily with rivastigmine nasal spray would deliver comparable rivastigmine exposure and efficacy as a 6–9.7 mg d^–1 oral dose and a 10 cm² transdermal patch, respectively (e.g. 9.7 mg oral × F _oral 0.602 = 5.84 mg ≈ 5.84 / 2.0 = 2.92 sprays twice‐daily) 6, 10.

At the estimated dose of the nasal spray that is comparable to the 10 cm² transdermal patch, the cholinergic burden exposure will be about 55–60% of the patch (e.g. 9.7 mg / 18 mg = 0.54; e.g. 5.84 mg / 9.5 mg = 0.61) 6. Removing undesirable cholinergic burden by using rapid‐onset dosing during waking hours has good potential to allow the patient to move further up the dose–response curve 17. Whilst further clinical studies are required, rivastigmine nasal spray has potential benefits compared to oral capsules and transdermal patches that include improved efficacy, tolerability and convenience. For example, it has potential to provide extra cognition benefit 7, 18, 19, less NVD and fatigue, improved sleep 20 and easier dosage increases.

A limitation of this study was the different rivastigmine doses administered for the measurement of absolute bioavailability. This was necessitated by bradycardia safety concerns if an equivalent 3.1 mg IV dose was administered 9, 10. The metabolic pathway for the elimination of rivastigmine is integrally related to enzyme carbamylation as it is the principal step required for cholinesterase inhibition 4. This conversion of rivastigmine to its main phenol metabolite (NAP226–90) follows Michaelis–Menten kinetics 10. However, for the 1 mg IV and 3.1 mg nasal doses administered in this study, no significant effect on absolute bioavailability measurement would have occurred because for all practical purposes rivastigmine displays linear pharmacokinetics up to a 6 mg day^–1 oral dose and at higher doses exhibits nonlinear pharmacokinetics 4, 10.

With absolute bioavailability confirmed and good tolerability observed for rivastigmine nasal spray in this single dose study, a future multiple dose pharmacokinetic and safety study in AD patients would determine rivastigmine exposure at higher doses where nonlinear pharmacokinetics have been shown previously for the oral capsule and transdermal patch 4, 16. With an escalating dose design, this study would also provide useful safety information about the prospective adverse events profile of the nasal spray as most of the adverse events for rivastigmine arise during dosage titration.

In conclusion, the rivastigmine nasal spray had superior absolute bioavailability compared to historical values for oral capsule and transdermal patch determined by other researchers. It had rapid onset of action, low NAP226–90 to rivastigmine exposure ratio and a favourable safety and tolerability profile. The ability to achieve adjustable, individual, twice‐daily dosing during waking hours has good potential to minimise undesirable cholinergic burden and sleep disturbances whilst delivering an effective dose for the treatment of dementia associated with Alzheimer's and Parkinson's disease.

Competing Interests

Both authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare T.M. and B.S. had support from Lachesis Biosciences Pty Ltd as sponsor for the study for the submitted work, T.M. was employed by and held stock for Lachesis Biosciences Pty Ltd in the previous 3 years. There are no other relationships or activities that could appear to have influenced the submitted work.

Supporting information

Appendix S1 Further clinical methods and results

Figure S1 Semi‐logarithmic plots of mean (±SD) plasma concentration‐time profile of rivastigmine and NAP226‐90 after administration of 1 mg rivastigmine as a 30 min constant i.v. infusion and 3.126 mg rivastigmine nasal spray, n = 8

Supporting info item

Morgan, T. M. , and Soh, B. (2017) Absolute bioavailability and safety of a novel rivastigmine nasal spray in healthy elderly individuals. Br J Clin Pharmacol, 83: 510–516. doi: 10.1111/bcp.13133.