The intention of a pharmaceutical company is to develop new, efficient products quick and with a minimum of costs. Compared to in vitro methods, animal experiments in general consume much more time and resources (costs as well as time to the market) thanin vitro methods. Therefore, the use of whole animal models depends primarily on the judgement of their efficacy in the screening process, but the willingness to incorporatein vitro methods in general is high and is furthered by new developments such as high-throughput screening. Nevertheless,in vitro tests might be politically promoted by increasing their costs (quality controls, requested housing conditions) and duration (time to start of an experiment, sequential performance).Which models are favoured by industry to include them in a screening process: They have to be based on our most recent understanding of the respective disease, well characterised to allow interpretation of results and require only limited development time. All these aspects argue in favour of collaboration between industry and academia, where our understanding of pathophysiology is generated and mechanism based models are developed and characterised. However, technology transfer towards industry represents a bottle-neck for industrial use of these newin vitro models. New platforms to promote this transfer should be developed in order to bring together developer and user of novel in vitro systems and promote demonstration projects. Financing of such collaborations is not the key problem (the development of a single drug makes up to 500 million $) but the dilemma of publication of results: The development advantage compared to competitors depends on the exclusive use of novel models. The protection of intellectual property rights and the public interest in spreading alternatives to animal experiments must be balanced, e.g. by delayed but indispensable publication or advantages for companies employing alternatives in the regulatory approval process for a new drug.
Quality control of therapeutic drugs (except hormones and blood products) represents a minor field of animal consumption with the exception of pyrogenicity testing. Despite considerable progress due to the introduction of the Limulus assay which represents the most successful in vitro alternative in use so far. However, some limitations of this in vitro test might be overcome in the near future by the currently validated human whole blood assay. During the last few years considerable progress has been made in the replacement (and deletion) of animal tests required for the potency and safety testing of hormones. This has been made possible by biotechnical production methods, by better-defined products, and because physico-chemical methods can be used for the potency testing of these products. In general, the better defined a drug is, the easier chemical, physical orin vitro techniques can be used for batch control. Control authorities should therefore urge the use of highly standardised components.

Immunobiologicals (vaccines, immunoglobulins and -sera) are considered to be the most cost-effective tools in the prevention of infectious diseases. Their importance will further increase due to various eradication programmes of the WHO and EU and the emergence of new infectious diseases or the re-emergence of diseases as diphtheria and tuberculosis. The production and quality control of immunobiologicals are regulated by monographs and guidelines, which are issued by international or national Pharmacopoeias (e.g. Ph. Eur.), international organisations (e.g. WHO, O.I.E.) and international regulatory bodies (e.g. EMEA). Their purpose is to assure the quality of the product, i.e. its safety and potency. It is estimated that 10 millions of laboratory animals are world-wide used for the production and quality control of immunobiologicals, of which 80% are needed for the safety and potency testing of the finished product (batch control).
In recent decades, the use of Three Rs principles has been recognised by the above mentioned organisations and various national competent authorities and been incorporated into general monographs and guidelines. Several tests with questionable relevance have been deleted from Ph. Eur. monographs (e.g. abnormal toxicity test, extraneous agents testing of viral vaccines for carnivores) or are now carried out during production. Reduction of the number of animals used could be achieved by introducing single-dilution tests. A large number of immunochemical tests have been developed, which could completely or partly replace the use of animals for potency testing, however, only a few have been validated so far (e.g. ToBI and ELISA for potency testing of human and veterinary tetanus vaccine; ELISA for potency testing of erysipelas vaccine). Regulatory acceptance of validated alternative methods is still a critical step. In particular, the period between successful validation and the implementation appears to be far too long. Reasons for this could be the slow process of multinational agreement to revise pharmacopoeial monographs and guidelines, and the time-consuming and expensive production of sufficient reference material (antigen, sera etc) for the new test systems.
The shift in the quality control concept from reliance on final batch testing to the concept of consistency of production offers the opportunity to reduce the numbers of animals being used and promote the use of alternative methods. Emphasis is put on a combination of in vitro tests, which could make it possible to monitor batch-to-batch consistency. This new concept of quality control is already in place for the new well-defined vaccines. In most cases, non-animal methods are used for monitoring consistency at critical steps in the production and testing of a vaccine. Whether the concept of consistency of production could be also applied to the conventional, less-defined products, should be investigated.
Only little progress has been achieved with regard to international harmonisation. Most of the manufacturers produce for the world market, so harmonisation of the requirements or mutual recognition of tests would help to reduce the use of animals.
There is agreement that for the time being animals will still be needed for the development of vaccines in order to gain best knowledge on the disease, the pathogen and the specific immune response, including: pathogenesis, identification of the protective antigens, the way the antigen is processed, the dynamics of the immune response, the induction of memory, and the selection of the best adjuvant. With regard to routine batch release of conventional products, a number of Three Rs approaches are already available and should further be developed and validated. Whereas routine batch release of new products should be based onin vitro methods already established during their development.