The potential of the biotechnology industry to provide blockbuster products is clearly demonstrated by examples like Enbrel (for the treatment of rheumatoid arthritis in adults and of active polyarticular-course juvenile chronic arthritis), Remicade (for the treatment of rheumatoid arthritis and Crohn's disease), and MabThera (for treatment of stage III-IV follicular lymphoma and CD20 positive diffuse large B-cell non-Hodgkin's lymphoma). The EvaluatePharma database1 indicates that at least 265 biotechnology products are currently marketed globally, more than doubling the figure from 1995. It is clear that there are many more blockbuster products to come, with an additional 100 biotechnology products currently in Phase III/preregistration, and a further 1000 reported to be in active development.

With this exponential growth in biotechnology product development, it follows that there is a corresponding increase in biotech clinical trial experience. Nevertheless, biotech clinical development continues to be a challenging exercise, particularly early on in anticipating the regulatory requirements for gaining approval to conduct a trial and in latter-stage clinical requirements for marketing approval.

In biotechnology forums, it is sometimes said that biotechnology product development by small- to medium-sized companies (called small- and medium-sized enterprises or SMEs in the EU) will be leaner and more efficient than for traditional "big pharma" drug development. Although many accept this concept, the reality is that this hypothesis is yet to be fully tested. Of the more than 265 biotechnology products marketed globally, the vast majority have come from big pharma, and a significant proportion of the product candidates have come from SMEs. Due to the prohibitive cost of clinical development, however, the clear majority of products were partnered or licensed to big pharma. In the future, we anticipate that biotech product development by SMEs will continue to be leaner and more efficient than big pharma. For broad applications, the cost of clinical development will continue to require SMEs to partner with big pharma. But for rare and serious orphan diseases, there is an opportunity for SMEs to fulfill the promise of independent, lean, and more efficient drug development.

In this article, we aim to provide some insight into the clinical development of biotech products by outlining the regulatory requirements and highlighting some of the "hot" topics to consider when designing and implementing a biotech product development strategy. The focus of this article is biotech product development in the European Union (EU). However, with the continuing growth of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), the principles discussed below have global significance.

Progress in EU regulatory guidance As more experience is gained with biotech product development, more and more guidance documents are being produced by both the EMEA ( http://www.emea.eu.int/index/indexh1.htm|~www.emea.eu.int/index/indexh1.htm ) and the ICH ( http://www.ich.org/UrlGrpServer.jser?@_ID=276&@_TEMPLATE=254|~www.ich.org/UrlGrpServer.jser?@_ID=276&@_TEMPLATE=254 ). From the EU perspective, two of the more significant developments include the publication of European Commission Directives 2001/20/EC² and 2003/63/EC.³

Table 1. Areas to consider for the first clinical trial application

Harmonization of clinical trials regulations across the EU. Directive 2001/20/EC was published on 4 April 2001 and was expected to be implemented into Member State legislation by 1 May 2004. Given the existence of the ICH E6 Good Clinical Practice guideline, the purpose of this directive is to harmonize the EU regulatory approaches to the clinical development of medicines as much as possible, and in doing so help to ensure the confidence of EU citizens in medicine development. Some predict that the implementation of this directive will have serious repercussions on biotech product development. It certainly has the potential to increase the regulatory burden prior to initiating a clinical trial, particularly for early clinical development and investigator-sponsored studies. However, a more harmonized EU regulatory process should also help to make the requirements for conducting multicentric international EU clinical trials more transparent. Time will tell whether the goals of the directive are achieved.

In accordance with Directive 2001/20/EC, all investigational medicinal products (IMP) in clinical development, even those in early development, will need to demonstrate Good Manufacturing Practice (GMP) in concordance with the phase of development, which will be a significant challenge for many independent investigators and start-up biotech companies. An additional regulatory procedure will also be introduced. Before submitting any documents to an ethics committee or to the national regulatory authority (defined in the legislation as a "competent authority"), it will be necessary to register with the EUDRACT database. This database is intended to facilitate both the exchange of information between the trial sponsor and the competent authority, the National Regulatory Authority, and between competent authorities of different Member States. To register in the EUDRACT database, it will be necessary for the sponsor to provide administrative information as well as information on the IMP, the trial protocol, and the envisioned countries and sites where the trial will take place. A unique EUDRACT identification number will then be allocated for each clinical trial entered by a sponsor, and this number will be requested for Ethics Committee and competent authority submissions. One benefit of this additional EUDRACT process is that the system may prefill for the sponsor the subsequent Ethics Committee and National Regulatory Authority trial application forms.

Once a sponsor has obtained the EUDRACT identification number, Ethics Committee and competent authority approval is still required. In practice, there are still likely to be differences between Member States, but overall it is expected that the process will be better harmonized. The applications for each will be very similar, including:

• Administrative data
• Protocol
• Investigator's Brochure
• IMP Dossier (all Quality, Nonclinical and Clinical data on the IMP, and a integrated summary with an overall benefit/risk assessment)
• Subject-related information (patient informed consent, recruitment methods)
• Protocol-related information (where is the trial conducted in EU and outside EU, any other trials conducted with the IMP)
• Manufacturing documents (GMP, TSE, Viral Safety, Certificate of Analysis)
• Financial data (insurance, compensation, agreements, and contracts).

Time will tell what impact Directive 2001/20/EC will have on biotech product development. While the goal is to better harmonize the regulation of clinical trials in the EU and help to ensure the safety of trial participants-both significant benefits-it remains to be seen whether it will place unrealistic demands on the heart of the biotech product pipeline, the independent investigator and biotech incubators.

Table 2. Overview of nonclinical testing of three approved monoclonal antibodies

Additional guidance on marketing approval requirements in the EU. While seeking marketing approval can seem a distant target, even in early product development, it is wise to consider what will need to be included in the marketing application, when this data will be generated, and how it will be presented. Directive 2001/83/EC, published on 6 November 2001, is the legislative basis for the regulation of marketing products in the EU.

Annex 1 of this directive provides detailed description of the content and format requirements of a marketing application. Directive 2003/63/EC was published on 25 June 2003 in order to update aspects of Directive 2001/83/EC. This update was in part motivated by the development of the ICH Common Technical Document, but also to address advancements in biotechnology that have led to the development of products based on antibodies, genetically modified organisms, and cell therapy. It is a significant step forward, answering many questions on what data is typically required and where it should be included in the marketing application dossier for a biotechnology-derived product. Review of Directive 2001/83/EC is ongoing, with additional revisions expected to be finalized by the European Parliament during the first half of 2004 (updates on the progress of the new legislation are published at http://pharmacos.eudra.org/F2/review/index.htm|~pharmacos.eudra.org/F2/review/index.htm ).

While both of these documents help to clarify the regulatory requirements for conducting a clinical trial and for clinical development of a product, biotech product developers are still frequently frustrated by the expressions "case by case" and "scientific-based approach." While this uncertainty can be unsettling, as illustrated in the following lessons, it does provide opportunities for biotech product developers to negotiate with the regulatory authorities using scientifically based arguments.

Biotech clinical trial application lessons Below are some examples of issues that can often cause delays in the clinical development of biotech products and are likely to become increasingly scrutinized by regulators as Directive 2001/20/EC is implemented.

Lesson #1-Data requirements for conducting the first clinical trial.

One of the more difficult decisions when developing a biotech product candidate is the compilation of the first clinical trial application. While there are some guidance documents available, such as the ICH S6 Safety Studies for Biotechnological Products, there is a need for a "case-by case" and "scientific-based approach." In our experience, there are two areas that most concern regulators, as indicated in Table 1.

Devising an appropriate nonclinical development program can be challenging for biotech products, because the required studies depend on the nature of the product and the condition it is being developed to treat. This is the basis for the dreaded expressions "case-by-case" and "scientific-based approach." While this scenario is challenging, it can also provide opportunities. A common mistake is to perform the traditional battery of nonclinical testing thinking that performing additional unnecessary studies is preferable to not performing enough. However, such an approach is not recommended because performing unnecessary tests such as toxicity testing in an irrelevant species can provide misleading and damaging data. Table 2 provides an overview of the nonclinical testing of three approved monoclonal antibodies and shows that a tailored "case-by-case" and "scientific-based approach" is acceptable to regulators. Each of the three products had minimal pharmacokinetic, pharmacodynamic, and toxicity programs since it was felt that a full traditional approach was not appropriate.

The key to Lesson #1 is to create a development plan based on sound science, which is then validated with leading experts in the relevant fields. This development plan can then be further validated with regulatory authorities, either by meetings with national authorities or by seeking scientific advice (or protocol assistance if the product has an orphan indication) from the EMEA. This approach ensures that your product is represented in regulatory submission in the best possible manner and therefore helps to avoid regulatory delays.

Lesson #2-Demonstrating viral safety. The demonstration of viral safety is becoming increasingly important for biotech products, with some EU regulatory authorities, such as the French agency (AFSSaPS), currently requiring a specific "Viral Safety Application" to be submitted before conducting a biotech clinical trial.

Viral safety is particularly pertinent to vaccine development, where the active substances traditionally are isolated viral or bacterial antigens, or more recently have been based on recombinant proteins or live attenuated viruses/viral particles. In these cases, the risk of propagating foreign viruses and of incomplete activation of the intended viral-based active substance must be addressed. The host cells may have a latent or persistent infection; for instance, a retrovirus that may be transmitted vertically from one cell generation to the next, and which may be expressed intermittently as an infectious virus.4 Adventitious viruses could also be introduced in the production process by the use of contaminated animal products. Special precautions therefore have to be followed to prevent the contamination of the final product with potential dangerous impurities, including retroviral contaminants and the possibility of contamination with DNA.

Taking into account the above considerations, evidence has to be provided that viruses are either inactivated or eliminated during the process of production. Different strategies to demonstrate the elimination of foreign viruses can be used depending on whether the vaccine contains a live virus, inactivated virus, or viral or bacterial antigens. Typically, the ability to remove/inactivate viruses during the production process is shown by spiking the drug substance before critical purification steps with a known number of different viruses. The efficacy of the purification step is then shown by the reduction of viral count. Frequently, small nonenveloped viruses, large enveloped RNA viruses, and large DNA viruses are used in such experiments.⁵

While such tests and experiments have only an indirect influence on clinical development, as the design of clinical studies is not influenced by these investigations, the viral safety program may have a major impact on the timing of the clinical development. To avoid regulatory delays, the viral safety studies should be performed early during process development, prior to the start of the clinical development of the product. In accordance with Directive 2001/20/EC, all countries in the EU have a responsibility to ensure that each clinical trial product lot is evaluated with respect to its manufacture and planned indication prior to its use in a trial. Experience shows that the issue of viral safety will be carefully evaluated by the authorities.

Lesson #3-Environmental risk assessment. Genetic engineering is obviously an important foundation of biotech products, and the regulation of such technology in the EU is the subject of considerable debate. Fortunately, the use of this technology in the development of medicinal products is currently less contentious than in agriculture. However, an assessment of the environmental risk posed by the manufacture and use of genetically engineered medical products must still be carefully considered. For example, many developmental vaccines are based on attenuated or genetically modified organisms (GMOs), and special consideration has to be given to the release of these organisms.^4,6-8 In these circumstances, the manufacturer must provide data demonstrating the risk a GMO may represent in nature, including evidence that the live organisms do not revert or recombine to de novo pathogenic strains.

An excellent case study is a live oral cholera vaccine, which is approved for marketing in a number of countries, including Switzerland, Australia, East Asia, and Latin America. During the development of this product, the excretion and survival of the vaccine strain in the environment was investigated in order to address potential environmental risk concerns. Studies demonstrated that up to 12% of vaccinees shed the vaccine strain at very low titres (6 logs lower than what the vaccine contains). The capacity of the vaccine strain to survive in the environment was subsequently found to be similar to the wild-type strain.⁹ As the wild-type strain of V. cholerae is known to enter a "viable but not culturable" (VBNC) status^10,11 it was necessary to demonstrate the vaccine strain does not have an increased ability to go into the VBNC status (i.e., that a "super" strain has not been inadvertently created).

Serologically, the vaccine strain cannot be differentiated from the wild-type strain. Therefore, a marker gene was inserted into the genome of the vaccine strain.^12,13 The environmental risk assessment included evidence that the inserted mercury resistance gene did not represent a risk as the gene: 1) is expressed by Shigella bacteria, 2) was stably inserted into the genome of the vaccine strain, and 3) no evidence was found that co-culture of the vaccine strain with other bacteria could result in the transfer of the marker gene to other bacteria strains.

Genetic stability with respect to the pathogenic capacity of the vaccine strain was also demonstrated. More than 90% of the cholera toxin gene was excised by genetic engineering. First, all transfection experiments performed in vitro showed that the vaccine strain has no capacity to revert to the toxin-carrying strain, and can therefore not spread the disease in parts of the world where it is not endemic. Secondly, the genetic characterization of shed bacteria revealed no capacity to take over genetic information from other bacteria growing in the intestine.¹⁴

Unlike viral safety (described in Lesson #2), environmental risk assessment has a direct influence on the clinical development of vaccines containing live bacteria or viruses. Clinical studies must be designed to allow the evaluation of the shedding kinetics and the vaccine bacteria/virus should be tested genetically after passage through the intestine. Evidence must be provided to demonstrate the ability of the shed vaccine bacteria/virus to survive in the external environment and the subsequent potential impact on the external environment.

Regulatory guidance and case studies do help. However, biotech drug development will always be a challenging task because rarely will a development be repeated. New challenges and obstacles will inevitably arise with each development, even though this environment does provide opportunities for those who respect the regulatory requirements and apply strong scientific and ethical principles. The continued application of biotechnology to address orphan diseases provides a true opportunity for the biotechnology industry to fulfill the promise of lean and efficient drug development.

References 1. EvaluatePharma Integrated Pharmaceutical Company Information Service, http://www.evaluatepharma.com|~www.evaluatepharma.com/ .

2. Directive 2001/20/EU. OJ 2001; L 121:34-44 (available at http://pharmacos.eudra.org/F2/eudralex/vol-1/home.htm|~pharmacos.eudra.org/F2/eudralex/vol-1/home.htm ).

3. Directive 2003/63/EC. OJ L 159: 46-94 (available at http://pharmacos.eudra.org/F2/eudralex/vol-1/home.htm|~pharmacos.eudra.org/F2/eudralex/vol-1/home.htm ).

4. CPMP/SWP/4447/00 draft: Note for guidance on environmental risk assessment of medicinal products for human use.

5. CPMP/BWP/268/95: Note for guidance on virus validation studies: The design, contribution and interpretation of studies validating the inactivation and removal of viruses.

6. Directive 90/220 EEC. OJ 1990; L 117: 241-262.

7. Directive 2001/18/EC, OJ 2001, L 106: 1-38.

8. http://www.emea.eu.int/htms/human/presub/q20.htm|~www.emea.eu.int/htms/human/presub/q20.htm .

9. S.J. Cryz Jr., J. Kaper, C. Tacket, J. Nataro, M.M. Levine, "Vibrio cholerae CVD 103-HgR 103-HgR Live Oral Attenuated Vaccine: Construction, Safety, Immunogenicity, Excretion and Non-target Effects," Dev Biol Stand, 84, 237-244 (1995).

10. S. Chaiyanan, A. Huq, T. Maugel, R.R. Colwell, "Vaibility of the Nonculturable Vibrio cholerae O1 and O139," Syst Appl Microbio, 24 (3) 331-341 (2001).

11. M.S. Islam, Z. Rahim, S. Begum, S.M. Moniruzzaman, A. Umeda et al., "Association of Vibrio cholerae O1 with the Cyanobacterium, Anabaena sp., Elucidated by Polymerase Chain Reaction and Transmission Electron Microscopy," Trans R. Soc Trop Med Hyg, 93 (1) 36-40 (1999).

12. D. Favre D, M.M. Struck, S.J. Cryz Jr., J.F. Viret, "Further Molecular Characterization and Stability of the Live Oral Attenuated Cholera Vaccine Strain V. cholerae CVD 103-HgR103-HgR," Vaccine, 14 (6) 526-531 (1996).

13. E. Studer and U. Candrian, "Development and Validation of a Detection System for Wild-type Vibrio Cholerae in Genetically Modified Cholera Vaccine," Biologicals, 28 (3) 149-154 (2000).

14. J.B. Kaper, J. Michalski, J.M. Ketley, M.M. Levine, "Potential for Reacquisition of Cholera Enterotoxin Genes by Attenuated Vibrio cholerae Vaccine Strain CVD 103-HgR," Infect Immun, 62 (4) 1480-1483 (1994).