What s New with Impurities in Pharmaceuticals?

What s New with Impurities in Pharmaceuticals? Southern California Pharmaceutical Discussion Group January 15, 2015 Bernard A. Olsen, Ph.D Olsen Pharmaceutical Consulting, LLC bolsen@comcast.net

ICH Q3 Impurities Q3A Drug Substances 1995 (step 4), R1 (2002), R2 (2006) Q3B Drug Products 1996 (step 4), R1 (2003), R2 (2006) Q3C Residual Solvents 1997, R1-5 (2002, 2005, 2009, 2011) Most ICH guidelines on impurities in drug substances and drug products are >15 years old What else is there to say? 2

Filling the Gaps M7 Genotoxic Impurities Step 4 (June 2014) - changes from EMA and FDA guidance Q3D Elemental Impurities Step 4 (Dec. 2014) - USP <232>, <233> Other gaps? Revisions needed? 3

ICH M7 Genotoxic Impurities Filling the ICH Q3 A/B gap for impurities that are expected to be unusually potent, producing toxic or pharmacological effects at a level not more than ( ) the identification threshold. - Identification of "unusually potent" impurities not described - No threshold of concern given EMA* guideline and FDA** draft guidance: Threshold of Toxicological Concern (TTC), 1.5 mg/day *http://www.ema.europa.eu/docs/en_gb/document_library/scientific_guideline/2009/09/wc500002 903.pdf **http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/uc m079235.pdf 4

Assessing Impurities ICH M7 All impurities (actual and potential), where the structures are known, should be evaluated for mutagenic potential.. 5

Classifying impurities PhRMA paper recommendation*, adopted in M7 All identified or predicted impurities should be classified into one of five classes: Class 1. Known to be genotoxic and carcinogenic Class 2. Known to be genotoxic but with unknown carcinogenic potential Class 3. With a unique alerting structure and of unknown genotoxic potential Class 4. With an alerting structure related to the parent API Class 5. With no structural alert GTI? Ordinary ICH impurity *Müller et al., Reg. Tox. Pharmacol. 44, 198-211 (2006); 6

Alerting Structures examples Müller et al. 7

Assessing Impurities ICH M7, cont. Is an impurity potentially genotoxic? Search databases and literature for carcinogenicity and bacterial mutagenicity data in order to classify impurity as Class 1, 2, or 5 When data are not available: Use Structure-Activity Relationships (SAR) that focus on bacterial mutagenicity predictions. This could lead to a classification into Class 3, 4, or 5. 8

(Q)SAR/in silico assessments Two (Q)SAR prediction methodologies that complement each other should be applied. - Expert rule-based - Statistical-based - follow Organisation for Economic Co-operation and Development (OECD) validation principles - review with expert knowledge Absence of structural alerts from 2 Q(SAR) predictions = normal impurity 9

Toxicity Prediction Software Freely available software Caesar models Lazar OncoLogic PASS Commercially available software ACD/Tox Suite BioEpisteme HazardExpert MDL QSAR MultiCASE TOPKAT q-tox EPI Suite OECD QSAR Application Toolbox Toxtree T.E.S.T ADMET Predictor DEREK Leadscope Molcode Toolbox OASIS TIMES ToxAlert CSGenoTox Review of Software Tools for Toxicity Prediction, M. F. Gatnik and A. Worth, European Commission Joint Research Centre, Institute for Health and Consumer Protection https://eurl-ecvam.jrc.ec.europa.eu/laboratories-research/predictive_toxicology/doc/eur_24489_en.pdf Industry Survey: Dobo et al., Reg. Tox. Pharmacol. 62 (2012) 449 455 10

Ames testing Negative result classifies compound as normal ICH impurity and overrides a positive in silico prediction for genotoxicity Test on the isolated impurity preferred vs. impurity in drug substance; 250 μg/plate needed for compound of interest GLP studies expected but test article characterization may not comply fully; exceptions also allowed for compounds difficult to prepare or isolate 11

ICH M7 - TTC-based acceptable limits Acceptable Daily Intakes* for an Individual Impurity, mg/day Clinical trials or marketed product Single Dose < 14 days 1 mo. 3 mo. 6 mo. 12 mo. >1 10 years >10 years to lifetime M7 ** ** 120 20 20 20 10 1.5 EMA 120 60 60 30 10 5 1.5 (marketed) 1.5 *Compound-specific risk assessments to derive acceptable intakes should be applied instead of the TTC-based acceptable intakes where sufficient carcinogenicity data exist. **Clinical trials of up to 14 days class 3 impurities can be treated as normal impurities 12

Mitigating factors in application of TTC Indication life-saving therapy Exposure from other sources, e.g., foods or endogenous metabolism (e.g., formaldehyde) Reduced life expectancy Late onset but chronic disease Limited therapeutic alternatives 13

Strategies to Address GTIs Modify synthesis to remove compounds of concern or move them as early in the synthesis as possible Purification provide rationale and/or data to demonstrate that GTI has negligible risk of being in drug substance Specification commit to analytical testing and acceptance limit at intermediate (higher levels?) or drug substance (staged TTC levels) Degradation product GTI packaging and storage to prevent formation, implement specification through shelf-life 14

ICH M7 Control Options for Process Impurities Option 1 Specification for impurity in drug substance with acceptable limit. Option 2 Specification at precursor with drug substance acceptable limit. Option 3 Specification at precursor with higher limit. Fate and purge data and associated process controls needed that assure the level in the drug substance is below the acceptable limit. Probability of regulatory acceptance Option 4 No routine testing. Compelling fate and purge knowledge with sufficient confidence that the level of the impurity in the drug substance will be below the acceptable limit. 15

Example GSK, pazopanib HCl DMS=dimethylsulfate includes monohydrate isolation before salt formation? III and VII designated as API starting materials D.Q. Liu, T.K. Chen, M.A. McGuire, A.S. Kord, J. Pharm. Biomed. Anal. 50 (2009) 144-150 16

Strategy Develop TTC-level LC-MS method for GTIs Analyze materials throughout process to show lack of carry-through (up to 79 batches of API) Perform impurity rejection studies to show process capability Establish tests for GTIs at levels higher than TTC using LC methods at starting materials or intermediates Note: Pazopanib HCl (Votrient) is a tyrosine kinase inhibitor approved for the treatment of renal cell carcinoma 17

Impurity rejection efficiency - DMS DMS introduced 2 steps back from starting material III 20 batches of III showed <1.7 ppm DMS 79 batches of API showed <1.7 ppm DMS 50,000 ppm DMS spiked in III <1.7 ppm in V >29,000 fold rejection in step 1 followed by 4 subsequent steps 0.1% acceptance criterion set for DMS in starting material III IS ROUTINE TESTING FOR DMS NEEDED? 18

Impurity rejection efficiency compound II 79 batches of API showed <1.7 ppm II 16 batches of III showed <24 ppm II 50,000 ppm in III 670 ppm in V 75 fold rejection 670 ppm in V 23 ppm in VI 29 fold rejection 23 ppm in VI <1.7 ppm in IX >13 fold rejection 0.1% limit set for II in III 19

Higher Upstream limits for VI and VIII limit of NMT 0.1% set for VIII in VII limit of NMT 0.6% set for VI in IX 20

Genotoxic degradation products Identify potential degradation product GTIs during predictive stress testing Does degradation product GTI form with time under normal storage conditions? Can degradation product GTI formation be prevented through formulation design, packaging, or storage conditions? Establish specification if necessary 21

Risk of Producing an Alerting Structure from Drug Degradation* Analysis of over 1100 known degradation products from more than 350 drugs suggests that degradation of drugs may lead to unique structure alerting functional groups in about 5-8% of the degradation products. Roughly 50% or less of these alerting structures can be expected to be Ames positive An average of 8-9 major deg products are observed in stress testing for a typical drug, so most drugs will have zero or one deg product for follow-up as a potential GTI *S.W. Baertschi et al., Stress Testing and Degradation-Derived Genotoxic Impurities: Scientific, Practical and Regulatory Considerations, Conference on Small Molecule Science, August 2, 2011, Chapel Hill, NC 22

ICH M7- Control for degradants Is degradant relevant? check accelerated stability study data (e.g., 40 C/75% relative humidity, 6 months) kinetically equivalent shorter term stability studies at higher temperatures in the proposed commercial package may be used to determine the relevance of the degradation pathway prior to initiating longer term stability studies. S.W. Baertschi et al., Stress testing as a predictive tool for the assessment of potential genotoxic degradants, in Pharmaceutical Stress Testing, 2 nd Ed., S.W. Baertschi, K.M. Alsante, R. A. Reed, 2011, Informa, London. 23

ICH M7 Other Considerations Not applied to products for advanced cancer indications (see ICH S9) Not applied to drug substances that are themselves genotoxic Will be applied to changes in existing authorizations if new or greater levels of previous impurities are present Assess potentially genotoxic impurities which may be present at levels below the Q3 A/B ID thresholds (same as current guidelines) Previous data from similar compounds may be used with justification to discharge risk 24

Elemental Impurities USP <232> Limits USP <233> Procedures replacing <231> Heavy Metals ICH Q3D 25

Disclaimer Information in this presentation related to USP Elemental Impurities is from publically-available sources. The speaker does not represent official USP positions or policy on Elemental Impurities or any other topic. 26

Average % Recoveries USP <231> Heavy Metals <231> has been in use for many years. What s the problem? 120 100 80 60 40 20 0 USP Results ICP-MS Results Pb As Se Sn Sb Cd Pd Pt Ag Bi Mo Ru In Hg Elements Lewen, N. et al., J. Pharm. & Biomed. Anal. 35 (2004) 739-752) 27

USP key issue elemental impurities <232> Elemental Impurities Limits <233> Elemental Impurities Procedures Revisions proposed in PF 40(2) http://www.usp.org/usp-nf/key-issues/elemental-impurities Implementation date when chapters apply to drug product monographs: December 1, 2015 BUT, what about ICH Q3D? January 14, 2015: USP is announcing plans to establish January 1, 2018 as the new date of applicability of General Chapters <232> Elemental Impurities Limits and <2232> Elemental Contaminants in Dietary Supplements. 28

Heavy metals limits USP proposal PDE = permitted daily exposure 29

Limits, cont. 30

Options for compliance Drug product analysis (Q3D option 3) Daily Dose PDE measured value (μg/g) maximum daily dose (g/day) Summation option (add metals present in each component, Q3D option 2b) Daily Dose PDE [Σ M 1(CM WM)] DD M = each ingredient used to manufacture a dosage unit CM = element concentration in component (drug substance or excipient) (μg/g) WM = weight of component in a dosage unit (g/dosage unit) DD = number of units in the maximum daily dose (unit/day) Individual component option (Large volume parenterals only) API and excipients meet limits given in Table 1 for LVP components 31

Testing If, by validated processes and supply-chain control, manufacturers can demonstrate the absence of impurities, then further testing is not needed. 32

USP Updates stay tuned October 2014 expert panel recommendation: Limits be revised to align with the ICH Q3D Step 4 document to the extent possible. Separately, USP is considering potential adjustments to the elemental impurities implementation timeline as specified in General Notices 5.60.30 (Dec.1, 2015) based on developments related to the anticipated ICH Q3D Step 4 document. http://www.usp.org/usp-nf/key-issues/elemental-impurities 33

ICH Q3D Guideline for Elemental Impurities Focus is on risk assessment for occurrence of and limits for elemental impurities Step 4 guideline published December 16, 2014 http://www.ich.org/fileadmin/public_web_site/ich_ Products/Guidelines/Quality/Q3D/Q3D_Step_4.pdf 34

ICH Q3D: Risk-based assessment of need for control of metal impurities 35

ICH Q3D Risk assessment Identify: Identify known and potential sources of elemental impurities that may find their way into the drug product. Evaluate: Evaluate the presence of a particular elemental impurity in the drug product by determining the observed or predicted level of the impurity and comparing with the established PDE. Control: Summarize and document the risk assessment. Identify if controls built into the process are sufficient or identify additional controls to be considered to limit elemental impurities in the drug product. 36

Sources of Elemental Impurities 37

ICH Q3D: Limits for Elemental Impurities 5.0 2.5 3.4 5.0 5.0 5.0 15 15 1.9 15 1.5 1.2 120 12 1.2 600 60 6.0 100 10 1.0 100 10 1.5 100 10 1.5 100 10 1.5 100 10 1.5 100 10 1.5 180 90 7.6 1300 130 13 2.9 Red = USP <232> PDEs Option 1: Assume 10 g/day dose. If all components meet PDE concentration, they may be used in any proportion Option 2a: Use the actual dose to calculate PDE concentration. If all components meet the PDE, they may be used in any proportion. Option 2b: Use the amounts of each component present and data on metals present to set limits for individual components. Option 3: Drug product analysis with limits based on daily dose. 38

Q3D Other Limit Considerations When PDEs are necessary for other routes of administration, the concepts described in the guideline may be used to derive PDEs. - Consider local effects, bioavailability, quality considerations Higher PDEs may be permitted for: - Intermittent dosing; - Short term dosing (i.e., 30 days or less); - Specific indications (e.g., life-threatening, unmet medical needs, rare diseases). 39

ICH Q3D Implementation Application of Q3D to existing products is not expected prior to 36 months after publication of the guideline by ICH. December 2017? Will USP implementation timing be revised to December 2017? January 1, 2018 Q3D implementation plan 21 Oct 2014 - Training materials - FAQ document http://www.ich.org/fileadmin/public_web_site/ich_products/guidelines/quality/q3d/q3d_iwg_final_concept_paper_ October_21_2014.pdf http://www.ich.org/fileadmin/public_web_site/ich_products/guidelines/quality/q3d/q3d_iwg_final_business_plan_ October_21_2014.pdf 40

USP <233> Method Proposal Procedure 1 can be used for elemental impurities generally amenable to detection by inductively coupled plasma atomic (optical) emission spectroscopy (ICP AES or ICP OES). Procedure 2 can be used for elemental impurities generally amenable to detection by inductively coupled plasma mass spectrometry (ICP MS). 41

Performance-based method Analysts are free to select a method/procedure that works for their samples The method selected may include plasma spectrochemistry, atomic absorption spectroscopy, OR ANY OTHER METHOD that displays adequate accuracy, sensitivity and specificity. 42

Other ICH Impurity Questions Other gaps? Thresholds and limits Impurity control in - multisource products - starting materials - reagents Depth of impurity investigations 43

Products not covered by ICH Excluded product types: biological/biotechnological (ICH Q6B) peptide (PhEur general monograph 2034) oligonucleotide radiopharmaceutical fermentation product and semi-synthetic products (EMA guideline for antibiotics) herbal products crude products of animal or plant origin 44

Extractable and Leachable Impurities Multitude of guidelines, but lack of harmonized expectations, focus has been on inhaled and nasal products Metered Dose Inhaler (MDI) and Dry Powder Inhaler (DPI) Drug Products; FDA Draft Guidance for Industry Nasal Spray and Inhalation Solution, Suspension, and Spray Drug Products Chemistry, Manufacturing, and Controls Documentation; FDA Guidance for Industry CHMP Guideline on the Pharmaceutical Quality of Inhalation and Nasal Products PQRI: Safety Thresholds and Best Practices for Extractables and Leachables in Orally Inhaled and Nasal Drug Products 45

Threshold limit disconnects Drug substance process impurity: 0.2% in a drug substance with 0.5 mg maximum daily dose requires identification and qualification Patient exposure - Total daily intake (TDI) of impurity: 1 mg TDI is less than 1.5 mg/day limit for a potentially genotoxic impurity Residual solvent: A drug substance with benzene at 2 ppm meets Q3C(R5) limit, but for a dose of 2 g/day, TDI is 4 mg A drug substance with benzene present at 5 ppm exceeds limit, but for a dose of 2 mg/day, TDI is 0.01 mg D.J. Snodin, S.D. McCrossen, Guidelines and pharmacopoeial standards for pharmaceutical impurities: Overview and critical assessment, Reg. Tox. Pharmacol. 63 (2012) 298 312 46

Degradation impurities threshold concerns Degradation impurity in drug substance Dose: 1 mg 10 mg 100 mg ID threshold (TDI) 1 mg 10 mg 100 mg Qual threshold (TDI) 1.5 mg 15 mg 150 mg Degradation impurity in drug product Dose: 1 mg 10 mg 100 mg ID threshold (TDI) 5 mg 20 mg 200 mg Qual threshold (TDI) 10 mg 50 mg 200 mg Should thresholds for degradation impurities in a drug substance be consistent with those for the drug product? 47

Other Threshold Considerations Should more latitude in application of Q3 thresholds be allowed based on: -Chronic vs. limited-duration therapy -Indication -Population Should in silico, literature, or other structure-based rationale be acceptable for impurity qualification in lieu of additional animal studies? 48

Risk assessment for impurities in multi-sourced materials (starting materials, reagents) What method(s) are used to prepare the material? What impurities could be introduced with the material? What is the likelihood QC methods will detect new impurities? Does supplier s change control for manufacturing changes evaluate the potential for new impurities? Does buyer s change control evaluate potential for new impurities from a different supplier? 49

Example: Impurity from reagent Repaglinide synthesis + + DCC hydrolysis Impurity derived from cyclohexylamine in DCC K.V.S.R. Krishna Reddy et al., J. Pharm. Biomed.Anal. 32 (2003) 461-467 50

DCC Quality? 74 suppliers (52 from China) Quality range: 98.0-99.5% Unknown control of cyclohexylamine impurity Switching suppliers without knowledge of the impact of cyclohexylamine as an impurity could generate a new impurity in the drug substance Do suppliers control cyclohexylamine levels? What levels are acceptable for regalinide process? Would the QC impurities test for repaglinide detect the cyclohexylamine-derived impurity? 51

Ambiguity in Impurity Investigation How much is enough? Q3A(R2) summarise the actual and potential impurities most likely to arise during the synthesis, purification, and storage of the new drug substance. This discussion can be limited to those impurities that might reasonably be expected based on knowledge of the chemical reactions and conditions involved. Potential Impurity: An impurity that theoretically can arise during manufacture or storage. It may or may not actually appear in the new drug substance.?????? How are most likely to arise and might reasonably be expected interpreted? Theoretical impurities can encompass an unreasonably large number of compounds, especially if genotoxicity thresholds are considered. 52

In Practice Focus investigation on potential, not hypothetical/theoretical impurities Theoretical/hypothetical: based on in cerebro or in silico predictions of synthetic byproducts or degradation chemistry Potential: used in process, found during process development,or formed as major degradation products during stress testing Actual or relevant: impurities present in drug or with a high likelihood of being present *S. Baertschi et al., Conference on Small Molecule Science UNC-Chapel Hill, NC, August 2, 2011 53

Summary Despite ongoing questions, ICH impurity guidelines have provided harmonized expectations for drug development. Changes being effected: Genotoxic impurities Elemental impurities Opportunities for improvement: guidance for other product types greater flexibility when justified reexamination of threshold rationale and inconsistency 54

Thank You! 55