INTRODUCTION:

Paroxetine hydrochloride,trans(−)-3-[(1,3-benzodioxol-5-yloxy)methyl]-4-(4-fluorophenyl) piperidine, is an antidepressant^[1] developed by Smithkline beecham and also used in treatment of Obsessive compulsive disorder under brand name Paxil®. Synthesis and characterization of impurities in drug substance is a current topic of interest as per ICH guidelines ^[3]. Impurity limits in final API are getting stringent and regulatory authorities are focusing in depth towards control of process related impurities. Quality of drug substance /API need to be ensured by following QbD principles par se ^[4]. Information related to characterization of impurities is an integral part of DMF filling process in highly regulated markets of US, Europe and Japan.

The prediction of formation of impurities in a particular process needs an in depth knowledge and understanding of manufacturing process of intermediates and drug substance and sound scientific principles involved based on reaction mechanisms of organic chemistry. Such a proactive approach would be an added advantage towards building QbD in the process and would assure quality of drug substance consistently during commercial manufacturing. It has been stated that quality of product is not ensured by mere in process controls/ testing, but needs to be inbuilt in the process^[5].The aim of the presnt study was to synthesis and characterize two novel impurities in the process of Paroxetine as it is necessary to identify impurities above 0.1%.^{[6, 9]}

RESULTS AND DISCUSSION:

Synthesis of Paroxetine hydrochloride hemihydrates^[7]involves multi step organic transformations [Scheme 1].Typical sequence may begin with resolution of racemic intermediate 4-(4-fluorophenyl)-3-hydroxymethyl-N-methyl-piperidine “carbinol” to get chirallly pure (-) carbinol, Stage-I: (-) carbinol is converted into “in situ mesylate” ^[2] or tosylate and then condensed with 3,4-methylenedioxyphenol “sesamol” in presence of strong alkali to get an advanced intermediate N-Methyl Paroxetine( NMP). Stage-II: NMP on further demethylation using carbamate chemistry generates free base of Paroxetine and treatment with HCl finally gives Paroxetine Hydrochloride API.

Scheme1

Stage-III A: To avoid environmental pollution

Deprotection^[6] of N-Methyl Paroxetine [Scheme 3] by using phenyl chloroformate [PAR-IIIA] intermediate which on alkaline hydrolysis in Toluene and sodium hydroxide flakes at reflux temperature generates sodium phenolate as side product in process effluents. This side product can not be thrown into effluent due to tighter environmental pollution norms by local as well as international environmental protection agencies. Sodium phenolate is non-biodegredable and hence it is difficult to treat in ETP. It is toxic to cell cultures in ETP/ aquatic animals like fish.^[11]. Hence this route for manufacturing was abandoned. Environmental pollution concerns lead us to replace phenyl chloroformate with 1-Chloroethyl chloroformate. 1CECF is more environmental friendly and does not produce any hazardous by products except carbon dioxide gas during ethanolysis. Hence keeping in mind the green chemistry aspects, we could successfully utilize 1CECF reagent.

Stage-III B: Synthesis of dimer type impurity at RRT 1.18

1-Chloroethyl chloroformate reagent is more environmental friendly as it does not generate non-biodegradable / toxic byproducts.

Reaction was completed in Toluene at reflux over 8-10 hours time. Toluene was recovered completely to get an oily mass [ carbamate intermediate PAR-III B in situ], which was treated with methanol solvent [ Scheme 4 reaction mechanism] However an unknown impurity was observed in final API at RRT1.18 in HPLC chromatogram. This was very difficult to remove by purification in Acetone solvent. The probabale mechanism for formation of impurities is depicted in Scheme 5 .This was due to in situ generated dimethyl acetal^[8][Scheme 4], which is highly reactive species in acidic medium. It gets protonated due to HCl present in the reaction medium and 3,4- methylene dioxy phenol moiety( sesamol part) being highly electron rich attacks on dimethyl acetal thus generating an intermediate having mono methyl acetal. Another molecule of Paroxetine attacks in similar manner to finally form the molecular structure depicted in following Scheme 5 as dimer. This is having a methyl group at 1.37ppm as doublet in ¹HNMR and a hydrogen at 4.55ppm as quartet coming from dimethyl acetal which originates from demethylating reagent 1-CECF.

More amount of impurity was generated deliberately by adding concentrated HCl and impurity was isolated in acetone in pure form for spectral characterization.

It was characterized by ¹H NMR, ¹³C NMR and MS. ¹H NMR showed additional two prominent peaks when compared with ¹H NMR signals of Paroxetine^[10]. A signal showing doublet at 1.37 ppm and integration of 3 protons and another signal showing quartet and integration of 1 proton was observed at 4.55ppm. Mass spectrum showed [M+H] at 664 which indicated two molecules of Paroxetine. Typical signal of a quartet and doublet helped us to predict and propose above structure of Impurity at RRT 1.18.

Applying QbD principles to process development:

If we rationalize how impurity at RRT 1.18 was formed?, then we understand advantage of Ethanol solvent over that of Methanol solvent. In this impurity synthesis, it was observed that methanol being a small size molecule generated “in situ” dimethyl acetal, which undergoes reaction with two molecules of Paroxetine under highly acidic environment under experimental conditions. However, when slightly bigger size molecule of ethanol generated “in situ” diethylacetal which probably does not react as fast as dimethylacetal and does not generate impurity in detectable levels in HPLC. Hence use of ethanol solvent proved beneficial with respect to impurity levels in final API ensuring highly pure quality of final API complying as per ICH guidelines. The use of other alcoholic solvents like Isopropanol, propanol which are cheaper than Ethanol for doing the final stage demethylation by alcohololysis is being explored to reduce cost of manufacturing of Paroxetine HCl API further.

Here we would like to emphasis that how QbD principles were applied in this process development. Firstly we identified the exact route cause of impurity formation, understood the mechanism of impurity formation and then we applied knowledge of organic chemisty to control/ stop the formation impurities. We have demonstrated that quality can be built in the process and “ right first time” approach can be adopted to avoid failures and bitter surprises during scale up in manufacturing plant. Such type of strategy would also reduce overall cost of manufacturing an API and make one the most competitive in current commercial environment. It is necessary that we are able to sustain the price pressure in generic market globally and remain in API market for long time doing business ( First in, last out generic business strategy).

EXPERIMENTAL:

Synthesis of ether type impurity at RRT 1.04:

(-) carbinol ( 2.23g , 0.01 moles) was dissolved in sulpholane at 30°C and Potassium hydroxide flakes were added. Reaction mixture was maintained under nitrogen blanket and heated to 55°C in water bath for 15-20 minutes. Mesylate intermediate( 3.0g , 0.01 moles) was added to reaction mixture at once and reaction mixture was heated for additional 2 hours. An aliquot was withdrawn and HPLC results showed an impurity 66.96% at RRT 1.04 with respect to N-Methyl Paroxetine. Reaction mixture was heated further for 2 hours and HPLC analysis showed an impurity 72.35% at RRT1.04.

Isolation of impurity: Reaction mixture was quenched with 45 ml distilled water and light yellow precipitate was observed. It was filtered on Buchner funnel and washed with1% KOH solution ( 15 ml x 3 times) and then with distilled water ( 15 ml x 3 times). Wet cake was suck dried and then dried in vacuum oven at 55-60°C. Dry wt: 3.0g ( HPLC Purity: 97.16%).

¹H NMR in CDCl₃ ( 300 MHz) 1.76( m, 6H),1.82(m, 4H), 2.25( m, 2H), 2.35( s, 6H),2.808( d, 4H), 2.93(dd, 2H), 3.023(dd, 2H), 7.065( m, 4H), 6.95( m, 4H),

Mass spectra m/z: [M+2]: 430

Synthesis of dimer type impurity at RRT 1.18:

N-methyl paroxetine( 15.0g, 0.045moles) was refluxed in 150ml toluene to remove residual water by azeotropic distillation using Dean-Stark water separator. 1-CECF(7.17 g, 0.050 moles) was diluted with 15 ml toluene and added dropwise into reaction mixture and heated at reflux for 3 hours. Reaction completion was monitored by TLC (10% Methanol:Chloroform) till disappearance of starting material i.e. N-methyl paroxetine. After completion of reaction, toluene was recovered under vaccum in rotavapor and carbamate intermediate i.e. an oily mass obtained was dissolved in 150 ml methanol and stirred at 40-45°C for 3 hours. Reaction was monitored by TLC for disappearance of in situ carbamate intermediate. Conc. HCl solution 25 ml was added (pH of reaction mixture was highly acidic), and further heated at 40-45°C for 3 hours. A white precipitate of product was observed. Methanol was removed under vaccum and product was filtered using minimum methanol and recrystallised by using Methanol: water (1:1) mixture and dried in vaccum oven till constant weight. Dry wt: 10 g ( HPLC purity 98.56%)

¹H NMR in DMSO-d6 ( 300 MHz): 1.37( d, 3H), 2.85( m, 2H), 2.97( m, 2H), 3.32( m, 4H), 3.48( m, 8H), 3.54(m, 4H), 3.81( m, 4H), 4.55( q, 1H), 5.88( s, 2H), 5.93( s, 2H), 6.30( s, 1H), 6.43( s, 1H), 6.58( s, 1H), 6.69( d, 2H, J=3Hz), 6.91(s, 1H), 7.01( d, 2H, J=9Hz), 7.17-7.29( m, 4H), 9.2-9.6( s, 2H, exchangeable with D₂O)

Mass spectra m/z: [M+1]:685

CONCLUSION:

We have demonstrated that using knowledge of reaction mechanism, formation of impurities can be explained and it is possible to control such impurities by adopting necessary suitable process modifications like change of reagent, solvent and rate of addition etc. These are desirable to achieve better quality products at affordable cost and hence have long term commercial gaining impact in entire product life cycle. The strategy of “First in last out” of generic manufacturers should adopt QbD principles for long term sustainability in generic market place.

ACKNOWLEDGEMENT:

We are grateful to the Management and Principal of Ismail Yusuf College for guidance, constant encouragement and support to carry out this research work.

REFERENCES:

[1]Drugs Fut., 11, 112-115 (1986).]

[2] Neal g. Anderson, OPRD, 8,260-265, 2004 for “in situ mesylate”.

[3] ICH Q9- quality risk management guideline.

[4] Zadeo Cimarosti, Fernado Brava, OPRD, 14,993-998, 2010

[5] European Medicines Agency(EMEA), Impurities, Residual solvents VICH, Topic Report GC18(web:http://www.emea.europa.eu/ pdfs/vet/vich/050299en.pdf)(2001)

[6] U S Food and Drug Administration ( FDA-CDER), Innovation and continuous improvement in pharmaceutical manufacturing

(Web:http://www.fda.gov/cder/gmp/gmp2004/manufsciwp.pdf) 2008.

[7] US Patents US3912743, US4007196, US7721723, WO 0104113A2

[8] R.A.Olofson, Jonathan T Martz, J. Org Chem, 1984, 49, 2081-2082. demethylation mechanism

[9] Berridge J C, Impurities in drug substances and drug products: new approaches to quantification and qualification. J Pharma Biomed Anal 1995, 14,7-12.

[10] Nirmala Munigela, Spectral Characterization of Degradation impurities of Paroxetine hydrochloride hemihydrates, Scientia Pharmaceutica, 2008, 76, 653-661.

[11]For phenol pollutant Nahed S. Gad and Amal S. Saad, Global Veterinaria 2 (6): 312-319, 2008.

Received on 20.01.2012 Modified on 13.02.2012

Asian J. Research Chem. 5(3): March 2012; Page 329-335