ICH Q3D: Practical implementation and the role of excipient data in a risk based approach Dr Andrew Teasdale Andrew.Teasdale@astrazeneca.com
Overview areas covered Why share data? What data already exists? How can this be augmented? What s the strategic intent of the database? Contributing data to the database / current status Vision for how the database could be used to facilitate risk assessments
Why Share Data? ICH Q3D is predicated on the evaluation of risk, this is made of 3 factors RISK = PROBABILITY x Severity x Detectability We know the severity Defined PDEs. We have detectability ICP / XRF DATA either newly generated or Historical data informs us as to the probability. Sharing data thus allows us to make informed judgement during the IDENTIFY and EVALUATE PHASES
Why Share Data? Q3D itself comments specifically on this: SECTION 5 - Information for this risk assessment includes but is not limited to: data generated by the applicant, information supplied by drug substance and/or excipient manufacturers and/or data available in published literature. SECTION 5.5. The data that support this risk assessment can come from a number of sources that include, but are not limited to: Prior knowledge; Published literature; Data generated from similar processes; Supplier information or data; Testing of the components of the drug product; Testing of the drug product.
What data already exists? How can this be augmented? Container Closure Systems THEORETICAL RISK Especially in the case of liquid formulations there is risk of metals leaching out of CCS into the formulation WHAT DOES THE DATA SAY? Materials in Manufacturing and Packaging Systems as Sources of Elemental Impurities in Packaged Drug Products: A Literature Review PDA J Pharm Sci Technol January/February 2015 69:1-48; Section 5.3 Probability of elemental leaching into solid dosage forms is minimal and does not require further consideration in the risk assessment
Why Share Data? Q3D Case Studies use of first principles approach based on existing data exemplified.
What data already exists? How can this be augmented? EXCIPIENT STUDIES Study involved: Some 200+ samples Examined 24 elements SUMMARY OF RESULTS Little evidence of substantial levels of even the big 4/Class 1 (ubiquitous?) in mined excipients Pb seen in TiO2 but levels <10ppm, variability not significant. Pb also seen in Zn Stearate. Cd levels in Magnesium hydroxide / Calcium carbonate exceed Option 1 limits levels need to fail to an option 2 limit before serious concern THIS IS 200 SAMPLES WHAT IF WE COULD COLLATE DATA FROM 2000+ SAMPLES?
What data already exists? How can this be augmented? The data to be shared is the analytical data generated to establish the levels of trace metals within batches of excipients used in the manufacture of pharmaceuticals. Data = Knowledge Potential to facilitate more scientifically driven elemental impurities risk assessments and reduce unnecessary testing as part of the elemental impurities risk assessment efforts. More data = More Knowledge
What s the strategic intent of the database? Become the primary source of EI data for excipients that drives initial risk assessment (c.f. the Jenke paper for packaging components & EIs) Publish key findings with the intention of de-risking commonly used excipients Compare / contrast with data published generated by FDA.
Building the database How has the database been built? How much data is in it? Lhasa designed and developed the Elemental Impurities database based on Vitic Nexus TM platform Approved by the consortium in December 2015 Initial round of donations was received beginning of 2016 The database was first released at the end of March 2016 The Elementals database v2016.1.0 contains the following number of records: 52 records in the Excipient table. 123 records in the Elementals table. V2016.2.0 just released now contains 157 excipients 757 result records
Building the database Procedure/process for organizations to share their in-house data Template defined to allow error free parsing of data. Data anonymised and checked by Lhasa Limited.
Building the database Data quality requirements Extensive discussions relating to data requirements Validation protocol generated Extent of Validation recorded + Digestion Conditions No difference between data donated and data published in peer review journal in terms of vindication of data Sub Class A Compare a matrix matched blank to your lowest standard, making sure there is no significant contribution compared to your lowest standard Minimum 5 point calibration R = >0.995 ~ >R 2 = 0.990 Minimum of 2 spikes one at the top and one at the bottom of the quantitative liner range spike recoveries are between 70-150% Governed by Accuracy and Range data. 6 replicate aspirations of a standard or spiked sample either together or taken throughout the analysis giving %RSD 20% or spike sample or standard tested at the start and end of the run give the same measurement ± 20% or a 5 point calibration gives an R value of 0.995 Minimum N=3 replicate spikes within the Range of the method, The spikes can be at the same level or different levels where the response factors give 20% RSD As long as test solutions and spikes are prepared within 24 hours of each other solution stability is assumed as long as all other parameters are met. Equivalent concentration in ug/g in sample of your lowest spike Equivalent concentration in ug/g in sample of your lowest and highest spike Estimate LOD by taking the Std Dev of 6 blank measurements, multiplying by 3.3 and dividing this by the slope of your calibration line. Sub class B Compare a matrix matched blank to your lowest standard, making sure there is no significant contribution compared to your lowest standard Minimum 3 point calibration R = >0.990 ~ >R 2 = 0.980 Minimum of 1 spike within the quantitative liner range spike recoveries are between 50-150% Governed by Accuracy and Range data. 6 replicate aspirations of a standard or spiked sample either together or taken throughout the analysis giving %RSD 20% or sample tested at the start and end of the run give the same measurement ± 30% or a 5 point calibration gives an R value of 0.990 Minimum of 2 spikes one at the top and one at the bottom of the quantitative liner range spike recoveries are between 50-150% As long as test solutions and linearity standards are prepared within 48 hours of each other solution stability is assumed as long as all other parameters are met. Equivalent concentration in ug/g in sample of your lowest standard Equivalent concentration in ug/g in sample of your lowest and highest standard Estimate LOD by taking the Std Dev of 6 blank measurements, multiplying by 3.3 and dividing this by the slope of your calibration line.
Building a Database Is all of the data for lactose and how will sufficient diversity of materials and suppliers be managed? ListNo CarlMrozListName Total 1 Magnesium stearate 23 2 Microcrystalline cellulose 41 3 Lactose 32 4 Starch 14 5 Cellulose derivatives 18 6 Sucrose 9 7 Povidone 15 8 Stearic acid 3 9 Dibasic calcium phosphate 18 10 Polyethylene glycol 6 Number of results 45 40 35 30 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 The content of the database will be actively managed Clear commitment from members to generate data if gaps are identified
How is use of the database envisioned? At EMA meeting in April EFPIA presented a series of Case Studies
Oral Solid Dose Several Excipients used in the formulated product. What data are available? Component Functionality Amount % in per 400 coated mg tablet tablet (mg) Type (Excipient) Core API Drug substance 400.00 62.64 Hypromellose 2910 Binder 21.70 3.40 Plant Microcrystalline Diluent Cellulose 37.20 5.83 Plant Lactose Diluent Monohydrate 111.50 17.46 Animal Crospovidone Disintegrant 43.40 6.79 Synthetic Magnesium Lubricant stearate 6.20 0.97 Mineral Coating Hypromellose 2910 Film-former 11.16 1.75 Plant Titanium dioxide Pigment 5.55 0.87 Mineral Triacetin Plasticiser 1.49 0.23 Synthetic Blue Aluminium Colorant Lake #2 0.37 0.06 Mineral Blue Aluminium Colorant Lake #1 0.03 0.005 Mineral Number of materials FDA External DB Internal Lactose 6 3 Hypromellose 2910 6 (not defined as 2910) MCC 14 6 Crospovidone 17 (povidone) Magnesium Stearate 1 7 9 Titanium Dioxide 7 Blue Aluminium Lake #1 1 Blue Aluminium Lake #2 e Lactose is the main excipient others <10% 8 3 Database should contain substantively more data for common excipients
Excipient data Maximum level seen (ppm) Number of materials As Cd Hg Pb V Ni Co FDA Extern DB Intern FDA Extern DB Intern FDA Extern DB Intern FDA Extern DB Intern FDA Extern DB Intern FDA Extern DB Intern FDA Extern DB Intern FDA Extern DB Lactose 6 3 <0.23 <0.03 <0.08 ND <0.5 ND <0.08 ND <2 ND <3 ND <0.8 ND Intern Hypromellose 2910 6 8 0 <0.03 0 <0.1 0 <0.3 0.01 <0.1 0.02 ND 0.64 2.09 0.01 <1 MCC 14 6 <1.0 ND <0.2 ND <0.5 ND <0.2 <0.1 <2 ND <3 <1 <0.8 ND 0.2 (actual number above LOQ) Crospovidone 17 3 0.02 ND 0 ND 0 ND 0.06 ND 0.02 ND 0.1 ND 0.1 ND Magnesium Stearate 1 7 9 0.02 <0.23 0.09 0 <0.2 <0.1 0 <0.5 <0.3 0.01 <0.2 <0.1 0 <2 1.7 0.16 <5 1.5 0 <0.8 <1 0.5 (actual number above LOQ) Titanium Dioxide 7 0.36 0.07 0.04 5.74 5.95 0.48 0.04 Blue Aluminium Lake #1 1 0 0.01 0.03 0.03 0.26 1.58 0.01
Excipient data Reflection on significance No appreciable traces of Class 1 or Class 2a elements in low risk Component Functionality Amount per 400 mg tablet (mg) % in coated Type (Excipient) tablet excipients Lactose Povidone MCC Mg Stearate Ni seen at 1.5ppm NB less than 1% of the formulation Titanium dioxide Is this significant? 6ppm Pb / 6ppm V Core Drug substance API 400.00 62.64 Hypromellose 2910 Binder 21.70 3.40 Plant Microcrystalline Diluent Cellulose 37.20 5.83 Plant Diluent Lactose Monohydrate 111.50 17.46 Animal Crospovidone Disintegrant 43.40 6.79 Synthetic Lubricant Magnesium stearate 6.20 0.97 Mineral Coating Hypromellose 2910 Film-former 11.16 1.75 Plant Titanium dioxide Pigment 5.55 0.87 Mineral Triacetin Plasticiser 1.49 0.23 Synthetic Blue Aluminium Lake Colorant #2 0.37 0.06 Mineral Blue Aluminium Lake Colorant #1 0.03 0.005 Mineral
Excipient data Reflection on significance Component Category Quantity (mg/form) Dose "x" form (mg/day) Arsenic in component ug/g As ug in daily dose of formulation Lead in component ug/g Pb ug in daily dose of formulation Mercury in component ug/g Hg ug in daily dose of formulation Cadmium in component ug/g Cd ug in daily dose of formulation Vanadium in component ug/g V ug in daily dose of formulation Cobalt in component ug/g Co ug in daily dose of formulation Nickel in component ug/g Ni ug in daily dose of formulation Dosage Form : x = 1 Total Bio Acc Total Bio Acc Total Bio Acc Total Bio Acc Total Bio Acc Total Bio AccTotal Bio Acc Total Bio AccTotal Bio Acc Total Bio Acc Total Bio Acc Total Bio AccTotal Bio Acc Total Bio Acc Active Synthetic 400 400 0.00 0.00 0.00 0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Hypomellose Synthetic 32.9 32.9 0.03 0.00 0.00 0.1 0.00 0.00 0.30 0.01 0.00 0.10 0.00 0.00 0.03 0.00 0.00 1.00 0.03 0.00 2.09 0.07 0.00 MCC Plant derived 37.2 37.2 1.00 0.04 0.00 0.2 0.01 0.00 0.50 0.02 0.00 0.20 0.01 0.00 2.00 0.07 0.00 0.80 0.03 0.00 3.00 0.11 0.00 Lactose Animal 112 112 0.23 0.03 0.00 0.1 0.01 0.00 0.50 0.06 0.00 0.08 0.01 0.00 2.00 0.22 0.00 0.80 0.09 0.00 3.00 0.34 0.00 Crospovidone Synthetic 43.4 43.4 0.02 0.00 0.00 0.1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.10 0.00 0.00 0.10 0.00 0.00 TiO2 Mineral 5.5 5.5 0.36 0.00 0.00 5.9 0.03 0.00 0.04 0.00 0.00 0.07 0.00 0.00 5.95 0.03 0.00 0.04 0.00 0.00 0.48 0.00 0.00 Mg Stearate Mineral 6.2 6.2 0.23 0.00 0.00 0.2 0.00 0.00 0.50 0.00 0.00 0.20 0.00 0.00 1.70 0.01 0.00 1.00 0.01 0.00 1.50 0.01 0.00 Al Lake 1 Mineral 3 3 0.00 0.00 0.00 0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 Triacetin Synthetic 1.5 1.5 0.00 0.00 0.00 0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Al Lake 2 Mineral 0.3 0.3 0.00 0.00 0.00 0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Total Dosage Form weight 642 642 Total element As 0.07 0.00 Pb 0.06 0.00 Hg 0.09 0.00 Cd 0.02 0.00 V 0.34 0.00 Co 0.16 0.00 Ni 0.53 0.00 Permissible Limits As Pb Hg Cd V Co Ni Formulation Q3D Q3D Q3D Q3D Q3D Q3D Q3D Oral PDE 15 5.0 30 5 100 50 200 Parenteral PDE 15 5.0 3 2 10 5 20 inhaled PDE 2 5.0 1 2 1 3 5 Based on data from database all predicted to be ~1% or less of PDE
Challenges to using first principles The data set is limited! - True but plan to develop a critical mass. Mined excipients will always show variability - Potentially true. Component Dosage Form : Category Quantity (mg/form) Dose "x" form (mg/day) Arsenic in component ug/g As ug in daily dose of formulation Lead in component ug/g Pb ug in daily dose of formulation x = 1 Total Bio Acc Total Bio Acc Total Bio Acc Total Bio Acc Active Synthetic 400 400 0.00 0.00 0.00 0.0 0.00 0.00 Hypomellose Synthetic 32.9 32.9 0.03 0.00 0.00 0.1 0.00 0.00 MCC Plant derived 37.2 37.2 1.00 0.04 0.00 0.2 0.01 0.00 Lactose Animal 112 112 0.23 0.03 0.00 0.1 0.01 0.00 Crospovidone Synthetic 43.4 43.4 0.02 0.00 0.00 0.1 0.00 0.00 TiO2 Mineral 5.5 5.5 0.36 0.00 0.00 1000.0 5.50 0.00 Mg Stearate Mineral 6.2 6.2 0.23 0.00 0.00 0.2 0.00 0.00 Al Lake 1 Mineral 3 3 0.00 0.00 0.00 0.0 0.00 0.00 Triacetin Synthetic 1.5 1.5 0.00 0.00 0.00 0.0 0.00 0.00 Al Lake 2 Mineral 0.3 0.3 0.00 0.00 0.00 0.0 0.00 0.00 How much impact would this have in the case of an excipient such as TiO2? 1000ppm Pb / Hg? Pb overall just exceeded RISK = PROBABILITY x Severity x Detectability Total Dosage Form weight 642 642 Total element As 0.07 0.00 Pb 5.52 0.00
So what can we learn from the database? v2016.2.0 just released now contains 157 excipients 757 result records Examination of Top 5 excipients 1 1 1 1 2A 2A 2A Cd Pb As Hg Co V Ni MAGNESIUM STEARATE Max 0.20 0.20 1.00 0.50 0.80 2.00 5.00 Min 0.02 0.05 0.02 0.01 0.03 0.01 0.14 Mean 0.08 0.10 0.25 0.13 0.19 0.57 1.32 MICROCRYSTALLINE CELLULOSE Max 0.20 0.20 1.00 0.50 0.80 2.00 3.00 Min 0.003 0.01 0.02 0.01 0.02 0.01 0.03 Mean 0.04 0.07 0.19 0.11 0.18 0.43 0.68 LACTOSE Max 0.20 0.21 0.23 0.50 0.80 2.00 3.00 Min 0.003 0.04 0.01 0.01 0.03 0.01 0.03 Mean 0.07 0.08 0.11 0.12 0.15 0.33 0.47 STARCH Max 0.10 0.10 0.20 0.10 0.10 0.15 0.30 Min 0.02 0.05 0.02 0.01 0.03 0.01 0.03 Mean 0.03 0.07 0.14 0.04 0.06 0.11 0.21 CELLULOSE DERIVATIVES Max 0.20 0.20 0.20 0.20 0.20 0.56 1.04 Min 0.02 0.02 0.02 0.01 0.01 0.01 0.09 Mean 0.05 0.08 0.11 0.05 0.07 0.16 0.34 Option1 Oral 0.5 0.5 1.5 3 5 10 20 Option1 Oral 30% 0.15 0.15 0.45 0.9 1.5 3 6 Levels uniformly low sensible to apply 30% limit on top of Option 1?
So what can we learn from the database? What about common mined excipients? E.g. calcium phosphate 1 1 1 1 2A 2A 2A Cd Pb As Hg Co V Ni ANHYDROUS DIBASIC CALCIUM PHOSPHATE Max 0.20 0.20 1.00 0.20 0.60 10.00 21.84 Min 0.05 0.10 0.10 0.01 0.07 0.04 0.27 Mean 0.11 0.16 0.38 0.08 0.31 2.06 7.72 DIBASIC SODIUM PHOSPHATE Max 0.20 0.20 0.41 10.00 0.20 0.20 3.06 Min 0.004 0.01 0.10 0.003 0.03 0.01 0.05 Mean 0.04 0.07 0.17 1.46 0.08 0.08 0.72 DIBASIC CALCIUM PHOSPHATE DIHYDRATE Max 0.04 0.27 0.37 0.02 0.51 0.02 15.97 Min 0.04 0.10 0.20 0.01 0.34 0.01 13.47 Mean 0.04 0.21 0.28 0.01 0.41 0.01 14.52 SODIUM CHLORIDE Max 0.20 0.20 0.20 0.20 1.00 0.20 1.00 Min 0.01 0.01 0.05 0.05 0.01 0.05 0.04 Mean 0.05 0.10 0.08 0.13 0.15 0.09 0.17 Option1Oral 0.5 0.5 1.5 3 5 10 20 Option1Oral30% 0.15 0.15 0.45 0.9 1.5 3 6 Couple of examples where level exceeds Option 1 limit. Ni in anhydrous calcium phosphate Mercury in Sodium phosphate UNLIKELY TO ULTIMATELY TO POSE A RISK
Pharmacopoeial notifications element specific testing KEEP Current tests provide valuable data that can be used as part of the risk assessment. Removing such tests may mean no data. Replacing tests with ICP could drive new limits. DELETE Specific testing to specific limits is inconsistent with Q3D These tests are ineffective and inefficient. The limits are quality as well as safety limits.
Pharmacopoeial notifications element specific testing Calcium Phosphate Common Filler Monograph recently revised Tests for 3 elements 1. Arsenic (2.4.2, Method A): maximum 10 ppm, determined on 2 ml of solution S. i.e. a wet chemistry limit test. 2. Barium. To 0.5 g, add 10 ml of water R and heat to boiling. While stirring, add 1 ml of hydrochloric acid R dropwise. Allow to cool and filter if necessary. Add 2 ml of a 10 g/l solution of dipotassium sulfate R and allow to stand for 10 min. No turbidity is produced. i.e. a Turbidity test 3. Iron (2.4.9): maximum 400 ppm. Dilute 0.5 ml of solution S to 10 ml with water R. i.e. another we chemistry limit test Three separate tests for 3 metals not really very efficient. Are these tests informative? Do they add value?
Pharmacopoeial element specific testing SUBST_ID SUPPLIER Co Os V Rh Ru Pd Pb Ni Fe Mn Sb Li Cu Cr Ba Mo Tl Hg Cd As Anhydrous dibasic calcium phosphate XA0081 0.60 LLOQ 0.07 0.06 LLOQ 0.29 0.18 21.8 1145 81.2 0.68 LLOQ 0.45 1.54 5.39 2.26 LLOQ 0.01 0.08 0.42 Anhydrous dibasic calcium phosphate XA0010 0.4 tested 1.5 tested tested LLOQ LLOQ 1 tested tested tested tested tested tested LLOQ LLOQ LLOQ tested tested tested Anhydrous dibasic calcium phosphate XA0011 0.4 tested LLOQ tested tested LLOQ 0.2 1 tested tested tested tested tested tested tested tested tested LLOQ LLOQ LLOQ Anhydrous dibasic calcium phosphate XA0478 0.36 LLOQ 0.06 0.04 LLOQ 0.22 0.14 13.5 604.0 75.0 0.68 LLOQ LLOQ 1.49 3.6 2.23 LLOQ LLOQ 0.07 0.25 Anhydrous dibasic calcium phosphate XA0476 0.60 LLOQ 0.04 0.06 LLOQ 0.29 0.10 20.6 860.2 79.5 LLOQ LLOQ LLOQ 1.40 4.8 1.68 LLOQ LLOQ 0.08 0.35 Anhydrous dibasic calcium phosphate XA0477 0.59 LLOQ 0.06 0.05 LLOQ 0.28 0.17 21.8 1145.2 76.9 0.237 LLOQ 0.448 1.54 5.4 1.83 LLOQ LLOQ 0.08 0.30 Anhydrous dibasic calcium phosphate SW0172 tested 1.7 0.18 1.00 0.07 tested tested detecte d Anhydrous dibasic calcium phosphate XA0010 Anhydrous dibasic calcium phosphate SW0174 tested 1.8 tested 0.8 0.17 0.9 tested 0.18 0.9 tested tested tested 8 detecte d 0.08 detecte d 0.05 detecte d Anhydrous dibasic calcium phosphate QW0356 LLOQ tested 5.7 tested tested tested LLOQ 2.03 tested tested tested tested tested tested LLOQ LLOQ 0.18 tested tested tested Anhydrous dibasic calcium phosphate BX0760 LLOQ tested LLOQ tested tested tested 0.1 0.27 tested tested tested tested tested tested tested tested tested LLOQ LLOQ LLOQ Dibasic calcium phosphate dihydrate XA0079 0.51 LLOQ 0.02 0.04 LLOQ 0.18 0.27 16.0 1665 37 0.35 LLOQ LLOQ 0.99 LLOQ 0.19 LLOQ 0.018 0.04 0.37 Dibasic calcium phosphate dihydrate XA0481 0.40 LLOQ 0.01 0.02 LLOQ 0.17 0.23 15.5 891 31 LLOQ LLOQ LLOQ 0.81 LLOQ 0.19 LLOQ LLOQ 0.04 0.37 Dibasic calcium phosphate dihydrate XA0482 0.43 LLOQ LLOQ LLOQ LLOQ 0.14 0.10 13.9 1338 26 0.29 LLOQ LLOQ 0.95 LLOQ LLOQ LLOQ LLOQ 0.04 0.20 Dibasic calcium phosphate dihydrate XA0484 0.43 LLOQ LLOQ LLOQ LLOQ 0.14 0.10 13.9 1338 26 0.29 LLOQ LLOQ 0.95 LLOQ LLOQ LLOQ LLOQ 0.04 0.20 Dibasic calcium phosphate dihydrate XA0479 0.34 LLOQ 0.02 0.04 LLOQ 0.18 0.26 13.5 719 37 0.35 LLOQ LLOQ 0.65 LLOQ 0.16 LLOQ 0.018 0.04 0.24 Dibasic calcium phosphate dihydrate XA0483 0.34 LLOQ 0.02 0.04 LLOQ 0.18 0.26 13.5 719 37 0.35 LLOQ LLOQ 0.65 LLOQ 0.16 LLOQ 0.018 0.04 0.24 Dibasic calcium phosphate dihydrate XA0480 0.40 LLOQ 0.01 0.022 LLOQ 0.17 0.23 15.5 891 31 LLOQ LLOQ LLOQ 0.81 LLOQ 0.19 LLOQ LLOQ 0.04 0.37
Pharmacopoeial element specific testing As limit test data from database shows although in some batches levels <1ppm Barium levels <10ppm (Limits shown below). Set against these limits what value does this test provide? Perhaps the most interesting of all! Iron limit 400ppm yet data derived from ICP shows that levels > 400ppm limit. Is this test therefore meaningful?
Conclusions The feasibility of sharing excipient elemental impurity data has been successfully demonstrated Pooling and publishing data can surely help to improve the ease with which risk assessments can be completed Ultimately it will give a much better picture of which materials represent a more significant risk than others Indicate where the risk is real & where it is negligible Reduce the amount of testing that is needed to be done moving forward to support implementation We typically expect that the EI database to be seen as key supportive information that is used routinely in conjunction with some product specific test data in the risk assessment.