Organisation mondiale de la santé (OMS) · Technical Documents

WHO international standard for endotoxin: report of an international collaborative study to evaluate three preparations of endotoxin for their suitability to serve as the third international standard for bacterial endotoxin

31 décembre 2012 Organisation mondiale de la santé

Le texte intégral est hébergé par l’organisation qui le publie. lawenc.com indexe les métadonnées et renvoie vers la source officielle.

Texte intégral

WHO/BS/2012.2193 and working document QAS/12.501 ENGLISH ONLY EXPERT COMMITTEE ON BIOLOGICAL STANDARDIZATION Geneva, 15 to 19 October 2012 EXPERT COMMITTEE ON SPECIFICATIONS FOR PHARMACEUTICAL PREPARATIONS Amsterdam, 9-12 October 2012

WHO International Standard for endotoxin Report of an international collaborative study to evaluate three preparations of endotoxin for their suitability to serve as the third international standard for bacterial endotoxin Stephen Poole, Trusha Desai, Lucy Findlay, Alan Heath National Institute for Biological Standards and Control (NIBSC), Potters Bar, Herts EN6 3QG, UK Mary Crivellone, Walter Hauck, Michael Ambrose, Tina Morris United States Pharmacopoeia (USP) Eriko Terao, Jean-Marc Spieser, Karl-Heinz Buchheit, Guy Rautmann, Arnold Daas European Directorate for the Quality of Medicines & Healthcare (EDQM) This document has been prepared for the purpose of inviting comments and suggestions on the proposals contained therein, which will then be considered by the Expert Committee on Biological Standardization (ECBS) and the Expert Committee on Specifications for Pharmaceutical preparations (ECSPP). Comments MUST be received by 01 October 2012 and should be addressed to the World Health Organization, 1211 Geneva 27, Switzerland, attention: Quality Safety and Standards (QSS). Comments may also be submitted electronically to the Responsible Officer: Dr Jongwon Kim at email: kimjon@who.int © World Health Organization 2012 All rights reserved. Publications of the World Health Organization are available on the WHO web site (www.who.int) or can be purchased from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: bookorders@who.int). Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press through the WHO web site (http://www.who.int/about/licensing/copyright_form/en/index.html). The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

WHO/BS/2012.2193 and working document QAS/12.501 Page 2 All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The named authors alone are responsible for the views expressed in this publication.

WHO/BS/2012.2193 and working document QAS/12.501 Page 3

Summary An international collaborative study was organised jointly by the World Health Organization (WHO)/National Institute for Biological Standards and Controls (NIBSC), the US Pharmacopeia (USP) and the European Directorate for the Quality of Medicines & HealthCare (EDQM/Council of Europe) for the establishment of harmonised replacement endotoxin standards for these 3 organisations. Thirty-five laboratories worldwide, including Official Medicines Control Laboratories and manufacturers enrolled in the study. Three candidate preparations (10/178, 10/190 and 10/196) were produced with the same material and same formulation as the current reference standards with the objective of generating a new (3rd) IS with the same potency (10,000 IU/vial) as the current (2nd) IS. The suitability of the candidate standards to act as the reference standard in assays for endotoxin performed according to compendial methods was evaluated. Their potency was calibrated against the WHO 2nd International Standard (IS) for Endotoxin (94/580). Gelation and photometric methods produced similar results for each of the candidate preparations. Overall, these results were in line with those generated for the establishment of the current preparations of reference standards. Accelerated degradation testing of vials stored at elevated temperatures supported the long-term stability of the 3 candidate preparations.

Introduction The control of parenteral pharmaceutical products for bacterial endotoxins is a procedure fully harmonised between the European Pharmacopoeia (Ph. Eur.), the US Pharmacopeia (USP) and the Japanese Pharmacopoeia (JP) [1, 2, 3]. Reference preparations used in these assays are, since the collaborative study for the establishment of the 2nd IS for Endotoxin in 1996, one of the truly harmonised standards prepared from a common starting material, evaluated in a wide international collaborative study and adopted as the international standard and as compendial standards. The present study, initiated in 2007, aimed at establishing replacement preparations for the current WHO IS, Ph. Eur. BRP and USP reference standard, stocks of which are dwindling. An international collaborative study was carried out to calibrate 3 candidate preparations against the WHO 2nd IS for Endotoxin (94/580) using official pharmacop(o)eial methods (gelation and photometric assays). The starting material for the production of the candidate preparations was a bulk endotoxin material kindly donated by the US Food and Drug Administration–Center for Biologics Evaluation and Research (FDA–CBER). The same starting material was used for the current and previous lots of reference standard endotoxins: WHO 1st and 2nd IS for Endotoxin (84/650 and 94/580, respectively), the Ph. Eur. Endotoxin standard BRP preparations 3 and 4, the USP Reference Standards lots F and G series (G, G-1, G2B274 and G3E069) as well as the FDA reference lots EC-1 through EC-6 [5, 6]. The candidate standards were filled at NIBSC in October 2010 and had been preliminarily evaluated for suitability by the laboratories of NIBSC, USP and EDQM. The calibrant for the collaborative study described below study was the 2nd IS for Endotoxin, which had previously been shown at NIBSC, USP and EDQM to yield equivalent results to the current standards of the USP and Ph. Eur.

WHO/BS/2012.2193 and working document QAS/12.501 Page 4

Materials and methods Candidate WHO endotoxin standards The starting material for the candidate 3rd IS was originally isolated from Escherichia coli (Braude strain) group O113:H10:K negative [4] donated by the US Food and Drug Administration–Center for Biologics Evaluation and Research (FDA–CBER). Production of the candidate preparations To ensure that the fills were as closely similar as was practicable in terms of the number of units of endotoxin per vial, multiple pilot productions and half-scale and full-scale trial fills were conducted in 2009 and 2010 to determine the mass of endotoxin/vial required to yield the target of 10,000 IU/vial, the current IS being assigned 10,000 IU/vial. The excipients contained in the candidate standards were the same as those contained in the 1st and 2nd IS for Endotoxin (94/580), USP Lot F/FDA EC-5, USP Lot G series (G, G-1, G2B274 and G3E069)/FDA EC-6 and the Ph. Eur. Endotoxin BRP preparations 3 and 4, there being no stability issues with this formulation. Three candidate preparations (10/178, 10/190 and 10/196) were produced in 2010 from a common bulk solution of endotoxin. For each candidate preparation, an aliquot of the endotoxin bulk solution and the excipients concentrate were combined and stirred for 1 hour at 2-8°C. The solution was brought to the final concentration with cold (2-8°C) water for injection and stirred overnight at room temperature. Filling into vials was performed at room temperature, with continuous stirring of the solution. Freeze-drying was performed over 4 days for each lot. Each vial contained the residue after freeze-drying of 1.0 mL of a solution that contained: 1.2 μg E. coli O113:H10:K negative endotoxin, 10 mg lactose and 1 mg polyethylene glycol 8000. The main specifics of the preparations are shown in Table 1. Collaborative study participants Thirty-five laboratories from 18 countries from Europe, North America, Asia (1 Australia, 1 Austria, 1 Belgium, 1 Brazil, 1 Canada, 1 China, 2 Denmark, 1 France, 4 Germany, 1 Italy, 5 Japan, 2 Korea, 1 The Netherlands, 1 Norway, 1 Portugal, 1 Sweden, 2 Switzerland, 7 USA and the Council of Europe/EDQM) took part in the study. These laboratories included Official Medicines Control Laboratories/regulatory institutes (19) and manufacturers (therapeutics: 8; reagents: 8). Throughout the report laboratories are referred to by an arbitrarily attributed code number; that code number is not reflected in the order of listing of the laboratories below. Study design Participants were provided with 30 vials altogether: 6 vials of IS (94/580) and 6 vials of each of 4 test preparations, candidate standards A, B, C, D, where D was a coded duplicate of B. The standard for all assays was the WHO 2nd International Standard (IS) for Endotoxin (94/580, 10,000 IU = EU/vial). Each assay was to include dilutions of (reconstituted) vials of the IS and all 4 test preparations (candidate standards) using a prescribed dilution scheme. The test preparations (candidate standards) were all filled at a nominal 10,000 IU/EU vial and were to be tested at the same nominal concentrations and at the same number of replicates as the IS, i.e. the test preparations were to be treated exactly as if they were the IS itself. Participants performing semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assays were provided with 3 vials of the IS and 3 vials of each of the 4 test preparations (candidate standards), each to be assayed twice, once using freshly reconstituted vials and once using vials within 2 weeks of their reconstitution. Thus, 15 vials in total (3 vials of IS and 3 vials of each of the 4 test preparations) were each to be assayed twice in total in LAL gelation assays. The

WHO/BS/2012.2193 and working document QAS/12.501 Page 5 protocol to be used was to be in accordance with published compendial procedures (Ph. Eur. general text 2.6.14., USP General Chapter <85>, JP general test 4.01) and the assays were to be performed with the LAL reagent routinely used by the laboratory, and having a sensitivity of 0.03 or 0.06 EU/mL. Participants performing quantitative Limulus Amoebocyte Lysate (LAL) Photometric Assays (Chromogenic/Turbidimetric) were provided with 3 vials of the IS and 3 vials of each of the 4 test preparations (candidate standards), each to be assayed twice, once using freshly reconstituted vials and once using vials within 2 weeks of their reconstitution. Thus, 15 vials in total (3 vials of IS and 3 vials of each of the 4 test preparations) were each to be assayed twice in total in LAL photometric assays. The protocol to be used was to be in accordance with official and harmonized compendial procedures (Ph. Eur. general text 2.6.14., USP General Chapter <85>, JP general test 4.01).

Statistical analysis All reported raw data were analysed at the USP using SAS software. For gelation data: The relative potency was determined for each of the Laboratory-Sample-Vial-Day combinations. These data were examined for unusual values, i.e. values corresponding to a relative potency outside 50%-200%. For photometric data: Linear regression of ln(results) vs. ln(concentration) for all combinations of Laboratory, Assay (when a laboratory provided results for multiple photometric assays), Sample or Standard, Day and Vial, were computed. The results were examined for unusual residual values. Next the remaining R2 values were examined for outliers. Furthermore, data failing the suitability condition specified in the USP General Chapter <85> that the R of the standard curve should be at least 0.98, corresponding to R2>0.96 [2] were excluded. The calculated potencies per assay were later provided to the EDQM as SAS datasets to prepare supplementary tables and figures. The raw data were not re-analysed but some additional statistics were calculated on the basis of the tabled potencies: geometric means across laboratories, Huber’s robust means (with k=1.5) to reduce the influence of extreme values, and other supplementary statistics.

Results Assay data returned Thirty-one laboratories in 17 countries contributed data within 2 months of receipt of samples as requested. Four laboratories reported results after the deadline: the results from these four laboratories were not included in the statistical evaluation and the assignment of the potency to the candidate preparations but their data was evaluated later for comparison with results from the 31 laboratories. The data from the four laboratories are given below as: Addendum: Data from Additional Laboratories. Twenty-four laboratories reported data using the gelation method. Twenty-three laboratories returned photometric assay results, using one or more photometric (chromogenic and/or turbidimetric) methods. Seventeen laboratories returned chromogenic assay results, 13 laboratories returned turbidimetric assay results. Seven laboratories provided results from more than 1 photometric assay for a total of 31 photometric assays.

WHO/BS/2012.2193 and working document QAS/12.501 Page 6 Assay validity: data excluded from analysis Gelation data from 1 laboratory (Lab 27) were invalid and not included in the analysis. One laboratory (Lab 14) provided gelation results from freshly reconstituted material only and 1 laboratory (Lab 3) provided data for 4 days, all but the first of which were treated as “later” days, assays. Two laboratories (Labs 10 and 12) provided gelation assay data that appeared to be entered as the transpose of what was intended; this was corrected in the statistical analysis data file. Two laboratories (Labs 17 and 26) marked some photometric data as outliers in the data sheets submitted. As the values were evidently out of any reasonable range, the laboratory’s judgement was accepted and these few points were not included in the analyses. Photometric results with undetermined endpoints reported as “greater (>) than some value” were treated as missing. One set of photometric results from Lab 22 was conducted more than 2 weeks after reconstitution and was not used. One laboratory (Lab 18) used only 2 dilutions for the photometric assay. Since the harmonised compendial procedure calls for at least 3 dilutions and since 2 dilutions does not permit assessment of lack of fit, this laboratory’s photometric data were not used. Two laboratories (Labs 16 and 17) performed a non-pharmacopoeial photometric assay using recombinant LAL (Factor C). These data were not included in the overall analysis. This left 23 laboratories with usable gelation data and 22 with usable photometric assay data. Potency estimates from gelation assays The relative potency was determined for each of the 564 Laboratory-Sample-Vial-Day combinations. After examination of these data for unusual values, 558 log relative potencies were further analysed. Some seventy per cent of results correspond exactly to a determined potency of 10,000 IU/vial. Four values corresponded to an absolute log relative potency greater than 0.7 (corresponding approximately to a relative potency outside the 50-200% interval) and these data were not used in further analyses. By inspection, typical data were that the samples and standard would either become negative on the same dilution or at most 1 dilution later. The extreme relative potencies correspond to results differing by 2 or more dilutions. Table 4 provides a complete overview of the potency estimates in IU/vial for each Lab-DaySample-Vial combination. The values vary from 2,500 IU/vial (Lab 23, Sample B) to 38,750 IU/vial (Lab 21, Sample C). The two-sided paired t-test of ln-transformed potencies (mean of 3 vials) showed no significant differences between values obtained on Day 1 and on Day 2 or later (P<0.528). Table 5 shows the geometric mean per laboratory and sample. Histograms of these values are provided in Figure 1. The values range from 8,476 IU/vial (Lab 21, Sample B) to 17,311 IU/vial (Lab 5, Sample C). The geometric mean across laboratories was 10,414, 10,739, 10,768 and 10,937 IU/vial for the respective samples (A-D). It was noted that the somewhat large values from Lab 5 may have resulted in a slight overestimation of the potencies for Samples B, C and D. To reduce the influence of extreme values Huber’s robust mean (k=1.5) was also calculated. This gave 10,250, 10,598, 10,509 and 10,648IU/vial for the respective samples. Potency estimates from photometric assays Linear regression of ln-transformed results vs. ln-transformed concentration for all combinations of Laboratory, Assay, Sample or Standard, Day and Vial, were computed. A total of 1019 regressions were examined for unusual residual values. There were no standardised residuals

WHO/BS/2012.2193 and working document QAS/12.501 Page 7 greater than 4.0 in magnitude, compared with the 1% point for Grubb’s test of 4.9. Thus, no values were excluded as outliers based on this analysis. Next, the 1019 R2 were examined. There was 1 value less than 0.88 compared to a next lowest value of 0.92. This was for one sample and these data were excluded from further analyses. Two sets of data failed the condition that the R be at least 0.98 (or R2 >0.96) for the IS and were excluded. When the data were reanalysed with and without the assumption of parallelism (equal slopes), 2 values (of 806 regressions) of the ratio of slopes (Sample/Standard) were found to be outside 0.8-1.25. No data were excluded. Eight values (of 806) of the estimated relative potency (assuming parallelism) fell outside 50-200%. Of these 8 values, 4 were from 1 laboratory and 3 from a second laboratory. All 8 values were excluded from the final calculations. Table 6 provides a complete overview of the potency estimates in IU/vial for each Lab-DaySample-Vial combination. Also shown is whether a chromogenic or turbidimetric method was used. The values vary from 4,203 IU/vial (Lab 22, Sample A, chromogenic) to 24,310 IU/vial (Lab 32, Sample D, turbidimetric). The two-sided paired t-test of ln-transformed potencies (mean of 3 vials) showed no significant differences between values obtained on Day 1 and on Day 2 or later (P<0.304). Table 7 shows the geometric mean per laboratory and sample. Histograms of these values are provided in Figure 2. Results obtained with the chromogenic method are displayed in light-grey boxes and the turbidimetric method in dark-grey boxes. The values range from 7389 IU/vial (Lab 15, Sample D, turbidimetric) to 17,702 IU/vial (Lab 32, Sample D, turbidimetric). The geometric mean across laboratories using the chromogenic method was 9,798, 10,333, 10,426 and 10,696 IU/vial for the respective samples (A-D). For the turbidimetric method the respective values were 10,292, 10,641, 10,986 and 11,229 IU/vial for samples A-D. No significant difference between the chromogenic and turbidimetric methods were found with the unpaired two-sided t-test (P<0.246, 0.482, 0.301, 0.326 respectively), so it was considered appropriate to pool values from both methods for the overall mean which yielded 10,017, 10,470, 10,673 and 10,904 IU/vial respectively. To reduce the effect of extreme values, notably from Labs 1 and 32, Huber’s robust mean (k=1.5) was also calculated. This gave 10,120, 10,404, 10,449 and 10,730 IU/vial respectively. Gelation versus photometric assays An analysis of variance of the pooled set of ln-transformed potencies (4 samples times 52 determinations) showed no significant difference between samples (P=0.06) or the 2 methods (P=0.30). It was therefore considered justified to calculate the mean potency per sample/preparation based on the pooled set of 52 results per sample. The overall potencies of the 3 preparations are shown in Table 8. Two-way comparisons between preparations The mean values per laboratory were plotted against each other in Figure 3. Each plot shows 1 of the 6 possible pairs. The two-sided paired t-test, as well as the sign test showed a significant difference between Sample A and samples B-D (P<0.05 for each pair). No significant difference was observed between Samples B, C or between Samples C and D, but the difference between the identical preparations B and D was significant (P=0.04). This slight difference may not be important, however, as the confidence limits of the final potency estimates do overlap. Although Samples B, C and D appear to be slightly more potent than Sample A, the difference may be considered unimportant.

WHO/BS/2012.2193 and working document QAS/12.501 Page 8

Stability Accelerated thermal degradation (ATD) studies for the candidate preparations are in progress at NIBSC. Twenty vials of each fill were stored at each of the following temperatures: -70°C, 20°C, +4°C, +20°C, +37°C, +45°C and +56°C. Samples were tested in a photometric chromogenic assay. The table below shows the estimated endotoxin activity for samples stored at +56°C as a percentage of that at -20°C, for periods from 10 to 17 months. Each value is based on two independent assays, using reversed plate layouts (to minimise plate-positional effects). There was no detectable degradation after 10 months at +56°C. There was an apparent drop in activity for preparation C (10/196) after 17 months. However, the pattern of results suggests that it is unlikely that there would be such a genuine drop between 15 and 17 months for a material that has proved highly stable at +56°C for 15 months, and assay variability may be a major contributor to the observed value. Summary of ATD results: +56°C as % of -20°C +56°C as % of -20°C 10 months 15 months 104.7 93.1 101.3 103.9 100.2 97.3

Fill A (10/178) B, D (10/190) C (10/196)

17 months 94.4 99.9 81.7

Applying the “rule of thumb” that degradation rates will double with every 10°C increase in storage temperature, 10 months at +56°C with no detectable degradation would be equivalent to around 150 years at -20°C. Even if there were a genuine drop to 95% after 17 months, this would be equivalent to over 250 years at -20°C before an equivalent loss was apparent for samples stored at -20°C. The data summarised above and in Tables 2(a)-2(e) below indicate that all three preparations are highly stable. Assessment of the stability of samples stored at temperatures ranging from 4°C to 45°C will continue.

Instructions for Use The draft Instructions for Use to accompany this reference material are provided in Appendix 3.

Participant feedback The participants have agreed the recommendation to establish preparation 10/178 as the third international standard for bacterial endotoxin with an assigned unitage of 10,000 IU/vial.

Discussion and Conclusions The objective of this collaborative study was to calibrate replacement standards for the WHO Endotoxin IS, Ph. Eur. Endotoxin standard BRP batch 4 and USP Endotoxin Reference Standard (lot G3E069). In order to ensure continuity of unitage, the potency values of 3 candidate preparations were evaluated against the current WHO 2nd IS for Endotoxin (94/580), using compendial gelation or photometric assays. Results from the accelerated thermal degradation study indicated that the 3 candidate preparations are highly stable and fit for purpose to serve as reference standard endotoxin.

WHO/BS/2012.2193 and working document QAS/12.501 Page 9 There was no significant difference between potencies obtained on freshly reconstituted vials and vials stored at 2 – 8 °C for up to 14 days after reconstitution, indicating that the reconstituted solution is stable for up to 2 weeks in the refrigerator. This allowed the data for freshly reconstituted vials and vials stored at 2 – 8 °C for up to 14 days after reconstitution to be pooled for the calibration of the endotoxin potency of the candidate preparations. Seventy per cent of the gelation assay results corresponded exactly to a potency of 10,000 IU/vial, consistent with the objective of the study to generate a new standard with a potency of 10,000 IU/vial. The mean values (Huber’s robust mean) for the 3 preparations (A, B, C [B=D]) were slightly larger, at 10, 250, 10,598 and 10,509 IU/vial respectively for gelation assays. The 2 photometric methods (chromogenic and turbidimetric assays) gave similar results to each other. This allowed the calculation of an overall potency for the photometric assays for each of the 3 preparations: 10,120, 10,404 and 10,449 respectively for Samples A, B, C [B=D]. The comparison of gelation and photometric assays was unbiased by the number of laboratories performing the test: 23 laboratories provided useable gelation data and 22 laboratories provided useable photometric assay data. The semi-quantitative gelation assay values trended higher for the 3 preparations than the quantitative photometric assays. This was most likely reflective of the gelation method itself, and could result from an insufficient number of dilutions surrounding the end point (2-fold dilution steps) with the gelation assay. The difference between gelation and photometric assay results was, however, not statistically significant, so that an overall potency value for all assays could be calculated for each preparation: 10,190 (95% CL 9,927–10,461), 10,588 (95% CL 10,252–10,935) and 10,715 (95% CL 10,289–11,159) IU/vial respectively for A, B, C]. This showed that the 3 candidate preparations, which shared the same endotoxin starting solution, are suitable for all applications (gelation and photometric assays). There was an approximately 3% difference, some 300 IU, in the calculated potency/vial between preparations B and D where D was in fact the coded duplicate of preparation B. This finding, together with the overlapping 95% CLs given above (and summarised in Table 8) puts the small, non-statistically different, differences noted above into context. The statistical comparisons presented in this report indicate that the potency of the candidate standard preparation A (10/178), at 10,250 IU in (semi-quantitative) gelation assays (with 70% of values being exactly 10,000 IU), 10,120 in (quantitative) photometric assays, and 10,190 overall (gelation + photometric, 95% CL 9,927–10,461) is a little closer to the target 10,000 IU/vial of the current IS than the 2 other preparations. That said, the confidence intervals for the potencies of the 3 preparations overlapped and the ranges of values were sufficiently similar for all three preparations to be considered equivalent for their intended use and be assigned a common value of 10,000 IU/vial. This potency assignment, to one significant value, follows the precedence of the assignment for the WHO 2nd IS in 1996 [5, 6]. In that former study, the geometric mean of gelation assays was 9,600 (CL: 9,300-10,300) IU/vial. The geometric mean for all photometric assays was 11,700 (CL: 11,000-12,400) IU/vial, and the combined mean of all assays was 10,400 (CL: 9,900-10,900) IU/vial. The potency assignment endorsed by the WHO for the 2nd IS was 10,000 IU/vial. In the present study the gelation and photometric assays gave much more similar results, as noted above. In order, as much as possible, to avoid drift during the calibration of future replacement standards, NIBSC, the EDQM, the USP and the participants in the collaborative study recommend that preparation 10/178 (candidate A) be established by ECBS as the new WHO International Standard (3rd IS) for endotoxin with an assigned unitage of 10,000 IU/vial.

WHO/BS/2012.2193 and working document QAS/12.501 Page 10 For information, preparation (10/190) was established in December 2011 as the USP Endotoxin RS (Lot H0K354) with an assigned content of 10,000 EU/vial. Vials of this preparation (10/190) have been presented to the US-FDA for establishing Endotoxin EC-7. Preparation 10/196 will be kept at USP for future use. Preparation 10/178 has been shared with EDQM and will be presented at the Ph. Eur. Commission in June 2012 for adoption as the Ph. Eur. Endotoxin Standard Biological Reference Preparation (BRP) batch 5 with an assigned content of 10,000 IU/vial. Addendum: Data from Additional Laboratories Data were received from an additional four participants after the study deadline had passed. These results were not available for the above analysis. They are presented here as an addendum for completeness. Three of the laboratories provided data from gelation assays, and four from photometric assays, with one laboratory providing data from both chromogenic and turbimetric methods. The results from the individual gelation assays are shown in Table A1. Laboratory 20 used 2 replicates per dilution while the other two laboratories used 4 replicates. All used doubling dilution series. Laboratory 28 provided two sets of results, one using tubes (28T), the other using microtitre plates (28P). Laboratory 28 obtained identical gelation results for the IS and all samples, resulting in estimates of 10,000 IU for all samples, with both the tube and plate methods. Laboratory 20 also had identical results (10,000) except for sample A in one assay that was one dilution step lower (5000). The results from laboratory 33, based on their own calculations, were a little more variable. The laboratory geometric means are shown in table A2. The photometric assays were analysed as parallel line assays, relating the log transformed assay response to log concentration, using the EDQM CombiStats package. Laboratory 33 returned raw assay response data for sample A – D, but only details of fitted standard curves for the standard. The results provided from their own calculations were therefore used. The results from individual photometric assays are shown in table A3. The results from laboratory 6 were highly variable. They used 10-fold dilution steps, which are not ideal for quantitative estimation. They also used an identical plate layout for each assay. It is possible that there were plate effects affecting the assay results. Sample A, which had estimates closest to 10,000 IU, was closest to the IS on the plate. The results from laboratory 20 were also highly variable. They did not use consistent dilution series across the assays, and often samples were tested at only one or two dilutions. This is not ideal for quantitative estimation using the parallel line method. Laboratory 28 used two-fold dilutions, and the calculated results were all highly consistent, and close to 10,000 for all samples in all assays. Laboratory 33 used ten-fold dilutions, and had more variable results. The laboratory geometric means are shown in Table A4. Apart from laboratory 28, and the gelation assays from laboratory 20, the results from the additional participants were quite variable. However, overall, there is no evidence that would indicate a need to modify the consensus values obtained from the main collaborative study reports. Table A1: Gelation Assays: IU/vial - Individual assay results Lab 20 Day 1 2 Vial 1 10000 10000 Sample A Vial 2 Vial 3 5000 10000 10000 10000 Vial 1 10000 10000 Sample B Vial 2 Vial 3 10000 10000 10000 10000 Vial 1 10000 10000 Sample C Vial 2 Vial 3 10000 10000 10000 10000 Vial 1 10000 10000 Sample D Vial 2 Vial 3 10000 10000 10000 10000

WHO/BS/2012.2193 and working document QAS/12.501 Page 11 28P 28T 33 1 2 1 2 1 2 10000 10000 10000 10000 6000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 5000 10000 10000 10000 10000 10000 7000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 7000 10000 10000 10000 10000 10000 8000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 12000 10000 10000 10000 10000 6000 5000 10000 10000 10000 10000 8000 7000 10000 10000 10000 10000 10000 12000

Table A2: Gelation Assays: IU/vial – Geometric mean results by laboratory Lab 20 28P 28T 33 Sample A 8910 10000 10000 8370 Sample B 10000 10000 10000 9490 Sample C 10000 10000 10000 9980 Sample D 10000 10000 10000 8000

Table A3: Photometric Assays: IU/vial - Individual assay results Lab 6 20 28 Method Chrom Turb Chrom Turb 33 Chrom Day 1 2 1 2 1 2 1 2 1 2 Vial 1 9365 7392 10994 10109 10178 14680 16070 9530 10090 Sample A Vial 2 8455 9569 12862 4883 10036 9991 12120 14480 9570 6380 Vial 3 9874 12611 22738 8973 10061 9958 7660 12860 9450 7990 Vial 1 4580 3763 11460 12632 10248 10154 7660 6870 10600 11600 Sample B Vial 2 5095 5594 9785 6448 10097 10098 9390 13880 11280 10120 Vial 3 5178 6698 20526 7705 10013 10132 11280 11810 11370 11390 Vial 1 7645 6629 16004 31062 10168 10141 10510 8540 10710 10290 Sample C Vial 2 7316 7607 9710 12472 10088 10028 14440 21690 10750 9580 Vial 3 7957 10418 31708 13260 10071 10037 15280 15670 10750 7770 Vial 1 5813 4765 14105 30928 10091 9983 10140 8400 10850 11870 Sample D Vial 2 6025 6243 12344 6511 10033 9940 23250 19380 10690 9960 Vial 3 6656 8053 18260 8521 9984 9990 12670 13420 11150 10250

Note: Lab 20 labelled assays 1 – 6, but did not specify vial or day. Results above assume 1-6 represents vial 1 days 1, 2 etc.

Table A4: Photometric Assays: IU/vial – Geometric mean results by laboratory Lab 6 20 28 33 Method Chrom Turb Chrom Turb Chrom Sample A 9417 10709 10055 13782 8826 Sample B 5073 10632 10124 10122 11050 Sample C 7849 17135 10089 14319 9782 Sample D 6183 13268 10003 14525 10800

Proposal On the basis of the results of this collaborative study and on the results of the stability study, it is recommended to adopt preparation 10/178 (preparation A in the collaborative study) as the 3rd international standard for endotoxin with an assigned unitage of 10,000 IU/vial.

Implementation plan

The availability of the 3rd IS for endotoxin as a replacement for the 2nd IS for endotoxin will be made clear on the NIBSC standards web site and the report of the collaborative study will be published in an international scientific journal.

Acknowledgements We wish to thank all participants for contributing valuable data to the study, Dr Paul Matejtschuk and his staff of the Protein Sciences Group, NIBSC, for lyophilisation developmental work and Dr Paul Jefferson and his staff of the Centre for Biological Reference

WHO/BS/2012.2193 and working document QAS/12.501 Page 12 Materials, NIBSC, for lyophilising the preparations and sample despatch. We wish to thank Dr A. Bristow for his advice and support of the Biological Standardisation Programme and Dr L. Soares (Infarmed, Portugal) and Dr Y. Cortez (Afssaps, France) for providing results for the pretesting of the 3 candidate preparations. The contribution of Mrs M. Fernandez (EDQM/Biology Section of the Laboratory Department) is acknowledged. The EDQM contribution to this study was under the aegis of the Biological Standardisation Programme (BSP) of the Council of Europe and the European Commission. The project (coded BSP111) was coordinated by Mr JM. Spieser, Dr KH. Buchheit and Dr E. Terao (EDQM/DBO). Ms S. Woodward ensured excellent secretarial support at EDQM/DBO. The USP contribution was coordinated by Drs. Mary Crivellone, Walter Hauck, Michael Ambrose and Tina Morris with review and approval from the USP Analytical Microbiology Expert Committee (USP 2010-2015 Council of Experts) and support from the USP Reference Standard Production Department, particularly Andrea Iwanik, as well as the USP Biologics & Biotechnology Laboratory.

References

[1] Bacterial endotoxins, general chapter 2.6.14. Ph. Eur. 7th Edition. Strasbourg, France: Council of Europe; 2012(vol. 1). [2] Bacterial Endotoxins Test, General Chapter <85>, USP 35-NF 30. Rockville, USA: United States Pharmacopeial Convention; 2012. [3] Bacterial endotoxins test, general test 4.01. JP XVI. Tokyo, Japan: Ministry of Health, Labour and Welfare. 2011. [4] Rudbach JA, Akiya FI, Elin RJ et al. Preparation and properties of a national reference endotoxin. J Clin Microbiol 1976; 3(1):21-5. [5] Poole S, Gaines Das RE. Report of the collaborative study of the candidate second international standard for endotoxin as agreed by participants. Ref: BS/96.1830 Rev.1. WHO Expert Committee on Biological Standardization; 1996. [6] Poole S, Dawson P, Gaines Das RE. Second international standard for endotoxin: calibration in an international collaborative study. J Endotoxin Res 1997;4(3):221-31.

Abbreviations BRP: Biological Reference Preparation; BSP: Biological Standardisation Programme; CBER: Center for Biologics Evaluation and Research; Chrom: Chromogenic; CL: Confidence Limits; DBO: Department of Biological Standardisation, OMCL Network & HealthCare; ECBS: Expert Committee on Biological Standardization; EDQM: European Directorate for the Quality of Medicines & HealthCare; EU: Endotoxin Unit; FDA: US Food and Drug Administration; GCV: Geometric Coefficient of Variation; GM: Geometric Mean; HPA/NIBSC: Health Protection Agency/ National Institute for Biological Standards and Control; IS: International Standard; IU: International Unit; JP: Japanese Pharmacopoeia; LAL: Limulus Amoebocyte Lysate; OMCL: Official Medicines Control Laboratory; Ph. Eur.: European Pharmacopoeia; USP: United States Pharmacopeia; Turb.: Turbidimetry; WHO: World Health Organization.

WHO/BS/2012.2193 and working document QAS/12.501 Page 13

Table 1 Summary of fill details Preparation 10/178 (Sample A)

Preparation 10/190 (Sample B & D)

Preparation 10/196 (Sample C)

Date of fill N. of vials filled Mean fill mass

September 16, 2010 25,651 0.9995 g CV=0.1506 % n=260 0.15%

October 7, 2010 25,644 0.9996 g CV=0.1920 % n=264 0.19%

October 21, 2010 26,026 0.9990 g CV=0.1790 % n=268 0.18%

Imprecision of the filling (coefficient of variation) Mean residual moisture

0.3659% CV=23.55% n=30 0.55% CV=17.94% n=30 Pre-filled = 0 Post-filled = 0 Post-Freeze-Dried = 0 Pre-filled = 0 Post-filled = 0 Post-Freeze-Dried = 0

0.5246% CV=20.07% n=30 0.71% CV=15.95% n=30 Pre-filled = 0 Post-filled = 0 Post-Freeze-Dried = 0 Pre-filled = 0 Post-filled = 0 Post-Freeze-Dried = 0

0.3060% CV=16.72% n=30 0.66% CV=14.25% n=30 Pre-filled = 0 Post-filled = 0 Post-Freeze-Dried = 0 Pre-filled = 0 Post-filled = 0 Post-Freeze-Dried = 0

Mean oxygen headspace

Microbial analysis Bacterial Colony Count (Cfu/mL, n=4 vials)

Microbial analysis Mould/Yeast Colony Count (Cfu/mL, n=4 vials)

Table 2(a) Accelerated Temperature Degradation for Endotoxin Candidate Standards: potencies (in IU) of accelerated degradation samples relative to the standard Endotoxin curve, and the potencies (in %) of the samples stored at +56°C relative to those stored at -20°C for 10 months. Assay 1 2 Geomean 3 4 Geomean 5 6 Geomean Fill A (10/178) -20°C 10,223 9,872 10,046 B, D (10/190) 11,102 9,984 10,528 C (10/196) 10,798 10,057 10,421 +56°C 10,313 10,736 10,522 10,457 10,868 10,660 10,001 10,911 10,446 +56 as % of -20°C 100.9 108.8 104.7 94.2 108.9 101.3 92.6 108.5 100.2

The mean potency estimates of the fill samples stored at -20°C and +56°C relative to the Endotoxin standard curve range (current IS) from around 10,000 to 10,500 EU for the -20°C samples and 10,400 to 10,700 EU for the +56°C samples. For each of the 3 preparations, there was no observed drop in potency between the samples stored at +56°C and those stored at -20°C after storage for 10 months.

WHO/BS/2012.2193 and working document QAS/12.501 Page 14 Table 2(b) Accelerated Temperature Degradation for Endotoxin Candidate Standards: potencies (in IU) of accelerated degradation samples relative to the standard Endotoxin curve, and the potencies (in %) of the samples stored at +56°C relative to those stored at -20°C for 15 months. All data. Assay 1 2 Geo mean 3 4 Geo mean 5 6 Geo mean Fill 10/178 -20°C 12064 10141 11061 10126 10073 10099 11099 10376 10732 +56°C 10444 10300 10171 10003 11515 10732 9854 11402 10600 +56°C as % of -20°C

92.0

10/190

106.3

10/196

98.8

Potencies for the samples stored at -20°C and +56°C were calculated relative to the current IS using a parallel line analysis. For each fill, two plates were used, reversing the order the test samples were placed on the plate. There was some evidence that the outer columns of the plate were giving faster reaction times than those from the replicates of the same samples in more central columns.

Table 2(c) Accelerated Temperature Degradation for Endotoxin Candidate Standards: potencies (in IU) of accelerated degradation samples relative to the standard Endotoxin curve, and the potencies (in %) of the samples stored at +56°C relative to those stored at -20°C for 15 months. Repeat of parallel line analysis excluding the two outer columns. Assay 1 2 Geo mean 3 4 Geo mean 5 6 Geo mean Fill -20°C 11317 10761 11035 9641 10926 10263 10680 11239 10956 +56°C 10834 9737 10271 10816 10506 10660 10424 10896 10657 +56°C as % of -20°C

10/178

93.1

10/190

103.9

10/196

97.3

WHO/BS/2012.2193 and working document QAS/12.501 Page 15 Table 2(d) Accelerated Temperature Degradation for Endotoxin Candidate Standards: potencies (in IU) of accelerated degradation samples relative to the standard Endotoxin curve, and the potencies (in %) of the samples stored at +56°C relative to those stored at -20°C for 17 months. All data. Assay 1 2 Geo mean 3 4 Geo mean 5 6 Geo mean Fill -20°C 10840 9663 10235 11058 9609 10308 11256 9155 10151 +56°C 9413 9986 9696 9733 9858 9795 8265 8333 8299 +56°C as % of -20°C

10/178

94.7

10/190

95.0

10/196

81.8

Table 2(e) Accelerated Temperature Degradation for Endotoxin Candidate Standards: potencies (in IU) of accelerated degradation samples relative to the standard Endotoxin curve, and the potencies (in %) of the samples stored at +56°C relative to those stored at -20°C for 17 months. Repeat of parallel line analysis excluding the two outer columns. Assay 1 2 Geo mean 3 4 Geo mean 5 6 Geo mean Fill -20°C 10252 10727 10487 10553 10527 10540 10645 9662 10142 +56°C 10126 9682 9902 10484 10585 10534 8765 7842 8291 +56°C as % of -20°C

10/178

94.4

10/190

99.9

10/196

81.7

WHO/BS/2012.2193 and working document QAS/12.501 Page 16 Table 3 Methods used by participating laboratories Lab 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 Gelation X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Chromogenic X X X X Turbidimetric X X X X

X X

WHO/BS/2012.2193 and working document QAS/12.501 Page 17 Table 4 Overview of results per laboratory (gelation assays; IU/vial) Lab 1 2 Day 1 2 1 2 1 2 3 4 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 Sample A Vial 1 Vial 2 Vial 3 10000 10000 11892 10574 10000 10000 10000 10000 10000 14142 10000 16818 10000 10000 10000 10000 10000 20064 8409 10000 10000 10000 10000 10000 7071 10000 10000 10000 10000 10000 10000 13919 8476 14142 20000 10000 10000 10000 10000 11892 10000 10000 10000 10000 8409 10000 11892 11892 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 12910 10000 10000 20064 14142 10000 10000 10000 10000 10000 14142 10000 10000 10000 10000 10000 10000 16685 10000 3536 7071 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 16818 10000 10000 8409 11892 11362 11362 10000 10000 8409 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 13919 5161 10000 20000 10000 20000 10000 10000 14142 10000 10000 10000 10000 10000 10000 10000 Sample B Vial 1 Vial 2 Vial 3 10000 11892 10000 10574 10000 10000 10000 10000 10000 14142 10000 16818 10000 10000 14669 10000 10000 10000 10000 10000 10000 10000 10000 10000 7071 10000 10000 10000 10000 10000 10000 10000 8476 20000 20000 10000 10000 10000 10000 10000 10000 20000 10000 10000 10000 10000 11892 10000 10000 10574 10000 10000 10000 10000 10000 10000 8409 16818 20000 11892 10000 14669 10000 10000 14165 11892 8409 10000 10000 10000 10000 14142 10000 10000 10000 10000 10000 10000 6089 5161 10000 14142 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 20000 10000 11180 10574 10000 10000 10000 10000 10000 20000 16818 10000 8409 20000 12910 16667 10000 14165 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 13919 10000 2500 14142 20000 20000 10000 10000 14142 10000 10000 10000 10000 14142 10000 10000 Sample C Vial 1 Vial 2 Vial 3 11892 11892 10574 11180 10000 10000 10000 10000 10000 14142 20000 16818 11892 11892 10000 6000 10000 20064 10000 11892 10000 10000 10000 10000 7071 20000 10000 10000 10000 10000 10000 38750 11798 5000 20000 10000 10000 10000 10000 11892 5000 14142 10000 10000 10000 7071 10000 11892 10000 10574 10574 10000 8412 10000 10000 10000 8409 20000 20000 14142 10000 10000 10000 4984 14165 14142 10000 10000 10000 10000 10000 14142 10000 10000 10000 10000 10000 10000 8476 10000 14142 14142 5000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10574 10000 10000 10000 10000 10000 20000 20000 10000 4204 5000 11362 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 13919 10000 20000 20000 20000 20000 10000 10000 20000 10000 10000 10000 10000 14142 10000 10000 Sample D Vial 1 Vial 2 Vial 3 11892 10000 10000 10000 10000 10000 10000 10000 10000 14142 11892 16818 10000 10000 10000 10000 10000 10000 8409 11892 10000 10000 10000 10000 7071 20000 10000 10000 10000 10000 10000 10000 8476 20000 20000 10000 10000 10000 10000 11892 10000 14142 10000 7071 10000 10000 11892 10000 10000 10000 10574 10000 10000 10000 10000 10000 10000 20000 20000 14142 11892 16667 7692 4984 14165 16818 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 11798 10000 14142 14142 10000 10000 10000 10000 10000 10000 11892 10000 10000 8409 10000 10000 14142 14142 10574 10574 10000 10000 10000 10000 10000 20000 20000 10000 11892 10000 20000 16667 10000 14165 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 5000 20000 20000 20000 10000 10000 20000 10000 10000 10000 10000 14142 10000 10000

4 5 7 9 10 11 12 13 14 16 18 19 21 23 24 25 29 32 34 35

WHO/BS/2012.2193 and working document QAS/12.501 Page 18 Table 5 Overview of mean results per laboratory (gelation assays; IU/vial) Laboratory 1 2 3 4 5 7 9 10 11 12 13 14 16 18 19 21 23 24 25 29 32 34 35 GM GCV * Huber’s robust mean (K=1.5) Sample A 10293 10389 10000 11554 10905 10000 10889 12613 10000 10000 10000 10000 10000 10000 10000 10595 10595 11225 10000 10905 10000 9715 10293 10 414 6.2 10 250 Sample B 11554 10476 10000 11554 14557 11225 12910 11231 10000 10000 10000 10000 10000 10000 10000 8476 11225 12599 10000 10595 11225 10595 10293 10 739 11.2 10 598 Sample C 10905 10574 9857 11554 17311 8655 9381 10597 10905 10000 10000 10000 11225 10000 10000 13243 14142 11225 10000 10293 10595 10595 9439 10 768 14.8 10 509 Sample D 11554 10283 10000 11892 15874 11225 12739 10000 10905 10000 10000 8909 11225 10000 10000 10000 14142 12599 10000 11554 10905 9715 10293 10 937 13.4 10 648

N.B. There co-exist 2 different definitions of the GCV in the published literature. They are not equivalent and caution is advised

when comparing the GCV’s from different studies. The definition used in this report is GCV=(10v-1)0.5‧100%, whereas some other publications use the definition GCV=(10s-1) ‧ 100%. In these equations s2 = v is the sample variance of the log10transformed potencies.

WHO/BS/2012.2193 and working document QAS/12.501 Page 19 Figure 1 Histograms of mean results per laboratory (gelation assays; IU/vial, Huber’s robust means) Sample A 32 25 19 18 16 14 13 12 11 7 34 3 10000 7500 8000 8500 9000 9500

35 23 21 2 1 10500

29 24 9 5 4 11000 11500 12000

10 12500 13000 13500 14000 14500 15000 15500 16000 16500 17000 17000 17000 17000 17500 17500 17500 17500

Sample B 25 19 18 16 14 13 12 11 3 10000 9000 9500

21 7500 8000 8500

35 34 29 2 10500

32 23 10 4 7 1 11000 11500 12000

24 9 12500 13000 13500 14000

5 14500 15000 15500 16000 16500

Sample C 25 19 18 14 13 35 12 9 3 10000 9000 9500

7 7500 8000 8500

34 32 29 10 2 10500

24 16 11 1 4 11000 11500 12000 12500

21 13000 13500

23 14000 14500 15000 15500 16000 16500

Sample D 25 21 19 18 13 32 12 16 10 35 11 29 24 14 34 3 2 7 1 4 9 10000 10500 11000 11500 12000 12500 13000 13500 7500 8000 8500 9000 9500

23 14000 14500 15000 15500

5 16000 16500

Numbers in the boxes are the laboratory codes.

WHO/BS/2012.2193 and working document QAS/12.501 Page 20 Table 6 - Overview of the mean results per laboratory (photometric assays; IU/vial) Lab Method Chrom 1 Turb 2 Turb Day 1 2 1 2 1 2 1 2 3 4 1 2 3 4 1 2 1 2 1 2 1 2 1 1 2 2 1 2 1 1 2 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 Vial 1 9267 9913 13167 10691 9863 10571 10055 10820 12131 11424 9795 9913 10786 10530 9947 10380 8766 9094 8752 9123 10830 10568 9680 10188 10299 10951 9982 10263 10400 10448 9773 10105 9984 10179 8815 10118 8568 8812 11584 10224 10518 9629 10967 9524 6830 9704 11173 10802 10734 8723 8139 8040 10024 10897 11715 16683 8288 8609 8791 10293 10476 9939 10274 10442 9740 Sample A Vial 2 10356 10159 21914 20343 11062 10793 10062 9867 12151 10732 9844 9718 10338 10507 9985 9294 9483 10045 9580 11218 10125 10678 10660 10220 10497 10372 10423 11135 11450 11247 11814 11506 9341 9956 9210 10222 9003 9085 10286 10418 5444 4203 9616 12012 10449 8606 9529 8875 11913 11820 7784 6581 8018 8914 7872 7662 9097 8215 8196 11028 9686 10120 9737 9864 9812 Vial 3 12357 11525 8092 9676 12629 12990 9985 11709 12019 12345 10026 10437 11176 10408 9906 10492 9560 9039 11546 12250 10878 10632 10323 10548 10940 10250 10054 10581 10922 11102 10881 11463 9149 9682 9730 9555 8785 10513 11753 12881 11068 9208 11323 10194 8773 10783 9517 10811 11441 7729 9123 8974 8936 10328 11042 17184 7648 7293 8284 9732 9862 10564 10515 10141 10223 Vial 1 10181 10974 16419 14942 11197 11169 10068 10526 13781 10616 9404 9625 12745 9971 10506 11210 8982 9307 9139 10564 10806 9266 9340 10362 10326 10493 9831 10489 11999 11806 10597 10866 9181 9157 8215 10203 8351 7901 12738 12367 12006 14353 10176 11895 10024 8509 11320 8717 11277 12324 9124 8840 8860 10951 9357 10323 21900 9568 8856 9729 9831 10433 9076 10874 9833 10330 Sample B Vial 2 10385 10032 14209 14308 10471 10995 10538 10697 12197 10839 10304 10432 10388 10212 10458 9314 9271 9747 10775 11378 11145 11480 11617 10803 10305 11246 10904 10471 10968 10741 11665 11636 9706 9750 8946 10405 7727 10027 12525 11782 10316 14482 11242 12768 11037 9503 8963 8457 11050 11554 8458 7350 8050 8679 10477 7595 8449 8880 9094 10820 9916 10729 10091 10219 12006 Vial 3 13768 11412 9812 17316 12722 13134 9290 12186 12405 13511 10639 10621 12199 10156 10449 11282 11155 9005 9524 11462 11517 11645 9628 9660 10911 10204 9504 11026 11547 11469 11834 12649 9495 10184 9663 9628 9077 10116 13370 14649 13224 9793 11153 9509 8406 10093 10140 10303 10873 8325 8998 9518 8956 11644 7331 17182 11077 8008 9114 8940 9524 10656 10256 10120 9997 Vial 1 10172 11178 17933 16248 10775 11162 10007 10606 11240 11039 9461 9929 10065 11781 10924 11778 9464 9313 10948 10470 11371 11297 9716 11146 10866 10344 10632 11277 10956 11227 10309 10775 9613 9032 9773 10444 9335 8703 12721 13315 10884 11421 10244 13437 9708 7378 13832 9269 10671 10867 9264 8867 8139 13683 13306 12873 17802 12427 8548 9812 10086 10916 8251 8437 8137 7925 Sample C Vial 2 11405 11205 16054 15793 10469 10580 9147 8958 10213 10165 8990 9084 9507 9626 11014 9893 10009 10297 9996 10187 10766 11512 10698 10774 10709 11093 10271 10678 10392 10503 11487 11956 9080 9696 9214 10006 10158 11072 12316 12213 10957 14005 12961 14792 11172 8676 9308 8204 7842 8195 8137 7268 7870 8667 12052 13627 12225 9194 8826 11718 10004 11765 10176 10713 11376 Vial 3 11473 11747 12527 17723 11185 14250 8768 12075 10681 13566 11100 10868 12104 10981 9398 10127 11176 8940 9016 10445 11598 10520 10718 10250 10799 10862 9722 10220 11167 11060 12075 11878 10291 10913 9362 9990 6586 8427 13105 15565 13036 11053 13118 11538 9222 10716 10740 10746 10392 8151 9623 9102 8853 13252 11940 23848 12775 8127 8328 10062 8784 11018 11252 11052 10736 Vial 1 12658 11842 20679 17599 10229 9631 11095 10550 12361 14296 10010 10656 11329 12562 11028 12503 11090 10704 10270 9272 11295 11156 10621 11630 10258 10584 11014 11483 9991 10569 10555 10855 8281 8500 8377 10479 6089 7008 13483 13727 12738 13040 10490 12227 8532 9067 10619 9477 11323 9696 10442 9341 9007 11616 10339 10149 23665 16755 8636 9673 10677 10914 9289 10407 8642 10041 Sample D Vial 2 12332 11623 16072 19195 10175 10268 10634 12230 13362 13506 10475 11093 12154 12015 11060 8256 10550 11393 10338 10274 11288 11466 10473 10901 10814 10555 10570 11145 11126 11395 11727 11945 8285 9253 9273 11192 6402 8477 12710 12326 10255 14180 13475 15883 17847 10684 9604 9398 9541 9175 8842 7675 8942 9044 10381 18067 13113 9016 11489 10501 12115 10297 8260 9868 10494 Vial 3 13023 11583 13791 19114 11982 9884 13292 12327 12737 11263 11549 12640 12009 11183 11742 13669 10091 8912 10867 11106 10901 9888 10421 10941 10710 10171 10208 10963 10681 11930 12123 9580 10434 9802 9857 7691 9135 14944 16560 14074 11084 15680 10731 9015 10407 10515 9070 6811 8558 9971 9909 9789 16091 11704 24310 13475 8227 8369 10344 9604 9817 8998 8888 8769

Chrom 3 Turb

Chrom 4 Turb 8 10 Chrom Turb

Chrom 11 Turb

Chrom

13 14 15 17 22 23 24 26 27

Turb Chrom Turb Turb Chrom Chrom Chrom Chrom Chrom Chrom

30 Turb 31 32 Chrom Turb Chrom 34 Turb Chrom 35 Turb

Chrom.: chromogenic assay. Turb: turbidimetric assay

WHO/BS/2012.2193 and working document QAS/12.501 Page 21 Table 7 Overview of the mean results per assay (photometric assays; IU/vial) Lab 1 3 4 8 11 12 14 22 23 24 26 27 30 31 34 35 1 2 3 4 10 11 13 15 17 30 32 34 35 GM GCV

Method Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Chromogenic Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Turbidimetric Chromogenic (N=16)

Sample A 10547 11070 9993 10328 10405 10909 9596 7709 10409 8976 9899 11243 7971 10277 8217 10187 13038 11264 10281 9322 10615 10399 9708 9106 11345 8792 10402 10168 10034 9798 11.7 10292 10.4 10017 11.2 10120

Sample B 11059 11311 10516 10436 10388 11467 9480 12776 11127 9456 9564 11213 8494 9712 8932 10262 14278 11576 10520 9550 10944 10356 9572 8816 12874 9126 11695 9892 10394 10333 10.6 10641 13.9 10470 12.0 10404

Sample C 11185 10461 10492 10159 10658 11134 9789 11998 12506 9508 10201 9699 8515 12671 8788 10055 15940 11337 10243 9840 11170 10456 9749 8930 13161 9214 14964 10222 9884 10426 11.3 10986 18.2 10673 14.7 10449

Sample D 12165 12117 10874 9966 10642 11136 9790 12773 12975 10611 9990 9168 9092 11546 9173 9481 17587 10428 11451 11195 11201 10754 9022 7389 13888 9676 17702 10666 9424 10646 12.0 11229 25.2 10904 18.8 10730

GM GCV

Turbidimetric (N=13)

GM GCV

Huber’s robust mean (k=1.5)

Pooled (N=29)

GM: geometric mean; GCV: geometric coefficient of variation (There co-exist 2 different definitions of the GCV in the published literature. They are not equivalent and caution is advised when comparing the GCV’s from different studies. The definition used in this report is GCV=(10v-1)0.5‧100%, whereas some other publications use the definition GCV=(10s-1) ‧100%. In these equations s2 = v is the sample variance of the log10-transformed potencies.

WHO/BS/2012.2193 and working document QAS/12.501 Page 22 Figure 2 Histograms of mean results per laboratory (photometric assays; IU/vial, Huber’s robust means) Sample A 32 11 10 3 31 23 11 27 8 12 17 1 3 2 10500 11000 11500 12000 12500

34 22 30 7500 8000 8500

35 34 30 13 35 15 4 26 24 14 4 9000 9500 10000

1 13000 13500 14000 14500 15000 15500 16000 16500 17000 17000 17000 17000 17500

Sample B

13 4 31 30 26 15 24 30 34 14 34 10000 7500 8000 8500 9000 9500

35 11 3 35 11 8 4 10500

10 27 23 1 11000

32 2 12 3 11500 12000 12500

17 22 13000 13500 14000

1 14500 15000 15500 16000 16500 17500

Sample C

35 34 4 3 35 30 13 26 15 27 14 30 34 24 8 10000 7500 8000 8500 9000 9500

11 11 10 4 12 31 3 1 2 22 23 17 10500 11000 11500 12000 12500 13000 13500 14000 14500

32 15000 15500

1 16000 16500 17500

Sample D

15 7500 8000 8500

13 34 34 35 26 2 30 30 14 24 27 35 8 11 10000 10500 9000 9500

11 10 4 12 3 3 4 31 1 11000 11500 12000 12500

23 22 13000 13500

17 14000 14500 15000 15500 16000 16500

32 1 17500

Numbers in the boxes are the laboratory codes. Chromogenic assays are shown in light-grey. Turbidimetric assays are shown in dark-grey.

WHO/BS/2012.2193 and working document QAS/12.501 Page 23 Table 8 Overall potency of the 3 candidate batches (gelation and photometric assays) Potency (IU/vial) Sample A B C D B&D Estimate 10 190 10 588 10 715 10 900 10 743 95% CL (9927 - 10 461) (10 252 - 10 935) (10 289 - 11 159) (10 414 - 11 410) (10 356 - 11 145)

WHO/BS/2012.2193 and working document QAS/12.501 Page 24 Figure 3 Two-way comparison between batches (IU/vial) Sample B versus Sample A 18000 18000

Sample C versus Sample B 17000 16000 15000 14000 Sample C

17000 16000 15000 14000 Sample B

13000

12000 11000 10000 9000

8000 7000

Sample A

Sample B

Sample C versus Sample A 18000 18000

Sample D versus Sample B 17000 16000 15000 14000 Sample D

17000 16000 15000 14000 Sample C

13000

12000 11000 10000 9000

8000 7000

Sample A

Sample B

Sample D versus Sample A 18000 17000 18000 17000

Sample D versus Sample C

16000 15000 14000 Sample D

13000

12000 11000 10000 9000

8000 7000

Sample A

Sample C

WHO/BS/2012.2193 and working document QAS/12.501 Page 25

Appendix 1 Participants (in alphabetical order of country) 1. 2. 3. 4. 5. 6. 7. Dr D. Pullirsch, Austrian Agency for Health and Food Safety (AGES), Austria Dr C. Rolls, Dr A. Smith, Therapeutic Goods Administration (TGA), Australia Mr O. Carabin, Mrs G. Waeterloos, Scientific Institute of Public Health (IPH), Belgium Prof S. Dalmora, Universidade Federal de Santa Maria, Brazil Mrs N. Fortin, Health Canada, Canada Dr H. Gao, National Institutes for Food and Drug Control (NIFDC), China Dr G. Rautmann, Mrs M. Fernandez, European Directorate for the Quality of Medicines & HealthCare (EDQM), Council of Europe 8. Mrs K. Vorup, Leo Pharma A/S, Denmark 9. Dr U.B. Westergaard, Statens Serum Institut (SSI), Denmark 10. Dr Y. Cortez, Agence nationale de sécurité du medicament et des produits de santé (ANSM), France 11. Dr S. Deutschmann, Dr H. Kavermann, Roche Diagnostics GmbH, Germany 12. Dr A. Ruland, Pyroquant Diagnostik GmbH, Germany 13. Dr I. Spreitzer, Paul Ehrlich Institut (PEI), Germany 14. Dr P. Weidner, Acila AG, Germany 15. Dr C. von Hunolstein, Dr M. Boccanera, Dr L. Campitelli, Istituto Superiore di Sanità (ISS), Italy 16. Dr K. Sugiyama, National Institute of Health Sciences (NIHS), Japan 17. Dr T. Nakatani, Lonza Japan Ltd, Japan 18. Dr Y. Nakagawa, Japanese Pharmacopoeia Reference Standards Laboratory, Pharmaceutical and Medical Device Regulatory Science Society of Japan (PMRJ), Japan 19. Dr T. Oda, Seikagaku Biobusiness Corporation, Japan 20. Ms A. Takaoka, Wako Pure Chemical Industries Ltd, Japan 21. Mr H. Min, Charles River Laboratories, Korea 22. Dr J. Joung, Korea Food and Drug Administration (KFDA), Korea 23. Dr S. R. Andersen, Norwegian Medicine Agency (NoMA), Norway 24. Dr L. Soares, Infarmed IP, Portugal 25. Dr K. Erlandsson-Persson, Medical Products Agency (MPA), Sweden 26. Mr P. Lang, Dr H. Rockstroh, Hoffmann-La Roche AG, Switzerland 27. Dr P. Bruegger, Novartis Pharma AG, Switzerland 28. Dr B. de Vries, Dr J. Berger, National Institute for Public Health and the Environment (RIVM), The Netherlands 29. Dr M. Dawson, Dr B. Markley, Associates of Cape Cod Inc., USA 30. Dr R. Deschenes, Amgen Inc., USA 31. Dr J. Schultz, Charles River Laboratories Inc., USA 32. Mr J.L. Kenney, US Food and Drug Administration (FDA), USA 33. Dr G. Johnson, Lonza Walkersville Inc., USA 34. Dr J. Mudrick, Medimmune Inc., USA 35. Dr M. Ambrose, Dr M. Crivellone, Dr T. Morris, Unites States Pharmacopeial Convention, USA

WHO/BS/2012.2193 and working document QAS/12.501 Page 26

Appendix 2 Collaborative study protocol and data sheets The standard for all assays will be the WHO 2nd International Standard (IS) for Endotoxin (94/580, 10,000 IU = EU/vial). Each assay will include dilutions of (reconstituted) vials of the IS and all 4 test preparations (candidate standards) using a dilution scheme that will be provided: please see the data sheets provided. The test preparations (candidate standards) were all filled at a nominal 10,000 IU/EU vial and are to be tested at the same nominal concentrations and at the same number of replicates as the IS, i.e. the test preparations are to be treated exactly as if they were the IS itself. Participants performing semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assays: 3 vials of the IS and 3 vials of each of the 4 test preparations (candidate standards) will each be assayed twice, once using freshly reconstituted vials and once using vials within 2 weeks of their reconstitution. Thus, 15 vials in total (3 vials of IS and 3 vials of each of the 4 test preparations) will each be assayed twice in total in LAL gelation assays. The protocol used is to be in accordance with published Pharmacopoeial methods (e.g. Ph. Eur. general text 2.6.14., USP general test <85>, etc.) and the assays are to be performed with the LAL reagent routinely used by the laboratory, and having a sensitivity of 0.03 or 0.06 EU/mL. Participants performing quantitative Limulus Amoebocyte Lysate (LAL) Photometric Assays (Chromogenic/Turbidimetric): 3 vials of the IS and 3 vials of each of the 4 test preparations (candidate standards) will each be assayed twice, once using freshly reconstituted vials and once using vials within 2 weeks of their reconstitution. Thus, 15 vials in total (3 vials of IS and 3 vials of each of the 4 test preparations) will each be assayed twice in total in LAL photometric assays. The protocol used is to be in accordance with published Pharmacopoeial methods (e.g. Ph. Eur. general text 2.6.14., USP general test <85>, etc.). N.B. 30 vials altogether (6 vials of IS and 6 vials of each of the 4 test preparations) will be provided for the above assays.

WHO/BS/2012.2193 and working document QAS/12.501 Page 27 Data Sheet for semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assay Number 1/2/3 to be carried out using freshly reconstituted vials DATE (MM/DD/YY) _________________ LABORATORY NAME __________________________ LYSATE MANUFACTURER AND LOT # __________________________ _______ LAL SENSITIVITY: _______ * NEGATIVE CONTROL WAS: ( - ) OR ( + ) [CIRCLE ONE] Assay Number 1/2/3 ANALYST NAME __________________________________________

GELATION ASSAY

This assay to be carried out using a freshly reconstituted vial of the international standard (IS) coded 94/580-IS and a freshly reconstituted vial of each of the test preparations coded A, B, C, and D. Reconstitute a vial of the current international standard for endotoxin 94/580-IS with 5 ml LAL reagent water. Vortex thoroughly (30 min). This will generate the endotoxin standard stock solution (2000 IU [=EU]/ml). Reconstitute one vial each of A, B, C and D with 5 ml LAL reagent water. Vortex each vial thoroughly (30 min). This will generate the stock solutions of preparations A, B, C and D. Starting with the endotoxin standard stock solution of the IS (2000 IU [=EU]/ml), prepare a dilution series for testing according to your organisation’s SOP that is compliant with your local Pharmacopoeia. Starting with the stock solutions of preparations A, B, C and D, prepare dilution series of each for testing exactly as was done for the endotoxin standard stock solution, i.e. exactly the same dilution steps. (This means, treating the stock solutions of preparations A, B, C and D exactly as if they were the (2000 IU [=EU)]/ml) endotoxin standard stock solution.) Carry out the semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assay according to your organisation’s SOP that is compliant with your local Pharmacopoeia and complete the result sheet below with the results obtained. Complete the data sheet below. N.B. There is room on the data sheet below for up to 4 replicates of each of up to 8 dilutions of each stock solution to be tested but you should test the number of replicates of the number of dilutions specified in your organisation’s SOP that is compliant with your local Pharmacopoeia. The objective is for your assay to identify the dilution of each stock solution at which (+) results change to (–) results. For the dilutions of stock solution of 95/580-IS tested please also give the endotoxin concentration in IU/ml. (Please note that 1 IU = 1 EU.)

WHO/BS/2012.2193 and working document QAS/12.501 Page 28 Assay Sheet for semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assay Number 1/2/3 to be carried out using freshly reconstituted vials (N.B. 1 IU = 1 EU) 94/580-IS (10,000 IU/vial) Dilutions of stock Replicate #1 Replicate #2 Replicate #3 Replicate #4 solution tested (e.g. Result (+ or Result (+ or Result (+ or Result (+ or 1:100,000, etc.) -) -) -) -) 1. ( IU*/ml) 2. ( IU*/ml) 3. ( IU*/ml) 4. ( IU*/ml) 5. ( IU*/ml) 6. ( IU*/ml) 7. ( IU*/ml) 8. ( IU*/ml) Vial A Dilutions of stock Replicate #1 Replicate #2 Replicate #3 Replicate #4 solution tested (same as Result (+ or Result (+ or Result (+ or Result (+ or for IS) -) -) -) -) 1. 2. 3. 4. 5. 6. 7. 8. Vial B Dilutions of stock Replicate #1 Replicate #2 Replicate #3 Replicate #4 solution tested (same as Result (+ or Result (+ or Result (+ or Result (+ or for IS) -) -) -) -) 1. 2. 3. 4. 5. 6. 7. 8. Vial C Dilutions of stock Replicate #1 Replicate #2 Replicate #3 Replicate #4 solution tested (same as Result (+ or Result (+ or Result (+ or Result (+ or for IS) -) -) -) -) 1. 2. 3. 4. 5. 6. 7.

WHO/BS/2012.2193 and working document QAS/12.501 Page 29 8. Vial D Dilutions of stock solution tested (same as for IS) 1. 2. 3. 4. 5. 6. 7. 8.

Replicate #1 Result (+ or -)

Replicate #2 Result (+ or -)

Replicate #3 Result (+ or -)

Replicate #4 Result (+ or -)

WHO/BS/2012.2193 and working document QAS/12.501 Page 30 Data Sheet for semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assay Number 4/5/6 to be carried out using stock solutions with 2 weeks of their reconstitution DATE (DD/MM/YY) _________________ LABORATORY NAME __________________________ LYSATE MANUFACTURER AND LOT # __________________________ _______ LAL SENSITIVITY: _______ * NEGATIVE CONTROL WAS: ( - ) OR ( + ) Assay Number 4/5/6 ANALYST NAME __________________________________________ DATE OF RECONSTITUTION OF VIALS (DD/MM/YY) ___________________________ [CIRCLE ONE]

GELATION ASSAY

This assay, Number 4/5/6, is to be carried out using the stock solutions of the international standard (IS) coded 94/580-IS and the stock solutions of the test preparations coded A, B, C, and D that were first used in Gelation Assay Number 1/2/3 and then stored at +4 degrees C for not more than 2 weeks after their reconstitution. Starting with the previously stored endotoxin standard stock solution of the IS (2000 IU [=EU]/ml), prepare a dilution series for testing according to your organisation’s SOP that is compliant with your local Pharmacopoeia. Starting with the previously stored stock solutions of preparations A, B, C and D, prepare dilution series of each for testing exactly as was done for the endotoxin standard stock solution, i.e. exactly the same dilution steps. (This means, treating the stock solutions of preparations A, B, C and D exactly as if they were the (2000 IU [=EU)]/ml) endotoxin standard stock solution.) Carry out the semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assay according to your organisation’s SOP that is compliant with your local Pharmacopoeia and complete the result sheet below with the results obtained. Complete the data sheet below. N.B. There is room on the data sheet below for up to 4 replicates of each of up to 8 dilutions of each stock solution to be tested but you should test the number of replicates of the number of dilutions specified in your organisation’s SOP that is compliant with your local Pharmacopoeia. The objective is for your assay to identify the dilution of each stock solution at which (+) results change to (–) results. For the dilutions of stock solution of 95/580-IS tested please also give the endotoxin concentration in IU/ml. (Please note that 1 IU = 1 EU.) Assay Sheet for semi-quantitative Limulus Amoebocyte Lysate (LAL) Gelation Assay Number 4/5/6 94/580-IS (10,000 IU/vial) Dilutions of stock Replicate #1 Replicate #2 Replicate #3 Replicate #4 solution tested (e.g. Result (+ or Result (+ or Result (+ or Result (+ or 1:100,000, etc.) -) -) -) -)

WHO/BS/2012.2193 and working document QAS/12.501 Page 31 1. ( IU*/ml) 2. ( IU*/ml) 3. ( IU*/ml) 4. ( IU*/ml) 5. ( IU*/ml) 6. ( IU*/ml) 7. ( IU*/ml) 8. ( IU*/ml) Vial A Dilutions of stock solution tested (same as for IS) 1. 2. 3. 4. 5. 6. 7. 8. Vial B Dilutions of stock solution tested (same as for IS) 1. 2. 3. 4. 5. 6. 7. 8. Vial C Dilutions of stock solution tested (same as for IS) 1. 2. 3. 4. 5. 6. 7. 8. Vial D Dilutions of stock solution tested (same as for IS) 1. 2. 3.

Replicate #1 Result (+ or -)

Replicate #2 Result (+ or -)

Replicate #3 Result (+ or -)

Replicate #4 Result (+ or -)

Replicate #1 Result (+ or -)

Replicate #2 Result (+ or -)

Replicate #3 Result (+ or -)

Replicate #4 Result (+ or -)

Replicate #1 Result (+ or -)

Replicate #2 Result (+ or -)

Replicate #3 Result (+ or -)

Replicate #4 Result (+ or -)

Replicate #1 Result (+ or -)

Replicate #2 Result (+ or -)

Replicate #3 Result (+ or -)

Replicate #4 Result (+ or -)

WHO/BS/2012.2193 and working document QAS/12.501 Page 32 4. 5. 6. 7. 8.

WHO/BS/2012.2193 and working document QAS/12.501 Page 33 PHOTOMETRIC ASSAY (Chromogenic/Turbidimetric) DATE (MM/DD/YY) _________________ LABORATORY NAME __________________________ LYSATE MANUFACTURER AND LOT # __________________________ _______ LAL SENSITIVITY: _______ * NEGATIVE CONTROL WAS: ( - ) OR ( + ) [CIRCLE ONE] Assay Number 1/2/3 ANALYST NAME __________________________________________

This assay to be carried out using a freshly reconstituted vial of the international standard (IS) coded 94/580-IS and a freshly reconstituted vial of each of the test preparations coded A, B, C, and D. Reconstitute a vial of the current international standard for endotoxin 94/580-IS with 5 ml LAL reagent water. Vortex thoroughly (30 min). This will generate the endotoxin standard stock solution (2000 IU [=EU]/ml). Reconstitute one vial each of A, B, C and D with 5 ml LAL reagent water. Vortex each vial thoroughly (30 min). This will generate the stock solutions of preparations A, B, C and D. Starting with the endotoxin standard stock solution of the IS (2000 IU [=EU]/ml), prepare a dilution series for testing according to your organisation’s SOP that is compliant with your local Pharmacopoeia. Starting with the stock solutions of preparations A, B, C and D, prepare dilution series of each for testing exactly as was done for the endotoxin standard stock solution, i.e. exactly the same dilution steps. (This means, treating the stock solutions of preparations A, B, C and D exactly as if they were the (2000 IU [=EU)]/ml) endotoxin standard stock solution.) Carry out the semi-quantitative Limulus Amoebocyte Lysate (LAL) Photometric Assay according to your organisation’s SOP that is compliant with your local Pharmacopoeia and complete the result sheet below with the results obtained. N.B. There is room on the data sheet below for up to 3 replicates of each of up to 5 dilutions of each stock solution to be tested but you should test the number of replicates of the number of dilutions specified in your organisation’s SOP that is compliant with your local Pharmacopoeia. The objective is obtain curves for dilutions of 95/580-IS and preparations A, B, C and D that will allow a parallel line analysis of the data to be carried out. Please state the readout for the assay and the units in which it is measured: Readout………………………………………units………………………………………… Assay Sheet for semi-quantitative Limulus Amoebocyte Lysate (LAL) Photometric (Chromogenic or Turbidimetric) Assay Number 1/2/3 to be carried out using freshly reconstituted vials (N.B. 1 IU = 1 EU)

WHO/BS/2012.2193 and working document QAS/12.501 Page 34 94/580-IS (10,000 IU/vial) Dilutions of stock solution tested (e.g. 1:100,000, etc.) 1. ( IU*/ml) 2. ( IU*/ml) 3. ( IU*/ml) 4. ( IU*/ml) 5. ( IU*/ml) *: 1 IU = 1 EU Negative control: water Vial A Dilutions of stock solution tested (e.g. 1:100,000, etc.) (Dilutions same as for IS) 1. 2. 3. 4. 5.