Elusive impurities â evidence versus hypothesis. Technical and

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

Review

Elusive impurities – evidence versus hypothesis. Technical and regulatory update on alkyl sulfonates in sulfonic-acid salts David John Snodin Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.8b00397 • Publication Date (Web): 18 Apr 2019 Downloaded from http://pubs.acs.org on April 18, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

Elusive impurities – evidence versus hypothesis. Technical and regulatory update on alkyl sulfonates in sulfonic-acid salts Corresponding author: David J Snodin, Xiphora Biopharma Consulting, UK Email: [email protected]

Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

1

Organic Process Research & Development 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

TOC Figure Synthesis of Sulfonic-Acid Salts Proton transfer reaction (equimolar reactants)

Base

+

AlkOH solvent

RSO2OH

BaseH+

+

RSO2O-

Hypothetical side reaction

+

AlkOH

Base

X

RSO2OH

H2O

+

RSO2OAlk

Key: Alk = methyl, ethyl or isopropyl; R = methyl, phenyl or p-tolyl Typical hypothetical impurities O H 3C

O

O

S

O

H 3C

S

H 3C

O

O

O

O

S

O

S

O

O

S O

O O

S

O

O

O O

O

O

O O

S

S

O O

O

S

O

O


2

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract There is a widespread and long-standing perception that drug substances presented as sulfonic-acid salts pose an inherent risk to patients owing to the potential presence of mutagenic alkyl-sulfonate impurities. Extensive and indisputable kinetic, mechanistic and experimental evidence indicates that this is not a sustainable position. Sulfonic-acid-salt formation normally involves addition of a sulfonic acid to an equimolar amount of the base form of a drug substance dissolved in a suitable solvent (most frequently ethanol or other protic solvent). In such systems no alkyl-sulfonate impurities are generated via an esterformation side-reaction owing to essentially instantaneous base protonation and neutralization of the acidic component. Carryover of impurities in sulfonic-acids is considered highly unlikely given the use of pharma-grade reagents and the significant inbuilt purge factors based on dilution and the high solubility of alkyl-sulfonates in alcohol solvents. Efforts to create a dialogue with the main drug regulatory agencies with a view to reversing current approaches based on disproved hypotheses have been largely unproductive. One exception is the EMA (European Medicines Agency) which undertook an internal evaluation resulting in the conclusion that the formation of alkyl-sulfonate impurities is unlikely but could not be totally excluded. Requests to EMA to provide evidence supporting this position were all unsuccessful. Although both EMA and the European Pharmacopoeia have proposed “risk assessment” as an alternative to analytical testing, to date there has been no progress on defining appropriate parameters. Key words: Sulfonic-acid salts; alkyl-sulfonate impurities; mutagenic; protonation; nucleophilicity; impurity purging; regulatory agencies.


3


Page 4 of 36

Elusive impurities – evidence versus hypothesis. Technical and regulatory update on alkyl sulfonates in sulfonic-acid salts Data must be used to tell the truth…Only relevant and accurate data is useful. Hans Rosling, Factfulness, 2018

Introductiona,b For many years the prevailing wisdom in drug regulatory agencies and pharmacopoeias worldwide has been that a drug substance presented as a sulfonic-acid salt (eg mesilate, besilate or tosilate) is associated with a potential safety concern relating to the possible presence of mutagenic sulfonic-acid-ester impurities [such as methyl methanesulfonate (MMS), ethyl methanesulfonate (EMS) or isopropyl methanesulfonate (IMS) – Figure 1]. Other sulfonic-acid counterions that are less commonly used include camsilate, edisilate, esilate, isethionate and napsilate. Alkyl esters of sulfonic acids can act as biological alkylating agents and generally test positive for mutagenic potential in bacterial reverse mutation assays (Ames’ tests)1. On the other hand aryl esters of sulfonic acids are nonalkylating (similar to halo-aryls such as bromobenzene2 and chlorobenzene3) since nucleophilic substitution at an sp2-hybridized aryl carbon atom is highly unfavoured except where the aromatic ring is activated (by the presence of nitro groups for example) or when a strong nucleophile is employed under forcing conditions4. Use of a phrase such as “aryl or alkyl sulfonates”5 in relation to genotoxic potential could be misleading, possibly suggesting that aryl esters of sulfonic acids are similar to alkyl sulfonates in terms of their mutagenic properties. Terminology should be sufficiently precise to distinguish between alkyl and aryl esters of sulfonic acids be they alkyl sulfonic acids (such as methanesulfonic acid - MSA) or aryl sulfonic acids (such as p-toluenesulfonic acid, p-TSA). The aryl-sulfonate ovicide chlorfenson (4-chlorophenyl 4-chlorobenzenesulfonate; CAS no 80-33-1) tests negative for mutagenicity in the Ames’ assay6 and is noncarcinogenic in a mouse lifetime bioassay7. A further example is the herbicide ethofumesate [(±)-2-ethoxy-2,3-dihydro-3,3dimethylbenzofuran-5-yl methanesulfonate; CAS no 26225-79-6] which is nonmutagenic in multiple strains of Salmonella typhimurium and is devoid of genotoxic potential in a range of other relevant in-vitro and in-vivo assays8. In addition, phenolic compounds are unlikely to be used as solvents in the synthesis of sulfonic-acid salts and so the theoretical presence of aryl-sulfonate impurities can be effectively discounted. Figure 1: Methyl, Ethyl and Isopropyl Methanesulfonate (MMS, EMS and IMS) O H 3C

S

O

Alk

O

Alk = Me: MMS; Alk = Et: EMS; Alk = iPr: IMS.

Spelling conventions for sulfonate esters (mesilate, tosilate, besilate) follow those of the European Pharmacopoeia. The alternative spelling (eg mesylate, etc) is employed only if it occurs in a direct quotation from a published article. b Abbreviations are defined at first use a


4

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In 2000 a notice in Pharmeuropa [which is administered by EDQM (European Department for Quality of Medicines)] voiced concerns regarding the potential for formation of alkylmesilate impurities during the synthesis of a mesilate-salt drug substance, and that “assurance is needed that the esters are not present in unacceptable quantities in medicinal products”. Four years later EDQM established the so-called “Production Statement” (Box 1) as a formal requirement of the Ph.Eur for monographed drug substances presented as mesilate salts9. Box 1: Original Ph.Eur Production Statement for Mesilate-Salt Drug Substances The production method must be evaluated to determine the potential for formation of alkyl mesilates, which is particularly likely to occur if the reaction medium contains lower alcohols. Where necessary, the production method is validated to demonstrate that alkyl mesilates are not detectable in the final product.

Lower alcohols such as methanol, ethanol and isopropanol are often considered the most suitable solvents for carrying out sulfonic-acid-salt-formation reactions and the wording of the Production Statement strongly hints (but without any supportive experimental evidence) that a side-reaction could occur between the solvent and methanesulfonic acid (MSA) to generate alkyl-mesilate impurities. [For future reference this is called the “side-reaction hypothesis”]. The text goes on to say that: “Tests for alkyl mesilates will not be included in monographs. The methods required to demonstrate the absence of alkyl mesilates at a suitable limit of detection are not suited for routine use and validation of the process is a better approach”. In practice, applicants were expected to provide analytical data on every batch of mesilate salt, possibly with skip-testing when no alkyl mesilates had been detected in numerous previous batches. Shortly after the initial Pharmeuropa announcement, but before establishment of the Production Statement, EDQM commissioned research at the University of Strathclyde on the analysis of alkyl mesilates in mesilate-salt drug substances. The results are presented in: S Ahmad, Investigation of mesylate salts for the presence of mesylate esters; MSc thesis 2002, University of Strathclyde, Glasgow, UK. Samples of mesilate salts listed in the European Pharmacopoeia were first examined to identify those containing traces of alcohols; subsequently those with alcoholic residues were analyzed for the presence of the mesilate esters by a direct-injection GC (gas chromatography) method. Of 11 mesilates tested (bromocriptine, dihydroergotamine, pefloxacin, betahistidine, phentolamine, dihydroergocristine, codergocrine, pergolide, dihydroergotoxine, dihydralazine) 8 contained either methanol or ethanol. Salts shown to contain either methanol or ethanol were analyzed for the presence of the methyl or ethyl esters, as appropriate. No esters were detected in the batches examined, limits of detection for both methyl and ethyl esters (MMS and EMS respectively) being 10 ppm. The research project, which appears to have been largely predicated on the side-reaction hypothesis, provided no evidence that supported its theoretical basis. Nevertheless, EDQM went on to establish the Production Statement and has never placed data from the Ahmad thesis in the public domain (or any other reliable analytical Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

5


Page 6 of 36

data). The project supervisor confirmed the negative results but commented that the esters are “too unstable to really run very well through a GC”10 and suggested NMR (nuclear magnetic resonance) as an alternative technique. [Since that time GC techniques seem to have improved in relation to reproducibility and analyte stability; see for example the publication by Sitaram et al, 201111]. Concerns over alkyl mesilates were not restricted to EDQM; in its 2004/2005 Pharmacology Review of the Pfizer drug Cardura XL12, which contains doxazosin mesilate (Figure 2) as the drug substance, FDA (US Food and Drug Administration) raised the following issue: “A new concern within CDER (Center for Drug Evaluation and Research) is the possible production of genotoxic impurities which can be produced when free base drugs are converted to a mesylate salt. These process impurities are known genotoxic agents…Further, IARC (International Agency for Research on Cancer) Monographs have reported that XXX (redacted), the one of primary interest for this submission, has been shown to cause cancer in mice followed (sic) s.c. administration (lung tumors), and in mice and rats following i.p. administration (lung and kidney tumors). Recommendations: The Sponsor needs to quantitate the XXX in-process impurity….” Although the name of the actual XXX impurity is redacted from the publicly-available version of this review, it is clear from the context (particularly the reference to the IARC Monograph) that the impurity of concern is EMS. The issue was picked up in the Chemistry Review13 and the applicant responded by providing analytical data for 12 batches of doxazosin mesylate drug substance and for the same number of batches of Cardura and Cardura XL using the lower limit of detection as a trigger-point for concern. Although no quantitative data are cited it seems clear that no residues of EMS were detected since PharmTox recommended approval in November 2004 and formal approval was achieved on 22nd February 200514. A 2008 public assessment report from the Dutch Medicines Evaluation Board (MEB) on an application for doxazosin mesilate of strengths 2 and 4 mg comments on the use of n-butanol as a solvent in the synthesis of doxazosin mesilate; no n-butyl mesilate impurity was detected in the drug substance15. The active substance concerns a mesilate-salt. A requirement for evaluation of possible presence of alkyl mesilates has been introduced the last years in Ph.Eur monographs of mesilate salts. If the reaction medium contains lower alcohols there is a possibility that alkyl mesylates are formed. Therefore, two post approval commitments were made. The MAH committed to develop an adequate, analytical method to perform the nbutyl mesilate determination. The MAH (Market Authorisation Holder) committed to provide data demonstrating the absence of alkyl mesilate in the drug substance. Both post-approval commitments have been solved. The MAH committed to develop an adequate, analytical method to perform the n-butyl mesilate determination. The method was developed and is found suitable to apply a limit test for n-butyl mesilate, methyl mesilate and ethyl mesilate at 10 ppm. The method has been demonstrated to be specific for the three alkyl mesilates. The MAH committed to provide data demonstrating the absence of alkyl mesilate in drug substance. A method was developed and is found suitable to apply a limit test for n-butyl mesilate, methyl mesilate and ethyl mesilate at 10 ppm. Tests with 6 batches demonstrated that none of the three tested alkyl mesilate has been found above the limit of 10 ppm.

Assay data on the 8 mg strength produced similar negative results for alkyl mesilates16.


6

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2: Doxazosin Mesilate (CAS no 77883-43-3) O O

N N

O

N

O H3 C

S

N

OH

OCH3

OCH3 NH2

O

Technical Update Detailed scientific information on alkyl-sulfonate impurities in sulfonic-acid salts has already been published1 and what follows is intended to be a brief review of the relevant evidence. Side-Reaction Hypothesis Figure 3 shows a reaction scheme illustrating the notion that if a sulfonic-acid salt is synthesised by addition of an equimolar amount of a particular sulfonic acid to the base form of a drug substance using a protic solvent medium, then formation of alkyl sulfonate impurities is likely to ensue as a side-reaction. Similar concerns are expressed by Balaji & Sultana, 201717: Recently, increased attention has been paid to the health risks associated with even minute levels of mesylate esters such as methyl methanesulfonate (MMS), ethyl methanesulfonate (EMS), isopropyl methanesulfonate (IMS) and hexyl methane sulfonate (HMS) in drugs because of their potent mutagenic, carcinogenic, and teratogenic effects. These mesylate esters can be derived from excess starting materials during pharmaceutical drug synthesis or are produced as by products from the reaction between methane sulfonic acid (frequently used as a counter ion) and alcohols/acid chlorides (commonly used as reaction media in developed and industrialized processes).

Figure 3: Illustration of Side-Reaction Hypothesis of Alkyl-Sulfonate Impurity Formation

Information now available indicates that there are five key factors (Box 2) involved in the generation of an alkyl sulfonate in a binary system containing a sulfonic acid and a protic (alcoholic) solvent.


7


Page 8 of 36

Box 2: Five Key Factors Involved in the Generation of an Alkyl Sulfonate from a Sulfonic Acid and an Alcoholic Solvent 1. 2. 3. 4. 5.

Acid strength of reaction medium; Reaction temperature and contact time; Presence or absence of water in the reaction medium; Mechanism: nucleophilic displacement of hydroxonium ion from protonated alcohol by sulfonate anion; Reaction rate: extremely slow owing to feeble nucleophilicity of sulfonate anion

The mechanism of alkyl sulfonate formation was first predicted in 200618 and confirmed experimentally in 2009 by Teasdale et al19. The reaction proceeds initially by alcohol protonation followed by displacement of the protonated hydroxyl group by sulfonate anion. Strongly acidic conditions are required to produce a meaningful concentration of protonated alcohol. Whereas a hydroxyl group is extremely resistant to nucleophilic displacement under neutral conditions, alcohol protonation can produce a hydroxonium moiety which is a reasonably good leaving group. However, since sulfonate anion is an extremely poor nucleophile owing to delocalization of negative charge over three oxygen atoms18 the subsequent reaction to form an alkyl sulfonate proceeds extremely slowly, as shown in Figure 4. [See Teasdale et al19 for a more detailed description of the mechanism and kinetics.] The molar rate constants for both alkyl mesilate and alkyl tosilate formation are approximately 10−7 s−1 at 70 °C under anhydrous conditions and the reaction rate is highly temperature-sensitive with an approximate quadrupling for each increment of 10 °C. A molar solution of anhydrous MSA or p-TSA in ethanol maintained at 30 °C for 2 h is predicted to contain approximately 3 ppm alkyl sulfonate1. Addition of even small amounts of water retards the rate of alkyl sulfonate formation; the extent of conversion is reduced by two-thirds in the presence of 5% water. A 1 M solution of p-TSA monohydrate in 96% ethanol would contain 50−60 g/L water, so maintaining such a solution at 30 °C for 2 h would be expected to produce ca 1 ppm ethyl tosilate1. Figure 4: Mechanism of Formation of an Alkyl Sulfonate from a Sulfonic Acid and an Alcohol


8

Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


What happens if an equimolar amount of a sulfonic acid is added to the free-base form of an amine-containing drug substance in ethanol solution? The formation of amlodipine besilate (Figure 5) from benzenesulfonic acid and amlodipine base dissolved in ethanol is used as an example. Figure 5: Amlodipine Besilate (CAS no 111470-99-6) Cl O H3 C

O

O H3 C

O N H

O

CH3

O

S

O

OH NH2

As shown in Figure 6 the pKa values for the conjugate acids of amlodipine and ethanol are approximately 9 and -2 respectively20,21 Thus, protonation of the primary amino group of amlodipine will be favoured 1011-fold over ethanol protonation, and since the rate of protonation of the base form of amlodipine in a protic solvent is diffusion-controlled, saltformation is effectively instantaneous with no protonation of the solvent. Figure 6: pKa Values for Conjugate Acids of Amlodipine and Ethanol Protonated Amlodipine Protonated Ethanol Structure R-NH3+ pKa 9

Et-OH2+ -2

Drug substances that are secondary or tertiary amines are less basic but generally have pKas of ≥6.0. For example, the sterically-hindered aromatic base 2,6-lutidine (2,6dimethylpyridine; CAS no 108-48-5) has a pKa of 6.622. Using an ethanolic solution of equimolar amounts of this base and MSA, no EMS formation was detected over 12 h at 70 °C23. Based on kinetic parameters for this reaction and assuming a 2% molar excess of MSA, generation of only 1.6 ppm EMS in solution was predicted to occur over 12 h at 15 °C. [It should be noted that adding a slight molar excess of sulfonic acid has no effect on the yield of sulfonic-acid salt and is completely unnecessary.] Experimental data indicate that at 70°C using 2,6-lutidine in the presence of a 2% molar excess of MSA only 0.004% conversion to EMS was detected (compared to approximately 0.25% in the absence of base)23. This provides a clear illustration of the importance of acid molarity in relation to alkyl-sulfonate production in that a 50-fold reduction in MSA concentration (from 1M to 0.02M) resulted in a 625-fold decrease in EMS production. During the synthesis of sulfonic-acid salts by addition (with adequate mixing and temperature control) of an equimolar amount of sulfonic acid to an organic base dissolved in a protic solvent, all added sulfonic acid will be neutralized and become unavailable for sulfonic-acidester formation. Experimental confirmation of this statement is available in relation to the commercial production of nelfinavir mesilate involving the addition of an equimolar amount of ultrapure pharma-grade MSA (containing 4100 for MMS, EMS and IMS respectively29. These values are probably underestimates given that the LOQ for the method of analysis was 1 ppm, evidence in other cases indicating efficient purging by precipitate washing and/or recrystallization to low/sub-ppm levels of alkyl sulfonates28. In a more typical case where pharma-grade MSA (containing 200 °C) and high thermal stability. Contaminated nelfinavir mesilate 250 mg tablets stored at 25 °C showed a loss of 9%/month in EMS content24Error! Bookmark not defined.. If high-purity MSA is employed for mesilate-salt synthesis it is considered extremely unlikely that any toxicologically-significant residues of alkyl mesilates will remain in the reaction solvent following product deliquoring. Under these circumstances limited recycling of the reaction solvent should not present any risk of mesilate-salt contamination. In any case, decisions on solvent recycling need to be taken case-by-case and based on GMP principles. No examples sulfonic-acid-salt drug substances or drug products with generation of alkyl-sulfonate degradation products during storage could be found in the literature.


12

Page 13 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In conclusion, all of the available kinetic, mechanistic and experimental evidence points to the fact that in a GMP-compliant synthesis of a sulfonic-acid salt using the type of precautions mentioned by Gerber and Toelle24 the presence of alkyl-sulfonate impurities can be excluded with a high degree of confidence. The key contributory factors are summarized in Box 3. Box 3: Synthesis of a Sulfonic-Acid Salt by Addition of an Equimolar Amount of a Sulfonic Acid to an Organic Base Dissolved in a Protic Solvent    

  



Protonation of base versus solvent favoured by a factor of ≥108 Base protonation is diffusion-controlled and so is essentially instantaneous In these circumstances there is no scope for a side-reaction between sulfonic acid and protonated solvent to produce alkyl-sulfonate impurities The salt-alcohol-interaction hypothesis is disproved as a source of alkyl sulfonates on the basis if mechanistic data (it would require the displacement of a hydroxyl group by sulfonate anion in neutral conditions) and by published experimental results (for example on the routine synthesis of nelfinavir mesilate) The only relevant source of pre-formed alkyl sulfonates is MMS in technical-grade MSA GMP compliance would mandate the use of pharma-grade MSA in which alkyl sulfonates are normally < 5 ppm Any traces of alkyl-sulfonates originating from reagents are eliminated owing to the large inbuilt purge factors (based on solvent dilution, preferential solubility in the reaction solvent and washing of precipitated sulfonic-acid salt with suitable solvents) Recrystallization of the sulfonic-acid salt using a solvent such as ethanol will produce a drug substance of high purity free from alkyl sulfonates

Analytical Aspects Drug Substances Numerous reports of analytical methods for detection of alkyl sulfonates have been published as previously described1. The rationale for assay development is always based on speculation or assertion regarding the actual or potential presence of alkyl sulfonates in sulfonic-acid-salt drug substances, with no supporting evidence. For example, Sarat et al, 201041, state the following: Recently the potential health hazards of trace amounts of mesylate esters, including methyl methane sulfonate (MMS) and ethyl methane sulfonate (EMS), in pharmaceuticals have attracted the attention of regulatory authorities. These mesylate esters are known to be potent mutagenic, carcinogenic and teratogenic compounds14. There (sic) presence in the pharmaceutical products may be the results of leftover starting materials, or formed as by-products between methanesulfonic acid (often used as a counterion) and alcohols often uses as a manufacturing process. Although official guidelines have not been established the concentration of these compounds are expected to be controlled at a level of less than or equal to 1.0 μg/g . Therefore, it is of great importance to develop analytical methods that are sensitive enough and meet all the regulatory requirements.


13


Page 14 of 36

Possibly the principal driving force leading to this plethora of publications is the opportunity for analytical chemists and instrument manufacturers to showcase their expertise and equipment capabilities given that analytical standards for the various sulfonate esters are readily available. A typical example relates to an application note on the HPLC/MS analysis of alkyl besilates and tosilates42; the introduction states: “…sulfonic acids are often used as the final salt form of the drug substance due to improved chemical properties or bioavailability. The presence of any residual alcohols from synthetic reaction or recrystalization steps may result in the formation of alkyl esters of the sulfonic acids”. It is clear that the authors are unaware of evidence disproving the salt-alcohol interaction hypothesis and that no alkyl sulfonate impurities are generated when a sulfonic-acid salt is purified by recrystallization from a solvent such as ethanol. Another instrument manufacturer43 states the following: Residual genotoxic impurities, and particularly alkyl esters of alkyl or aryl sulfonic acids, have been, and probably remain, a significant safety concern to drug regulators. Since the sulfonate moiety is readily displaced by a variety of nucleophiles, such esters can act as DNA alkylating agents in biological systems and have been shown to exert genotoxic effects in bacterial and mammalian cells. Chemicals like Ethane sulfonic acid used in the process of pharmaceutical synthesis are likely to generate Ethane Sulfonic acid ester with alcohols as reaction by-products. These Ethane sulfonic acid esters (Esylates) that include Methyl Ethane Sulfonate and Ethyl Ethane Sulfonate are Potential Genotoxic Impurities (PGI's) and are of great concern to pharmaceutical manufacturers. A TTC (Threshold of Toxicological Concern) based acceptable intake of a mutagenic impurity of 1.5 μg per person per day is considered to be associated with a negligible risk and can, in general, be used for most pharmaceuticals as a default value, to derive an acceptable limit for control.

The above comment is considered particularly misleading in relation to its reiteration of the disproved side-reaction hypothesis as well as implying the need for highly sensitive assays with limits based on the generic lifetime TTC (Threshold of Toxicological Concern) rather than less-than-lifetime (LTL) and/or compound-specific limits (for individual alkyl mesilates). The continuing appearance of new articles on methods of analysis of alkyl sulfonates may lend credibility to the notion that detectable residues of such esters can be found in sulfonicacid salts, although it is noteworthy that in the majority of publications no numerical data on levels of alkyl-sulfonate impurities are provided. Some recent examples include publications by Khile et al, 201744 on alkyl tosilates in brinzolamide, Kakasaheb et al, 201545 on MMS in zidovudine and Zhang et al, 201646 on MMS and EMS in imatinib mesilate. Drug Products The Commission of the European Union (EU) and the Council of Europe decided in 1994 to create a network of official medicines control laboratories (OMCLs) with the aim of collaboration on the quality control of marketed medicinal products for human and veterinary use. In 1995 EDQM took on this responsibility and subsequently set up the OMCL network and laboratories in this network have been employed by EDQM to provide analytical back-up on surveys of impurities in drug products containing sulfonic-acid-salt APIs47.


14

Page 15 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In 2012 Wollein & Schramek48 reported on a screening assay using a GC-MS technique for MMS, EMS and IMS in mesilate-salt drug products and methyl and ethyl besilate in products containing benzenesulfonic-acid APIs. Test samples were sourced from pharmaceutical wholesalers. Powdered drug products (plus internal standard) were extracted with n-hexane for 15 minutes and after centrifuging the supernatant was injected directly to the GC-MS system. The limit of quantitation (LoQ) for alkyl sulfonates was set at 0.04 ppm based on the API content of the drug product. Alkyl mesilates were all < LoQ for bromcriptine mesilate capsules and doxazosin mesilate tablets. Similar negative findings were obtained for three neat APIs: bromocriptine mesilate, doxazosin mesilate and trimipramine mesilate. [Their results contradict introductory remarks invoking the side-reaction hypothesis as a source of alkyl sulfonates.] The authors reflect on their sample-extraction technique as follows: “The unpolar solvent n-hexane provides major advantages in inhibiting API extraction and assures longer stability of the analytes”. These comments probably relate to a previous technique using the considerably more polar solvent acetonitrile (Wollein & Schramek, 201149). The latter publication reported on the use of a screening assay for the analysis of MMS, EMS and IMS in 12 drug products containing mesilate-salt APIs. The analytical technique employed acetonitrile extraction of powdered drug product, centrifugation and direct injection of the supernatant into GC/MS equipment. Data for MMS (in ppm based on weight of mesilate-salt drug substance) are shown in Table 2 for 10 drug products. For reasons already discussed any alkyl mesilates detected in a drug product are expected to originate from the drug substance. Data on two additional drug products are not listed; levels of alkyl mesilates in dihydroergotoxin were < LoD and values for desferoxamine were difficult to interpret owing to uncertainty over the MDD and pack size of the drug product. Mean values have been calculated when more than one data point is reported for the same drug/strength. By applying the Cimarosti MMS purge factor of 4100 the implied concentration of MMS in MSA has also been determined resulting in a range of 0.06-231% (based on the reasonable assumption that any MMS present in the drug substance would originate from the MSA reagent). For EMS and IMS the majority of values were at or close to the LoD. In general there was reasonable concordance between drug products showing high values of MMS with those reported to contain EMS or IMS. Using 10 ppm alkyl sulfonate in the drug substance as a level of concern, only 5 drug products exceeded this value for EMS and only one for IMS. By contrast, MMS was >10 ppm for 29 samples (out of a total of 32 analyzed). For EMS the highest concentration was 242 ppm (in dihydroergotamine) with an implied content of 116% EMS in MSA. For IMS dihydroergotamine again contained the highest concentration of 278 ppm, representing an implied concentration of 114% in MSA. Table 2: Data from Wollein & Schramek, 2011, on MMS in 10 Drug Products containing Mesilate-Salt APIs Drug (no of samples)

Dose strength (mg)

Dose as mesilate salt (mg)

Reported MMS content (µg)

MMS in drug substance (ppm)

**Implied MMS concentration in MSA (%)

Doxazosin (4) Doxazosin (5) Doxazosin (4)

1.0 2.0 4.0

1.21 2.42 4.84

0.145 0.168 0.173

119.8 69.4 35.7

49.1 28.5 15.4


15


Page 16 of 36

Doxazosin (2) 8.0 9.68 0.228 23.6 9.7 Bromocriptine (2) 2.5 2.87 0.122 42.5 17.4 Bromocriptine (2) 5.0 5.74 0.143 24.9 10.2 Bromocriptine (1) 10.0 11.47 0.361 31.5 12.9 Pergolide (1) 0.05 0.065 0.037 569 230.8 Pergolide (1) 0.25 0.33 0.033 100 41.0 Pergolide (1) 1.0 1.3 0.086 66.2 27.1 Dihydroergotamine (1) 1.0 1.16 0.287 247.4 101.4 *Dihydroergotamine 1.0 1.16 0.468 403.4 165.4 (1) Rasagiline (1) 1.0 1.56 0.024 15.4 6.3 Reboxetine (1) 1.0 1.29 0.202 39.1 16.0 Eprosartan (2) 600 735.6 0.682 0.93 0.38 Trimipramine 50 66.3 5.208 79.6 32.6 Maprotiline 50 67.3 0.724 10.8 4.4 Saquinavir (2) 500 571.6 0.085 0.15 0.06 *Third sample < LoD. **Multiply previous column by 4100 (purge factor calculated from data in Cimarosti et al, OPRD, 2010)

Table 3: Doxazosin Mesilate: MMS content vs tablet strength based on data from Wollein & Schramek, 2011 Tablet Strength (mg) MMS content (µg) Mean content (µg) Ratios for MMS content

1.0

2.0

4.0

8.0

0.143, 0.134, 0.121, 0.180 0.145 1.0

0.154, 0.138, 0.178, 0.162, 0.210 0.168 1.2

0.121, 0.140, 0.203, 0.227 0.173 1.2

0.227, 0.228 0.228 1.6

An additional check can be made by evaluating the relationship between MMS content and tablet strength, which has been performed for doxazosin mesilate tablets (Table 3). Assuming that MMS is present in doxazosin drug substance at a reasonably stable concentration it would be expected that the absolute amount present in drug products would be proportional to tablet strength, which is shown not to be the case. If the authors were minded to posit the side-reaction hypothesis to explain the detection of MMS in almost all samples, this could be countered by: (a) methanol, being considerably more polar and more expensive than ethanol, is not a particularly suitable solvent for sulfonic-acid-salt synthesis and (b) no analysis of solvent residues was undertaken in order to validate such an explanation. Overall, the data from Wollein & Schramek, 2011, are considered to have many hallmarks of artefacts given that: MMS was claimed to be present at >10 ppm in 29/32 (90.6%) of products evaluated whereas no other researchers have reported finding MMS at > LoD or LoQ for any mesilate-salt APIs; the Wollein & Schramek data are truly unique in this respect and this fact alone should have triggered a follow-up evaluation of provisional data generated using a screening assay. It is inconceivable that MMS concentrations up to 570 ppm (in pergolide API) would be acceptable to manufacturers and/or regulatory agencies; no attempt was made to verify the results by contacting manufacturers or to cross-validate the assay methods using acetonitrile or n-hexane as extractants; purge factor considerations indicate Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

16

Page 17 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


that the MSA reagent would have needed to be heavily contaminated with MMS (in some cases to >100%) to have produced such high concentrations in the APIs; the results for doxazosin mesilate (24-120 ppm MMS) are discordant with those determined using n-hexane extraction (500 °C) are hydrogen cyanide and methane52, indicating that thermal decomposition at temperatures > 200 °C can produce methylating species (such as the methyl cation or radical) which could react with mesilate anion to produce MMS. Detection of EMS and IMS might be explained by transesterification of MMS in the GC equipment by residual ethanol and isopropanol respectively in the drug products. Following receipt of the evaluation shown in Table 2, Dr Wollein failed to respond to a number of communications inviting him to defend the published results. There is a precedent for artefact formation of alkyl sulfonates during GC analysis. As part of the description of official Ph.Eur method of analysis for alkyl besilates in besilate salts (Pharmeuropa 201453) it is noted that: “This method (head-space GC/MS) is not suitable for clopidogrel besilate since it was observed that methyl benzenesulfonate was obtained during the gas chromatography analysis as an artefact originating from degradation.” It should be noted that the Ph.Eur method employed a derivatization technique and so the mechanism of artefact formation is likely be different from that proposed for the Wollein & Schramek findings. Toxicological Update According to the provisions of ICH M7 (R1)54 generic limits for a particular mutagenic impurity (MI) are dependent mainly on the duration of exposure. For exposure >10 years the limit is set at the current lifetime TTC of 1.5 µg/day with higher limits for exposures of 1 month (120 µg/day), 1-12 months (20 µg/day) and 1-10 years (10 µg/day). Since virtually all alkyl sulfonates have been shown to act as bacterial mutagens in the Ames’ assay1, these limits will apply generally except in relation to MMS, EMS and IMS for which compoundspecific values can be determined based on in-vivo toxicological data. Alkyl esters of sulfonic acids act as chemical and biological alkylating agents. EMS has been shown to alkylate cellular, nucleophilic sites via a mixed SN1/SN2 reaction mechanism. While ethylation of DNA occurs principally at nitrogen positions in the bases, because of the partial SN1 character of the reaction, EMS is also able to produce significant levels of alkylation at oxygens such as the O6 of guanine and in the phosphate groups attached to DNA55. Highly effective negative-charge delocalization has the dual effect of making sulfonate anion a good leaving group and, as previously mentioned, an extremely poor nucleophile. By contrast, alkyl esters of carboxylic acids, even those containing a good leaving group such as ethyl trifluroacetate (CAS 383-63-1), test negative for bacterial mutagenicity56.


17


Page 18 of 36

MMS: MMS has been tested for carcinogenicity in the mouse and the TD50 value listed in the CPDB (Carcinogenic Potency Database) is 31.8 mg/kg/day57. Thus a lifetime AI (acceptable intake) of 31.8 µg/day can be determined as per ICH M7 (R1) (by extrapolation to a carcinogenic risk of 1 in 105). Although MMS is mutagenic in the Ames’ assay, data from the ToxTracker58 assay indicate that the principal mode of action is oxidative stress rather than DNA damage. EMS: Müller et al, 200959, established a threshold dose of 25 mg/kg/day EMS in terms of genotoxicity in a transgenic-mouse 28-day toxicity study. This was also considered to be the NOEL (no-observed-effect level) based on a variety of considerations such as the absence of effects in bone marrow, liver, and GI-tract tissue. Müller & Gocke60 employed the threshold dose as a metric for determination of a PDE (permitted daily exposure) using principles set out in ICH Q3C [and later adopted in ICH M7 (R1)]. Application of an overall assessment factor of 12000 leads to a PDE of 104 µg. More recently Bercu et al, 201861, have argued that a PDE of 1.0 mg is more appropriate: “It was noted by the authors (Müller & Gocke) that the adjustment factors might be too conservative. The adjustment factor F4 used by Müller & Gocke for severity of toxicity was equal to 10, since mutagenicity used here to predict carcinogenicity is considered a severe toxicity. However it was pointed out that since mutagenicity of EMS shows a non-linear dose relationship this would also be true for the carcinogenicity and that a lower F4 adjustment factor could be applied. Reducing the total adjustment factor to 1200, would result in a PDE of 1 mg/day.” IMS: Coffing et al, 201562, reported a NOEL (no-observed-effect level) of 0.25 mg/kg/day for IMS in a 28-day in-vivo Pig-a mutation assay in the rat. The authors determined a PDE of 2.5 µg using ICH Q3C methodology and an overall assessment factor of 5000. Since the mutagenicity dose-response is, like EMS, nonlinear, similar arguments as for EMS are considered to apply in relation to the F4 factor for serious toxicity. Thus an updated PDE of 25 µg can be determined using an overall assessment factor of 500. The recommended lifetime limits for the three alkyl sulfonates plus comparative mutation frequencies (in TA100 –S9 taken from Hakura et al, 198463) are shown in Table 4. Clearly EMS is the least potent ester and IMS the most potent, although the available in-vivo data do not reflect the significant in-vitro potency differences. It should be noted however that three different in-vivo techniques were employed for potency determination. [Unfortunately no invitro data on mutation frequencies were reported for alkyl tosilates or besilates.] Table 4: MMS, EMS, IMS: Comparative Mutation Frequencies (MFs) and Recommended Lifetime Limits Metric MMS EMS IMS *Relative MF at 1 mM 1.0 0.00045 25 *Relative MF at 10 mM 1.0 0.072 47 **Lifetime limit AI = 32 µg/day PDE = 1 mg PDE = 25 µg *Mutation frequency in Salmonella typhimurium TA100, -S9Error! Bookmark not defined.; **See text for explanations

Regulatory Aspects Following the suggestion of the side-reaction hypothesis in Pharmeuropa in 2000, it has been possible to track regulatory developments on alkyl-sulfonate impurities by following statements by EMA (European Medicines Agency) and EDQM/Ph.Eur (European Pharmacopoeia as well as using freedom-of-information (FOI) requests. Virtually no Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

18

Page 19 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


information or feedback could be obtained from other major regulatory agencies, for example FDA, HC (Health Canada) or TGA (Therapeutic Goods Administration, Australia). An account of key regulatory developments is presented below. Regulatory Actions immediately following Viracept EMS Contamination Incident in 2007 In May 2007 the F Hoffmann-La Roche company became aware of contamination of Viracept (nelfinavir mesilate) by EMS. The alarm was raised initially by patient complaints of a noxious odour of the tablets. It was soon quite clear that the Viracept contamination incident24 was unique and dependent on the combined effect of two factors: the presence of high levels of preformed EMS contaminant in MSA caused by an operational GMP failure and spray drying of the entire reaction mixture in order to isolate nelfinavir mesilate API. Nevertheless regulatory agencies instigated several actions based on the assumption (for which there was no evidence) that alkyl-sulfonate impurities could be present in routinely synthesised sulfonic-acid salts. Both Swissmedic64and EMA65 launched surveys of approved products containing sulfonic-acid-salt APIs in October 2007 and January 2008 respectively. Questions for MAHs (Market Authorisation Holders) were similar in both cases and focused on:    

Appropriate toxicologically-based limits for alkyl-sulfonate impurities Availability of suitable validated methods of analysis Appropriate raw-material specifications Risk analysis based on various hypotheses: o Side-reaction o Reagent-impurity o Salt-alcohol interaction (particularly during wet granulation prior to tabletting) o Retention of alkyl sulfonates in recycled solvents

No report on the results of the survey has been issued by either EMA or Swissmedic. However, in 2017 EMA confirmed in response to an FOI request (ASK-16726) that no companies reported toxicologically significant levels of alkyl-sulfonate impurities in the 2008 survey. Swissmedic, in response to an FOI request, provided a document indicating that responses on 72 preparations had been received, although all information on levels of alkylsulfonate impurities was redacted. The document announcing the EMA survey is still available on its website in spite of the fact that all of the proposed justifications for suspecting the presence of alkyl-sulfonate impurities have been disproved both by the response to ASK16726 and the evidence presented above in the Technical Update. It is considered that by withholding information on the outcome of the survey and continuing to display the January 2008 document EMA is perpetuating mistaken perceptions on alkyl-sulfonate impurities. A request to remove the document in question or to add a comment on the outcome of the survey was ignored. Also in 2008 the European Pharmacopoeia Commission (EPC) appointed the Mesilate Working Party (MSLWP; now listed as “dormant”66; chaired by Prof J Midgley) whose terms of reference67 appear to be entirely analytically-focused: To provide support and advice in case of questions raised related to general methods drafted by the working party ie 2.5.37. Methyl, ethyl and isopropyl methanesulfonate in methanesulfonic acid, 2.5.38. Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

19


Page 20 of 36

Methyl, ethyl and isopropyl methanesulfonate in active substances, 2.5.39. Methanesulfonyl chloride in methanesulfonic acid, 2.5.40. Methyl, ethyl and isopropyl toluenesulfonate in active substances, 2.5.41 Methyl, ethyl and isopropyl benzenesulfonate in active substances.

In 201268 Prof Midgley indicated a potentially broader remit for the MSLWP including a review of “the need for, or revision of the wording of, a Production Statement in the monographs for mesylate salts”. A press release from EDQM in 201669 referred to completion/termination of the work of the MSLWP in relation to the introduction of analytical methods for alkyl tosilates and alkyl besilates: “Publication of chapter 2.5.41 terminated the work of the Mesilate Working Party. This working party had been appointed by the Ph. Eur. Commission in 2008 to assist users in determining mutagenic impurities potentially present in mesilate-, besilate- or tosilate salts of active substances.” It thus appears that the extension of Ph.Eur requirements to alkyl-tosilate and -besilate impurities was highly dependent on the output of the MSLWP with the underlying assumption that such impurities could be present. It is not clear whether the MSLWP reviewed evidence on actual levels of alkyl-sulfonate impurities (for example from Certification of Suitability - CEP - applications, OMCL surveys, EMA 2008 survey). Moreover the MSLWP was in an ideal position to perform laboratory-scale simulations of sulfonic-acid-salt syntheses to evaluate the critical factors (if any) that lead to the formation/presence of alkyl-sulfonate impurities. Since the side-reaction hypothesis has been disproved1 (which appears to have been accepted by Prof Midgley/ EDQM68) the principal concern seems to relate to sub-standard starting materials68. In practice this affects only MSA since BSA and p-TSA, which are synthesised by sulfonation (using oleum) of benzene and toluene respectively, contain no mutagenic impurities or precursors of mutagenic impurities. This notion is supported by the fact that the MSLWP did not elaborate methods of analysis for these aromatic sulfonic acids. Thus, the MSLWP may or may not have confirmed the results published by Cimarosti et al25 regarding the large inherent purge factors for potential impurities (such as MMS) in MSA. Overall, reliance on the MSLWP conclusions imposes a non-transparent justification concerning the need for a “Production” section in Ph.Eur monographs for sulfonic-acid salts and in relation to methods of analysis for alkyl sulfonates. For this reason requests were made to EDQM on three occasions (in 2015, Q90910; in 2016, Q103168; in 2017, Q104165) for access to MSLWP reports. All requests were denied because these reports are classified as “working documents” and so are considered confidential. Key text from the 2017 response to Q104165 (shown in Box 4) cites data from the 2011 Wollein & Schramek publication, which, as discussed above, shows evidence of analytical-artefact formation.

Box 4: Part content of final refusal by EDQM to release MSLWP reports (February 2017) As you certainly know, the work of the MSL working party is based on the Viracept case and the related extensive literature, information obtained from EMEA (as it was previously called), and requests from National authorities to the Ph. Eur. Commission. You will also remember your own review article in Readers' tribune of Pharmeuropa of 2012 and the follow-up letter from Prof Midgley, the chair of the MSL working party, which is fully supported by the group and EDQM. It appears that studies performed by OMCLs, e. g. as published by Wollein et al in Pharm. Ind. (2011), 574-80, demonstrate the presence of alkylsulfonates in certain finished Elusive Impurities – Evidence vs Hypothesis products. ACS Paragon Plus Environment

20

Page 21 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Internal EMA review of alkyl-sulfonate impurities A letter from the EPC secretariat (signed by Dr Susanne Keitel, Director of EDQM and Mrs Cathie Vielle, secretary to EPC) was sent to Dr J-L Robert (Chair of both QWP and the EPC) on 27.07.15 as follows: “A new article by Snodin et al indicates that the current paradigm of genotoxicity on alkyl sulfonate (sic) and current regulatory positions may no longer be justified. In order to have a better understanding of the overall impact on the Ph.Eur but also on the assessment of marketing authorisation applications and applications for a certificate of suitability, the European Pharmacopoeia Secretariat seeks the opinion of the Joint CHMP/CVMP Quality Working Party (QWP) on this matter and more precisely on the question whether the current approach needs to be changed.” [CHMP = Committee for Medicinal Products for Human Use; CVMP = Committee for Medicinal Products for Veterinary Use]. In fact both SWP (Safety Working Party) and QWP were involved. The only publicly available comment on the outcome of the review is contained in the CVMP minutes of the meeting on 4-5 November 201570: “The Committee endorsed the QWP response to the EDQM request for an opinion on new information on alkyl sulfonates. The QWP reviewed the article from Snodin et al. QWP acknowledges the scientific rationale in this article and that the formation of alkyl sulfonates is very low and very much depends on the reaction conditions. This makes the presence of these mutagenic impurities at toxicologically significant levels unlikely. However, as the presence and formation of these alkyl sulfonates cannot be totally excluded, QWP proposes the following approach: marketing authorisation holders should justify via Risk Assessment that alkyl sulfonates are not expected to be present for their product, which may be sufficient.”

Several FOI requests (ASK-17940, 20704 and 20709) were made to EMA in order to obtain more details of the review. Heavily redacted extracts from SWP and QWP minutes were made available; the latter contained a comment closely similar to that reported in the CVMP minutes although SWP minutes were more informative (Box 5). The formal response to EDQM by Dr J-L Robert (provided by EMA; EDQM refused to disclose any information on the basis that correspondence between EMA and EDQM is confidential) is shown in Box 6. [A new chair of the EPC, Dr Tobias Gosdschan, was elected in March 201671.] Box 5: Comments on alkyl sulfonates in SWP minutes Publication on alkyl sulfonates in sulfonic-acid salts 30.06.15: Peter explained that the assumption of the presence of genotoxic impurities was erroneous. Will be discussed further in the QWP. M7 quality experts informed. 29-30.09.15: Very short discussion. Overall SWP is in agreement with the paper. Feedback from QWP expected (can these impurities be found in MPs?)


21


Page 22 of 36

Box 6: Main text of letter (02.10.15) from Dr J-L Robert (QWP Chair) to Dr S Keitel (Head of EDQM) Subject: QWP response to EDQM request for opinion on new information on alkyl sulfonates Quality Working Party had a look at the article from Snodin et al. We acknowledge the scientific rationale in this article and that the formation of alkyl sulfonates is very low and very much depends on the reaction conditions. This makes the presence of these mutagenic impurities at toxicologically significant levels unlikely. However, as the presence and formation of these alkyl sulfonates cannot be totally excluded, QWP proposes the following approach: MAHs should justify via Risk Assessment that alkyl sulfonates are not expected to be present for their product, which may be sufficient. Depending on the outcome of this RA supportive analytical data might or might not be required. With regards to the Ph.Eur texts (and in particular the presence of a Production section in monographs), QWP consider that amendment of the current texts is not necessary.

Requests to EMA for information (as per the FOIs noted above) on a variety of ancillary issues were made and the main points are summarised in Table 5. Regarding point 4, EMA’s position is that it does not create a competing interest to be an established expert at both EMA and EDQM. However, in this case Dr Robert was a rather special expert given his dual roles as chair of key committees in both organizations. Moreover, although there may have been no formal conflict of interest, the QWP review cannot be considered independent of EDQM. Several requests were made to EMA regarding clarification of point 8 on the precise nature of a risk assessment that would obviate the need for analytical verification regarding the absence of alkyl-sulfonate residues, but in all responses no information was provided. The main text of EMA’s final response is shown in Box 7.


22

Page 23 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 5: Questions and responses relating to EMA review of alkyl sulfonates # Question Summary of EMA response/Comment 1 2 3 4

Single source of evidence supplied to SWP/QWP to support review? SWP and QWP opinions appear to be in conflict? Why no reflection paper for public consultation? Potential conflict of interest for Dr J-L Robert who was chair of both QWP and EPC?

5

Co-operation between EDQM and EMA

6

Lack of transparency regarding decisionmaking process?

7

Comment to the effect that “the presence and formation of these alkyl sulfonates cannot be totally excluded”? Risk-assessment/management parameters?

8

Only the Snodin/Teasdale article was provided. No provision of archival data available to EMA and EDQM. SWP later agreed with QWP’s approach. Not a guideline so no public consultation. EMA policy on conflicts of interest mentions only direct or indirect relationships with pharmaceutical industry. The fact that an EU expert is both an expert at EMA and EDQM is not seen as a competing interest. The significant extent of co-operation could, in this case, be considered akin to collusion Outcome of review mentioned in CHMP/CVMP minutes. Heavily redacted SWP/QWP minutes obtained using FoI shed no light on how conclusions were reached. No evidence provided by EMA. Not specified.

Box 7: Main text of EMA’s final response to queries on QWP review of alkyl sulfonates (ASK-20709) As previously clarified, the EDQM had requested the Quality Working Party’s opinion on the matter, providing only the publication from Snodin et al. It should be noted that the Quality Working Party (QWP) acknowledged the scientific rationale in the article, that the formation of alkyl sulfonates is very low and very much dependant (sic) on the reaction conditions. This makes the presence of these mutagenic impurities at toxicologically significant levels unlikely (as stated in your article). However, as the presence and formation of these alkyl sulfonates cannot totally be excluded, the QWP has proposed the following approach: MAHs should justify via a Risk Assessment that alkyl sulfonates are not expected to be present for their product, which may be sufficient. Depending on the outcome of this Risk Assessment, supportive analytical data may or may not be required. •

Please note that the opinion of all the experts who participated in the discussion was taken into account.

•

The QWP has taken a precautionary approach and will at this point maintain this position

Proposal to modify Ph.Eur texts on alkyl-sulfonate impurities Any proposal to modify existing or draft Ph.Eur texts has to be made via one’s regional pharmacopoeia; in the case of a UK resident it is the British Pharmacopoeia (BP). Consequently a formal proposal was made to the BP in July 2015 to suppress the following three monographs: 2.5.38. Methyl, ethyl and isopropyl methanesulfonate in active substances, Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

23


Page 24 of 36

2.5.40. Methyl, ethyl and isopropyl toluenesulfonate in active substances, 2.5.41 Methyl, ethyl and isopropyl benzenesulfonate in active substances. Monographs relating to the purity of MSA were not included in the proposal; it was emphasised that these should be retained: 2.5.37. Methyl, ethyl and isopropyl methanesulfonate in methanesulfonic acid, 2.5.39. Methanesulfonyl chloride in methanesulfonic acid. It is understood that the UK delegation tabled the proposals at an EPC meeting in November 2015 and it appears that a decision to reject was made at the March 2016 EPC meeting. Representatives of all national pharmacopoeias, except the UK delegation, voted to reject the proposals and it was left to the EPC secretariat (provided by EDQM) to “come up with a justification”72. Only a short excerpt of the meeting minutes (shown in Box 8) was released by BP. Box 8: EPC justification for rejection of proposals to suppress Ph.Eur monographs on alkyl sulfonates in sulfonic-acid salts A request had been received from the UK delegation requesting the suppression of general chapters 2.5.38, 2.5.40 and 2.5.41 for the control of potentially genotoxic alkyl sulfonates in active substances. This was based on data which seemed to show that during the synthesis of certain mesilates, besilates and tosylates the impurities could not be formed and therefore the general chapters and the corresponding production statements in a number of individual monographs should be suppressed. The request was discussed by QWP and the Presidium. Further information had been sought, i.e. publication on the readers’ tribune in Pharmeuropa and a reply from the chair of the MSL working party, in 2012. A publication from a German OMCL indicated that genotoxic alkyl sulfonates have been detected in certain finished products, though at low levels below the TTC (threshold of toxicological concern). These impurities may stem from the use of sub-standard starting materials, even when they did not occur during synthesis. The QWP recommended not to suppress the general chapters and the production statements.

There are major concerns over the scientific validity of this justification including: 

  



the German OMCL publication cited refers to the Wollein & Schramek 2011 Pharmind article which, as discussed in a previous section, is considered to be unconvincing based on the likelihood of analytical artefacts; Even if the Pharmind publication were valid to some extent it deals only with mesilate esters in drug products, no data being presented on tosilates or besilates; Reference is made to “substandard starting materials” even though a specific request was made for retention of monograph 2.5.37 on MSA purity; The fact that Ph.Eur has not developed monographs that evaluate the purity of TSA and BSA implies that there are no concerns regarding carryover of impurities from these reagents, thus further undermining the EPC justification; Even if the reported concentrations of alkyl mesilates were considered genuine, potential patient exposures at the maximum daily dose of each drug substance would have been below the lifetime limits shown in Table 4 (and well below any limits for which the LTL – less than lifetime – principle can be applied)


24

Page 25 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Given these criticisms it appears that the official justification provided in Box 8 could be considered as a less-than-rigorous scientific rationale for a policy decision, which was acknowledged by the UK delegation. However, in order to preserve goodwill with the majority of EPC representatives, the UK delegation was unwilling to raise objections, and so without the support of the BP no further challenge to the EPC is possible. Current status of Ph.Eur provisions In a 2016 press release (Box 9) EDQM provided an updated version of the production section that is included in every Ph.Eur monograph for a sulfonic-acid-salt drug substance. [At the time of writing Ph.Eur contains monographs on 14 mesilate-salt drug substances: betahistine, bromocriptine, codergocrine, deferoxamine, dihydroergocristine, dihydroergotamine, doxazosin, ergoloid, imatinib, pefloxacin dihydrate, pergolide, phentolamine, saquinavir and ziprasidone trihydrate. Only one tosilate-salt drug substance is monographed, sultamicillin dihydrate, although four besilates are included: amlodipine, atracurium, cisatracurium and clopidogrel. In addition there are three monographs (2.5.38, 2.5.40 and 2.5.41 on methods of analysis for alkyl mesilates, tosilates and besilates, respectively. In February 2019 EDQM’s Certification database73 listed a total 74 valid CEP authorizations issued from 2009-2019 for sulfonic-acid salts (33 for mesilate salts, 40 for besilate salts and one for a tosilate salt)]. Additional information is provided in the 2016 press release on the use of risk assessment and risk management to justify not undertaking product testing for alkyl-sulfonate impurities. As discussed previously1, and summarised in the initial sections of this article, detailed, precise and extensive mechanistic, kinetic and experimental information is available which demonstrates that no alkyl-sulfonates are formed or carried over as reagent impurities Box 9: Ph.Eur statement on monographs for sulfonic-acid salts in 2016 press release “It is considered that [XXX esters] are genotoxic and are potential impurities in [name of the API]. The manufacturing process should be developed taking into consideration the principles of quality risk management, together with considerations of the quality of starting materials, process capability and validation. The general method [2.5.XX] is available to assist manufacturers.” The purpose of these Production sections is to alert users of the risk related to the potential presence of such mutagenic impurities in mesilate-, besilate- or tosilate salts of active substances. Marketing Authorisation Applicants are not obliged to perform the testing when they can justify via risk assessment that alkyl sulfonates are not expected to be that as part of the synthesis of a sulfonic-acid salt under GMP conditions no present in their product. The final decision on the evaluation of the Marketing Authorisation Application and especially whether “the manufacturing process [has been] developed taking into consideration the principles of quality risk management, together with considerations of the quality of starting materials, process capability and validation” lies with the Competent Authorities.

Against this background QWP has asserted that formation of alkyl-sulfonate impurities is unlikely but cannot be totally excluded, yet has declined on several occasions to suggest any mechanism explaining how this might come about or provide any experimental evidence Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

25


Page 26 of 36

showing unexpected-impurity formation. [Note that this is different from EDQM’s justification for the continued implementation of the production section of affected monographs based on the notion of reagent-impurity carryover to the API.] Assuming this highly unlikely impurity-formation hypothesis remains the “official” QWP position, how is an applicant able to perform a satisfactory risk assessment (without undertaking impurity assays) if it is believed that an unpredictable “rogue” element of the synthesis can lead to the formation of alkyl-sulfonate impurities? [Given that the chemical mechanisms for alkylsulfonate formation rely on simple chemical concepts and have been established to a high degree of certainty, there appears to be minimal scope for alternative mechanisms.] Risk assessment and/or risk management have been advocated by both QWP (Boxes 6 & 7) and Ph.Eur (Box 9) although, in order for Industry and regulatory agencies to maintain a consistent approach, it is critical to establish the key factors ensuring the absence of alkylsulfonate impurities. However, as shown in Table 5, no guidance from QWP was initially provided, nor did anything emerge in the next 2.5 years. Consequently in mid-2018 a riskmanagement paradigm (shown in Table 6) was proposed to both QWP and EDQM (via the BP); this was rejected by QWP74 and no acknowledgement or reply has been received from BP or EDQM. So it seems that both QWP and BP/EDQM are minded to ignore their own recommendations to implement risk assessment/risk management procedures. It should be stressed that all of the factors listed in Table 6 fall under the general heading of GMP and so are really applicable only if a manufacturer is contemplating synthesis of a sulfonic-acid-salt in a non-GMP environment. It is also important to emphasise that all known mechanisms of alkyl-sulfonate formation are taken into consideration. Assuming that a GMP-based risk assessment/risk management is implemented as described in Table 6 the need to confirm “process validation” is unclear. Would regulators demand alkyl-sulfonate assay data on the final drug substance in order to confirm “validation” in spite of the statement in Boxes 6, 8 and 9 that applicants are not obliged to perform testing? Discussion and Conclusions The current wording of the Ph.Eur monographs relating to analytical methods for alkyl sulfonates in sulfonic-acid salts strongly implies an ongoing belief in a solvent-sulfonic acid interaction leading to the formation of alkyl-sulfonate impurities. Thus, nearly two decades have elapsed since the side-reaction hypothesis was conceived and unfortunately there remains an actual or perceived regulatory requirement for implementing analytical controls on (hypothetical) alkyl-sulfonate impurities. For instance, Guo et al, 201475, state: “..although in the majority of cases the high level of regulatory concern over the potential presence of sulfonate esters in APIs is largely unwarranted, the ability to detect genotoxic impurities at low concentrations is still important in the development of APIs.” And in the US context of test materials for clinical trials, Olsen et al, 201776, indicate: The need to control alkyl sulfonate esters is an example of a typical early phase regulatory expectation. Despite ongoing debate about the safety liabilities of these potential impurities or the lack of probability that they would be present, in the authors' experience, specification controls will be expected for these impurities, even at Phase 1. Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

26

Page 27 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


It should be emphasized that neither of these publications provides examples of positive findings for alkyl-sulfonate impurities. A further recent case concerns the discussion of impurities in the review of dabrafenib mesilate by the Japanese agency (PMDA)77: “The level of methanesulfonate alkyl ester can be controlled ….throughout the manufacturing process by setting the control level of alcohol in acetone and content of methanesulfonate alkyl ester in methanesulfonate”. Thus, it appears that PMDA mistakenly believes that the presence of alcohol in acetone (used as a reaction or recrystallization solvent) could lead to the formation of alkyl-mesilate impurities. In terms of regulatory aspects, EDQM was not deterred from introducing the Production Statement in 2004 by evidence from an MSc project indicating that no alkyl-mesilate impurities could be detected in several drug substances containing residues of short-chain alcohols. The existence of the MSc thesis was revealed by a staff member from EDQM alarmed that this critical information had not been disclosed in the run-up to the introduction of the Production Statement. It would have been good practice for EDQM to commission further assays on mesilate-salt drug substances in order to ascertain the validity of the sidereaction hypothesis; whether this action was taken is unclear, again owing to non-disclosure by EDQM. Introduction of the Production Statement against a background of hypothesis rather than evidence is considered to be the first in a line of critical actions (or inactions) by EDQM and/or EMA that have thwarted a science-based evaluation of sulfonic-acid-salt impurities. Subsequent key regulatory factors are considered to be (in chronological order): 

 



2008: Non-disclosure by EMA of the negative outcome of the survey of approved sulfonic-acid-salts initiated in January 200865. [Establishment of the MSLWP would surely have been compromised or questioned had this information been made publicly available]. 2015: Conclusion of internal and largely confidential SWP/QWP review based on assertion rather than evidence. 2015-2017: Extension of controls to tosilate and besilate salts justified by EDQM on the basis of confidential reports of MSLWP. All attempts to access these reports have been denied. 2017: EPC rejection of request to suppress the “production” part of Ph.Eur monographs on sulfonic-acid-salt drug substances justified by reference to data from a publication49 that is considered unreliable owing to analytical artefacts.

As a follow-up to the EPC decision noted above to maintain the production statement for sulfonic-acid-salt monographs FOI requests were made to EDQM in March 2017 (Q104165) and April 2017 (Q105237). A number of questions were asked on various relevant issues including: apparent lack of due diligence on 2011 Pharm Ind publication (based on the analysis provided in this article); justification for suppression of data from 2002 MSc thesis by Ahmad when introducing the Production Statement for monographs on mesilate-salt drug substances; mandate of the MSLWP; anonymised data held by EDQM on impurities in MSA Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

27


Page 28 of 36

and sulfonic-acid salts. EDQM initially responded that the MSLWP would be reconstituted in order to address these questions. However it was noted one year later that this working party was still listed as “dormant” on the EDQM website, and so a request for clarification was made in March 2018 (Q113498). EDQM replied that “we will come back to you in September 2018”. Then in July 2018 EDQM provided an additional response to Q105237 and Q113498: “..we will need to investigate the related documentation or consult our experts further. We will provide you with a more detailed reply once this is complete.” In response to a further request for an update (Q121065) sent in early February 2019 EDQM indicated that: “we will reply to you soon”. At the time of writing (April 2019) over 2 years have elapsed without any factual response. It seems extraordinary that EDQM is unable, or unwilling, to answer simple questions in a timely manner on the evidentiary basis for such an important policy that has been in place (at least for mesilate salts) for nearly 15 years. The critical mechanistic and kinetic data demonstrating the absence of alkyl-sulfonateimpurity formation under normal conditions of sulfonic-acid-salt synthesis were generated entirely as part of an Industry initiative, the work being carried out at the independent Product Quality Research Institute (PQRI)78. On the two occasions when initiatives were undertaken by regulatory bodies, presumably with a view to substantiating existing hypotheses (2002 MSc thesis at University of Strathclyde and 2008 EMA survey of licensed sulfonic-acid-salt APIs), the unanticipated negative results appear to have been suppressed. Contacting key personnel in other regulatory agencies (FDA, Health Canada, Therapeutic Goods Administration in Australia) produced no responses (or even acknowledgements of receipt). Also noteworthy is the fact that, to date, no regulatory body has attempted to provide an evidence-based rebuttal of the narrative on alkyl-sulfonate impurities articulated in early 2015 by Snodin & Teasdale1. An absence of response from regulatory agencies was noted by Hadlington who contacted three agencies including FDA and EMA79. Tracking the timeline for the implementation of controls for alkyl-sulfonate impurities in different territories suggests that EDQM/Ph.Eur played a lead role by initiating concerns followed by promulgation of the need for compendial and other regulatory controls. It is now apparent that EDQM’s original rationale (in the form of the side-reaction hypothesis) was effectively a false alarm, and that EDQM has failed to provide evidence supporting continuation of controls on mesilate salts and extension of controls to tosilates and besilates, as well as denying requests for access to information on exactly what rationale was used by the MSLWP to justify such controls. Moreover, both EMA and EDQM seem to have withheld the release of negative evidence on this topic (for example from CEP applications) and have also employed administrative measures to maintain the regulatory status quo and block any further scientific evaluation. Also disappointing is the absence of follow-up by EMA/EDQM on their proposals (first made in late 2015) to employ “risk assessment” as an alternative to performing analytical testing for the presence of alkyl-sulfonate impurities (see Boxes 6, 7 and 9). Opinions on what constitutes a convincing risk assessment are likely to differ and so this type of procedure is considered viable only if a suitable template is available. Since EMA and Elusive Impurities – Evidence vs Hypothesis ACS Paragon Plus Environment

28

Page 29 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


BP/EDQM have repudiated or ignored proposals for such a template (based on the key factors in Table 6) the status of risk assessment versus analytical testing is unclear. One way forwards would be for Industry to undertake for a limited period both risk assessment and analytical testing in parallel in order to demonstrate the viability of the former. Maybe evidence will eventually displace hypothesis on this long-running regulatory conundrum? Table 6: Key factors in relation to risk assessment/management of sulfonic-acid-salt drug substances Factor Reaction conditions

Comment -

Reagent purity

-

Product isolation

-

Solvent recycling Storage stability

-

GMP

-

Equimolar proportions of base-form of API and sulfonic acid (or slight undercharge of sulfonic acid) Ensure that there is no sulfonic-acid carryover from previous synthetic steps to salt-forming stage of synthesis Charge sulfonic acid directly from supplier’s container Addition of acid to base, not vice versa Steady rate of addition with adequate stirring to avoid hot-spots Monitoring of pH during addition of final 10% of acid reagent Appropriate type and volume of solvent (ethanol as default) Appropriate reaction time and temperature Addition of water to reaction mixture not necessary (likely to reduce yield of sulfonic-acid salt) Pharma-grade MSA strongly preferable with specified limits for MMS, EMS, IMS and MsCl (with reference to Ph.Eur general methods 2.5.37 and 2.5.39, or other similar analytical methods) For p-TSA and BSA there are no known mutagenic impurities (or precursors of mutagenic impurities) although these reagents should be of adequate purity Both p-TSA and BSA are available as hydrates which are acceptable for use but the presence of water may reduce yields owing to increased solubility of sulfonic-acid salts in aqueous solvents (as noted above) Deliquoring by filtration/centrifugation strongly preferred Solvent-washing of filtered precipitate of sulfonic-acid salt also advisable; suitable solvents include cold ethanol and ethyl acetate Spray drying is suitable only if sulfonic-acid salt forms a solvate (as for nelfinavir mesilate); use of ultra-pure MSA is necessary in this case Recrystallization can be employed to obtain the highest purity sulfonicacid salt with ethanol as the default solvent Some degree of solvent recycling should produce no problems if pharma-grade reagents are employed Sulfonic-acid reagents should be stored in a manner that precludes interaction with an alcohol, for example using a supplier’s (disposable) plastic container Sulfonic-acid salts generally exhibit good storage stability No evidence or plausible mechanism for alkyl sulfonate formation during storage of bulk sulfonic-acid salts at moderate temperatures Synthesis of pharmaceutical sulfonic-acid salts should be undertaken according to GMP following ICH Q7 and/or Ph.Eur Substances for Pharmaceutical Use.


29


Page 30 of 36

Acknowledgements Thanks are due to Andrew Teasdale and Michael Snodin for their helpful comments on draft versions of this article. Conflict of Interest The author declares that he has no conflict of interest in relation to the topic of this manuscript. Funding The author is an independent pharmacotoxicology consultant and no external funding has been received to support the preparation of this article.


30

Page 31 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References Snodin D, Teasdale A (2015) Mutagenic Alkyl-Sulfonate Impurities in Sulfonic Acid Salts: Reviewing the Evidence and Challenging Regulatory Perceptions. Org Process Res Dev, 19, 11, 1465-1485. https://pubs.acs.org/doi/abs/10.1021/op500397h. Accessed 4th June 2018 1

National Toxicology Program. Testing Status for Bromobenzene. https://ntp.niehs.nih.gov/testing/status/agents/ts-10060-g.html. Accessed 5th June 2018 2

Chlorobenzene. MAK assessment (1995). https://onlinelibrary.wiley.com/doi/pdf/10.1002/3527600418.mb10890e0012. Accessed 5th June 2018 3

Haloalkanes and haloarenes. Chemistry Unit 10, 281-314. NCERT (National Council of Educational Research and Training), 2015. http://ncert.nic.in/ncerts/l/lech201.pdf Accessed 4th June 2018 4

Guidance 18.2.6. Impurities relating to mesilates, tosilates, (di)isetionate and besilates. Therapeutic Drugs Administration, Australia. https://www.tga.gov.au/book/182-related-impurities-drug-substances-and-drugproducts. Accessed 12th April 2019. 5

Moriya M et al (1983) Further mutagenicity studies on pesticides in bacterial reversion assay systems. Mutat Res. 116(3-4):185-216. https://www.sciencedirect.com/science/article/abs/pii/0165121883900599?via%3Dihub. Accessed 12th April 2019. 6

Carcinogenic Potency Database. p-Chlorophenyl-p-chlorobenzene sulfonate. https://toxnet.nlm.nih.gov/cpdb/chempages/p-CHLOROPHENYL-pCHLOROBENZENE%20SULFONATE.html. Accessed 4th June 2018 7

FAO Specifications and Evaluations for Agricultural Pesticides (2007) Ethofumesate. http://www.fao.org/fileadmin/templates/agphome/documents/Pests_Pesticides/Specs/Ethofumesate07.pdf. Accessed 4th June 2018 8

Elder DP et al (2010) The utility of sulfonate salts in drug development. J Pharm Sci. 99(7):2948-61. http://c3162802.workcast.net/10211_The_Utility_of_Sulfonate_Salts_in_Drug_Developments_2011106154629. pdf. Accessed 4th June 2018 9

10

Dr DG Watson, private communication, January 2016.

Sitaram C et al (2011) Determination of Alkyl Methanesulfonates in Doxazosin Mesylate by Gas Chromatography-mass Spectrometer. Indian J Pharm Sci. 73(1):107-10. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3224401/. Accessed 4th June 2018 11

Cardura XL (doxazosin mesylate). Pharmacology/Toxicology Memo to File. US Food and Drug Administration, November, 2004. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2005/021269s000_PharmR.pdf. Accessed 4th June 2018 12

Cardura XL (doxazosin mesylate extended release tablets). Chemistry Review, US Food and Drug Administration, November 2004. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2005/021269s000_ChemR.pdf. Accessed 4th June 2018 13

Drug Approval Package. Cardura XL. Approval date 2/22/2005. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2005/021269_carduraxl_toc.cfm. Accessed 5th June 2018 14

Doxazosine 2 and 4 mg tablets. Public Assessment Report of the Medicines Evaluation Board in the Netherlands. August 2008. https://db.cbg-meb.nl/mri/par/nlh-0896-001-002.pdf. Accessed 5th June 2018 15

Doxazosine 8 mg tablets. Public Assessment Report of the Medicines Evaluation Board in the Netherlands. March 2008. https://db.cbg-meb.nl/mri/par/nlh-1160-001.pdf. Accessed 5th June 2018 16


31


Page 32 of 36

Balaji N, Sultana S (2017) GC and GC-MS Detection of Alkyl Mesylates in Active Pharmaceutical Ingredients. Int. J. Pharm. Sci. Rev. Res. 46(2), 88-92. http://globalresearchonline.net/journalcontents/v462/16.pdf Accessed 7th June 2018 17

Snodin D (2006) Residues of genotoxic alkyl mesylates in mesylate salt drug substances: Real or imaginary problems? Regul Toxicol Pharmacol 45(1),79-90 https://www.sciencedirect.com/science/article/pii/S0273230006000286 Accessed 21st June 2018 18

Teasdale A et al (2009) Mechanism and Processing Parameters Affecting the Formation of Methyl Methanesulfonate from Methanol and Methanesulfonic Acid: An Illustrative Example for Sulfonate Ester Impurity Formation. Org. Process Res. Dev.13 (3), 429-433. 19

20

Amlodipine besylate. Drugbank. https://www.drugbank.ca/salts/DBSALT001054. Accessed 21st June 2018

21 pK Values for Organic and Inorganic Bronsted Acids at 25 oC. a https://owl.oit.umass.edu/departments/OrganicChemistry/appendix/pKaTable.html Accessed 21st June 2018

2,6-Lutidine. PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/2_6-Dimethylpyridine Accessed 21st June 2018 22

Teasdale A, Delaney EJ, et al. 2010. A Detailed Study of Sulfonate Ester Formation and Solvolysis Reaction Rates and Application toward Establishing Sulfonate Ester Control in Pharmaceutical Manufacturing Processes Org. Process Res. Dev.14 (4), 999–1007. https://pubs.acs.org/doi/abs/10.1021/op900301n?src=recsys&journalCode=oprdfk Accessed 21st June 2018 23

Gerber C, Toelle H-G. 2009. What happened: The chemistry side of the incident with EMS contamination in Viracept tablets. Toxicol Lett. 190 (3) 248-253. https://www.sciencedirect.com/science/article/pii/S0378427409001167?via%3Dihub Accessed 21st June 2018 24

Cimarosti Z, Bravo F, et al. 2010. Application of Quality by Design Principles to Support Development of a Control Strategy for the Control of Genotoxic Impurities in the Manufacturing Process of a Drug Substance. Org. Process Res. Dev 14 (4), 993–998. https://pubs.acs.org/doi/abs/10.1021/op900242x Accessed 21st June 2018 25

Methanesulfonic acid. Product Regulatory Datasheet. https://www.sigmaaldrich.com/content/dam/sigmaaldrich/docs/Sigma-Aldrich/General_Information/1/11317eqpd.pdf Accessed 22nd June 2018 26

Zhou J, et al. 2017. Determination of methyl methanesulfonate and ethyl methanesulfonate in methanesulfonic acid by derivatization followed by high‐performance liquid chromatography with ultraviolet detection. J Sep Sci. 40 (17) 3414-3421. https://onlinelibrary.wiley.com/doi/pdf/10.1002/jssc.201700543 Accessed 22nd June 2018 27

Elder D et al. 2012. An Approach to Control Strategies for Sulfonate Ester Formation in Pharmaceutical Manufacturing Based on Recent Scientific Understanding. Org. Process Res. Dev 16 (11) 1707–1710. https://pubs.acs.org/doi/abs/10.1021/op300216x. Accessed 22nd June 2018 28

29

Cimarosti, Z. Personal communication

Elder DP, Snodin DJ. 2009. Drug substances presented as sulfonic acid salts: overview of utility, safety and regulation. J. Pharm. Pharmacol.(61), 269-278. https://onlinelibrary.wiley.com/doi/epdf/10.1211/jpp.61.03.0001. Accessed 9th July 2018. 30

European Medicines Agency. 24th January 2008. Request to assess the risk of occurrence of contamination with mesilate esters and related compounds in pharmaceuticals. EMEA/44714/2008. http://www.ema.europa.eu/docs/en_GB/document_library/Other/2009/11/WC500015375.pdf. Accessed 9th July 2018. 31


32

Page 33 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Drug Regulatory Affairs (DRA) Newsletter Issue 3/2008. Contamination with mesilate esters and related compounds. https://www.apvmainz.de/fileadmin/dateiablage/Dokumente/Newsletter/Newsletter_DRA_3_2008.pdf. Accessed 8th July 2018. 32

Nassar MN et al. 2009. Ethyl methanesulphonate in a parenteral formulation of BMS-214662 mesylate, a selective farnesyltransferase inhibitor: Formation and rate of hydrolysis. Pharm Dev Technol, 14(6):672-677. https://www.tandfonline.com/doi/abs/10.3109/10837450902980262?journalCode=iphd20. Accessed 8th July 2018. 33

Ubben J et al. Alkyl mesylate formation using pharmaceutically relevant conditions. Eastern Analytical Symposium 2011, Abstract 68 in 2012 Abstract Book. 34

Baghernejad, B. 2011. Application of p-toluenesulfonic Acid (PTSA) in Organic Synthesis. Curr. Org. Chem.15(17) 3091-3097. http://www.ingentaconnect.com/content/ben/coc/2011/00000015/00000017/art00014?crawler=true. Accessed 8th July 2018 35

Kulkarni P. 2015. Methanesulphonic acid is green catalyst in organic synthesis. Orient J Chem 2015;31(1) 447-451. http://www.orientjchem.org/vol31no1/methane-sulphonic-acid-is-green-catalyst-in-organic-synthesis/. Accessed 8th July 2018 36

Mahindaratne MPD & Wimalasena K. 1998. Detailed Characterization of p-Toluenesulfonic Acid Monohydrate as a Convenient, Recoverable, Safe, and Selective Catalyst for Alkylation of the Aromatic Nucleus. J. Org. Chem.63 (9), 2858–2866. https://pubs.acs.org/doi/abs/10.1021/jo971832r?journalCode=joceah&. Accessed 8th July 2018 37

Stowers S et al. 2017. Analytical Methodology for Characterization of Reactive Starting Materials and Intermediates Commonly Used in the Synthesis of Small Molecule Pharmaceuticals. Am Pharmaceut Rev. August 15. https://www.americanpharmaceuticalreview.com/Featured-Articles/341315-AnalyticalMethodology-for-Characterization-of-Reactive-Starting-Materials-and-Intermediates-Commonly-Used-in-theSynthesis-of-Small-Molecule-Pharmaceuticals/ . Accessed 8th July 2018 38

Swissmedic circular dated 23 October 2007: Invitation to carry out a risk assessment for the presence of mesylate ester type impurities and related compounds in pharmaceutical preparations 39

Yang Q et al. 2009. Controlling the Genotoxins Ethyl Chloride and Methyl Chloride Formed During the Preparation of Amine Hydrochloride Salts from Solutions of Ethanol and Methanol. Org. Process Res. Dev., 13 (4), 786–791. https://pubs.acs.org/doi/abs/10.1021/op9000737. Accessed 8th July 2018 40

Sarat M et al. 2010. Low-level Determination of Residual Methyl Methane Sulfonate and Ethyl Methane Sulfonate in Pharmaceuticals by Gas Chromatography with Mass Spectrometry. E-Journal of Chemistry. 7(2), 629-635 41

Alden P, Jones M. Analysis of Benzenesulfonic Acid and p-Toluenesufonic Acid Esters in Genotox Monitoring using UPLC/UV-MS. Waters Corporation, Milford, MA, U.S. http://www.waters.com/webassets/cms/library/docs/720003412en.pdf. Accessed 10th July 2018 42

Hase P, et al. Analysis of Genotoxic Impurities like Ethane Sulfonic acid esters in pharmaceutical substance by GC-MS/MS. Shimaszu ASMS 2017. https://solutions.shimadzu.co.jp/an/n/en/gcms/ipo217128.pdf?return=https://www.shimadzu.com/an/literature/g cms/ipo217128.html. Accessed 10th July 2018 43

Khile AS et al. 2017. Determination of Genotoxic Alkyl Paratoluene Sulfonates in Brinzolamide Using RPLC. Der Pharma Chemica, 9(6):101-103. https://www.derpharmachemica.com/pharma-chemica/determinationof-genotoxic-alkyl-paratoluene-sulfonates-in-brinzolamide-using-rplc.pdf. Accessed 10th July 2018 44

45

Kakasaheb NA et al. 2015. Genotoxic impurity method development and validation by GCMS


33


Page 34 of 36

for the analysis of methyl methane sulfonate (MMS) in Zidovudine drug substance. Anal Chem Ind J. 15(5), 184-189. http://www.tsijournals.com/articles/genotoxic-impurity-method-development-and-validation-by-gcmsfor-the-analysis-of-methyl-methane-sulfonate-mms-in-zidovud.pdf. Accessed 10th July 2018 Zhang C et al. 2016. Determination of sulfonate ester genotoxic impurities in imatinib mesylate by gas chromatography with mass spectrometry. J Sep Sci. 39 (18): 3558-63. https://www.readbyqxmd.com/read/27461842/determination-of-sulfonate-ester-genotoxic-impurities-inimatinib-mesylate-by-gas-chromatography-with-mass-spectrometry. Accessed 10th July 2018 46

47

Wollein U, private communication, 2017

Wollein U, Schramek N. 2012. Simultaneous determination of alkyl mesilates and alkyl besilates in finished drug products by direct injection GC/MS. Eur J Pharm Sci. 45(1-2):201-204. https://www.sciencedirect.com/science/article/pii/S0928098711004131?via%3Dihub. Accessed 10th July 2018 48

Wollein, U., Schramek, N., 2011. Bestimmung von Alkylsulfonsäureestern in Mesilat- bzw. Besilatsalzhaltigen Arzneimitteln mittels GC/MS bzw. LC/MS. Pharmazeutische Industrie 73, 574–580. 49

Decomposition at the Melting Point, Dihydroergotamine Mesylate. Mettler Toledo. https://www.mt.com/gb/en/home/supportive_content/matchar_apps/MatChar_HB801.html. Accessed 10th July 2018 50

Ewing MB, Sanchez Ochoa JC. 2004. Vapor Pressures of Acetonitrile Determined by Comparative Ebulliometry. J. Chem. Eng. Data, 49 (3), 486–491. https://pubs.acs.org/doi/abs/10.1021/je034144f?journalCode=jceaax. Accessed 10th July 2018 51

Britt PF. 2002. Pyrolysis and Combustion of Acetonitrile. Report ORNL/TM-2002/113, Oak Ridge National Laboratory. https://info.ornl.gov/sites/publications/Files/Pub57226.pdf. Accessed 10th July 2018 52

53

Pharmeuropa, 26.2, 2014, p1

Assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals to limit potential carcinogenic risk. ICH Harmonised Guideline M7 (R1). http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Multidisciplinary/M7/M7_R1_Adden dum_Step_4_31Mar2017.pdf. Accessed 10th July 2018 54

55

Sega GA. 1984. A review of the genetic effects of ethyl methanesulfonate. Mutat Res. 134(2-3):113-42.

Ethyl trifluoroacetate. European Chemicals Agency Registration Dossier. https://echa.europa.eu/registrationdossier/-/registered-dossier/10724/7/7/2. Accessed 4th September 2018 56

Methyl methanesulfonate. CPDB. https://toxnet.nlm.nih.gov/cpdb/chempages/METHYL%20METHANESULFONATE.html. Accessed 10th July 2018 57

Hendriks G et al.2016. The Extended ToxTracker Assay Discriminates Between Induction of DNA Damage, Oxidative Stress, and Protein Misfolding. Toxicol Sci. 150(1),190–203. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5009621/pdf/kfv323.pdf. Accessed 10th July 2018 58

Müller L et al. 2009. Ethyl methanesulfonate toxicity in Viracept--a comprehensive human risk assessment based on threshold data for genotoxicity. Toxicol Lett. 12;190(3):317-29 59

Müller L, Gocke E.2009. Considerations regarding a permitted daily exposure calculation for ethyl methanesulfonate. Toxicol Lett. 12;190(3):330-2. https://www.sciencedirect.com/science/article/pii/S0378427409001623?via%3Dihub Accessed 2nd October 2018 60

Bercu JP et al. 2018. Potential impurities in drug substances: Compound-specific toxicology limits for 20 synthetic reagents and by-products, and a class-specific toxicology limit for alkyl bromides. Regul Toxicol 61


34

Page 35 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Pharmacol. 94:172-182. https://www.sciencedirect.com/science/article/pii/S0273230018300461. Accessed 2nd October 2018 Coffing SL et al. 2015. Evaluation of the in vivo mutagenicity of isopropyl methanesulfonate in acute and 28day studies. Environ Mol Mutagen. 56(3):322-32. https://onlinelibrary.wiley.com/doi/10.1002/em.21910. Accessed 2nd October 2018 62

Hakura A et al. 1984. Studies on chemical carcinogens and mutagens. XXVI. Chemical properties and mutagenicity of alkyl alkanesulfonates on Salmonella typhimurium TA100. Chem Pharm Bull (Tokyo). 32(9):3626-35. https://www.jstage.jst.go.jp/article/cpb1958/32/9/32_9_3626/_pdf/-char/en Accessed 2nd October 2018 63

Mesylate Ester Type Impurities Contained in Medicinal Products, Swissmedic Department for Control of the Medicinal Products Market, 23rd October 2007. 64

European Medicines Agency, Request to Assess the Risk of Occurrence of Contamination with Mesilate Esters and Other Related Compounds in Pharmaceuticals, EMEA/44714/2008, London, 24th January 2008. https://www.ema.europa.eu/documents/other/request-assess-risk-occurrence-contamination-mesilate-estersrelated-compounds-pharmaceuticals_en.pdf. Accessed 2nd October 2018 65

EDQM. 2017. Groups of experts and working parties. https://www.edqm.eu/sites/default/files/list_of_groups_of_experts_and_working_parties.pdf. Accessed 2nd October 2018 66

European Pharmacopoeia Commission. 2016. Terms of reference and profile for members of groups of experts and working parties. http://www.ogyei.gov.hu/dynamic/terms%20of%20referencesg1641comr1.pdf. Accessed 2nd October 2018 67

Midgley JM. 2012. Background note and comment on the article by Snodin et al. on Ph. Eur. Monographs on mesylate salts. Pharmeuropa Readers’ Tribune (April).11-12 68

EDQM Press Release, 25 February 2016, Strasbourg, France. Potential presence of mutagenic alkyl sulfonates in active substances. https://www.edqm.eu/sites/default/files/press_release_on_mutagenic_impurities_february_2016.pdf. Accessed 2nd October 2018 69

Committee for Medicinal Products for Veterinary Use (CVMP). 8 December 2015. EMA/CVMP/829802/2015. Minutes of the 4-6 November 2015 meeting. https://www.ema.europa.eu/documents/minutes/minutes-cvmp-meeting-4-6-november-2015_en.pdf Accessed 2nd October 2018 70

EDQM Press Release. 155th Session of the European Pharmacopoeia Commission (21-22 June 2016). 30 June 2016, Strasbourg, France. https://www.edqm.eu/sites/default/files/press_release_pheur_commission_155th_session_june_2016.pdf. Accessed 2nd October 2018 71

72

BP secretariat, private oral communication, January 2017.

EDQM Certification Database: https://extranet.edqm.eu/publications/recherches_CEP.shtml. Accessed 15th February, 2019 73

74

Gover, S. Quality Working Party Secretariat, 18th June 2018

Guo T, et al. 2014. Rapid and simultaneous determination of sulfonate ester genotoxic impurities in drug substance by liquid chromatography coupled to tandem mass spectrometry: comparison of different ionization modes. J Chromatogr A. 1355.73-9. 75


35


Page 36 of 36

Olsen, BA et al. 2018. Impurity investigations by phases of drug and product development. Trends Anal Chem. 101. 17-23. http://iranarze.ir/wp-content/uploads/2017/12/E5216-IranArze.pdf. Accessed 15 October 2018 76

Evaluation and Licensing Division, Pharmaceutical Safety and Environmental Health Bureau Ministry of Health, Labour and Welfare. February 24, 2016. Talfinlar Capsules. https://www.pmda.go.jp/files/000219617.pdf Accessed 15 October 2018. 77

Teasdale A.2015. Sulfonate Esters: How Real is the Risk. Summary of key findings from PQRI on the reaction between sulfonic acids and alcohols: http://pqri.org/wpcontent/uploads/2015/08/pdf/Sulfonate_Esters_Webinar.pdf . Accessed 29th October 2018 78

Hadlington S. 2015. Regulators have drug firms chasing phantom impurity, Chemistry World. 16 June. https://www.chemistryworld.com/news/regulators-have-drug-firms-chasing-phantom-impurity/8648.article. Accessed 26th November 2018 79


36

Elusive impurities â evidence versus hypothesis. Technical and

Recommend Documents

Elusive impurities â evidence versus hypothesis. Technical and