Bio-analytical Validated Method Development and Bioequivalence Study of Serotonin Receptor 5-HT4 Agonist Dihydrobenzofurancarboxamide Derivative used for Chronic Constipation Drug Prucalopride by LC-ESI-MS/MS in Indian Human Plasma

Pallab Mandal, Soumya Chakraborty, Rakesh Bera, Chiranjit Saha, Balaram Ghosh, Sourav Poddar, Sujata Barma, Sanmoy Karmakar, Tapan K Pal

Published Date: 2021-09-30

Pallab Mandal1, Soumya Chakraborty1, Rakesh Bera1, Chiranjit Saha1, Balaram Ghosh2, Sourav Poddar3, Sujata Barma4, Sanmoy Karmakar4, Tapan K Pal1*

1Bioequivalence Study Centre, Dept. of Pharmaceutical Technology, Jadavpur University, TAAB Biostudy Services, Kolkata, India

2Department of Pharmacology, Calcutta School of Tropical Medicine, Kolkata, India

3Department of Chemical Engineering, National Institute of Technology, Tiruchirappalli, India

4Department of Pharmaceutical Technology, Jadavpur University, Kolkata, India

*Corresponding Author:
Tapan K Pal Bioequivalence Study Centre, Dept. of Pharmaceutical Technology, Jadavpur University, TAAB Biostudy Services, Kolkata, India E-mail: profpaltk@gmail.com

Received Date: July 01, 2021; Accepted Date: July 07, 2021; Published Date: September 30, 2021

Citation: Mandal P, Chakraborty S, Bera R, Saha C, Ghosh B, et al. (2021) Bio-analytical Validated Method Development and Bioequivalence Study of Serotonin Receptor 5-HT4 Agonist Dihydrobenzofurancarboxamide Derivative used for Chronic Constipation Drug Prucalopride by LC-ESI-MS/MS in Indian Human Plasma. In Silico In Vitro Pharmacol Vol.7 No.5:1.

Visit for more related articles at Journal of In Silico & In Vitro Pharmacology

Abstract

Aim and objective: Clinically, prucalopride is used to treat persistent constipation since it is a serotonin receptor agonist. According to US-FDA and European Medicines Agency (EMA) requirements, a validated bioanalytical technique for prucalopride quantification from human plasma was developed and utilized in a comparative investigation of pharmacokinetics and bioequivalence of the drug.

Method: Prucalopride's protonated precursor ion was 368.1/196.0 m/z, while propranolol's was 260.1/116.0 m/z. When it came to plasma, it was extracted using liquid-liquid technology. It combined with 10 mM ammonium acetate as an aqueous solvent with an apparent pH of 1.20 and 1% formic acid in acetonitrile as an organic solvent at a flow rate of 0.5000 ml/min as the mobile phase. 0.25 ng/ml, 0.50 ng/ml, 1 ng/ml, 2 ng/ml, 4 ng/ml, 8 ng/ml, and 16 ng/ml is used as calibration concentrations.

Result: Peak Concentration Cmax was 4.98 ng/ml ± 0.95 ng/ml for reference preparation at 2.71 hour ± 0.33 hour. (Tmax) and 5.01 ± 1.05 ng/ml for test preparation at 2.64 hour ± 0.28 hour (Tmax).

Conclusion: Using this technique for comparative pharmacokinetics and bioequivalence studies has proven to be a highly selective, sensitive, high recovery, low ion suppression, repeatable, and cost effective approach.

Keywords

Prucalopride; LC-MS/MS; Bioanalytical method; Serotonin receptor agonist

Abbreviation

GABA: Gamma-Aminobutyric Acid; HT: Hydroxytryptamine; hERG: human Ether-Ago-Go Related Gene; USFDA: United States Food and Drug Administration; CPU: Clinical Pharmacological Unit; CDSCO: Central Drugs Standard Control Organization; ISTD: Internal Standard; DMSO: Dimethyl Sulfoxide; LLOQ: Lower Limit of Quantification; LQC: Low-Quality Control; MQC: Middle-Quality Control; HQC: High-Quality Control; EMA: European Medicine Agency; API: Active Pharmaceutical Ingredient; LOD: Limit of Detection; ESI: Electrospray Ionization; ME: Matrix Effect; LLE: Liquid-Liquid Extraction; MTBE: Methyl Tert-Butyl Ether

Introduction

Serotonin receptors, activated by the neurotransmitter serotonin, are G-protein coupled receptors and ligand-gated ion channels found in the peripheral and central nervous systems [1]. But these receptors modulate releases of many neurotransmitters and hormones like GABA, dopamine, acetylcholine, epinephrine or norepinephrine, glutamate, and hormone-like oxytocin, prolactin, vasopressin, corticotrophin, and substances P [2,3]. All 5-HT receptors have excitatory activity except 5-HT1 and 5-HT5, which have inhibitory activity [4]. The pharmacological functions are anxiety, appetite, GI motility, learning, memory, mood respiration [5,6]. Constipation is a symptom which causes divided into congenital, primary (intrinsic problems of colonic or anorectal function), and secondary (related to organic disease, systemic disease, or medications) [7-11]. Prucalopride is a dihydro benzofuran carboxamide derivative from the benzofuran family that is a serotonin 5-HT4 receptor agonist and has enterokinetic properties that are it has potent prokinetic activity [12]. This drug changes the colonic motility pattern via serotonin 5-HT4 receptor stimulation and stimulates colonic mass movements, providing the main propulsive force for defecation [13-15]. Prucalopride is well absorbed in the GI tract. It reaches maximum plasma concentration around 3 hrs after initial administration, an overdose may cause severe diarrhea, and bioavailability is over 90% and not influenced by the ingestion of food. Still, plasma protein binding is 30% [16-20]. After reaching maximum plasma concentration, prucalopride is mainly excreted by the urine (84%) and remaining excreted by feces [21].

Ultra-high-performance mass spectrometry was used to measure prucalopride in human plasma after a literature analysis revealed one or two techniques. There was no mention of a precise bioanalytical approach, mass spectrometric fragmentation of the medication, or a complete process for validating the results.

As part of the bioequivalence study conducted by Estradiol Valerate (EV), we developed and validated a fast, rapid, sensitive, and specific LC-MS/MS (API-4000) method for quantifying prucalopride in Indian human plasma. This method was successfully applied to several pharmacokinetic studies on healthy human volunteers.

Materials and Methods

Chemical and reagents

Acetonitrile purchased from Merck (MERCK India Ltd., Mumbai), isopropyl alcohol, formic acid was AR grade, and solvents were HPLC grade. Milli-Q water was used in the study (Elix, Milli-Q A10 Academic, Bedford, MA, USA) until a resistivity of 18.2 MΩ was achieved. The blank human plasma with EDTA-K3 anticoagulant was collected from the Clinical Pharmacological Unit (CPU) of TAAB Biostudy Services, Kolkata, and stored at 20°C until analysis.

Ethical clearance

The protocol of prucalopride study and related documents like volunteers informed consent form, case record form, subject information sheet submitted to the HURIP Independent Bioethics committee, Kolkata, India (Central Drugs Standard Control Organization (CDSCO) registration: ECR/103/Indt/WB/2013/RR- 19 which is valid up to 21st Nov, 2024) and the ethical clearance obtained before initiation of the study.

Drug information and study design

It is a randomized, open-label, balanced, laboratory blind, two treatment, two-period, two sequences, single-dose, truncated, two way, crossover, bioequivalence study of Prucalopride 2 mg tablet as a test with Pruvict 2 (Prucalopride 2 mg) tablet of Torrent pharmaceutical Ltd as a reference product, in healthy adult human subjects under fasting condition. The volunteers have received either test or reference product with 240 ml drinking water based on the randomization code in each clinical period as specified in Table 1.

Table 1. Randomization schedule.

Subject No. Period I Period II
1 A1 A2
2 A2 A1
3 A2 A1
4 A1 A2
5 A1 A2
6 A2 A1
7 A1 A2
8 A2 A1
9 A2 A1
10 A1 A2
11 A1 A2
12 A2 A1
13 A1 A2
14 A2 A1
15 A1 A2
16 A1 A2
17 A2 A1
18 A1 A2
19 A2 A1
20 A2 A1
21 A1 A2
22 A1 A2
23 A2 A1
24 A2 A1
A1:Reference preparation; A2:Test preparation

Test product Prucalopride 2 mg tablets were compared to a reference product under fasting conditions with the primary goal being to compare the rate and amount of absorption. Prucalopride 2 mg tablet will be administered orally to healthy adult human volunteers under fasting conditions with the secondary goal of monitoring safety and acceptability. Total 21 blood sampling time point were 0 hour (before drug administration) and 0.50 hours, 1.00 hours, 1.33 hours, 1.67 hours, 2.00 hours, 2.33 hours, 2.67 hours, 3.00 hours, 3.33 hours, 3.67 hours, 4.00 hours, 4.50 hours, 5.00 hours, 6.00 hours, 8.00 hours, 10.00 hours, 12.00 hours, 24.00 hours, 48.00 hours, 72.00 hours, which were collected from cubital vein and separated plasma by centrifugation and stored in -20˚C. 5 ml of blood taken at each time point. The demographic data of each volunteer is specified in Table 2.

Table 2. Demographic data of 24 volunteers.

Vol. No. Sex         Age Height (cm)      Weight (kg)    BMI(kg/m2)
1 M 28 156 52 21.37
2 M 42 156 55 22.6
3 M 39 160 60 23.44
4 M 34 163 65 24.46
5 M 42 165 58 21.3
6 M 24 179 65 20.29
7 M 24 170 56 19.38
8 M 32 166 59 21.41
9 M 42 163 65 24.46
10 M 30 158 53 21.23
11 M 28 173 61 20.38
12 M 40 165 52 19.1
13 M 30 155 51 21.23
14 M 22 167 53 19
15 M 29 163 52 19.57
16 M 35 173 72 24.06
17 M 45 157 56 22.72
18 M 35 158 60 24.03
19 M 42 166 62 22.5
20 M 41 155 52 21.64
21 M 30 154 56 23.61
22 M 38 167 56 20.08
23 M 27 165 55 20.2
24 M 27 164 57 21.19
Mean 33.58 163.25 57.62 21.63
S.D. 7.08 6.39 5.38 1.7

Bio-analytical method development by LC-MS/ MS (API-4000)

Prucalopride is a dihydro benzofuran carboxamide derivative from the benzofurane family, with chemical name 4-Amino-5-chloro- 2methoxybenzoic acid, 3-dihydro-N-[1-(3-methoxypropyl)-4- piperidinyl]-7-benzofurancarboxamide group and represented in Figure 1 [22]. It acts as an agonist of the 5-HT4 receptor. In addition, it acts as a prokinetic drug that stimulates these receptors and colonic mass movements, providing the primary propulsive force for defecation [23].

pharmacology-Prucalopride

Figure 1. Chemical structure of Prucalopride.

The exact mass of prucalopride is 367.16 (molecular weight 367.88) and the PKa value of prucalopride is 14.64 (conjugate base) and 8.98 (strongest basic) for benzofuran and piperidine heterocyclic structure. Benzofuran is a fragile basic compound depending on its PKa value -2.9 that is strongly conjugate basic and used as psychoactive substances which interacted with 5-HT receptors and activated 5-HT4 receptors. Piperidine is an azacycloalkane that is cyclohexane. A more substantial base replaces one carbon because the lone pair of electrons in piperidine is in an sp3 hybrid orbital and PKa value 11.28, which is used as an analgesic in prucalopride. So drug prucalopride itself a basic [24,25].

Due to lower protein binding, optimum conditions of plasma extraction and method development are required for a different type of pH value of the analyte. Therefore, propranolol is used as an Internal Standard (IS) for quantitation positive polarity to achieve adequate response for their simultaneous analysis. Moreover, the positive ionization mode is selective and highly sensitive for compounds with low electron affinity. Thus positive ionization mode was selected to fragment the analyte and IS to obtain intense and consistent product ions.

The protonated precursor ions (M+H)+ at 368.1 m/z (highest peak), 370.1 (2nd peak), were observed in Q1 MS in which selected parent ion 368.1 for prucalopride and characteristic product ions or fragment ions found in Q3 MS were at m/z 196.0, 278.8, 156.1. However, the most stable and consistent fragment ion selected was m/z 196.0 (M-C19H20N2O+H)+ for releasing nonanohydrazide and the product ion chemical name 2-[(4-chlorophenyl) methylideneamino]acetate, which monoisotopic mass 196.0 and displayed in Figure 2.

pharmacology-product

Figure 2. Parent (Q1) and product ion (Q3) scan of Prucalopride.

For the internal standard, the protonated precursor ions [M+H]+ at m/z 260.1 (highest peak) were observed in Q1 MS for propranolol, and characteristic product ions or fragment ions found in Q3 MS were m/z 116.0, 182.9. However, the most stable and consistent fragment ion selected was m/z 116.0[(M-C10H8O+H)+] forreleasingnaphthalen-2-ol and the product ion chemical name hexanamide monoisotopic mass 116.0. The chromatographic elution of the analytes on a Phenomenex Kinetex 5μ C18 100A 50 × 3 mm column was initiated as a rapid, susceptible, and rugged bioanalytical method covering the dynamic linear range. Mobile phase selection was necessary for the analysis of the drug depending on its pka value. Thus, the pH of the mobile phase, buffer concentration, and choice and proportion of diluents were necessary for chromatographic resolution with adequate response to achieve the desired sensitivity. Optimized instrumental (mass) parameters for prucalopride and IS are illustrated in Table 3.

Table 3. Optimized instrumental (mass) parameters for Prucalopride and IS.

Parameter(s) Value
Ionization mode MRM (positive)
Source temperature (ºC) 400
Dwell time per transition (msec) 100
Curtain gas (psi) 35
CAD gas (psi) 8
Ion spray voltage (V) 5500
Ion source gas 1 (psi) 55
Ion source gas 2 (psi) 45
Focussing potential (V) 400
Declustering potential (V) 23  (Prucalopride) and  30 (IS)
Entrance potential (V) 11
Collision energy (V) 39 (Prucalopride) and 29 (IS)
Collision cell exit potential (V) 15 (Prucalopride and IS)
Transition pair of Prucalopride  (analyte) 368.1/196.0
Transition pair of Propranolol (IS) 260.0/116.0

Initially, acetonitrile/methanol with one mM ammonium acetate buffer (pH 6.5) gave a response for prucalopride. However, the response was not reproducible. The signal of the lower limit of quantification concentration was changed when buffer concentration increased from 1 mM to 10 mM. Further, the chromatography was better with a higher response using an acetonitrile buffer than a methanol-buffer combination. At pH above 5.0 that is basic, the resolution of prucalopride was affected, which further deteriorated with an increase in PH. Thus, to achieve more excellent reproducibility and better chromatography, low pH buffers were tried.

The lower drug concentration’s reproducibility and peak shape were better in 1.0% formic acid, but the inadequate signalnoise ratio. So as a final solvent was used, 1% formic Acid was in Milli Q water and mixed with 10 mM ammonium acetate as an aqueous solvent having apparent pH 1.20 and 1% formic acid in acetonitrile as an organic solvent at a flow rate of 0.5000 ml/ min then a superior signal to noise ratio (≥ 22), and baseline resolution achieved. So the selected mobile phase was 1% formic acid in Milli Q water and mixed with 10 Mm ammonium acetate as an aqueous solvent, and 1% formic acid in acetonitrile as an organic solvent drug was totally basic, so for better ionization and more negligible matrix effect, the acidic solvent was necessary. For analysis of prucalopride, the bioanalytical method was performed by gradient technique in which 90% aqueous solvent was used for 0.01 min to 0.50 min. From 0.50 min to 3.50 min of total run time, 10% aqueous solvent was used, and from 3.50 min to rest of the entire run time (7.00 min) for washing purposes used 90% aqueous solvent was. The gradient curve is illustrated in Figure 3. For better intensity and a narrower peak, the acidic aqueous solvent runs for a long time, and in this timing, the analyte and IS eluted. The chromatographic elution time for prucalopride and IS (propranolol) was found at 1.76 and 1.82 min respectively, in a total run time of 7.00 min. The representative MRM chromatograms were showed in Figures 4-9.

pharmacology-curve

Figure 3. Gradient curve of method development.

pharmacology-chromatogram

Figure 4. Blank sample chromatogram of prucalopride.

pharmacology-IS

Figure 5. Blank IS sample chromatogram of Prucalopride.

pharmacology-LLOQ

Figure 6. LLOQ sample chromatogram of Prucalopride.

pharmacology-LOC

Figure 7. LOC sample chromatogram of Prucalopride.

pharmacology-MQC

Figure 8. MQC sample chromatogram of Prucalopride.

pharmacology-HQC

Figure 9. HQC sample chromatogram of prucalopride.

Plasma extraction procedure of prucalopride

The liquid-liquid extraction technique performed plasma extraction. 400 μl volume of plasma sample transferred to a 15 ml plastic Carson tube. Then 100 μl of internal standard propranolol (25 μg) spiked for getting 5 μg/ml of standard internal concentration in the final prepared sample and added 500 μl n-Hexane then vortex 1 min added 1.0 ml of Diethyl Ether and vortex for 1 min then 3 ml TBME added to the sample tubes. After that, the sample was vortex mixed for 5 min and then centrifuged at 5000 r.p.m for 10 min. Then obtained supernatant organic layer 3.5 ml was transferred to a 15 ml plastic Carson tube and evaporated to dryness at 400°C under a stream of nitrogen. Then the dried extract was reconstituted in 200 μl of diluents (mobile phase) acetonitrile: water (50:50) and vortex for 2 min, then taken into autosampler vial, and 10 μl aliquot injected into chromatographic system.

Stock solution and calibration standards preparation

Preparation of prucalopride and propranolol stock solution (W/V): Weighted about 1.0 mg of prucalopride and propanolol separately and dissolved in 1.0 ml DMSO in a separate tube then mixed well by a vortex. So this each of the two stocked solution concentrations was 1mg/ml and stored in the refrigerator at 2˚C-8˚C. This stock solution was used to prepare an intermediate concentrated solution of prucalopride and internal standard, respectively.

Preparation of calibration concentrations in plasma: To prepare calibration concentrations in plasma of prucalopride, use a liquidliquid extraction procedure to extract prucalopride from human plasma. So from prepared stock solutions of prucalopride and IS (propranolol) diluted with ACN: water 50: 50 (v/v) and prepared intermediate concentrations. The intermediate each of the corresponding concentrated solutions 500 μl of Prucalopride transfer into 500 μl blank human plasma to achieve calibration concentrated points 0.25 ng/ml, 0.50 ng/ml, 1.00 ng/ml, 2.00 ng/ ml, 4.00 ng/ml, 8.00 ng/ml, 16.00 ng/ml including LLOQ 0.25 ng/ ml, LQC 0.75 ng/ml, MQC 6.0 ng/ml, HQC 12.0 ng/ml.

Method validation: The method validation performed as per guidelines for bioavailability and bioequivalence studies published by Central Drugs Standard Control Organization (CDSCO) and EMA [26].

Results and Discussion

Method validation

Specificity, selectivity, and linearity: The prepared plasma calibration concentrations of prucalopride were 0.25 ng/ml, 0.50 ng/ml, 1.00 ng/ml, 2.00 ng/ml, 4.00 ng/ml, 8.00 ng/ml and 16.00 ng/ml and LLOQ was 0.25 ng/ml, LQC was 0.75 ng/ml, MQC and HQC were 6 ng/ml and 12 ng/ml respectively. The concentrations range of the proposed assay was linear, and it was 0.25 ng/ml to 16 ng/ml in plasma. The calibration curve of linearity was showed in Figure 10, which codes no LIN-2. The Limit of Detection (LOD) of lower concentrations of prucalopride in plasma was 0.10 ng/ ml. The mean regression value of three days linearity was 0.9985, and the %of CV was 0.16. Pre-study linearity of detector response with statistics represented in Table 4 and Table 5.

pharmacology-calibration

Figure 10. Plasma calibration curve of prucalopride.

Table 4. Pre-study linearity of detector response (n=3).

Linearity Concentration (ng/ml)
0.25 0.5 1 2 4 8 16
LIN 1 0.24 0.54 1 1.98 3.92 7.86 16.09
LIN 2 0.25 0.5 0.98 1.98 4.04 8.35 15.48
LIN 3 0.23 0.56 1.02 2.04 3.78 7.98 15.35
Average 0.24 0.533 1 2 3.913 8.063 15.64
S.D 0.01 0.031 0.02 0.035 0.13 0.255 0.395
% C.V. 4.167 5.728 2 1.732 3.325 3.168 2.526
Nominal % 96 106.67 100 100 97.83 100.79 97.75

Table 5. Pre-study linearity of detector response statistics (n=3).

Linearity Statistics
Linearitycode Slope (m) Intercept (c) R square
LIN 1 0.123 -0.00524 0.9991
LIN 2 0.118 0.002 0.9996
LIN 3 0.123 -0.00322 0.9967
MEAN 0.1213 Not Aplicable 0.9985
S.D. 0.0029 0.0016
C.V.% 2.38 0.16

Precision and accuracy: In the case of between-run precision and accuracy, the percent of coefficient of variation range was 3.825 to 5.516 for precision, and the percent of absolute bias was 101.33% for LLOQ, 102.58% for LQC, 99.99% for MQC, and 96.14% for HQC for accuracy In the case of within-run precision and accuracy, the percent of coefficient of variation range was 2.645 to 3.683 for precision, and percent of absolute bias range was 94.92 to 100.00 for LLOQ to HQC range accuracy. The obtained result of precision and accuracy illustrated in Table 6.

Table 6. Precision and accuracy (n=5).

Between run Within run
Sample Mean ± SD C.V.% Absolute bias (%) Mean ± SD C.V.% Absolute bias (%)
LLOQ(0.25 ng/ml) 0.253 ±0.014 5.516 101.33 0.250 ± 0.007 2.828 100
LQC(0.75 ng/ml) 0.769 ±0.032 4.182 102.58 0.750 ±0.020 2.667 100
MQC(6.00 ng/ml) 5.999 ±0.229 3.825 99.99 5.956 ±0.219 3.683 99.27
HQC(12.00 ng/ml) 11.537 ±0.533 4.616 96.14 11.390 ± 0.301 2.645 94.92

Stability: The result of the stability study illustrated in Table 7.

Table 7.Stability Study (Freeze thaw, Short term, Auto sampler, Bench
top stability, Long term stability)

Inj No. LQC(0.75ng/ml) MQC(6.00ng/ml) HQC(12.00ng/ml)
Freshly Thawed 1 0.81 6.06 11.73
2 0.81 6.37 11.03
3 0.8 6.13 11.61
4 0.82 6.09 12.32
5 0.78 5.85 11.63
Mean 0.8 6.1 11.66
Freeze Thaw Stability After 3 cyccle in 2-8˚ in freeze 1 0.75 6.6 12.37
2 0.69 6.19 11.72
3 0.78 5.49 11.17
4 0.82 6.58 12.62
5 0.84 5.38 11.44
Mean 0.78 6.05 11.86
  % Stability 96.52 99.15 101.71
Short term stability After 24 hours in 2˚C -8˚C in freeze 1 0.74 5.9 10.68
2 0.73 5.95 13.17
3 0.78 5.36 11.16
4 0.75 5.51 12.75
5 0.76 6.73 11.55
Mean 0.75 5.89 11.86
% Stability 93.53 96.56 101.7
Auto Sampler Stability After 24 hours in auto sampler (15˚C) 1 0.79 5.39 10.9
2 0.73 6.55 12.92
3 0.73 5.75 11.05
4 0.83 5.66 11.78
5 0.72 5.43 10.98
Mean 0.76 5.76 11.53
% Stability 94.53 94.36 98.82
Bench top stability After 24 hours in laboratory room temperature 1 0.79 5.62 11.12
2 0.78 6.16 11.09
3 0.79 6.5 11
4 0.79 5.7 11.56
5 0.7 5.37 11.03
Mean 0.77 5.87 11.16
% Stability 95.77 96.23 95.68
Long term Stability After 7 days in 2˚C-8˚C in freeze 1 0.73 6.35 11.76
2 0.73 6.16 12.23
3 0.78 5.97 11.35
4 0.8 5.92 12.89
5 0.8 6.11 13.14
Mean 0.77 6.1 12.27
% Stability 95.52 100.03 105.23

Short term stability: The stability samples of LQC, MQC and HQC analyzed after keeping the samples in the refrigerator at 2˚C-8˚C for 24 hours and compared these samples with fresh samples of the same concentration. The result of short term stability of prucalopride obtained from 93.53%- 101.70%

Freeze-thaw stability: The stability samples of LQC, MQC and HQC analyzed after keeping the samples in the refrigerator at 2˚C-8˚C for three cycles and compared these samples with fresh samples of the same concentration. The result of freeze-thaw stability of prucalopride obtained from 96.52%-101.71%

Auto-sampler stability: The stability samples of LQC, MQC and HQC analyzed after keeping the samples in an auto sample at 15˚C and compared with fresh samples of the same concentration. The result of autosampler stability of prucalopride obtained from 94.36%-98.82%

Benchtop stability: The stability samples of LQC, MQC and HQC analyzed after keeping the samples on the sample preparation bench at laboratory environmental room temperature and compared these samples with fresh samples of the same concentration. The result of benchtop stability of prucalopride obtained from 95.68%-96.23%

Long term stability: The stability samples of LQC, MQC and HQC analyzed after keeping the samples in the refrigerator at 2˚C-8˚C for seven days and compared these samples with fresh samples of the same concentration. The result of long term stability of prucalopride obtained from 95.52%- 105.23%

Matrix effect: The matrix effect of mass analysis of the analyte is the suppression of ionization of the analyte and IS in mass spectrometry. This parameter calculates by the peak area of the analyte and IS of the plasma extracted samples of LQC, MQC and HQC concentrations comparing with the unextracted samples of the same concentrations. The matrix factor of prucalopride was 0.92-0.95, and matrix effect ranges were 92.31%-94.82%, and for IS matrix factor was 0.95 and matrix effect was 94.79%-95.18%. The matrix effect of prucalopride illustrated in Table 8.

Table 8. Matrix effect (area) (N=5)

Sample Matrix effect IS Matrix effect Prucalopride
% of ME Matrix factor % of ME Matrix factor
LQC (0.75 ng/ml) 94.95 ± 3.66 0.95 ± 0.4 92.31 ± 2.32 0.92 ± 0.02
MQC (6.00 ng/ml) 94.79 ± 4.04 0.95 ± 0.4 94.82 ± 2.70 0.95 ± 0.03
HQC (12.00 ng/ml) 95.18 ± 2.62 0.95 ± 0.3 94.74 ± 3.86 0.95 ± 0.04

Recovery: Recovery of mass analysis of the analyte and IS calculated by the peak area of the analyte and IS of the plasma extracted samples of LQC, MQC, and HQC concentrations comparing with the unextracted samples of the same concentrations. The recovery of prucalopride was 100.43%- 111.83%, and for IS, it was 95.66%-106.02%. The recovery result of prucalopride is illustrated in Table 9 and Table 10.

Table 9. Recovery of IS.

INJ No. Diluent Sample In Plasma
L 0.75 ng/ml MQC 6.00 ng/ml H 12 ng/ml LQC 0.75 ng/ml MQC 6.00 ng/ml HQC 12 ng/ml
1 70132.46 68283.9 73732.73 75982.59 68029.45 67391.87
2 95865.56 67887.55 67707.4 65385.23 67865.24 88859.28
3 63211.14 94384.67 65609.54 81186.64 85221.24 74671.35
4 78147.19 75667.52 112860.53 85800.22 62345.28 88445.57
5 60634.57 79912.52 67686.94 81776.07 85926.6 66348.09
Mean 73598.18 77227.23 77519.43 78026.15 73877.56 77143.23
% Recovery 106.02 95.66 99.51

Table 10. Recovery of Prucalopride.

INJ No. Diluent Sample In Plasma
LQC 0.75 ng/ml MQC 6.00 ng/ml HQC 12.00 ng/ml LQC 0.75 ng/ml MQC 6.00 ng/ml HQC 12.00 ng/ml
1 5951.24 42134.83 92371.6 6120.5 44304.5 105546.93
2 8629.33 44551.84 85602.85 6193.08 42910.1 117985.14
3 6072.98 69771.72 85820.57 7908.51 56142.68 108367.84
4 6507.96 47394.4 148659.48 7975.65 44709.54 112446.6
5 4877.86 49325.84 100658.96 7630.86 66211.11 84466.55
Mean 6407.87 50635.73 102622.69 7165.72 50855.59 105762.61
% Recovery 111.83 100.43 103.06

Result of comparative pharmacokinetic study of prucalopride in human volunteers

From the analysis of unknown concentrations of prucalopride 2 mg in human plasma by LCMS/MS, itobserved that after administration of reference film-coated tablet of prucalopride 2 mg as a single dose in the fasting state reached maximum plasma concentration 4.98 ng/ml ± 0.95 ng/ml (Cmax) at the time of 2.71 hours ± 0.33 hours. (Tmax) and on the other hand, after administration of test preparation of film-coated prucalopride 2 mg in same volunteers, it observed that the maximum plasma concentration of prucalopride 2 mg reached 5.01 ng/ml ± 1.05 ng/ml (Cmax) at the time of 2.64 hours ± 0.28 hours (Tmax). The area under the curve of plasma concentration time of reference preparation was 30.34 ng.h/ml ± 8.07 ng.h/ml (AUC0-t) and 31.96 ng.h/ml ± 8.09 ng.h/ml (AUC0-∞), but in test preparation, it was 30.44 ng.h/ml ± 9.78 ng.h/ml (AUC0-t) and 31.82 ng.h/ml ± 10.10 ng.h/ml (AUC0-∞) shown in Figure 11. The elimination constant (Kel) value in reference preparation was 0.057 hour ± 0.014 hour. 1 hour and 0.058 hour ± 0.011 hour. (1 hour for test preparation). The elimination half life (T1/T2) of reference preparation was 12.84 hour ± 2.78 hour and 12.37 hour ± 2.28 hour for test preparation. From the above comparative pharmacokinetic study of prucalopride 2 mg, the relative bioavailability of the prucalopride test sample was 100.35% compared to the reference tablet prucalopride. The comparative pharmacokinetic parameters illustrated in Table 11.

pharmacology-pharmacokinetics

Figure 11. Mean pharmacokinetics plasma concentration time profile of prucalopride.

Table 11. Pharmacokinetic parameters of prucalopride (n=24).

Pharmacokinetic parameter Prucalopride
Reference preparation (A1) Test preparation (A2)
Cmax (ng./ml.) Mean 4.98 5.01
S.D. 0.95 1.05
Tmax (hour.) Mean 2.71 2.64
S.D. 0.33 0.28
AUC 0-t (ng. hour./ml.) Mean 30.34 30.44
S.D. 8.07 9.78
AUC 0-inf (ng. hour./ml.) Mean 31.96 31.82
S.D. 8.09 10.1
kel (hr.-1) Mean 0.057 0.058
S.D. 0.014 0.011
T1/2 (hr.) Mean 12.84 12.37
S.D. 2.78 2.28
Relative Bioavailability (%) 100% 100.35%

Result of statistical analysis

Statistical ANOVA test using SAS 9.1.3 Version-Grizzles Model by PROC GLM was applied to calculate Ln Cmax, AUC0-t and Ln AUC0-t values.

The results of Cmax, AUC0-t, AUC0-∞, Tmax, Kel, T1/2 of ANOVA for untransformed and logtransformed data of Cmax show that the parameters like Period and Treatment are not statistically significant at 5% level in both untransformed and log-transformed data. Still, subject and sequence are statistically significant at 5% in both untransformed and log-transformed data i.e. p<0.05.

The 90% confidence interval for the ratio (Test/Reference) of geometric means, based on the log-transformed data for Cmax, was found to be 97.12%-103.98% relative to Test Preparation with Reference Preparation, for AUC0-t found to be 97.95%- 101.04% relative to Test Preparation with Reference Preparation, for AUC0-inf found to be 97.53%-101.02% close to Test Preparation with Reference Preparation.

The 95% confidence interval for the ratio (Test/Reference) of geometric means, based on the log-transformed data for Cmax, was found to be 97.63%-101.36% relative to Test Preparation with Reference Preparation, for AUC0-t found to be 97.63%- 101.36% relative to test preparation with reference preparation, for AUC0-inf found to be 97.16%-101.38% close to test preparation with reference preparation.

Application of paired t-test that the Cmax (untransformed data) for test preparation and reference preparation is not statistically significant at 5% level and p-value=0.874. Both the trials are similar effectson the body. The summarised statistical analysis report shown in Table 12.

Table 12. Representation of class level information.

Class Level Information
Class Levels Values
Sequence 2 12
Treatment 2 12
Subject 24 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
Period 2 12
Number of Observations Read 48
Number of Observations Used 48

Dependent variable: Cmax in Tables 13-16.

Table 13. Annova for response of the “model.”

Source DF Sum of Squares Mean Square F Value pr>F
Model 25 39.13519167 1.56540767 4.73 0.0002
Error 22 7.27373333 0.33062424 - -
Corrected Total 47 46.408925 - - -

Table 14. Representation of R2 , Coefficient variable, Root MSE, and Cmax Mean.

R-Square Coeff Var Root MSE Cmax Mean
0.843269 11.51438 0.574999 4.99375

Table 15. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean Square F Value pr>F
Period 1 1.29363333 1.29363333 3.91 0.0606
Sequence 1 6.885675 6.885675 20.83  0.0002
Subject (sequence) 22 30.94625 1.40664773 4.25 0.0006
Treatment 1 0.00963333 0.00963333 0.03 0.866

Table 16. Fit statistics response for the model of type III.

Source DF Type III (SS) Mean Square F Value pr>F
Period 1 1.29363333 1.29363333 3.91 0.0606
Sequence 1 6.885675 6.885675 20.83 0.0002
Subject (sequence) 22 30.94625 1.40664773 4.25 0.0006
Treatment 1 0.00963333 0.00963333 0.03 0.866

Dependent variable: AUC0-t in Tables 17-20.

Table 17. Annova for response of the “model.”

Source DF Sum of Squares Mean Square F Value pr>F
Model 25 3549.0339 141.96136 21.23 <.0001
Error 22 147.08151 6.685523 - -
Corrected Total 47 3696.1155 - - -

Table 18. Representation of R2, Coefficient variable, Root MSE, and AUC0-t Mean.

R-Square Coeff Var Root MSE AUCt Mean
0.960206 8.50828 2.585638 30.38966

Table 19. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean Square F Value pr>F
Period 1 18.372297 18.372297 2.75 0.1116
Sequence 1 937.680901 937.680901 140.26 <.0001
Subject (sequence) 22 2592.848306 117.856741 17.63 <.0001
Treatment 1 0.132437 0.132437 0.02 0.8894

Table 20. Fit statistics response for the model of type III.

Source DF Type III (SS) Mean Square F Value pr>F
Period 1 18.372297 18.372297 2.75 0.1116
Sequence 1 937.680901 937.680901 140.26 <.0001
Subject (sequence) 22 2592.848306 117.856741 17.63 <.0001
Treatment 1 0.132437 0.132437 0.02 0.8894

Dependent variable : AUC0-∞ in Tables 21-24.

Table 21. Annova for response of the “model.”

Source DF Sum of Squares Mean Square F Value pr>F
Model 25 3629.820281 145.192811 14.12 <.0001
Error 22 226.288472 10.28584 - -
Corrected Total 47 3856.108753 - - -

Table 22. Representation of R2 , Coefficient variable, Root MSE, and AUCinf Mean.

R-Square Coeff Var Root MSE AUCinf Mean
0.941317 10.05244 3.207154 31.90425

Table 23. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean Square F Value pr>F
period 1 21.736016 21.736016 2.11 0.1602
sequence 1 904.770337 904.770337 87.96 <.0001
subject (sequence) 22 2703.092221 122.867828 11.95 <.0001
treatment 1 0.221707 0.221707 0.02 0.8846

Table 24. Fit statistics response for the “model” of type I.

Source DF Type III (SS) Mean Square F Value pr>F
Period 1 21.736016 21.736016 2.11 0.1602
Sequence 1 904.770337 904.770337 87.96 <.0001
Subject (sequence) 22 2703.092221 122.867828 11.95 <.0001
Treatment 1 0.221707 0.221707 0.02 0.8846

Dependent variable: Ln Cmax in Tables 25-28.

Table 25. Annova for response of the “model.”

Source DF Sum of Squares Mean Square F Value pr>F
Model 25 1.65321574 0.06612863 5.32 <.0001
Error 22 0.27326483 0.01242113 - -
Corrected Total 47 1.92648057 - - -

Table 26. Representation of R2, Coefficient variable, Root MSE, and LnCmax Mean.

R-Square Coeff Var Root MSE LnCmax Mean
0.858153 7.016677 0.11145 1.58836

Table 27. Fit statistics response for the “model” of type I

Source DF Type I (SS) Mean Square F Value pr>F
Period 1 0.04650698 0.04650698 3.74 0.066
Sequence 1 0.31598188 0.31598188 25.44 <.0001
Subject (sequence) 22 1.29064962 0.05866589 4.72 0.0003
Treatment 1 0.00007727 0.00007727 0.01 0.9378

Table 28. Fit statistics response for the “model” of type I.

Source DF Type III (SS) Mean square F Value pr>F
Period 1 0.04650698 0.04650698 3.74 0.066
Sequence 1 0.31598188 0.31598188 25.44 <.0001
Subject (sequence) 22 1.29064962 0.05866589 4.72 0.0003
Treatment 1 0.00007727 0.00007727 0.01 0.9378

Dependent variable: Ln AUC0-t in Tables 29-32.

Table 29. Annova for response of the “model.”

Source DF Sum of squares Mean square F Value pr>F
Model 25 5.0952268 0.20380907 21.12 <.0001
Error 22 0.2123498 0.00965226 - -
Corrected Total 47 5.3075766 - - -

Table 30. Representation of R2, Coefficient variable, Root MSE, and LnAUCt Mean.

R-Square Coeff Var Root MSE LnAUCt mean
0.959991 2.920634 0.098246 3.363856

Table 31. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean square F Value pr>F
Period 1 0.0158377 0.0158377 1.64 0.2136
Sequence 1 1.2210363 1.2210363 126.5 <.0001
Subject (sequence) 22 3.85589462 0.17526794 18.16 <.0001
Treatment 1 0.00245817 0.00245817 0.25 0.6188

Table 32. Fit statistics response for the “model” of type I.

Source DF Type III (SS) Mean square F Value pr>F
Period 1 0.0158377 0.0158377 1.64 0.2136
Sequence 1 1.2210363 1.2210363 126.5 <.0001
Subject (sequence) 22 3.85589462 0.17526794 18.16 <.0001
Treatment 1 0.00245817 0.00245817 0.25 0.6188

Dependent variable : Ln AUC0-∞ Tables 33-36.

Table 33. Annova for response of the “model.”

Source DF Sum of squares Mean square F Value pr>F
Model 25 4.68997647 0.18759906 14.5 <.0001
Error 22 0.28463085 0.01293777 - -
Corrected Total 47 4.97460732 - - -

Table 34. Representation of R2, Coefficient variable, Root MSE, and LnAUCinf Mean.

R-Square Coeff Var Root MSE LnAUCinf Mean
0.942783 3.330096 0.113744 3.415646

Table 35. Representation of R2, Coefficient variable, Root MSE, and LnAUCinf Mean.

Source DF Type I SS Mean square F Value pr>F
Period 1 0.014672 0.014672 1.13 0.2985
Sequence 1 1.035116 1.035116 80.01 <.0001
Subject(sequence) 22 3.633799 0.165173 12.77 <.0001
Treatment 1 0.006389 0.006389 0.49 0.4896

Table 36. Fit statistics response for the model of type III.

Source DF Type III (SS) Mean square F Value pr>F
Period 1 0.01467201 0.01467201 1.13 0.2985
Sequence 1 1.03511628 1.03511628 80.01 <.0001
Subject (sequence) 22 3.63379871 0.16517267 12.77 <.0001
Treatment 1 0.00638947 0.00638947 0.49 0.4896

Dependent variable : Tmax Tables 37-40.

Table 37. Annova for response of the “model.”

Source DF Sum of squares Mean square F Value pr>F
Model 25 4.14468333 0.16578733 16.9 <.0001
Error 22 0.21578333 0.00980833 - -
Corrected Total 47 4.36046667 - - -

Table 38. Representation of R2 , Coefficient variable, Root MSE, and Tmax Mean.

R-Square Coeff Var Root MSE Tmax Mean
0.950514 3.704627 0.099037 2.673333

Table 39. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean square F Value pr>F
Period 1 0.00240833 0.00240833 0.25 0.6251
Sequence 1 0.68163333 0.68163333 69.5 <.0001
Subject (sequence) 22 3.40323333 0.15469242 15.77 <.0001
Treatment 1 0.05740833 0.05740833 5.85 0.0243

Table 40. Fit statistics response for the model of type III.

Source DF Type III (SS) Mean square F Value pr>F
Period 1 0.00240833 0.00240833 0.25 0.6251
Sequence 1 0.68163333 0.68163333 69.5 <.0001
Subject (sequence) 22 3.40323333 0.15469242 15.77 <.0001
Treatment 1 0.05740833 0.05740833 5.85 0.0243

Dependent variable : Kel in Tables 41-44.

Table 41. Annova for response of the “model.”

Source DF Sum of Squares Mean Square F Value pr>F
Model 25 0.005737 0.00022948 2.93 0.0065
Error 22 0.00172067 0.00007821 - -
Corrected Total 47 0.00745767 - - -

Table 42. Representation of R2 , Coefficient variable, Root MSE, and Kel Mean.

R-Square Coeff Var Root MSE Kele Mean
0.769275 15.53809 0.008844 0.056917

Table 43. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean Square F Value pr>F
Period 1 0.00028033 0.00028033 3.58 0.0716
Sequence 1 0.00028033 0.00028033 3.58 0.0716
Subject (sequence) 22 0.00516433 0.00023474 3 0.0064
Treatment 1 0.000012 0.000012 0.15 0.699

Table 44. Fit statistics response for the model of type III.

Source DF Type III (SS) Mean Square F Value pr>F
Period 1 0.00028033 0.00028033 3.58 0.0716
Sequence 1 0.00028033 0.00028033 3.58 0.0716
Subject(sequence) 22 0.00516433 0.00023474 3 0.0064
Treatment 1 0.000012 0.000012 0.15 0.699

Dependent variable : T1/2 in Tables 45-51.

Table 45. Annova for response of the “model”.

Source DF Sum of Squares Mean Square F Value pr>F
Model 25 253.0316482 10.1212659 3.85 0.0011
Error 22 57.7888073 2.626764 - -
Corrected Total 47 310.8204555 - - -

Table 46. Representation of R2 , Coefficient variable, Root MSE, and Thalf Mean.

R-Square Coeff Var Root MSE Thalf Mean
0.814077 12.74987 1.620729 12.71173

Table 47. Fit statistics response for the “model” of type I.

Source DF Type I (SS) Mean Square F Value pr>F
Period 1 2.4430675 2.4430675 0.93 0.3453
Sequence 1 29.816345 29.816345 11.35 0.0028
Subject (sequence) 22 218.148158 9.9158254 3.77 0.0015
Treatment 1 2.6240777 2.6240777 1 0.3284

Table 48. Fit statistics response for the model of type III.

Source DF Type III (SS) Mean Square F Value pr>F
Period 1 2.4430675 2.4430675 0.93 0.3453
Sequence 1 29.816345 29.816345 11.35 0.0028
Subject (sequence) 22 218.148158 9.9158254 3.77 0.0015
Treatment 1 2.6240777 2.6240777 1 0.3284

Table 49. Statistical report.

Level of period N Cmax AUCt AUCinf LnCmax
Mean Std Dev Mean Std Dev Mean Std Dev Mean Std Dev
1 24 5.157917 1.059914 31.00834 9.286691 32.57718 9.621599 1.619488 0.211704
2 24 4.829583 0.915487 29.77099 8.582497 31.23133 8.610267 1.557233 0.192145

Table 50. Version-Grizzles model by PROC GLM.

Level of period N LnAUCt LnAUCinf Tmax Kele
Mean Std Dev

Mean

Std Dev Mean Std Dev Mean Std Dev
1 24 3.382021 0.348156 3.433129 0.343011 2.66625 0.31193 0.05933 0.016048
2 24 3.345692 0.329944 3.398163 0.313038 2.68042 0.30361 0.0545 0.007384

Table 51. The GLM procedure.

Level of  period N T half
Mean Std Dev

1

24 12.486125 3.2262967

2

24 12.937333 1.731682

Overall Conclusion

According to the criteria, it was sensitive, selective, had minimal ion suppression, was highly recoverable, and was repeatable. Aside from that, this technique successfully utilised to quantify prucalopride in unidentified volunteer plasma samples and an in-vivo pharmacokinetic and bioequivalence investigation. There were just a few articles on bioanalytical technique development, based on the literature review. LC-MS/MS API-4000 was utilised in this work to establish bioanalytical methods to quantify prucalopride in human plasma.

Conflict of Interest

None

Acknowledgement

The author is thankful to M/S, TAAB Biostudy Services, Kolkata-700032, India, for providing human plasma and delivering the necessary instrumental facilities. In addition, the authors acknowledge theDepartment of Pharmaceutical Technology, Jadavpur University, Kolkata, India, Calcutta School of Tropical Medicine, Kolkata 700073, India.National Institute of Technology, Tiruchirappalli, Government of India, India, for supporting this analysis.

Ethical Considerations

The protocol of prucalopride study and related documents like volunteers informed consent form, case record form, subject information sheet was submitted to the HURIP Independent Bioethics committee, Kolkata, India (Central Drugs Standard Control Organization (CDSCO) registration: ECR/103/Indt/ WB/2013/RR-19 which is valid up to 21-Nov, 2024) and the ethical clearance obtained before initiation of the study.

Safety Assessment

Clinical examinations were done at the time of screening, checkin and checked out from the clinical facility in each period. All vital signs like blood pressure, pulse ratemeasured, and wellbeing enquiry performed before check-in, before dosing, at 3 hours, 6 hours, 9 hours and 12 hours post-dose ( ± 60 minute) and check Fout of each period whenever felt necessary by the investigators or medical doctors Body temperature was monitored during the pre-dose and at the time of check out. In addition to the above well being, the assessment performed during each ambulatory sample collection. Clinical laboratory tests performed during screening and at the end of the study. The subjects did not report any adverse events and toxic effects at the time of the survey of prucalopride.

References

open access journals, open access scientific research publisher, open access publisher
Select your language of interest to view the total content in your interested language

Viewing options

Flyer image

Share This Article

paper.io

agar io

wowcappadocia.com
cappadocia-hotels.com
caruscappadocia.com
brothersballoon.com
balloon-rides.net

wormax io