API Assessment Common Deficiencies Isabel Ortega and Antony

API Assessment: Common Deficiencies Isabel Ortega and Antony Fake WHO Prequalification of Medicines Programme Assessors training, Copenhagen January 2012 1

Overview 1. API information. 2. What are we looking for. 3. A summary of deficiencies found in APIMFs assessed within PQP. 4. Common deficiencies. 2 API Common Deficiencies January 2012

API Information There are different options that can be used by applicants to submit API information. • Use of a prequalified API • Use of an EDQM CEP • Use of an APIMF • Provision of full API information in the product dossier. 3 API Common Deficiencies January 2012

API Information • Regardless of the option chosen the same information is required. • The API information assessed within PQP is compiled in the Common Technical Document format (CTD). • This is regardless of whether the API is pharmacopoeial or not. 4 API Common Deficiencies January 2012

API Information 5 API Common Deficiencies January 2012

Overview 1. API information. 2. What are we looking for. 3. A summary of deficiencies found in APIMFs assessed within PQP. 4. Common deficiencies. 6 API Common Deficiencies January 2012

Is the starting point of the manufacturing process acceptable? 7 API Common Deficiencies January 2012

Traditional manufacturing Raw materials Final API Traditionally, the API manufacturer manufactured the API from simple raw materials at their own facility. This meant it was safe to assume that the molecule used in the first step at the API manufacturer's factory was an appropriate API-SM. 8 API Common Deficiencies January 2012

Contemporary manufacturing Intermediate API-SM Raw Material Final API Increasingly, intermediates late in the synthesis are being purchased. This means the API-SM can not be assumed to be the molecule used in the first step at the API manufacturer's factory. 9 API Common Deficiencies January 2012

API Starting material Choice of API-SM ASSESSORS Simpler molecules Final API INDUSTRY 10 API Common Deficiencies January 2012

API Starting material API manufacturers prefer to have the API-starting material (APISM) defined as late in the synthesis as possible because: • It reflects the manufacturing steps undertaken at their factory under their control. • The API-SM is the point at which GMP applies to manufacture. • This can be financially advantageous when API manufacturers buy reaction intermediates from secondary manufacturers that do not operate under API GMP. 11 API Common Deficiencies January 2012

API Starting material The problem for assessors is: • Information on the preparation of a complex API from one or two steps makes determination of impurities in the API very difficult. • If the SM is complex it is hard to judge the acceptability of the API-SM specifications. • The guarantee of quality API is the result of good manufacture and control throughout all the steps (GMP). Comprehensive testing of the final API does not replace this. 12 API Common Deficiencies January 2012

Are the manufacturing process and process controls adequately described? 13 API Common Deficiencies January 2012

Manufacturing Process A detailed and complete flow diagram is required A detailed narrative description is required. API starting material API intermediate(s) Final API intermediate Crude API Final API 14 API Common Deficiencies January 2012

What are the potential impurities? 15 API Common Deficiencies January 2012

Potential impurities API SM impurities API starting material Reagents Solvents Catalysts Reaction by-products API intermediate Reagents Solvents Catalysts Reaction by-products Crude API Crystallisation solvent Degradation products 16 Final API Common Deficiencies January 2012

Potential impurities • Residue of the SM • Residue of the intermediate • Impurities in the SM • Reagents • Solvents • Catalysts • Reaction by-products • Degradation products 17 API Common Deficiencies January 2012

Can the potential impurities be carried over into the final API? 18 API Common Deficiencies January 2012

Are the potential impurities detectable by the analytical procedure? 19 API Common Deficiencies January 2012

Final API Specifications Control Analytical procedures Reference standards Stability 20 Container/closure system API Common Deficiencies January 2012

Are the API specifications acceptable? 21 API Common Deficiencies January 2012

Can a retest period and recommended storage conditions be established? 22 API Common Deficiencies January 2012

Overview 1. API information. 2. What are we looking for. 3. A summary of deficiencies found in APIMFs assessed within PQP. 4. Common deficiencies. 23 API Common Deficiencies January 2012

Summary of deficiencies • I have started to investigate commonly observed deficiencies with APIMFs assessed by the PQ programme. 24 API Common Deficiencies January 2012

Investigation overview • The initial assessment reports for 90 APIMFs were investigated (assessed between 2007 -2011). • This includes 45 HIV, 24 MA, 19 TB and 2 IN APIMFs. • For more than 70% of the products, an official compendial monograph was available at that time. A total of 2828 deficiencies classified by CTD section 25 API Common Deficiencies January 2012

Deficiencies by CTD section Number of deficiencies 26 API Common Deficiencies January 2012

% of deficiencies by CTD section Common deficiencies 27 API Common Deficiencies January 2012

Overview 1. API information. 2. What are we looking for. 3. A summary of deficiencies found in APIMFs assessed within PQP. 4. Common deficiencies. 28 API Common Deficiencies January 2012

Top 9 deficiencies (by CTD section) 1 - Manufacturing process (3. 2. S. 2. 2) 2 - Starting material quality (3. 2. S. 2. 3) 3 - Quality of all other materials (3. 2. S. 2. 3) 4 - Control of intermediates (3. 2. S. 2. 4) 5 - Polymorphism (3. 2. S. 3. 1) 6 - Impurities (3. 2. S. 3. 2) 7 - Specifications (3. 2. S. 4. 1 and 3. 2. S. 4. 5) 8 - Validation of analytical procedures 9 - Stability deficiencies (3. 2. S. 7. 1, 3. 2. S. 7. 2 and 3. 2. S. 7. 3) 29 API Common Deficiencies January 2012

1 - Manufacturing process (3. 2. S. 2. 2) 30 API Common Deficiencies January 2012

Information on the API preparation is not sufficiently detailed A detailed flow diagram of synthesis indicating chemical structures, molecular weights, solvents, reagents should be presented. When an APIMF is used this information should be provided in both the open and closed sections. A detailed narrative of each synthetic step should be presented, including: the types and quantities of reagents and solvents; reaction conditions; critical steps, in-process controls; and yields. • The description of the API manufacturing process represents the applicant’s commitment for the manufacture of the API. 31 API Common Deficiencies January 2012

The proposed API SM is too complex • It is often observed that the proposed API SM is merely the molecule that is used in the first manufacturing step at the API manufacture site. • This is not a sufficient justification. • The API SM should be a relatively simple molecule. In general it should be one or more synthetic steps from the final intermediate. 32 API Common Deficiencies January 2012

The proposed API SM is too complex There are exceptions to this rule. For instance • Artemisinin is accepted as a API SM despite it’s complexity because it is derived from a plant. • Aminobutanol is accepted as the API SM for ethambutol, despite being a one step synthesis, because both the API SM and AP are simple molecules. 33 API Common Deficiencies January 2012

The proposed API SM is too complex The point at which the API SM is introduced is the starting point of the application of GMP requirements. The API SM itself needs to be proposed and justified by the manufacturer and accepted as such by assessors. • In these cases, the applicant is requested to redefine the API-SM as a synthetic precursor one or more synthetic steps prior to the final key intermediate. • The manufacturer of the key intermediate should be included as an API intermediate manufacturing site and should be GMP compliant. 34 API Common Deficiencies January 2012

The information on reprocessing is not sufficiently detailed Steps where reprocessing may occur should be identified and justified. • Reprocessing is the repetition of a step of manufacture for a batch of API or intermediate that is considered substandard. • The frequency of reprocessing should be stated because excessive reprocessing indicates the manufacturing process is not under control. This is not in line with GMP. 35 API Common Deficiencies January 2012

The maximum batch size is not clearly stated The description of the process should indicate the scale of manufacture and the range for which the considered process may be used. It may be helpful to indicate the yield or yield range produced at each stage. • Often the intended scale (or range) of manufacture is not categorically stated. • The applicant should be able to provide evidence of manufacture at this scale (at a minimum batch data). 36 API Common Deficiencies January 2012

The maximum batch size is not clearly stated • We need to know what is the maximum batch size for which the applicant has experience with the defined method. • Also, if blending occurs it has to be confirmed that each batch incorporated into the blend is individually tested and found to meet specifications set for the final API prior to blending. 37 API Common Deficiencies January 2012

Solvents used are not clearly identified in the process All solvents used in the manufacture (including purification and/or crystallization step(s)) should be clearly identified. Quantities of solvents reflecting the representative batch scale for commercial manufacture should be stated. Solvents used in the final steps should be of high purity. • Specifications should be provided for all solvents. • The carry-over of all solvents into the final API needs to be assessed. 38 API Common Deficiencies January 2012

The information on recovery of solvents is not sufficiently detailed The recovery of materials, if any, should be described including the step they are taken and re-introduced into the process. Use of recovered solvents in the final steps of purification and/or crystallization is not recommended. • Often no comment is made as to whether the recovery of materials occurs or not. • The recovery of solvents is very common. • The use of multiple solvents and reagents for different purposes within a process leads to the formation of solvent mixtures. • The potential for cross contamination needs to be assessed (the impurity profile of the API should remain unaffected). 39 API Common Deficiencies January 2012

2 - Starting material quality (3. 2. S. 2. 3) 40 API Common Deficiencies January 2012

The information on API SM preparation is not sufficiently detailed or is missing. • Information on the API SM preparation is needed to determine the appropriateness of the specifications. • Detailed information is usually not required, because the SM should be relatively simple. • This information is often supplied in form of a flow-chart (synthetic transformation, reagents, and solvents). • If there are two or more suppliers of the API SM, information on the preparation of the SM from each supplier should be provided. 41 API Common Deficiencies January 2012

Information on the API SM manufacturer is missing For each SM, the name and manufacturing site address of the manufacturer should be indicated. If there are several manufacturers, it should be clarified whether the SM obtained from different sources is prepared by the same route of synthesis or if different routes are used. 42 API Common Deficiencies January 2012

The API SM specification is not satisfactory The SM specifications should include among others an identity test and tests and limits for assay and for specified, unspecified and total impurities. • The API SM impurity limits should be justified by the demonstrated ability to manufacture API from batches of SM with similar impurity levels. 43 API Common Deficiencies January 2012

The API SM specification is not satisfactory • Each API SM supplier does not have to have the same specifications but a single SM specifications should be defined by the API manufacturer and applied to all sources of SM. • The carry-over of impurities into the final API should be considered and discussed. The same applies for solvents, reagents, and catalysts used in the SM preparation, as needed. 44 API Common Deficiencies January 2012

3 - Quality of all other materials (3. 2. S. 2. 3) 45 API Common Deficiencies January 2012

The specifications for raw materials, solvents, reagents, catalysts are unsatisfactory. Information of the quality and control of these materials should be provided. Information demonstrating that materials meet standards appropriate for their intended use should be provided. • The specifications for all materials used should be provided. • They should include at least an identification test and assay limit. • Limits for benzene in toluene, methanol, ethanol and acetone are commonly requested as this can be an impurity. 46 API Common Deficiencies January 2012

The specifications for recovered solvents are not satisfactory. Recovery operations should be adequately controlled such that impurity levels do not increase over time. Use of recovered solvents in the final steps of purification and/or crystallization is not recommended. • Their specifications should include appropriate limits for assay, impurities and non-volatile matter (0. 1%). • Any lessening of quality of the recovered solvent limits needs to be justified. • It should be demonstrated that recovered solvents with impurity levels approaching the limits proposed affords API of an acceptable quality. 47 API Common Deficiencies January 2012

4 - Control of intermediates (3. 2. S. 2. 4) 48 API Common Deficiencies January 2012

The specifications for isolated intermediates are not satisfactory. Information on the quality and control of intermediates isolated during the process should be provided. The specifications should include tests and acceptance criteria for identity, purity (individual and total impurity limits) and assay, where applicable. • The limits chosen should be justified and demonstrated to be acceptable (in terms of carry-over into the final API). • Often the proposed limit for total impurities is too high. • Control of certain impurities in the intermediate may eliminate the need to control this impurity in the final API. 49 API Common Deficiencies January 2012

5 - Polymorphism (3. 2. S. 3. 1) 50 API Common Deficiencies January 2012

The information on API polymorphism is not sufficiently detailed Information on the potential forming polymorphs should be included. API manufacturers are expected to have adequate knowledge about the polymorphism of the APIs produced. 51 API Common Deficiencies January 2012

The information on API polymorphism is not sufficiently detailed Polymorphism known Polymorphism not known Highly soluble The polymorphic form The crystalline form produced API produced should be assigned. should be determined. Non-highly soluble API 52 The polymorphic form produced should be assigned. Stability of the form should be determined API Common Deficiencies January 2012 The crystalline form produced should be assigned. Crystallinity under different crystallisation condition should be investigated. Stability of the form should be determined

The information on API polymorphism is not sufficiently detailed • Polymorphic forms may possess different chemical and physical properties, that can have a direct impact FPP properties, such as stability, dissolution and bioavailability. • Unexpected appearance or disappearance of a polymorphic form may lead to serious pharmaceutical consequences. • We need to know what polymorph is consistently produced and if it is stable. 53 API Common Deficiencies January 2012

6 - Impurities (3. 2. S. 3. 2) 54 API Common Deficiencies January 2012

Insufficient discussion on potential impurities Regardless of whether a pharmacopoeial standard is claimed, a discussion should be provided of the potential and actual impurities arising from the synthesis, manufacture or degradation of the API. The discussion of pharmacopoeial APIs should not be limited to the impurities specified in the API monograph 55 API Common Deficiencies January 2012

Insufficient discussion on potential impurities • Monographs are developed based upon how the API was prepared historically. A particular manufacturer's manufacturing method may lead to unexpected impurities, due to a different route of synthesis, different reagents, etc. • When the absence of a certain impurity, reagent, solvent or catalyst has not been adequately demonstrated, a test and limit should be included in the API specification. 56 API Common Deficiencies January 2012

The API has not been investigated for the presence of potential impurities cited in the monograph (or draft) If there are identified impurities specified in an official compendial monograph that are not controlled by the proposed routine in-house analytical procedure, a justification for their exclusion from routine analyses should be provided. 57 API Common Deficiencies January 2012

The API has not been investigated for the presence of potential impurities cited in the monograph (or draft) • Some of the Ph. listed impurities can be excluded by the applicant with a justification that they are not potential impurities from the proposed route of synthesis. • If justifications are not acceptable and the routine in-house method is not capable of detecting the pharmacopoeial listed impurities, their absence can only be shown by testing several recent batches with the pharmacopoeial method. 58 API Common Deficiencies January 2012

A specific discussion on potential genotoxic impurities is missing The absence of known established highly toxic impurities (genotoxic) used in the process or formed as by-product should be discussed and suitable limits should be proposed. 59 • Certain chemical structures "alert structures" are considered to be genotoxic, except if also present in the API. • The most common situation that arises is the use of the sulphonic acid reagents that in the presence of alcohols like methanol or ethanol. They can form sulphonate esters (genotoxic). API Common Deficiencies January 2012

7 - Specifications (3. 2. S. 4. 1 and 3. 2. S. 4. 5) 60 API Common Deficiencies January 2012

The format of the API specifications is unsatisfactory Copies of the API specifications, dated and signed by authorized personnel (e. g. the person in charge of the quality control or quality assurance department) should be provided. • It is important that the manufacturer provides a signed, dated and version-numbered specification. This is to ensure traceability of testing requirements in case of GMP audits and future variations. • The API specifications should also include reference to analytical procedures and whether they are compendial or not (i. e. pharmacopoeial reference or in-house reference). 61 API Common Deficiencies January 2012

Absence of certain tests and limits in the API specification • It should include a limit and test for heavy metals. Heavy metals can also come from the process (e. g. , leaching from pipes and other equipment) or occur naturally (e. g. , from naturally derived plant or mineral sources). • The specification for a chiral API should include enantioselective tests for identity and purity (optical rotation and enantiomeric purity). • When the absence of certain residual solvent, reagent or catalyst has not been shown, a limit and test needs to be included in the API specification. 62 API Common Deficiencies January 2012

The proposed limits for impurities are unsatisfactory Specified impurities (=known impurities qualified or ≤ ICH qualification threshold; can be identified or unidentified i. e RRT but no structure) Any unspecified impurities (=unknown imp. ≤ ICH identification threshold) Total impurities (=sum of impurities > the ICH reporting threshold) 63 API Common Deficiencies January 2012

The proposed limits for impurities are unsatisfactory • When there is an official compendial monograph, the limits for specified impurities should comply the ones of the monograph and regardless of the related substances requirements a test for any unknown impurity and total impurities should be included. • ICH limits for unknown impurities are generally required (with exceptions), rather than the general limits of the monograph. 64 API Common Deficiencies January 2012

8 - Validation of analytical procedures 65 API Common Deficiencies January 2012

The validation of analytical procedures is insufficient or missing The objective of the analytical validation information is to demonstrate that the analytical procedures are suitable for their intended purpose • All non-pharmacopoeial methods must be fully validated, particularly assay, related substances and residual solvents methods, as outlined in ICH Q 2. • These methods should be able to separate/detect the potential impurities/solvents of the manufacturing process. • The pharmacopoeial methods should also be shown suitable for determination of impurities related to the manufacturer's specific route of synthesis and not covered by the monograph. 66 API Common Deficiencies January 2012

Cross-validation data are not submitted when in -house methods are used for controlling pharmacopoeial APIs If an officially recognized compendial standard is claimed an in-house method is used in lieu of the compendial method (e. g. for assay or for specified impurities), equivalency of the in-house and compendial methods should be demonstrated. 67 API Common Deficiencies January 2012

Cross-validation data are not submitted when in -house methods are used for controlling pharmacopoeial APIs By performing duplicate analyses of one sample by both methods or by spiking the API with impurities at concentrations equivalent to their specification limits. • This is the way to check that the Ph. listed impurities can be detected with the in-house methods and that the inhouse is superior or at least equivalent. 68 API Common Deficiencies January 2012

9 - Stability deficiencies (3. 2. S. 7. 1, 3. 2. S. 7. 2 and 3. 2. S. 7. 3) 69 API Common Deficiencies January 2012

Methods for assay or impurities methods are not shown stability-indicating Stress testing of the API can help identify the likely degradation products, which can in turn establish the degradation pathways and the intrinsic stability of the molecule and validate the stability indicating power of the analytical procedures used. • In-house analytical procedures for assay and impurities should be fully validated and should be shown stabilityindicating. • They should be able to detect significant changes in the quality of the product (i. e. increase of degradation products). 70 API Common Deficiencies January 2012

The primary stability data have not been conducted at 30ºC The required long-term storage conditions for APIs in the PQP is either 30ºC/65%RH or 30ºC/75%RH. Studies covering the proposed re-test period at these conditions will provide better assurance of the stability of APIs at the conditions of the supply chain corresponding to the WHO and PQP environments. • This is required unless it has been demonstrated that the API is inherently unstable at 30ºC/65%RH or 30ºC/75%RH. • If there is no data available at 30ºC then the API manufacturer is requested to commit to initiating long-term stability trials at 30ºC and to change the recommended storage conditions to stored below 30ºC. 71 API Common Deficiencies January 2012

The stability data provided do not support the proposed retest period A retest period should be derived from the stability information and should be displayed on the container label. Limited extrapolation of the real-time data from the long-term storage condition can be undertaken to extend the retest period. • There are minimum data required at the time of submission (6 months in accelerated and long-term conditions). • Most of the times the stability data included in the first submission do not support the proposed retest period and updated stability data are needed. 72 API Common Deficiencies January 2012

A storage condition is not proposed A storage statement should be established for display on the label based on the stability evaluation of the API. The WHO stability guideline includes a number of recommended storage statements to be used, when supported by stability studies. • The prequalification programme recognises the following conditions: store below 30ºC, store below 25ºC, and store between 2 - 8ºC. • In addition, the ancillary statements protect from humidity or protect from light should also be considered. 73 API Common Deficiencies January 2012

Questions? 74 API Common Deficiencies January 2012