Comprehensive UHPLC-MS screening methods for the analysis of triazolopyrazine precursor and its genotoxic nitroso-derivative in sitagliptin pharmaceutical formulation

A case study on Sitagliptin drug products and Sitagliptin/Metformin drug products concerning contamination with N-nitrosamines was performed using two newly developed analytical methods for determination of N-nitroso-triazolopyrazine (NTTP; 7-nitroso-3-(trifluoromethyl)-5,6,7,8-tetrahydro-[1,2,4]triazolo[4,3-a]pyrazine) and its precursor triazolopyrazine (3-(trifluoromethyl)-5,6,7,8-tetrahydro-[1,2,4]triazolo[4,3-a]pyrazine). The method for determination of triazolopyrazine was previously unpublished, the method for determination of NTTP was published only for analysis of active pharmaceutical ingredient Sitagliptin and not the drug forms. Solving the N-nitrosamine contamination is requested by regulatory authorities all over the world and thus is vital for all pharmaceutical companies. The solution always requires a sensitive analytical method. Both newly developed methods use liquid chromatography coupled with mass spectrometry (single quadrupole analyzer in case of triazolopyrazine and triple quadrupole analyzer in case of NTTP). Separation of triazolopyrazine was achieved on a column Acquity CSH C18 using a mobile phase consisting of aqueous ammonium formate buffered at pH 4.2 and acetonitrile. Detection was performed using positive electrospray and selected ion monitoring at m/z 193. Separation of NTTP was achieved on a column Acquity HSS T3 using a mobile phase consisting of 0.1 % formic acid in water and methanol. Detection was performed using positive electrospray and multiple reaction monitoring at transitions m/z 222.15→42.05 (collision energy 17 eV) and m/z 222.15→192.15 (collision energy 11 eV). Two issues specific to NTTP and triazolopyrazine previously not described in scientific literature were successfully troubleshooted. Spontaneous degradation of Sitagliptin to triazolopyrazine and methyl (R)-3-amino-4-(2,4,5-trifluorophenyl)butanoate was solved by using N,N-dimethylformamide as sample solvent during development of the method for quantitation of triazolopyrazine. A bad peak shape of NTTP due to the presence of rotamers of NTTP was successfully troubleshooted by increasing column temperature. Both methods were used during an optimization study of manufacturing of Sitagliptin and Sitagliptin/Metformin drug products. The goal of the study was to decrease NTTP content in the final drug product under the strict legislative limit set by Federal Drug Agency. The efficacy of several solutions was proven, but could not be fully disclosed due to Intellectual Property Protection policy of Zentiva. Instead, a brief review of recently published strategies to cope with N-nitrosamine contamination is presented.

N-Nitrosation in the absence of nitrosating agents in pharmaceuticals?

The possibility of N-Nitrosation in the absence of nitrosating agents was studied on model solutions and film coated tablets containing metformin. N-nitrosodimethylamine (NDMA) and N-nitrosation precursors (dimethylamine and nitrites) were determined using previously published fully validated analytical methods. Alternative routes to N-nitrosation were found. Dimethylamine can undergo an oxidation to nitrite in the presence of strong oxidants (e.g., H₂O₂), as was observed during wastewater treatment in several published works. The resulting nitrite can consecutively act as a nitrosating agent. We proved that the described reaction indeed leads to N-nitrosation (NDMA formation in case of dimethylamine precursor) in model solutions made of dimethylamine and H₂O₂. An experiment was designed in order to prove those reactions take place in dosage forms. Film coated tablets present a highly heterogenous system with several solid phases and low water activity, which is in stark contrast to the liquid wastewater, where this reaction was originally studied. Despite that, the described reaction took place even in the tablets, but only to a small degree. The amount formed via this alternative route corresponds to less than 10 % of the total formed NDMA. The pH optimum of this alternative route lies in the alkaline range which was confirmed by the determined NDMA concentration in model solutions. The solid phase system (i.e., tablets) was found to behave differently. The addition of Na₂CO₃ into the tablets during manufacture resulted in tablets without NDMA (c_NDMA < LOQ) even in batches spiked with both dimethylamine and H₂O₂. Thus, adjusting the pH of the solid dosage forms remains a sufficient measure of controlling N-nitrosamines in the product, even in product with limit amounts of oxidating agent (H₂O₂) and N-nitrosation precursor (dimethylamine).

The determination of two analogues of 4-(azidomethyl)-1,1'-biphenyl as potential genotoxic impurities in the active pharmaceutical ingredient of several sartans containing a tetrazole group

4'-(azidomethyl)-[1,1'-biphenyl]-2-carbonitrile (GTI-azide-1) and 5-(4'-(azidomethyl)-[1,1'-biphenyl]-2-yl)-1H-tetrazole (GTI-azide-2) are potentially genotoxic impurities that can be present at trace levels in the active pharmaceutical ingredients and drug products of sartans containing a tetrazole group. A method of high-performance liquid chromatography coupled with mass spectrometry, that allows the determination of those genotoxic impurities at sub-ppm level relative to the active pharmaceutical ingredient, was developed. The method utilises a very efficient liquid chromatograph Waters Acquity I-Class coupled with a highly sensitive tandem mass spectrometer Xevo TQ-XS. The separation was achieved on a column Acquity UPLC BEH Shield RP18 1.7 μm employing a linear elution gradient. The mass spectrometer was used with a heated electrospray ionization. The method was found to be sufficient in terms of sensitivity, linearity, precision, accuracy, selectivity and robustness and is easily applicable in the pharmaceutical quality control environment. The method allows for accurate quantification of both impurities GTI-azide-1 and GTI-azide-2 at levels below 1/10^th of the specification limit, which is crucial in the context of pharmaceutical analysis. The limit of quantification was determined to be 0.033 ppm and 0.025 ppm for GTI-azide-1 and GTI-azide-2, respectively.

Insight into the formation of N-nitrosodimethylamine in metformin products

An effective analytical method for the quantification of N-nitrosodimethylamine (NDMA) using a liquid chromatography coupled with tandem mass spectrometry was developed and applied to a process optimization study of the production of metformin film coated tablets in order to identify the key factors behind the NDMA formation in metformin products. The method uses a linear gradient elution with mobile phases 0.1 % formic acid in water for chromatography and methanol for chromatography and a column Acquity UPLC HSS T3 1.8 μm. The use of the tandem mass spectrometry in a positive ion mode with an atmospheric pressure chemical ionization allows for the use of an isotopically labelled internal standard and an external calibration standard. The method was validated according to the guidelines of International Council for Harmonization in terms of limit of detection and quantification, linearity, precision, accuracy and method selectivity. To further justify the effectiveness of the method, a comparison between two laboratories was performed using a linear regression testing. Both methods give comparable results. 469 samples of both metformin active pharmaceutical ingredient and film coated tablets were analysed and the key factors behind NDMA formation were identified. Hypotheses explaining the mechanism were formulated and confronted with measurements and scientific literature. Protective measures to prevent NDMA contamination in metformin products were drawn.

Fast HPLC Method Using Ion-Pair and Hydrophilic Interaction Liquid Chromatography for Determination of Phenylephrine in Pharmaceutical Formulations

A rapid procedure based on a direct extraction and HPLC determination with fluorescence detection of phenylephrine in pharmaceutical sachets that include a large excess of paracetamol (65 + 1, w/w), ascorbic acid (5 + 1, w/w), and other excipients (aspartame and sucrose) was developed and validated. The final optimized chromatographic method for ion-pair chromatography used an XTerra RP18 column, 3 µm particle size, 50 ⨯ 3.0 mm id. The mobile phase consisted of a mixture of acetonitrile and buffer (10 mM sodium octane-1-sulfonate, adjusted with H3PO4 to pH 2.2; 200 + 800, v/v), with a constant flow rate of 0.3 mL/min. The separation was carried out at 30C, and the injection volume was 3 µL. Fluorescence detection was performed at excitation and emission wavelengths of 275 and 310 nm, respectively. The mobile phase parameters, such as the organic solvent fraction (acetonitrile) in mobile phase as an organic modifier, the concentration of sodium octane-1-sulfonate as a counter-ion, temperature, and pH of mobile phase, were studied. As an alternative to ion-pair chromatography, hydrophilic interaction liquid chromatography (HILIC) was investigated using a Luna HILIC column, 3 m, 100 ⨯ 4.6 mm id. The mobile phase consisted of acetonitrile and buffer (5 M potassium dihydrogen phosphate, adjusted with H3PO4 to pH 2.5; 750 + 250, v/v) at a flow rate of 0.8 mL/min. The separation was carried out at 25°C, and the injection volume was 5 µL. The proposed method has an advantage of a very simple sample pretreatment, and is much faster than the currently utilized HPLC methods using gradient elution and UV detection. Commercial samples of sachets were successfully analyzed by the proposed HPLC method.

Identification and structure elucidation of a new degradation impurity in the multi-component tablets of amlodipine besylate

New unknown impurity at m/z 421.15 was observed during the accelerated stability analysis (40 °C/75% relative humidity) in the multi-component tablets of amlodipine besylate by reversed-phase ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS). UHPLC-MS and nuclear magnetic resonance (NMR) techniques were employed to identify and fully characterize the degradation compound. The degradation product was unambiguously identified as 3-ethyl 5-methyl 4-(2-chlorophenyl)-6-methyl-2-(morpholin-2-yl)-1,4-dihydropyridine-3,5-dicarboxylate and mechanism of its formation was proposed. It was confirmed that the degradation product was formed by the reaction of amlodipine with formaldehyde originating from the excipients present in the dosage form.

Esterification of Ibuprofen in Soft Gelatin Capsules Formulations-Identification, Synthesis and Liquid Chromatography Separation of the Degradation Products

Unknown impurities were identified in ibuprofen (IBU) soft gelatin capsules (SGCs) during long-term stability testing by a UHPLC method with UV detection and its chemical formula was determined using high resolution/accurate mass (HRAM) LC-MS. Reference standards of the impurities were subsequently synthesized, isolated by semi-preparative HPLC and characterized using HRAM LC-MS, NMR and IR. Two impurities were formed by esterification of IBU with polyethylene glycol (PEG), which is used as a fill of the SGCs, and were identified as IBU-PEG monoester and IBU-PEG diester. Two other degradants arised from reaction of IBU with sorbitol and sorbitan, which are components of the shell and serves as plasticizers. Thus, IBU sorbitol monoester (IBU-sorbitol) and IBU sorbitan monoester (IBU-sorbitan ester) were identified. An UHPLC method was further optimized in order to separate, selectively detect and quantify the degradation products in IBU SGCs.

Identification, characterization, synthesis and HPLC quantification of new process-related impurities and degradation products in retigabine

Two new impurities were described and determined using gradient HPLC method with UV detection in retigabine (RET). Using LC-HRMS, NMR and IR analysis the impurities were identified as RET-dimer I: diethyl {4,4'-diamino-6,6'-bis[(4-fluorobenzyl)amino]biphenyl-3,3'-diyl}biscarbamate and RET-dimer II: ethyl {2-amino-5-[{2-amino-4-[(4-fluorobenzyl) amino] phenyl} (ethoxycarbonyl) amino]-4-[(4-fluorobenzyl)amino] phenyl}carbamate. Reference standards of these impurities were synthesized followed by semipreparative HPLC purification. The mechanism of the formation of these impurities is also discussed. An HPLC method was optimized in order to separate, selectively detect and quantify all process-related impurities and degradation products of RET. The presented method, which was validated in terms of linearity, limit of detection (LOD), limit of quantification (LOQ) and selectivity is very quick (less than 11min including re-equilibration time) and therefore highly suitable for routine analysis of RET related substances as well as stability studies.

The formation of furfural compounds in selected saccharide- and polysaccharide-based pharmaceutical excipients

The acid hydrolysis of various selected saccharide- and polysaccharide-based pharmaceutical excipients under acid hydrolysis and the formation of degradation compounds were studied. New degradation products formed from these excipients were discovered. Liquid chromatography-mass spectrometry and nuclear magnetic resonance techniques were employed to identify and fully characterize these unknown compounds. The degradation products were identified as [(5-formylfuran-2-yl)methoxy]acetic acid, 5-[(propan-2-yloxy)methyl]furan-2-carbaldehyde, along with the previously identified 5-(methoxymethyl)furan-2-carbaldehyde. On the basis of the identification of these degradation products, a reasonable mechanism for their formation can be proposed. Temperature and pH affect the hydrolysis rates of saccharides and polysaccharides, which in turn affects the rate of formation of furfural compounds.

Fast HPLC method using ion-pair and hydrophilic interaction liquid chromatography for determination of phenylephrine in pharmaceutical formulations

A rapid procedure based on a direct extraction and HPLC determination with fluorescence detection of phenylephrine in pharmaceutical sachets that include a large excess of paracetamol (65 + 1, w/w), ascorbic acid (5 + 1, w/w), and other excipients (aspartame and sucrose) was developed and validated. The final optimized chromatographic method for ion-pair chromatography used an XTerra RP18 column, 3 microm particle size, 50 x 3.0 mm id. The mobile phase consisted of a mixture of acetonitrile and buffer (10 mM sodium octane-1-sulfonate, adjusted with H3PO4 to pH 2.2; 200 + 800, v/v), with a constant flow rate of 0.3 mL/min. The separation was carried out at 30 degrees C, and the injection volume was 3 microL. Fluorescence detection was performed at excitation and emission wavelengths of 275 and 310 nm, respectively. The mobile phase parameters, such as the organic solvent fraction (acetonitrile) in mobile phase as an organic modifier, the concentration of sodium octane-1-sulfonate as a counter-ion, temperature, and pH of mobile phase, were studied. As an alternative to ion-pair chromatography, hydrophilic interaction liquid chromatography (HILIC) was investigated using a Luna HILIC column, 3 microm, 100 x 4.6 mm id. The mobile phase consisted of acetonitrile and buffer (5 mM potassium dihydrogen phosphate, adjusted with H3PO4 to pH 2.5; 750 + 250, v/v) at a flow rate of 0.8 mL/min. The separation was carried out at 25 degrees C, and the injection volume was 5 microL. The proposed method has an advantage of a very simple sample pretreatment, and is much faster than the currently utilized HPLC methods using gradient elution and UV detection. Commercial samples of sachets were successfully analyzed by the proposed HPLC method.

Factors influencing the spontaneous reporting of adverse drug reactions--the experience of the Slovak Republic

The system of adverse drug reaction (ADR) monitoring in the Slovak Republic is greatly influenced by changes taking place in the current health care delivery system. Under-reporting and reporting biases due to selective ADR reporting provide very serious problems in pharmacovigilance. In the year 2001, the number of reported ADRs increased due to reports of angiotensin converting enzyme (ACE) inhibitor-induced cough; this coincided with limits being imposed on the use of angiotensin-II receptor antagonists. These factors provide an example of selective ADR reporting, deforming the quality of drug safety monitoring. They also indicate that administrative strategies can significantly affect spontaneous reporting activities.

Evaluation of ischemia-reperfusion injury by malondialdehyde, glutathione and gamma-glutamyl transpeptidase: lack of specific local effects in diverse parts of the dog heart following acute coronary occlusion

Alterations in the levels of glutathione, glutathione disulfide, malondialdehyde, and the activity of gamma-glutamyl transpeptidase in nonischemic and ischemic parts of the left ventricle and in the right ventricle were studied in canine hearts after occlusion of the left anterior descending coronary artery for 60 minutes and subsequent reperfusion for 20 minutes. Ischemia caused no significant change in malondialdehyde concentration and gamma-glutamyl transpeptidase activity in ischemic or nonischemic parts of the left ventricle, but it increased the activity of gamma-glutamyl transpeptidase in the continuously perfused right ventricle. Reperfusion of the ischemic areas of the left ventricle was accompanied by accumulation of malondialdehyde and an increase in gamma-glutamyl transpeptidase activity, not only in the reperfused and adjacent areas of the left ventricle, but also in the continuously perfused right ventricle. An increase in the level of glutathione disulfide and decrease in glutathione occurred in all parts of the myocardium during coronary occlusion; these changes were maintained in reperfusion. The findings indicate that the effects of acute occlusion and reperfusion of the left anterior descending coronary artery on myocardial concentrations of glutathione, glutathione disulfide and malondialdehyde or gamma-glutamyl transpeptidase activity are not confined to the local area.

[The effect of diltiazem on atrioventricular conduction in dogs]

The effect of diltiazem on atrioventricular conduction was studied in pentobarbital anesthetized dogs, conscious dogs, and atropine pretreated conscious dogs. In anesthetized dogs diltiazem administration resulted in diminished heart rate and aortal pressure, as well as prolongation of the functional refractory period of the AV node and of AV conductance. The effective refractory period of the atria and intraatrial conductance were not affected. The function of the AV node was comparable in the three groups studied. An anticholinergic action of diltiazem affecting the AV node has thus to be excluded. The recorded slight increase in heart rate in conscious dogs may be accounted for by the activation of baroreceptors induced by decreased aortal pressure.

Differences in the responses of the canine basilar and dorsal pedal artery to phenylephrine, clonidine and serotonin

The effects of phenylephrine, clonidine and serotonin were studied on isolated preparations of the canine basilar and dorsal pedal artery. Differences were mainly observed in the responses of the two types of arterial preparations to clonidine and phenylephrine. While in the dorsal pedal artery, alpha 1- and alpha 2-receptors were involved in the response to alpha-agonists, in the basilar artery contractions induced by alpha 2-receptor stimulation prevailed. Moreover, in the latter artery the contractile responses induced by each of the agonists used were more dependent on the presence of extracellular calcium than those in the dorsal pedal artery.

[Vasodilating effects of etimizol in vitro and its relation to purinergic receptors]

The paper is concerned with the effect of the breath analeptic agent etimizole on the smooth muscle of vessels under in vitro conditions from the viewpoint of its assumed interaction with purinergic receptors. Isolated preparations of the coronary artery and a. dorsalis pedis of the dog served as experimental material. Etimizole administered in a cumulative manner in dependence on concentration decreases the strength of contraction of the preparations contracted with K+ ions. This relaxant effect, similarly as the effect of adenosine, is not affected by a removal of the endothelium, can be inhibited with theophylline and potentiated with dipyridamole. Etimizole in higher concentrations inhibits the response of vascular preparations to electrical stimulation. In the concentration of 10(-4) mol.l-1 it potentiates the relaxant effect of adenosine, the experimental effects not differing from the theoretical curve for the addition of effects. The obtained results support the assumption that the vasodilating effect of etimizole is mediated prevalently by P1 purinergic receptors.

[Hemodynamic effects of etacizine and lidocaine after intravenous administration in dogs]

In acute experiments on anaesthetized dogs, the haemodynamic effect of etacizine (0.5 and 1.0 mg.kg-1 i.v.) and lidocaine (2.5 and 10 mg.kg-1 i.v.) were compared. Etacizine produced a decrease in heart rate and a temporary decrease in myocardial contractility. After the dose of 1 mg.kg-1 there occurred a temporary increase in peripheral vascular resistance. It did not affect aortal blood pressure, systolic and minute output volume of the heart. Lidocaine produced a decrease in heart rate, aortal blood pressure, systolic and minute output cardiac volume with a rapid return to the original values. Besides this negative effect it was possible to observe an increase in the minute output volume and peripheral vascular resistance with larger doses, which gives evidence of the central effect of the drug.

[Comparison of the effect of BL 343 Ac with beta-adrenolytics on heart muscle preparations]

The paper evaluates the effect of the newly synthesized substance BL 343 Ac--propyl-4[(2-hydroxy-3-isopropylamino)-propoxy]carbanilate hydrochloride--on mechanical parameters of the myocardium of isolated guinea pig atria. At high concentrations of the substance a negative inotropic and chronotropic effect and prolongation of the functional refractory period of the preparations were recorded. These effects were comparable to those of propranolol and metipranolol. Furthermore, a mild increase in the rate of the reserpinized atria after application of 10(-7) mol.l-1 BL 343 Ac suggests a possible intrinsic sympathetic activity of the substance. BL 343 Ac has an antiisoprenaline effect with a pA2 value comparable to that of propranolol and metipranolol. However the slope of the Schild plot was different from 1, indicating that the substance possesses also other than competitively beta-adrenolytic properties.

Heterogeneity in the response to histamine of isolated dog arteries

Arterial rings were used to evaluate the differences in the effect of histamine (HIS) on canine dorsal pedal (DPA) and coronary arteries. Moreover, the responses of intact coronary arteries (IA) to HIS were compared with those of arteries isolated from sham operated dogs (SA) and from dogs in the late postinfarction period (PA). DPA responded to HIS added cumulatively with contraction which was potentiated by cimetidine (CIM) and reversed by mepyramine (MEP). IA responded with relaxation potentiated by MEP and reversed by CIM. Under resting conditions, coronary arteries responded to a single concentration of HIS with contraction or relaxation (in IA, SA). In PA, only contractions were seen. Responses of PA were significantly higher than those of IA and SA. It appears that in PA the ratio of H1- and H2-receptors is changed.

Cardiac electrophysiologic actions of the hydrochloride of 1-methoxymethyl-2-(1-perhydroazepinyl)ethyl ester of 2-(n)-pentyloxycarbanilic acid in the anesthetized dog

Cardiac electrophysiologic effects of the hydrochloride of 1-methoxymethyl-2-(1-perhydroazepinyl)ethyl ester of 2-(n)-pentyloxycarbanilic acid (BK 129) were studied in pentobarbital anesthetized dogs. Doses of 0.5, 1 and 2 mg kg-1 i.v. produced a dose dependent slowing down of S-A and idioventricular pacemakers, increase in atrioventricular (A-V), atrial and ventricular refractoriness, and decrease in intra-atrial, A-V and intra-ventricular conduction. Retrograde ventriculoatrial conduction was affected only slightly. In dogs with A-V blockade BK 129 suppressed norepinephrine-induced acceleration of S-A and idioventricular rhythm and intra-ventricular conductivity. These results suggest that BK 129 may be an effective agent for treating cardiac arrhythmias.

Antiarrhythmic activity of the hydrochloride of 1-methoxymethyl-2-(1-perhydroazepinyl)ethyl ester of 2-(n)-pentyloxycarbanilic acid

Antiarrhythmic effects of the hydrochloride of 1-methoxymethyl-2-(1-perhydroazepinyl)ethyl ester of 2-(n)-pentyloxycarbanilic acid (BK 129) was studied in pentobarbital anesthetized dogs. The dose of 1 mg kg-1 i.v. suppressed ouabain-induced ventricular tachycardia for approx. 15 min. The minimal effective dosis was 0.72 +/- 0.09 mg kg-1 in comparison with 4.56 +/- 1.90 mg kg-1 of trimecaine and 6.80 +/- 0.86 mg kg-1 of quinidine. The doses of 1 and 2 mg kg-1 decreased ectopic activity 24-48 h after coronary artery ligation without complete suppression of extrasystoles. Epinephrine-induced arrhythmias were not prevented with the dose of 1 mg kg-1. The results demonstrate effectiveness of BK 129 in suppression of some types of arrhythmias.