Sulfonate Esters - How Real is the Risk?

Andrew Teasdale1*, Stephen C. Eyley1, Ed Delaney2, Karine Jacq3, Karen Taylor-Worth4, Andrew Lipczynski4, Van Reif5, David P. Elder6, Kevin L. Facchine7, Simon Golec

1 AstraZeneca, R&D Charnwood, Bakewell Road, Loughborough, Leicestershire, LE11 5RH, United Kingdom. 6 GlaxoSmithKline, Park Road, Ware, Hertfordshire, SG12 0DP, United Kingdom 2 Reaction Science Consulting, LLC, Princeton, NJ 08542, USA 7 GlaxoSmithKline, Five Moore Drive, Research Triangle Park, North Carolina 27709-3398, USA 3 Research Institute for Chromatography, Pres. Kennedypark 26, B-8500, Kortrijk, Belgium 8 Wyeth Research, 500 Arcola Road, Collegeville, PA 19426, USA 4 Pfizer Global Research and Development, Analytical R&D, Ramsgate Road, Sandwich, Kent CT13 9NJ, United Kingdom 9 F. Hoffmann-La Roche Ltd., Grenzacher Strasse, 4070 Basel, Switzerland 5 Schering-Plough, 556 Morris Avenue, Summit, NJ 07901-1330, USA

Introduction – Description of the issue Aims • There have been growing concerns expressed by regulators in relation to the potential generation of • To provide a sound scientific understanding of the formation and decomposition of genotoxic impurities as a result of interactions between strong acids and alcohols. sulfonate esters, • This has centred primarily on sulfonate esters, theoretical impurities resulting from interaction – under synthetically relevant conditions. between sulfonic acids and alcohols. • To understand the absolute levels of such impurities that can form under process-related conditions. O O – Optimal process conditions to minimize the sulfonate ester formation. R S OH R' OH + R S OR' + OH2 – Effective purge processes. O O • To place reputable, peer-reviewed science-based knowledge into the public domain R - alkyl / aryl – Methodologies for analyses and kinetic studies R'- alkyl - methyl / ethyl / isopropyl etc. – Teaching with regard process design to obviate/minimise ester formation

Design Space

Analytical Methodolgy Front of cube’ O 70C O CH3S OEt + H 0 –understanding the (strongly) acidic CH3S OH + EtOH 2 O system… O 1) Samples withdrawn over 2) Samples spiked ‘Mid and rear of cube’ time and treated with: with small amount of 3) samples heated for period of d5 EMS: F time (15 min at 105 deg C in published method) to effect understanding ‘salt-like’ systems… i - + – Pr2NEt F S Na O derivatization and insure equilibration within the + NaOH CH3S OCD2CD3 Headspace prior to assay All reactions carried out in solution F F O

Temp Lutidine F 40C Concentration Concentration of acid: >0.25M No added Base 4) Levels of Et PFTB and d5Et PFTB K. Jacq, et al (internal standard) analyzed by GC/MS: Stoicheometry of added base: 1:1 0% Water (v/v) 50% values vs time J.Pharm. Biomed. F F F SCD CD based upon Anal, 2008, 48(5), F SCH2CH3 2 3 These conditions are typical of standard process conditions ratio of Et 1339 F F F F F F PFTP to d5Et

Et PFTB d5Et PFTB PFTP area Reaction Systems counts •Commonly used 1 and 2 alcohols in combination the 2 most common sulfonic acids in terms of marketed salts –The methyl, ethyl and isopropyl esters of Methanesulfonic acid Ethyl Mesylate System –The methyl, ethyl and isopropyl esters of Toluenesulfonic acid. 1M MSA in EtOH, no added water Effect of Water (1M MSA, 70C) •Initially study Ethanol-Methanesulfonic acid system 0.45 –Followed by focused studies on other systems 0.4 70C 0.25 60C no water 0.35 5%w/w 50C 0.2 0.3 40C 25%w/w 0.25 0.15 66%w/w 0.2 0.1

% conversion % 0.15 % conversion % 0.1 0.05 O O 0.05 Reaction Mechanism 18 18 0 0 R2 S O-H + R1 O H R2 S O - + R1 O H 18 + 0 5 10 15 20 0 5 10 15 20 Reactions of (O ) labelled methanol H O O Time (hr) Time (hr) with MSA were analysed by CG-MS: reaction occurs through nucleophilic attack O Effect of Added Base (70C) Excess lutidine (green of the sulfonate anion on the protonated alcohol 18 H O H trace): Ester Undetectable 18 R2 S O-R1+ – O label appears in the WATER. 0.3 O 1M MSA, 70degC over background…

– Precludes mechanisms where the alcohol is the nucleophile 2% deficit lutidine • O18 label would have been found in the ester 0.2 9% excess lutidine LOQ: 0.001% conversion NB Solvolysis is a significant additional mechanism consuming ester to form the

ether and regenerate sulfonic acid conversion % 0.1 •This critical proton dependence underpins all the observed results 0 0 2 4 6 8 10 12 Time (hr)

Learning for Process Design… Similar findings for all systems studied Minimise (avoid) sulfonate ester formation by •Methanesulfonic acid (MSA) and p-Toluenesulfonic acid (pTSA) • Use an excess of the API base, or as near as possible to an exact stoichiometery. •Ethanol, Methanol, iso-propanol • If an excess of sulfonic acid is needed, use the minimum excess possible and conduct the salt formation and isolation steps at the lowest practical temperature. Comparison of MSA and pTSA • Include water in the salt formation and isolation procedures where possible 0.4 • Competition for proton. ETS, 70C ETS, 60C • Rapid hydrolysis rates relative to rates of ester formation. 0.3 ETS, 50C ETS, 40C • Avoid situations in which sulfonic acid and alcohol are mixed and stored before use. 0.2 EMS, 70C EMS, 60C

% conversion % EMS, 50C • If this is unavoidable then any solutions should be prepared at as low a temperature as possible and 0.1 hold times kept to a minimum. EMS, 40C 0.0 • If low level formation likely ensure efficient washing of cake. 0 5 10 15 20 Time (hr)

Conclusion • Based on the thorough understanding of the reaction between sulfonic acids and alcohols developed through the PQRI studies it is entirely possible and straight forward to control process conditions such that levels of sulfonate esters can be controlled to such low levels as to present no appreciable risk

• Ultimately this shows that sulfonic acids can be used under the right conditions without fear of risk