Safety and Efficacy Testing of Hormones and Related Products

The Report and Recommendations of ECVAM Workshop 91,2

Reprinted with minor amendments from ATLA 23, 699-712

Bernward Garthoff3, Coenraad Hendriksen4, Alain Bayol5, Daniéele Goncalves6, Andreas Grauer7, Renato de Leeuw8, Jan van Noordwijk9, Monique Pares10, Raffaella Pirovano11, Michael Rieth12, Hansjörg Ronneberger13, Jean-Marc Spieser14, Patrick Storring15, Klaus Vosbeck16, and Henry Weichert17
3Bayer AG, PF Centre Monheim, 51368 Leverkusen, Germany; 4National Institute of Public Health and Environmental Protection (RIVM), P.O. Box 1, 3720 BA Bilthoven, The Netherlands; 5Sanolfi Elf Bio Recherches, Labége Inolpole-Voie No. 1, BP 137, 31676 Labége Cedex, France; 6Dev. Analytique, Laboratoires CASSENNE 17, rue de Pontoise, BP 31, 95521 Cergy-Pontoise Cedex, France; 7Med. Klinik und Poliklinik der Universität Heidelberg, Abt. Innere Medizin I, Endokrinologie, Bergheimer Str. 58, 69115 Heidelberg, Germany; 8Dept. of Endocrinology, N.V. Organon, P.O. Box 20, 5340 BH Oss, The Netherlands; 9European Pharmacopoeia Commission, Taveernlaan 15, 3735 KA Bosch en Duin, The Netherlands; 10Agence du Médicament, Unité Pharmacologie-Toxicologie, 14 Rue Ecole de Pharmacie, 34000 Montpellier, France; 11Instituto di Ricerche Biomediche, "Antoine Marxer" SpA, Via Rives 1, 10010 Colleretto Giacosa (TO), Italy; 12Schering AG, PH-QS Biologische Qualitätsprüfung, Müllerstr. 178, P.O. Box 650311, 13353 Berlin, Germany; 13SGE Therapeutika, Forschung Pharmakologie/Toxikologie, Behringweke AG, P.O. Box 1140, 35001 Marburg/Lahn, Germany; 14European Pharmacopoeia Commission, B.P. 907, 67029 Strasbourg Cedex 1, France; 15National Institute for Biological Standards and Control (NIBSC), Blanche Lane, South Mimms, Potters Bar, Herfordshire EN6 3QG, UK; 16CIBA-Geigy AG, K-122.4.33, 4002 Basel, Switzerland; 17CTL CellTechnologie GmbH, Südstrasse 55, 04403 Böhlitz-Ehrenberg, Germany

1ECVAM - The European Centre for the Validation of Alternative Methods. 2This document represents the agreed report of the participants as individual scientists.

Address for correspondence: Dr Bernward Garthoff, Bayer AG, PF Center Monheim, 51368 Leverkusen-Bayerwerk, Germany

Address for reprints: ECVAM, TP 580, JRC Environment Institute, 21020 Ispra (VA), Italy

Preface

Preface

This is the report of the ninth of a series of workshops organised by the European Centre for the Validation of Alternative Methods (ECVAM). ECVAM's main goal, as defined in 1993 by its Scientific Advisory Committee, is to promote the scientific and regulatory acceptance of alternative methods which are of importance to the biosciences and which reduce, refine or replace the use of laboratory animals. One of the first priorities set by ECVAM was the implementation of procedures which would enable it to become well-informed about the state-of-the-art of non-animal test development and validation, and the potential for the possible incorporation of alternative tests into regulatory procedures. It was decided that this would be best achieved by the organisation of ECVAM workshops on specific topics, at which small groups of invited experts would review the current status of various types of in vitro tests and their potential uses, and make recommendations about the best ways forward (1).

The workshop on Safety and Efficacy Testing of Hormones and Related Products was held in Düsseldorf, Germany, on 25-27 November 1994. The aims of the workshop were to determine the current status and to outline future directions in this area. In particular, the possibilities for reducing the current use of laboratory animals in this area, and for promoting recent developments in alternative methods for testing such products, were discussed. The workshop was held under the chairmanship of Jan van Noordwijk (European Pharmacopoeia Commission, Netherlands) and was organised by Bernward Garthoff (Bayer, Germany) and Coenraad Hendriksen (RIVM, The Netherlands), both of whom are members of the ECVAM Scientific Advisory Committee. The participants were representatives from industry, research institutes, national organisations, and regulatory agencies.

Introduction

The safety and efficacy testing of hormones and related products are regulated primarily by pharmacopoeia requirements. In the European Union (EU), these testing requirements are described in the European Pharmacopoeia. A casual glance at the index of the European Pharmacopoeia might give the impression that there have been no advances regarding the replacement of bioassays and biological safety tests since the production of the first European Pharmacopoeia monographs in 1964. The tests for abnormal toxicity, depressor substances, and pyrogens, are still included. However, three developments are reducing the use of such tests in quality control of hormones and related products, and these should be reflected in future European Pharmacopoeia monographs:

  1. Synthetic chemistry and biotechnology processes mean that natural hormones are being replaced with synthetic or recombinant DNA (rDNA)-derived products of greater purity, thus eliminating the need for biological safety tests.
  2. Analytical chemistry (notably high performance liquid chromatography [HPLC]) and cell culture methods are eliminating the need for in vivo bioassays, or are limiting their use to the validation of the method of manufacture; production batches are monitored with physicochemical tests.
  3. Some collaborative studies relating to the establishment of a biological standard or reference preparation have included a physicochemical or cell culture method, to assess their value as a possible replacement for the in vivo bioassay. This has been undertaken by the World Health Organisation (WHO) and by the European Pharmacopoeia Commission. For example, the evaluation of in vitro methods is a regular feature of the European Pharmacopoeia Biological Standardisation Programme.

These important developments were discussed during the workshop, in addition to discussions on the specific contributions of the participants.

Guidelines for the Humane Treatment and Care of Animals Used in the Quality Control of Hormones and Other Biological Products

Although the number of opportunities to replace animal tests is increasing, it is believed that we cannot replace all animal tests in the near future. It is recommended that guidelines should be defined for the humane treatment and care of laboratory animal used in hormone quality control. These are presented in Appendix A. They express the concern of the scientific community for the welfare of experimental animals, and set guidelines for their humane treatment and for the implementation of the Three Rs concept. It is envisaged that these guidelines will act as a model for international regulatory agencies and national control authorities, to be adapted as appropriate. In addition, it is hoped that such bodies will specifically consider the design of animal tests, and the application of the Three Rs, when producing and reviewing monographs, product licences, and other requirements. Within their overall objective of requiring animal tests to be carried out to ensure the production of safe and effective biological products, they might wish to consider animal welfare issues by establishing appropriate administrative procedures or committees. This would provide reassurance of the importance such bodies attach to animal welfare issues.

Testing of Gonadotrophins

At present, the potency of gonadotrophin preparations is estimated in a 3 + 3 point assay by comparing, under specified conditions, their ability to increase the ovarian mass (in the case of follicle stimulating hormone [FSH]), or the mass of seminal vesicles (for luteinising hormone [LH] or chorionic gonadotrophin), of immature rats to the same extent as an international standard/reference preparation, calibrated in International Units (IU). Due to the complex structures of these glycoproteins, which exist as different isoforms having different in vitro and in vivo biological activities, replacement of the in vivo bioassay with in vitro tests alone is not feasible. Since the isohormone composition of gonadotrophin preparations is very important for their in vivo biological activity and efficacy, a combination of an in vitro assay with physicochemical tests for evaluating the isohormone composition is a possible alternative to the in vivo bioassay. This possibility has been explored using recombinant FSH of high purity; the results obtained so far indicate that this approach is worth investigating further, particularly for recombinant and urofollitropin preparations. However, such an approach is not considered to be feasible for menotropin preparations, because of their multi-component nature.

It is recommended that:

  1. Physicochemical tests for monitoring the isohormone composition are included in the quality control procedures for new gonadotrophin preparations. Additional studies need to be carried out to provide further evidence that in vivo bioassays for gonadotrophins can be replaced by combining an in vitro assay with a physicochemical test(s) for the isohormone analysis.
  2. Instead of an isohormone analysis, other procedures for quantification of the carbohydrate composition, such as lectin-binding assays, might be considered and should be evaluated.
  3. The possibility of reducing the number of animals used for bioassays should be evaluated by further investigating the precision of a 2 + 2 point assay procedure (2).

Testing of Urokinase

The current European Pharmacopoeia monograph has been amended to contain the requirement (under "Production") that the method of manufacture of urokinase must be designed to exclude the presence of vasoactive substances. Previously, the text included the test for vasoactive substances as a mandatory requirement. The test for vasoactive substances is a general method. Originally, the test was to be carried out using an anaesthetised cat, but the requirement was then changed so that a rabbit under prolonged anaesthesia was to be used. However, this has proved to be an unsatisfactory replacement.

It is recommended that:

  1. Under "Production", the text of the European Pharmacopoeia monograph be redrafted so that the manufacturer is required to demonstrate the absence of vasoactive products following the purification of urokinase. For this purpose, the test for vasoactive substances needs to be improved (for example: shorter observation period; use of an anaesthetic other than phenobarbital sodium). An exemption form performing the test on subsequent batches should be given, as long as the validated mode of manufacture has not been changed. (Since the workshop was held, the European Pharmacopoeia monograph has been revised so that, from 1 January 1996, manufacturers will be required to demonstrate the validity of their methods of production with respect to the minimisation/elimination of microbial and viral contamination and of the presence of vasoactive substances.)
  2. Manufacturers of urokinase should be asked to supply the European Pharmacopoeia Commission with information on batches reacting positively to the test for vasoactive substances. Such information is needed in order to decide whether this test could be deleted from the requirements at a future date.

Testing of Oxytocin and Oxytocin Solutions

The intention to replace in vivo bioassays for testing oxytocin and oxytocin solutions has been published in Pharmeuropa, the forum publication of the European Pharmacopoeia Commission, for comments (3). A set of physicochemical tests relying mainly on a highly discriminatory liquid chromatography procedure will be introduced. This proposal has been formulated on the basis of results obtained in a collaborative study, in which manufacturers and control laboratories have participated, and which was organised within the framework of the expert group on biologicals for the detection of impurities in drug substances. In response to an enquiry, manufacturers provided data on a large number of batches of oxytocin peptide and the bioassay. In addition, the expert group had access to the data published by Maxl & Siehr in 1989, which also supported this conclusion (4). This study clearly showed that the in vivo bioassays were more variable in their responses than the physicochemical tests, and that they are less reliable for the standardisation of oxytocin.

These was, nevertheless, reluctance on the part of a few experts to completely remove any reference to the bioassay. They would, as a minimum, be in favour of a confirmatory test based on in vitro tissue culture (perhaps carried out only periodically). In this context, the questions of availability of reagents and the costs of the methods were raised. Such tests are already required for demonstrating bioactivity, in addition to the general in vivo test undertaken during the development phase.

The experts supported the changes proposed by the European Pharmacopoeia Commission for routine batch-to-batch control analysis. It is hoped that the revised monograph will be finished and adopted by November 1995, and that is will come into force as soon as possible.

As with most of these hormones, clinicians are accustomed to prescribing the therapeutic dose in IU of biological activity. To prevent accidents, and to promote the planned change to mg dosages, the adoption of an IU to mg conversion factor becomes a prerequisite for replacement of the bioassay.

Testing of Calcitonin

At present, the biological activity of calcitonin is determined using the rat hypocalcaemia assay. This assay uses a large number of animals, is cumbersome and is not very precise. Alternative in vitro assays have been developed and published (5), which measure cAMP production in the human cancer cell line, T47D, as an endpoint. An intact cell assay and an assay based on membrane preparations have been validated with respect to most of the pharmacopoeial requirements, and these may be able to replace the in vitro assay. The assays were used in two WHO-sponsored collaborative studies (6,7).

Quality control of batch-to-batch preparations by using physicochemical methods is very precise; these methods are superior to bioassays in this respect (8). Physicochemical methods do not, however, give information on biological activity as required by the pharmacopoeiae; therefore, for example, variations in secondary structure, which might affect the biological activity (9), would not be detected. Also, data have been published which report a discrepancy between HPLC analysis and results obtained in in vitro bioassays when assessing calcitonin production batches subjected to thermal degradation (10). Thus, physicochemical methods and bioassays complement each other -- the physicochemical method for demonstrating consistency in production and the bioassay for assessing biological activity.

It is recommended that:

  1. The consistency of the production process for calcitonin should be verified by both physicochemical analyses (for example, HPLC) and in vitro bioassays (either based on intact cells or on membrane preparations) of batches of the active ingredient. It is expected that there will be no further need for the in vivo hypocalcaemia assay. Some of the participants thought that physicochemical analyses alone would suffice, and that no bioassay would be necessary once development was complete.
  2. International harmonisation of pharmacopoeial requirements is necessary if the animal test is to be replaced with in vitro bioassays, or if testing for biological activity is to be abandoned.
  3. Additional validation requirements (including interlaboratory studies), if any, for the in vitro assay systems should be identified by the European Pharmacopoeia COmmission. The suitability of an in vitro assay for testing salmon and eel calcitonin was demonstrated in the WHO collaborative study (7); this should be confirmed. Funding for this further testing is needed, and should be provided by ECVAM and the European Pharmacopoeia Commission.
  4. The differences between the results obtained using the HPLC system and the in vitro assay following thermal degradation of calcitonin samples needs to be explained. Comparative in vitro studies should be performed. The nature of the additional compounds detected by HPLC should be elucidated using other physicochemical methods.

Testing of Erythropoietin

Erythropoietin (EPO) is similar to other glycoprotein hormones in that it exists as many isoforms, which differ mainly in their glycosylation. The isoform composition of EPO may differ between products and batches, and may affect its efficacy. The extensive heterogeneity of EPO makes it impossible, as yet, to meet the purity criteria (absence of non-EPO contaminants) established for several other hormone preparations, or to characterise the preparations adequately by physicochemical methods alone.

It is recommended that:

  1. At present, the data available indicate that it will be neccesary to continue to use in vivo bioassays to ensure batch potency and consistency of EPO. This is because the in vitro assays available are not sufficiently sensitive to differences in glycosylation. However, efforts should be directed toward adoption of the normocythaemic mouse assay, rather than the posthypoix polycythaemic mouse assay, because it is less stressful for the animals used. In addition, potency estimates from in vivo bioassays should be restricted to the bulk substances, thereby substantially reducing the use of animals.
  2. For making progress in the future, it is recommended that research be undertaken into alternative methods for assessing batch consistency, specifically the use of in vitro assays in combination with determinations of the composition of EPO isoforms. Promising results for estimating isoform composition are being achieved using lectin-binding assays and charge-based separation methods. Further attention should also be focused on the use of sophisticated physicochemical techniques, such as high-pH anion exchange chromatography, for comparing the glycosylation patterns of different batches of EPO.

Testing of Heparin

The current monograph on heparin in the European Pharmacopoeia requires the manufacturer to use a mode of production that has been shown to eliminate the presence of depressor substances. This requirement replaced the test for depressor substances in the preceding version of the monograph. The change was made after a number of manufacturers had provided data showing that the test for depressor substances had given negative results with a large number of batches produced over a number of years. However, it may not be generally known that, for heparin, this test is no longer mandatory.

It is recommended that the European Pharmacopoeia Commission be asked to publish, for example, in Pharmeuropa, the fact that the test for depressor substances no longer appears to be necessary for heparin, on the basis of information currently available.

Testing for Depressor Substances (Other Than With Heparin)

A number of monographs on antibiotics and some biologicals (urokinase, etc.) make the test for depressor substances a mandatory requirement. In addition, some manufacturers conduct this test voluntarily. This practice is based on the occurrence of untoward reactions with initial batches. It is uncertain whether the mode of production has been improved to the extent that the risk of untoward depressor effects has become negligible.

It is recommended that the European Pharmacopoeia Commission be asked to undertake a survey among the manufacturers of these products, requesting them to submit information on the performance of the test for depressor substances with previous batches, for example, those produced over the past five years. The data received could then be used to decide whether the test for depressor substances could be deleted from the appropriate monographs, on a case-by-case basis. This survey could be analogous to that conducted in 1994 on results obtained using the test for abnormal toxicity with vaccines.

Testing of Cytokines and Related Biotechnological Products

For preclinical toxicity studies on cytokines and related biotechnological products, there are different testing requirements in the EU, the USA, and Japan. In Japan, these products are normally tested according to the requirements for medicines, unless it can be demonstrated that the performance of such tests is not possible or is clearly irrelevant. In the USA, a more pragmatic approach is adopted on a case-by-case basis. In the EU, the data required are documented in various test guidelines, but some European authorities ask for additional data (for example, on mutagenicity).

More harmonisation of testing requirements seems to be necessary, to enable rapid preclinical development on the basis of relevant toxicity studies. It is recommended that the decision on whether to conduct a particular test should be based on the characteristics of the substance in question (for example, species specificity, or the formation of neutralising antibodies in animals). The tests should include studies with single and repeated administration (normally up to four weeks), assessments of local tolerance, and a variety of pharmacological safety experiments in relevant species. To further reduce the use of laboratory animals, other toxicity studies, such as those for in vivo mutagenicity, reproductive toxicity or tumorigenicity, should only be required and carried out when they are clearly needed. A similar approach should be adopted for the toxicity testing of monoclonal antibodies (of either human or murine origin).

Testing of Recombinant and Other Biological Products

For growth hormone/somatotropin and insulin, there have been large collaborative studies which finally resulted in the decision to delete the requirement for animal testing for batch-to-batch control in the European Pharmacopoeia monographs. Unfortunately, the USA did not take this progressive move, maintaining the requirement for animal testing, albeit reducing the total number of animals which had to be used in a particular test (three animals per test). Japan adopted the European position with respect to animal testing for batch-to-batch control purposes.

Other recombinant products which are now being reviewed by regulatory authorities are products that either mimic, or substitute for, endogenous substances, factors or hormones (such as blood coagulation Factor VIII), and for which prolonged pharmacological and toxicological studies are not possible because of the formation of antibodies.

It is recommended that:

  1. Since animal testing to assess the efficacies of somatotropin and insulin is still required in the USA, there is an urgent need to review and discuss this issue at the International Conference on Harmonisation (ICH). It cannot be acceptable that tests in only three animals, which do not provide any meaningful results, should be a requirement. In addition to this issue being addressed through the ICH initiative, it should also be taken up by ECVAM in discussions with its counterpart organisations in the USA. No further research or other activities are required in this area.
  2. For other recombinant products, the decision on whether to conduct animal tests has to be taken on a case-by-case basis. In general, these products should be treated in the same manner as other biologicals and, therefore, in most cases only a limited amount of animal safety testing is applicable.
  3. With regard to glucagon, validation studies are currently in progress and no further action is needed at the present time.

The Test for Abnormal Toxicity

A number of monographs on biological products contain the test for abnormal toxicity as a mandatory requirement. This is based on the occurrence of untoward reactions with the first batches produced. It is unclear whether the mode of production has been improved to the extent that the risk of untoward toxic effects can now be considered to be negligible.

It is recommended that the European Pharmacopoeia Commission be asked to organise a survey among the manufacturers of these products, requesting them to submit information on the performance of the test for abnormal toxicity on previous batches, for example, those produced during the past five years. The data could then be used to decide whether the test for abnormal toxicity could be deleted from the appropriate monographs, on a case-by-case basis. This survey could be analogous to that conducted in 1994 on results obtained using the test for abnormal toxicity with vaccines.

Pyrogenicity Testing of Biological Products

There have been a few cases in the past in which the rabbit pyrogen test gave a positive result with a particular blood product, yet the administration of the product to patients did not result in untoward reactions. These results were obtained before the Limulus amoebocyte lysate (LAL) test became available. If such cases of an apparently false positive rabbit pyrogen test result were to occur again in the future, it would be extremely valuable to carry out LAL tests on such products and, if possible, to establish the cause of this apparently false positive result.

There are conflicting reports concerning the relationship between the LAL gelation test and quantitative LAL tests, such as the turbidimetric kinetic test. According to one manufacturer, the gelation test can be carried out on all products for which a test for the presence of bacterial endotoxins is required. On the other hand, there are claims that the gelation test cannot be carried out on some antibiotics, in contrast with the quantitative methods. This discrepancy needs to be resolved.

It is recommended that:

  1. For those products for which a rabbit pyrogen test is still required, efforts should be made to replace it with the LAL test. Validation of the LAL test should be carried out on a case-by-case basis. The guideline for the LAL test in the European Pharmacopoeia monograph (text revised in 1994) gives advice on the conduct of such validation studies.
  2. The US Pharmacopoeia standard for endotoxin has to be replaced in the near future. It is hoped that this opportunity will be taken to calibrate the preparation in IU, and that the relative potencies of the American, European and Japanese reference preparations of endotoxin will also be established in terms of IU.

Production of Murine Monoclonal Antibodies

Murine monoclonal antibodies can be produced either by ascites production or using in vitro techniques (11). There are ethical concerns with regard to the ascites production method, for animal welfare reasons.

It is recommended that:

  1. Monoclonal antibodies should only be produced using in vitro methods, except where it can be clearly demonstrated that they can only be produced in the mouse. This is not normally the case.
  2. Sometimes, it is problematic to produce small quantities of monoclonal antibodies for special purposes using the in vitro technique (for example, because of inexperience with the method, or due to high costs). Thus, universities and other scientific institutions should be encouraged to install specialised laboratories as service centres for the production of monoclonal antibodies using in vitro methods (as, for example, has been established in the Netherlands). These service units should have the equipment and qualified personnel needed to conduct this technique. Sponsorship should be sought from appropriate sources.

Validation Procedures

Some authors who wish to demonstrate the validity of an alternative method do so by plotting the results obtained with the new assay against the results from the traditional bioassays for the same test substance(s), and then calculating the correlation coefficient between the two sets of measurements (Figure 1). Bland & Altman (12) pointed out that this does not provide the most relevant information. What one wants to know is not how well the new method and the old one are correlated, but how close the agreement is between the results obtained with the two methods. Hence they proposed the following procedure:

  1. Calculate for each experimental unit the mean of the results of the new and the old methods; this average gives the closest approximation to the true value for this element.
  2. Calculate for each experimental unit the difference between the results obtained with the two methods, d (i.e. d = result of the new method - result of the old method).
  3. Plot the value of d against each value of the average. Ideally, the points will be scattered around a mean value of 0.

  4. Calculate the standard deviation (SD) of d, and the 95% confidence limits, d + 2SD and d - 2SD, and draw horizontal lines on the graph through these values and also through the mean value of d (for example, see Figure 2).
  5. The range (d - 2SD) to (d + 2SD) then indicates the range between the 95% of these values of d may be expected to lie.
  6. Calculate the standard error of the mean (SEM) of d using the formula:

    where S2 = the variance and n = the numberof values of d.

  7. The range (mean of d) - t x SEM to (mean of d) + t x SEM indicates the precision with which the mean of d has been determined, where t = student's t statistic with n - 1 degrees of freedom.

This procedure makes it possible to decide whether the agreement between the alternative and the traditional methods is sufficiently close. If possible, the results obtained with the alternative method and the traditional method should be plotted against theoretical expected values. For an illustration, see the publication on oxytocin in the December 1994 issue of Pharmeuropa (3).

This procedure has been applied to the results obtained by Grauer et al (5) with their in vitro assay for human calcitonin (Figure 2). The following values were deduced from the figure (expressed as a percentage of the standard preparation): n (number of observations) = 30; (mean difference) (d) = -1.87; SD = 8.42; d + 2SD = 14.97; d - 2SD = -18.71.

Figure 2: Comparison of Two Assays by Plotting Difference Data

The data are taken from Gauer et al. (5).

It is recommended that this procedure for comparing data from different methods is used whenever possible. ECVAM and the European Pharmacopoeia Commission should be asked to enforce the use of this method for presenting data from possible alternative procedures.

Summary of Conclusions and Recommendations

Animal Welfare Guidelines

  1. Guidelines should be defined for the humane treatment and care of laboratory animals used in the quality control of hormones and other biological products. In this respect, it is envisaged that the recommendations given in the Appendix to this report will act as a model for international regulatory agencies and national control authorities, for them to adapt as appropriate.

Testing of Gonadotrophins

  1. Physicochemical tests for monitoring isohormone composition should be included in the quality control procedures for new gonadotrophin preparations. Additional studies need to be carried out to determine whether in vivo bioassays for gonadotrophins can be replaced by using a combination of an in vitro assay with a physicochemical test(s) for the isohormone analysis.
  2. Instead of an isohormone analysis, other procedures for quantification of the carbohydrate composition, such as lectin-binding assays, might be considered and should be evaluated.
  3. The possibility for reducing the number of animals used for bioassays should be determined by investigating the precision of a 2 + 2 dose procedure (2).

Testing of Urokinase

  1. Since the workshop was held, the European Pharmacopoeia monograph relating to "production" has been revised, so that from 1 January 1996, manufacturers will be required to demonstrate the validity of their methods of production with respect to the minimisation/elimination of microbial and viral contamination and of the presence of vasoactive substances. For this purpose, the test for vasoactive substances needs to be improved (for example, by using a shorter observation period and/or an anaesthetic other than phenobarbital sodium). Exemption from performing the test on subsequent batches should be given, as long as the validated mode of manufacture has not been changed.
  2. Manufacturers of urokinase should be asked to supply the European Pharmacopoeia Commission with information on batches reacting positively to the test for vasoactive substances. Such information is needed in order to decide whether this test could be deleted from the requirements at some stage in the future.

Testing of Oxytocin and Oxytocin Solutions

  1. It is envisaged that the in vivo bioassays for testing oxytocin solutions will be replaced with a set of physicochemical tests which rely principally on a highly discriminatory HPLC method.
  2. Despite evidence that there is a good correlation between the HPLC assay for oxytocin peptide and the bioassay, and that the in vivo assays are more variable in their responses than the physicochemical tests, there is reluctance on the part of a few experts to completely delete reference to the bioassay from the relevant monograph. They would, as a minimum, favour the use of a confirmatory in vitro test (perhaps carried out only periodically).
  3. The changes proposed by the European Pharmacopoeia Commission for routine batch-to-batch control analysis are supported. It is hoped that the revised monograph will be finalised and adopted by November 1995, and will come into force as soon as possible.
  4. As with most of these hormones, clinicians are accustomed to prescribing the therapeutic dose in IU of biological activity. To prevent accidents, and to promote the planned change to mg dosages, the adoption of an IU to mg conversion factor becomes a prerequisite for replacement of the bioassay.

Testing of Calcitonin

  1. The consistency of the production process for calcitonin should be verified by both physicochemical analyses (for example, HPLC) and in vitro bioassays (either based on the intact cells or on membrane preparations) of batches of the active ingredient. It is expected that there will be no further need for the in vivo hypocalcaemia assay. Some of the participants thought that physicochemical analyses alone would suffice, and that no bioassay would be necessary once development had been completed.
  2. International harmonisation of pharmacopoeial requirements is necessary if the animal test is to be replaced with in vitro assays, or if testing for biological activity is to be abandoned.
  3. Additional validation requirements (including interlaboratory studies), if any, for the in vitro assay systems should be identified by the European Pharmacopoeia Commission. The suitability of an in vitro assay for testing salmon and eel calcitonin was demonstrated in the WHO collaborative study (7); this should be confirmed. Funding for this further testing is needed, and should be provided by ECVAM and/or the European Pharmacopoeia Commission.
  4. The difference between the results obtained using the HPLC system and the in vivo assay following thermal degradation of calcitonin samples needs to be explained.

Testing of Erthropoietin

  1. At present, the data available indicate that it will be necessary to continue to use in vivo bioassays to ensure batch potency and consistency of EPO. This is because the available in vitro assays are not sufficiently sensitive to differences in glycosylation. However, efforts should be directed toward adoption of the normocythaemic mouse assay, rather than the post-hypoxic polycythaemic mouse assay, because it is less stressful for the animals used. In addition, potency estimates from in vivo bioassays should be restricted to the bulk substances, thereby substantially reducing the use of animals.
  2. Research should be undertaken into alternative methods for assessing batch consistency, specifically the use of in vitro assays in combination with determinations of the composition of EPO isoforms. Promising results for estimating isoform composition are being achieved using lectin-binding assays and charge-based separation methods. Further attention should also be focused on the use of sophisticated physicochemical techniques, such as high-pH anion exchange chromatography, for comparing the glycosylation patterns of different batches of EPO.

Testing of Heparin

  1. The European Pharmacopoeia Commission should be asked to publish (for example, in Pharmeuropa) the fact that, on the basis of information currently available, the test for the presence of depressor substances no longer appears to be necessary for heparin.

Testing for Depressor Substances (Other Than with Heparin)

  1. The European Pharmacopoeia Commission should be asked to undertake a survey among the manufacturers of these products, requesting to submit information of the performance of the test for depressor substances with batches produced, for example, during the past five years. The data received could then be used to decide whether the test for depressor substances could be deleted from the appropriate monographs, on a case-by-case basis.

Testing of Cytokines and Related Biotechnological Products.

  1. Further international harmonisation of testing requirements seems to be necessary, to enable rapid preclinical development on the basis of relevant toxicity studies. It is recommended that the decision on whether to conduct a particular test should be based on the characteristics of the substance in question (for example, species specificity, or the formation of neutralising antibodies in animals). The tests should include studies with single and repeated administration (normally up to four weeks), assessments of local tolerance, and a variety of pharmacological safety experiments relevant species. To further reduce the use of laboratory animals, other toxicity studies, such as those for in vivo mutagenicity, reproductive toxicity or tumorigenicity, should only be required and carried out when they are clearly needed. A similar approach should be adopted for the toxicity testing of monoclonal antibodies (of either human or murine origin).

Testing of Recombinant and Other Biological Products

  1. Since animal testing to assess the efficacies of somatotropin and insulin is still required in the USA, but not in Europe or in Japan, there is an urgent need to review and discuss this issue at the ICH. It cannot be acceptable that tests in only three animals, which do not provide any meaningful results, should be a requirement. In addition to addressing the issue through the ICH initiative, it should also be taken up by ECVAM in discussions with its counterpart organisations in the USA. No further research or other activities are required in this area.
  2. For other recombinant products, the decision on whether to conduct animal tests has to be taken on a case-by-case basis. In general, these products should be treated in the same manner as other biologicals and, therefore, in most cases only a limited amount of animal safety testing is applicable.
  3. With regard to glucagon, validation studies are currently in progress and no further action is needed at the present time.

The Test for Abnormal Toxicity

  1. The European Pharmacopoeia Commission should be asked to organise a survey among the manufacturers of those biological products for which the test for abnormal toxicity is a mandatory requirement, requesting them to submit information on the performance of the test with batches produced, for example, during the past five years. The data submitted could then be used to decide whether the test for abnormal toxicity could be deleted from the appropriate monographs, on a case-by-case basis.

Pyrogenicity Testing

  1. For those products for which a rabbit pyrogen test is still required, efforts should be made to replace it with the LAL test. Validation of the LAL test should be carried out on a care-by-case basis. The guidelines for the LAL test in the European Pharmacopoeia monograph (text revised in 1994) gives advice on the conduct of such validation studies.
  2. The US Pharmacopoeia standard for endotoxin has to be replaced in the near future. It is hoped that this opportunity will be taken to calibrate the preparation in IU, and that the relative potencies of the American, European, and Japanese reference preparations of endotoxin will also be established in terms of IU.

Production of Murine Monoclonal Antibodies

  1. Monoclonal antibodies should only be produced using in vitro methods, except where it can be clearly demonstrated that they can only be produced in the mouse. This is not normally the case.
  2. Sometimes it is problematic to produce small quantities of monoclonal antibodies for special purposes using in vitro techniques (for example, because of inexperience with the method, or due to high costs). Thus, universities and other scientific institutions should be encouraged to install specialised laboratories as service centres for the production of monoclonal antibodies using in vitro methods (as, for example, has been established in The Netherlands). These service units should have the equipment and qualified personnel needed to conduct this technique. Sponsorship should be sought from appropriate sources.

Validation Procedures

  1. The procedure of Bland & Altmann (12) should be used for comparing data from different methods, whenever possible. ECVAM and the European Pharmacopoeia Commission should be asked to enforce usage of this method for presenting data from possible alternative procedures.

References

  1. Anon. (1994). ECVAM News & Views. ATLA 22: 7-11.
  2. van Noordwijk, J. (1989). Bioassays in whole animals. Journal of Pharmaceutical and Biomedical Analysis 7: 139-145.
  3. Anon. (1994). Note on the revision of the monographs Oxytocinum and Oxytocini solutions. Pharmeuropa 6: 377-386.
  4. Maxl, F. & Siehr, W. (1989). The use of high-performance liquid chromatography in the quality control of oxytocin, vasopressin, and synthetic analogues. Journal of Pharmaceutical and Biomedical Analysis 7: 211-216.
  5. Grauer, A., Raue, F., Reindel, H.H., Schneider, H.G., Schroth, J., Kabay, A., Brügger, P. & Ziegler, R. (1992). A new in vitro bioassay for human calcitonin: validation and comparison to the rat hypocalcemia bioassay. Bone and Mineral 17: 65-74.
  6. Blind, E., Raue, F., Kienle, P., Schroth, J., Grauer, A., Kabay, A., Brügger, P. & Ziegler, R. (1993). Development and validation of an assay to measure bioactivity of human calcitonins in vitro using T47D cell membranes. Analytical Biochemistry 212: 91-97.
  7. Zanelli, J.M., Gaines-Das, R.E. & Corran, P.H, (1993). Establishment of the second international standard for porcine and human calcitonins: report of the international collaborative study. Acta Endocrinologica 128: 443-450.
  8. Zanelli, J.M., Gaines-Das, R.E. & Corran, P.H. (1990). International standards for salmon calcitonin, eel calcitonin, and the Asu1-7 analogue of eel calcitonin: calibration by international collaborative study. Bone and Mineral 11: 1-17.
  9. Arvinte, T. & Drake, A.F. (1993). Comparative study of human and salmon calcitonin secondary structure in solutions with low dielectric constants. Journal of Biological Chemistry 268: 6408-6414.
  10. Buck, R.H. & Maxl, J. (1990). A validated HPLC assay for salmon calcitonin. Comparison of HPLC and biological assay. Journal of Pharmaceutical and Biomedical Analysis 8: 761-769.
  11. Falkenberg, F.W., Weichert, H., Krane, M., Bartels, I., Palme, M., Nagels, H-O. & Fiebig, H. (1995). In vitro production of monoclonal antibodies in high concentration in a new and easy-to-handle minifermenter. Journal of Immunological Methods 179: 13-29.
  12. Bland, J.M. & Altman, D.G. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. Lancet i: 8 February, 307-310.

Appendix A