International Harmonisation of PharmaceuticalsKEY ISSUES OF CONCERN FOR PUBLIC HEALTH
By John Abraham
This paper is about the international harmonisation of pharmaceuticals regulation. This process is important for Canadian citizens because it is likely to influence the standards of safety and effectiveness employed by the Canadian Government when deciding to approve pharmaceuticals for marketing. Such harmonisation may also influence Canadian citizens', or Canadian health care systems', access to affordable medicines. The main focus of this paper is the activities and implications of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), but some other aspects of harmonisation are also discussed.
The proceedings of the main ICH meetings are published, as are the actual ICH guidelines on pharmaceutical regulation. To date the participation of public health advocacy groups, the broader medical profession and the wider public have been excluded from these harmonisation processes. Rather the key participants have been the pharmaceutical industry associations and government drug regulatory agencies of North America, Europe and Japan, together with a few international bodies, such as the World Health Organisation (WHO). There is a danger that such harmonisation may reduce drug safety standards in Canada without proper democratic scrutiny and accountability.
The implications of this could be particularly problematic for women, who are often frequent and/or chronic users of medication for some conditions. For example, lowering the regulatory checks on whether drugs cause cancer (i.e. whether they are carcinogenic) might expose the many women who suffer from arthritis to new non-steroidal anti-inflammatory drugs (NSAIDs) that pose an unnecessarily elevated carcinogenic risk. (A number of NSAIDs have been found to be carcinogenic). Furthermore, as explained below, lowering the regulatory checks on the chronic toxicity of new drugs to non-rodents might expose women, who are not of childbearing potential, to greater risks in clinical trials.
This paper identifies specific areas of concern for public health and the public interest in the realisation of health. These are explained and discussed in depth later in the paper but, in summary, the crucial concerns for public health faced by Canada in relation to international harmonisation are as follows:
- The dangers of reducing the requirements for checking whether new drugs cause cancer in people (carcinogenicity testing) apparently endorsed by ICH.
- The dangers of reducing chronic toxicity testing without a defensible basis, thereby undermining public health, especially of women in clinical trials, and the credibility of proper drug regulation, in general, in favour of commercial interests.
- As regards women of childbearing potential, no progress seems to have been made in the harmonisation of the timing and co-ordination of non-clinical safety studies for the conduct of clinical trials. ICH appears to have simply accepted that the EU, the US and Japan all have different safety standards. This apparent lack of progress needs explanation.
- ICH guidelines, in general, and the ICH guideline on "General Considerations for Clinical Trials' (E8), do not examine harmonisation of inclusion of women in clinical trials. Guideline E8 states that, "in general, pregnant women should be excluded from clinical trials where the drug is not intended for use in pregnancy", and that "in general, women of childbearing potential should be using highly effective contraception to participate in clinical trials". This guideline also states that, as clinical trial phases proceed, "the populations tested should be broadened to reflect the target population" [ultimate patient population]. However, ICH guidelines do not address directly the inclusion of women in trials. For example, they do not mention the US National Institutes of Health (NIH) guideline that enough (non-pregnant) women should be included to allow for separate analysis of drug efficacy and safety by gender.
- The dangers of weak, flexible, regulations for management of clinical safety data regulatory requirements on pharmaceutical companies need to be precise and prescriptive so that both companies and regulators can be held publicly accountable for their conduct in relation to these regulations
- The danger of mistaking the rhetoric for the reality in connection with the motivations, rationale and implications of ICH. Much more independent critical scrutiny of the process is needed and unfortunately it is not obvious that this will be provided by regulators. This underlines the need for much more involvement from public health advocacy organisations, consumer organisations and academic scientists, who are entirely independent of the industry.
- The neglect of harmonisation of good labelling and advertising regulatory standards, which is very important for doctors, patients and other users of pharmaceuticals in the realisation of health.
- The danger of relinquishing the Bolar provision as this would threaten the health of the poor in Canada (and in developing countries).
- Given the scientific uncertainty surrounding much of the risk-benefit assessments involved in pharmaceuticals regulation, and given that much of the rhetoric surrounding harmonisation implies that ICH is in the public interest because it will deliver new, and needed, medicines to patients faster, Health Canada should revise the Food and Drugs Act to state explicitly that, when the interests of public health and the pharmaceutical industry diverge or conflict, the benefit of the scientific doubt will be awarded to the former, over and above the interests of the latter.
- The backward regulatory trend of reducing carcinogenicity testing should be reversed and future scientific developments scrutinised carefully and publicly, without undue haste. Specifically, the Therapeutic Products Programme (TPP) of Health Canada should not adopt (or revoke its adoption of) ICH guidelines S1a and S1b on carcinogenicity testing in the interests of patients and public health in Canada.
- The backward regulatory trend of reducing toxicity testing in order to accommodate the commercial interests of the pharmaceutical industry when such reduction has no scientific credibility in term of protecting public health also needs to be reversed. The TPP should not adopt (or revoke its adoption of) ICH guideline S4a on repeat dose toxicity in non-rodents in the interests of patients, especially women of non-childbearing potential, and public health in Canada.
- TPP regulations should state clearly that transnational pharmaceutical companies are responsible for all subsidiaries, and that poor communication between different parts of a company is not acceptable as mitigation for failing to report ADRs in a timely manner. TPP should propose that a statement to this effect is made explicit in the ICH guideline on "Clinical Safety Data Management: Definitions & Standards for Expedited Reporting" (E2a) before adopting it.
- TPP should recommend that there is a specified quantitative minimum percentage increase in the occurrence of expected, serious ADRs, above which pharmaceutical companies are required to report it to regulatory agencies. TPP should ensure that this alteration is made before adopting the guideline on "Clinical Safety Data Management: Definitions & Standards for Expedited Reporting" (E2a).
- TPP should adopt the guideline on "Clinical Data Safety Management: Periodic Safety Update Reports for Marketed Drugs" (E2c) only by also adopting the US Food and Drug Administration's (FDA's) standard of requiring periodic safety update reports (PSURs) every 3 months during the first 3 years of marketing. TPP should not harmonise with the EU or Japan on the timing of PSURs.
- Regarding women of childbearing potential, TPP should take a lead in raising the issue of how to develop the best regulatory practice concerning the timing and co-ordination of non-clinical safety studies for the conduct of clinical trials.
- Regarding inclusion of women in clinical trials more generally, TPP should introduce this issue into the harmonisation debate, taking the NIH guidelines as a starting point.
- The idea that the ICH process involves purely "technical" matters solely of concern to scientists is misleading and foolish. It is clear that ICH involves matters of risk-benefit assessment and, therefore, issues concerning "acceptable" risk (acceptable, that is, for patients and consumers). Given that this is the case, the harmonisation process should be much more democratic, in terms of public debate, scrutiny and accountability. In the context of Health Canada's legislative renewal, TPP should encourage public health advocacy groups, patients' organisations and the wider medical and health professionals to become involved in Canada's approach to ICH.
- Health Canada should initiate a process of international harmonisation of good labelling and advertising regulatory standards.
- Health Canada and other parts of the Canadian Government should resist the European Commission's legal challenge to the Bolar provision at the WTO.
1. Legislative Renewal in Canada
Health Canada, the Canadian government's department of health is currently involved in "updating" Canadian legislation on health. This process, known as legislative renewal, began in the Fall of 1998, when Health Canada began the first round of consultations aimed at identifying the issues and problems that new legislation needs to address in order to meet the health protection needs of Canadians in the 2st century. In particular, four Acts are to be "renewed", namely, the Quarantine Act, the Hazardous Products Act, the Radiation-emitting Devices Act and the Food and Drugs Act. In Canada, pharmaceutical products are currently regulated by the Therapeutic Products Programme (TPP) of Health Canada under the 1952 Food and Drugs Act. As this paper is concerned with pharmaceutical harmonisation, only the Food and Drugs Act is relevant here.
Health Canada's legislative renewal proposal for the Food and Drugs Act, drawing on the first round of national consultations, including discussions with 'stakeholder groups', provincial health officials, the Science Advisory Board and other federal departments, is currently being finalised for submission to the Cabinet Committee of the leading (governing) party. After Cabinet comments, the proposal will be presented for a second round of public consultations, probably in late 2000, seeking views on a detailed proposal for a new Food and Drugs Act but the exact timetable will be influenced by the timing of the next general election. The proposal for renewal of this Act will address overall policy direction for pharmaceutical regulation, including product safety, transparency of the regulatory process, protection of companies' commercial information and the need to "modernise" regulatory and enforcement powers. After the second round of public consultations the legislative renewal proposals will go back to the Cabinet Committee before the final stage, which is the parliamentary process.
The Food and Drugs Act governs the regulation of drug safety and efficacy in Canada. Hence, international harmonisation of pharmaceutical regulation could have implications for the renewal of the Act, and vice-versa. Current proposals on renewal of the Act do not specifically mention ICH, but the proposed renewal legislation could provide Health Canada with the authority to enter into international harmonisation agreements regarding requirements concerning the format or content of submissions for marketing authorisations of new drugs. It could also provide the authority to adopt regulations concerning matters, such as the implementation of international agreements and protocols, within federal jurisdiction. A regulation could incorporate, by reference, material produced by another government or government agency, an international body or another outside organisation, such as ICH.
Of course, the contents of the renewed Food and Drugs Act are not yet finalised or known. However, it seems clear that international harmonisation is likely to be embraced in the Act as a general principle, and as something that Health Canada will have the authority to implement at its discretion. It seems unlikely that the renewed Food and Drugs Act will specifically incorporate individual ICH guidelines. Rather it seems much more likely that the renewed Act will simply underline the authority of Health Canada, and hence of TPP, to adopt ICH guidelines at will. The consequence of this is that, although proposals for legislative renewal will go to public consultation and be debated in Parliament, specific regulatory standards, such as ICH guidelines may never be debated in public or by Parliament. In fact, the TPP has already adopted some ICH guidelines, suggesting that it already posseses the authority to do this under the existing Food and Drugs Act without any public consultation or parliamentary scrutiny.
2. History and Rationale of the ICH and International Harmonisation Process
The first thing to be said is that not everyone agrees about the history of the ICH process, although there is a considerable convergence of perspective and most of the differences of opinion appear to ones of emphasis.
Trade and international market competition were key motivating factors. In the mid to late 1980s, bi-lateral initiatives between the governments of the US and Japan were taken, including a determined objective on the part of the US to open up Japanese markets, including the pharmaceutical market. In response, the European Commission strengthened its resolve that there should be a single EU market which could compete with Japan and the US in research and development (R & D) and international trade negotiations. This had two effects: the project of international harmonisation of regulation within the EU was taken much more seriously at all levels of industry and government; and in 1988, the first 'mission' of government regulators and industry representatives from the pharmaceutical sector in Europe were sent to Japan to discuss bilateral harmonisation of regulation between Japan and the EU, so that Japanese markets might become more accessible to the European industry. However, given the importance of the US market, the European pharmaceutical industry was unenthusiastic about solely bi-lateral harmonisation with Japan. Consequently, the International Federation of Pharmaceutical Manufacturers' Associations (IFPMA) took responsibility for organising trilateral meetings between the industry and government regulators in the pharmaceutical sectors of the EU, Japan and the US, which became known as ICH during the 1990s (Interviews with the Association of the British Pharmaceutical Industry [ABPI] and European Commission).
In particular, the European Commission accepted the argument put forward by the European pharmaceutical industry that it was significantly constrained by the lengthy and differing drug registration procedures of the EU Member States (Cecchini 1988). According to the industry, the inconsistencies between national regulatory standards produced wasteful duplication in drug testing, driving up drug development costs. Thus, within the EU, the regulatory agencies and the national pharmaceutical trade associations examined ways of harmonising regulatory standards, especially guidelines for drug development. As the intra-EU 'experiment' in harmonisation grew in viability, and in the context of the inter-regional concerns about market competition mentioned above, a similar argument was applied more globally in the form of ICH, involving the industry representatives and regulators from the EU, Japan and the US as main, voting parties, and representatives from the Canadian regulatory agency, the World Health Organisation (WHO) and the European Free Trade Area (EFTA) as non-voting observers.
To date the work of ICH has focused on the harmonisation and 'rationalisation' of regulatory guidelines for new drug approval (also known as marketing authorisation). It has been concerned with prescription drugs, rather than generics and/or over-the-counter drugs. It is envisaged that regulatory agencies (especially those in the EU, Japan and the US) will adopt ICH guidelines as their regulatory standards, which pharmaceutical companies will then be expected to meet in their new drug applications. As will be explained below, the harmonisation of regulatory guidelines achieved by ICH has not been confined to the reduction of duplication in drug testing requirements, it has also involved the industry challenging the necessity of particular safety checks on new drugs, irrespective of duplication.
Regarding the rationale for ICH, it is also important to appreciate that in the ICH documentation, and at the opening session of the first ICH conference in Belgium in 1991, it was argued that the savings made by companies from harmonised regulations would further the delivery of innovative research yielding therapeutic benefit to patients:
It is only through continuous and painstaking research that new and better treatments can be found. Research costs money and a successful outcome to this conference would release funds currently spent on repeated 'defensive' research, and allow saved resources to go towards the development of new medicines through 'innovative' research, for the direct benefit of patients and consumers (Bangemann 1992:4).
The IFPMA is the formal secretariat for ICH so it is also instructive to examine their published declarations of the rationale for, and implications of, ICH. This argument relating the success of ICH to patient need was reiterated and further emphasised by the IFPMA in 1998 as follows:
The urgent need to rationalise and harmonise regulation was impelled by concerns over rising costs of health care, escalation of the cost of R&D and the need to meet the public expectation that there should be a minimum of delay in making safe and efficacious treatments available to patients in need [emphasis added] (IFPMA. 1998).
And again even more emphatically in 2000:
ICH clearly enhances the competitive position of those companies that choose to operate using its standards, as well as significantly benefiting both the regulators and the patients, who, most importantly, receive crucial new treatments sooner. In summary, harmonisation through ICH brings important, life-saving treatments to patients faster, while releasing the pharmaceutical companies' development funds to projects that will produce the ground-breaking treatments of the future (IFPMA 2000).
The meaning of these claims is clear-cut. The implication is that ICH is in the interests of patients and public health because it will enable the development of more innovative drugs more quickly, which are innovative in the sense of delivering greater therapeutic benefit to patients who need them. Indeed, a significant justification for government regulators' involvement in ICH is that the rationalisation process entailed will contribute to pharmaceutical product innovation in the public interest. Indeed, this is fully appreciated by the ICH secretariat as an informant at IFPMA explained:
The main reason the regulators can justify the time, effort, and expense of getting involved in ICH, is the promise that new and better medicines on a better scientific basis will reach the patient earlier and universally (Interview with IFPMA).
According to the IFPMA, Dr Fernand Sauer, Executive Director of the European Agency for the Evaluation of Medicines (EMEA) has concluded:
The impact of the ICH process can now be seen by regulators when reviewing new drug applications. Consistent submissions allow an early dialogue between EMEA, the FDA and our Japanese counterparts when required. As a consequence, patients throughout the world will have a quicker access to new and better medicines (Sauer quoted in IFPMA 2000: 12).
Other regulators at the Swedish Medical Products Agency (MPA), the EMEA and the EU's Committee for Proprietary Medicinal Products (CPMP) were not so optimistic about ICH's contribution to public health, but they did seem to be convinced that it offered an opportunity for greater innovation in the public interest (Interviews with CPMP, EMEA and MPA).
3. Canada's Role in the ICH Process
As noted above, the Canadian drug regulatory agency, now known as the Therapeutic Products Programme (TPP) of Health Canada has had 'observer' (non-voting) status, within ICH since its first major meeting in Brussels in 1991. This means that the agreement of the TPP (or its predecessor) has not been required for ICH guidelines to be established, but the Canadian regulatory agency has been permitted to participate in discussions at both the Steering Committee and Expert Working Groups of ICH. As Canada has only observer status, one might say that Canadian citizens are even one more step removed from the process than their American, European, or Japanese counterparts in terms of democratic accountability. Having said this there is no evidence that the Canadian drug regulatory agency has objected to any proposals put before ICH, which were not also objected to by one of the main parties, particularly the US Food and Drug Administration (FDA). There are some cases when both the FDA and the Canadian drug regulatory objected to proposals put before the ICH, but Canadian objections went no further than those of the FDA.
So far as the established ICH guidelines are concerned, the TPP has declared that it is 'committed to the principle of harmonisation and is an active participant in the ICH proceedings' (TPP 1996). The TPP has adopted most, if not all of the ICH guidelines, including safety guidelines on carcinogenicity testing and reproductive toxicology. (Also see Canada's Bolar provision under 6).
4. Recommendations that have been put forward
All ICH guidelines have the same status in the sense that a consensus has been reached over them by the six main parties (the regulatory agencies and the industry associations of the EU, Japan and the US). However, the degree and depth of consensus and harmonisation is greater in some than in others. This is often reflected in the extent of detailed prescription contained in the guidelines. Where the guidelines exhibit 'openness' (or more uncharitably, 'vagueness'), this suggests that some or all of the main parties wish to retain flexibility in their approach to drug development and regulation. Where they contain precise recommendations, this suggests agreement to move towards a particular regulatory standard.
Direct review of the ICH guidelines and interviews with pharmaceutical companies suggest that greater harmonisation, rationalisation and reduction in regulatory guidelines has been achieved in the areas of stability testing (contained under ICH Quality Guidelines) and non-clinical safety testing (known under ICH as Safety Guidelines) than in the area of clinical trials (known under ICH as Efficacy Guidelines, though clearly these are also relevant for safety). This view is consistent with results reported by IFPMA who recently stated:
The last ICH Utilisation Survey [of the industry] in 1997 reported utilisation of the Safety Guidelines was the highest of the three areas. This suggests significant savings in both animal resources and time in animal testing in drug development programmes. The international acceptability of studies with ICH study design was considerably increased (IFPMA 2000: 7).
There is also a miscellaneous guideline focusing on overall non-clinical safety testing, known as M3 (Non-Clinical Safety Studies for the Conduct of Clinical Trials for Pharmaceuticals). This paper cannot discuss all recommendations in detail so the main focus will be on those which raise the greatest concerns for public health at this stage.
4b. Quality Guidelines and Product Quality for Users
The ICH has produced 14 guidelines on Quality, focusing on product stability and impurities. Review of these guidelines together with interviews with regulators and scientists in pharmaceutical companies implies that the ICH guidelines on stability testing of conventional pharmaceuticals are likely to raise the standards of drug quality internationally by the use of humidity controls in drug development. As regards the biotechnology drug products quality standards are in flux so regulation needs to be vigilant on a case-by-case basis.
Within ICH, the pharmaceutical industry and regulatory agencies, "quality" denotes the biochemical quality of the drug. However, for users of medicinal products, the quality of product labelling and advertising are very important. To date ICH has not been concerned with these matters, but international harmonisation of good labelling and advertising quality is very important for doctors, patients and other users. This matter should not be neglected any longer.
4c. Safety Guidelines (i.e. toxicology)
Apart from M3, the ICH has produced 14 guidelines on non-clinical safety testing, including one on biotechnology drug products. This section will focus on the guidelines on carcinogenicity and repeat (chronic) dose toxicity because, contrary to what IFPMA (2000: 7) seems to claim, these involve more than solely the eradication of duplicative testing they imply a reduction in regulatory safety checks.
ICH Guidelines on Carcinogenicity Studies
The purpose of carcinogenicity testing is to determine whether a drug causes cancer in the experimental animals (usually rodents), and then to contribute to an evaluation of the drug's carcinogenic risk to humans. Animal carcinogenicity testing is important because cancers are often induced over long periods of the life-span and may manifest themselves some time after the carcinogenic exposure. Testing on animals permits long-term exposure (usually for 18 to 30 months), which is not duplicated in the clinical sphere (and nor should it be). We cannot, and should not, treat patients in clinical trials to test for carcinogenicity because (a) we would have to do so for many years and (b) this would defeat the object of trying to protect people from such exposure. Hence, there is no option but to work with non-clinical data.
According to the FDA, pharmaceuticals generally used for 3 months or more require carcinogenicity testing, while under the drug regulations of the EU and Japan such studies are required if patients take the drug continuously for at least 6 months or frequently in an intermittent manner so that the total exposure is similar to continual exposure of 6 months or more. Furthermore, it is expected that most pharmaceuticals indicated for 3 months treatment would also be likely to be used for 6 months. The ICH Expert Working Group on carcinogenicity testing reports that 'in an enquiry to a number of pharmaceutical research and regulatory groups, no cases were identified in which a pharmaceutical would be used only for 3 months' (ICH 1995a).
In the first instance, this is relevant to the exposure of patients using the drug in clinical trials. Under the ICH Guideline on "The Extent of Population Exposure to Assess Clinical Safety for Drugs Intended for Long-Term Treatment of Non-Life-Threatening Conditions" (E1), the Expert Working Group comments:
There is concern that, although they are likely to be uncommon, some adverse drug events (ADEs) may increase in frequency or severity with time or that some serious ADEs may occur only after drug treatment for more than 6 months. Therefore, some patients should be treated with the drug for 12 months (ICH 1994a).
In other words, when conducting clinical trials with new drugs intended to be used long-term to treat non-life-threatening illnesses, some serious and non-serious ADEs might not be detected properly, or at all, without trials of up to one-year duration. Indeed, for such drugs, clinical trial data on patients treated for 12 months must be submitted to the regulatory agencies prior to marketing approval in the US and Japan, though 'filing for approval will usually be possible based on the data from patients treated through 6 months' (ICH 1994a).
Yet the ICH 'Guideline on the Need for Carcinogenicity Studies of Pharmaceuticals' recommends that 'completed rodent carcinogenicity studies are not needed in advance of the conduct of large scale clinical trials, unless there is special concern for the patient population' (ICH 1995a: 2). Thus, even though it is acknowledged that in the EU, Japan and the US, there must be some clinical trial data of at least 6 months (12 months in Japan and the US) prior to marketing approval, and that the FDA requires carcinogenicity testing for drugs to be used for more than 3 months, the ICH recommends that no carcinogenicity testing needs to be completed prior to exposing patients to new drugs for more than 3 months (or even more than 6 months in Japan and the US) during clinical trials.
Moving on to the ICH recommendation regarding carcinogenicity testing prior to marketing approval. Historically, the regulatory requirements in the EU, Japan and the US have required long-term carcinogenicity testing in two rodent species (usually the rat and the mouse) prior to marketing approval. There are good reasons for this. Scientific studies with known human carcinogens have shown that convergent results in two animal species are better predictors of human carcinogenicity than one result in one species. Thus, if carcinogenicity tests are negative in both mice and rats, one can be more confident of safety in humans, and so on and so forth.
The ICH Expert Working Group on 'Testing for Carcinogenicity of Pharmaceuticals' decided that it was 'in keeping with the mission of ICH to examine whether this practice requiring long term carcinogenicity studies in two species could be reduced without compromising human safety' (ICH 1997a: 1). Consequently, the ICH Guideline on 'Testing for Carcinogenicity of Pharmaceuticals' outlines 'experimental approaches to the evaluation of carcinogenic potential that may obviate the necessity for the routine conduct of two long term carcinogenicity studies for those pharmaceuticals that need such evaluation' (ICH 1997a: 1). This guideline mentioned that consideration has been given to the relative individual contributions of rat and mouse carcinogenicity studies and whether the use of rats or mice alone would result in a significant loss of information on carcinogenicity relevant to human risk assessment. These issues were addressed by six surveys carried out by the International Agency for Research on Cancer (IARC), the FDA, the US Physician's Desk Reference (PDR), the CPMP, the Japanese Pharmaceutical Manufacturers' Association and the UK industry-sponsored Centre for Medicines Research (CMR). However, there is no mention of any definitive conclusion one way or the other. Apparently the surveys identified 'very few instances of mouse tumours being the sole reason for regulatory action concerning a pharmaceutical', but 'data from this species may have contributed to a "weight of evidence" decision and in identifying agents that caused tumours in two rodent species'. They also implied that 'of the compounds displaying carcinogenic activity in only one species, the number of "rat-only" compounds was double the number of "mouse-only" compounds, implying in a simplistic sense that the rat is more "sensitive" than the mouse'. Nevertheless, this also clearly implies that a significant number of carcinogenic pharmaceuticals could go undetected if tested solely in rats.
Despite this very weak basis for discarding the requirement for carcinogenicity testing in two species prior to marketing approval, this ICH guideline recommends that only one long-term rodent carcinogenicity needs to be conducted, plus one other study in a short or medium term in vivo rodent test system. In so doing, it is acknowledged that these short and medium term in vivo rodent test systems are still being investigated for the potential to act effectively as carcinogenicity tests (ICH 1997a: 2-3, 6). Meanwhile, the IFPMA (2000: 7) has declared that 'for carcinogenicity studies, only one long-term study plus one short or medium term study is needed (the latter are being evaluated with the results available at the end of 2000)'.
ICH Guideline on Repeat Dose Toxicity
This guideline was produced by the Expert Working Group on the 'Duration of Chronic Toxicity Testing in Animals'. Chronic toxicity testing in animals is important, though not as special as carcinogenicity testing in animals because there is a greater chance of detecting the former with clinical trials. Historically, in the EU, 6-month animal toxicity testing in non-rodents has been required prior to marketing approval, while in the US, such testing has been required for 12 months. This ICH Expert Working Group set about an investigation of whether there could be harmonisation of the testing duration down to 6 months if studies of 12 months did not detect any toxicity that had not already been identified after 6 months. However, 'in a number of cases there were findings observed by 12 months, but not by 6 months' (ICH 1997b: 1).
The Expert Working Group 'concluded that these [toxicity findings] would, or could have been detected in a study of 9 months duration' (ICH 1997b: 1). In other words, they concluded that new, unforeseen toxicities were being detected after 6 months of chronic testing in non-rodents. It is not at all clear how the conclusion was reached that the toxicities detected at 12, but not 6, months could, or would, be detected after 9 months. The scientific basis for that inference is notable by its absence.
Nevertheless, this ICH Guideline recommends that chronic toxicity testing in non-rodents should be for 9 months in the EU, Japan and the US. Moreover, even though the Expert Working Groups' investigation discovered that unforeseen toxicities were being detected after 6 months of chronic testing in non-rodents, the EU regulatory authorities decided to continue to accept such studies of only 6 months duration (ICH 1997b: 2).
This development could also have some adverse implications for women who are not of childbearing potential. Under the ICH guideline on 'Non-Clinical Safety Studies for the Conduct of Clinical Trials for Pharmaceuticals' (M3), 'women not of childbearing potential may be included in clinical trials without reproduction toxicity studies provided the relevant repeated dose toxicity studies have been conducted'. However, the relevant repeat dose toxicity study in non-rodents is in dispute and ICH has 'non-resolved' this dispute by harmonising safety testing downwards, and hence the safeguards for such women on clinical trials have been reduced.
4d. Efficacy (and Clinical Safety) Guidelines
In general these guidelines are very open and leave regulators and industry with a lot of flexibility when considering the specifics of particular drugs in particular therapeutic categories. This makes sense given that there are also regulatory standards applied by national regulatory agencies concerning particular therapeutic classes of drugs. These ICH guidelines represent good practice in clinical efficacy evaluation and detection of adverse drug reactions (ADRs). However, the WHO had produced equally demanding guidelines on good clinical practice several years before (Interview with WHO). Moreover, there are a few areas worthy of closer examination.
ICH Guidelines on Clinical Safety Data Management: Definitions & Standards for Expedited Reporting (E2a) and Clinical Data Safety Management: Periodic Safety Update Reports for Marketed Drugs (E2c)
These guidelines are concerned with requirements on pharmaceutical companies to report adverse drug events (ADEs) and ADRs to regulatory agencies in a timely manner. Unfortunately the guidelines do not make clear that companies bear full responsibility for the conduct of any foreign subsidiaries. Transnational pharmaceutical companies too often attempt to justify failures to provide regulatory agencies with timely ADR reports because of poor communication between different parts of the company. The ICH guidelines should make clear that both the practice and the attempts to justify it in this way are unacceptable.
For "expected", serious ADRs, it is recommended that "an increase in the rate of occurrence, which is judged to be clinically important," should be reported to regulators (ICH 1994b). Similarly, regarding reporting of ADRs to marketed drugs to regulators, ICH commented:
Increase in the frequency of reports for known ADRs has traditionally been considered as relevant new information. Although attention should be given in the periodic safety update report (PSUR) to such increased reporting, no specific quantitative criteria or other rules are recommended. Judgement should be used in such situations to determine whether the data reflect a meaningful change in ADR occurrence or safety profile (ICH 1996a).
These recommendations permit too much flexibility, which is unnecessary. Unfortunately there have been cases of huge increases in the occurrence of known ADRs, but they have not been reported in a timely manner by the pharmaceutical companies on the grounds that they did not reflect a "meaningful change in ADR occurrence or safety profile". However, that judgement by scientists in industry was not shared by regulators. Leaving the matter to subjective judgement in this way is unsatisfactory and has been proven to be so in the past. The Expert Working Groups on these two guidelines should rethink this recommendation and provide a clear-cut minimum quantitative percentage increase in ADR occurrence warranting a report to regulatory agencies (e.g. 50 per cent increase).
It is also notable that the FDA requires quarterly PSURs during the first 3 years of marketing, while the EU and Japan require PSURs only every 6 months. Yet no attempt appears to have been made to harmonise these requirements upwards to the FDA's standards so as to maximise protection of public health. This is a weakness within the Guidelines on Clinical Data Safety Management: Periodic Safety Update Reports for Marketed Drugs, which ought to be rectified.
ICH Guideline on Ethnic Factors in the Acceptability of Foreign Clinical Data
This guideline makes clear that "the acceptability of the foreign clinical data component of the complete data package depends upon whether it can be extrapolated to the population of the new region" (ICH 1997: 2). The provisions for bridging data on this matter seem to be sensible developments, which have been carefully constructed to take account of the various safety and efficacy issues.
ICH Guideline on Studies in Support of Special Populations: Geriatrics
Importantly this guideline stipulates the following general principle:
Drugs should be studied in all age groups, including the elderly, for which they will have significant utility. Patients entering clinical trials should be reasonably representative of the population that will later be treated by the drug (ICH 1993).
Most of the recommendations in this guideline are concerned with protecting the safety of geriatric subjects and patients and represent very good clinical practice. Despite the major importance of these recommendations, the guidelines states:
It is recommended that exemptions from the guideline be determined in advance either by sponsors or, where feasible, by the sponsor and drug regulatory authorities (ICH 1993).
This statement is unclear and needs to be challenged. It needs to be clear that it is the regulatory authorities who determine whether or not an exemption is acceptable, and that failure by sponsors to gain approval for such exemption would jeopardise marketing approval if the guideline has not been followed. Pharmaceutical companies (i.e. sponsors) should make the case for exemption, if that is their wish, but should not be permitted to determine whether such an exemption is justified.
5. Future Steps and Possible Outcomes
5a. The Delivery of Therapeutic Innovations to Benefit Patients and Public Health
As mentioned above a future promise of ICH made by IFPMA is that this international harmonisation will produce time and cost savings in the drug development process which will lead to the faster delivery of innovative drugs to patients who need them. As indicated above, this rhetoric now appears to have been adopted by the Executive Director of EMEA. The reality, however, may be rather different and interested parties need to be aware of this. Interviews with a small sample (9) of pharmaceutical companies revealed a very high and striking consistency of responses on the "delivery of therapeutic innovations issue".
The interviews were in-depth. The motivations, rationale and benefits of ICH were discussed during which the industry representatives had the opportunity to mention spontaneously the delivery of therapeutic innovations to patients. If this did not occur, then the issue was raised with respondents directly. If they still did not see how ICH could contribute to the delivery of therapeutic innovations to patients, the arguments put forward by IFPMA were spelt out to them to see if they agreed with them.
In these interviews, representatives of industry repeatedly emphasised the commercial importance of harmonisation for reducing R & D costs, but mentioned innovation merely as an afterthought, if at all. Representatives of a British firm explained the industrial perspective as follows:
We need to have, as a pharmaceutical industry, one global development package covering quality, safety and efficacy because there are indeed costs going up. In research and development we have to cover the costs of the ones that fall by the wayside and in addition what's happening is that the pharmaceutical sector is suffering from tight constraints both in volume and in price wherever you are in the world (Interview with UK company 1).
Similarly, informants at a Swedish pharmaceutical company viewed the purpose of ICH as:
to avoid duplicate testing and then to reduce costs - thats one - and then the second is to speed up the registration process and because of that you get earlier registration which then, of course, is beneficial for the company (Interview with Swedish company1)
In fact, informants from industry were frank in admitting that the main point of the ICH process for them was to reduce R & D costs by shedding various aspects of drug testing prior to licensing by regulators. For example, one Director of Regulatory Affairs at a German company stated:
Obviously the industry has an interest in trying to reduce the amount of time one spends in developing products and the best way of doing that is to reduce the requirements for long-term animal testing (Interview with German company1).
Another industry informant commented:
I believe there are some significant advantages [from ICH]. Obviously the cost one is a direct one in terms of the number of studies that we do will decrease which is clearly a benefit. The speed at which we can develop drugs may also increase, which would again be a benefit (Interview with German company2).
One scientist from a British pharmaceutical firm typified the perspective from the industry interviewees as follows:
I'm not sure whether it will influence innovation at all. If you look at the innovation within the industry then it may actually have hours to produce more compounds so that's innovation in that sense. I think the biggest advantage [with ICH] is the time. They [pharmaceutical companies] will try and take more compounds through [the regulatory process]. If the costs to take one drug through are less then to save money you take more drugs through (Interview with UK company2).
Moreover, these perspectives were confirmed by representatives of the major industry trade associations involved in the ICH process. A representative of the European Federation of the Pharmaceutical Industry Associations (EFPIA) commented:
The main aims and objectives are to avoid duplication of effort, to harmonise internationally so that duplication of resources can be avoided, a reduction of animal testing which is also a possible duplication of resources .and basically, to try and speed up the process of registration [licensing] in the three international regions (Interview with EFPIA).
Notably, the concern to save costs by an acceleration of the drug testing and licensing was not accompanied by any mention of (or commitment to) diverting these costs towards drug innovation to meet patients' needs. Indeed, when asked if one purpose of ICH was to further drug innovation, the informant interviewed at the ABPI replied that it was not:
No, the purpose of harmonisation was to actually reduce development times because the duplication required, particularly for reproduction studies, meant that now we had reproduction studies having to be replicated in Japan - taking another 18 months to 2 years, so that added to the development time for Japan or increased your expense in development in that you had to run two studies in parallel - one for Europe and one for Japan. At that time in fact there were three different protocols for reproduction studies - the US, the European and the Japanese. It's [ICH] about bringing down those development times and subsequently the review times to the shortest possible time. (Interview with ABPI).
In the more private context of an interview, even the interviewee from IFPMA, conceptualised drug innovation as somewhat of an afterthought within the ICH process:
Well certainly the speed of getting new drugs on the market - this is one of the objectives of ICH, is that the new medicine should get on to the market quicker. One hopes that a spin-off is going to be that the innovative process will be stimulated. Obviously if registration, research development and registration, are more efficient then you can have more drugs moving down the pipeline. I wouldn't claim for ICH that they're the prime movers in, in making innovation more efficient (Interview with IFPMA).
Regulators involved in ICH justify their involvement in the process by optimism towards its potential contribution to the delivery of therapeutic drug innovation which patients need, but there is little or no deliberate management of knowledge and expertise in the process designed to bring this about. Rather, the realisation of this potential is left to the commercial logic of pharmaceutical companies. Moreover, these regulators' optimism about the relationship between ICH and drug product innovation may be a triumph of hope over reality. The interviews drawn upon in this paper suggest that this should be regarded as a serious concern. At least some in industry conceive of ICH as a cost-cutting exercise via the reduction of drug testing required by regulatory authorities. According to some working in pharmaceutical firms, the savings gained through the ICH process are likely to generate the development and licensing of more medicines more quickly, but not necessarily any drug innovation. More "me-too" drugs of commercial value to companies may make their way through the regulatory process than before ICH in any given time period, but drug innovation in the public interest might not feature in this at all. While this problem cannot be predicted with any certainty, it does at least warrant critical consideration.
5b. Carcinogenicity Testing
The future prospects in this area depend partly on scientific development and partly on the political will of regulatory agencies to take proper steps to protect public health and the health of patients in clinical trials. It is clearly the hope of IFPMA that the studies currently evaluating short and medium in vivo animal carcinogenicity tests will show that they are valid ways of detecting carcinogens. If this occurs, industry and regulators will be encouraged to abandon long-term carcinogenicity studies in two rodent species. However, these evaluative studies need to be examined thoroughly and in the light of full public scrutiny. All data sets on this matter must be in the public domain, free of charge, so that full scrutiny is ensured. They also need to be examined in the light of all the other evidence concerning the validity of the conventional two-species long-term battery of tests.
As indicated by interviews with expert scientists from industry and industry-sponsored research institutions, a key issue is whether or not the new short and medium term in vivo animal tests can detect carcinogens which are not already detected by mutagenicity tests. In other words, can they detect non-genotoxic carcinogens? Interviews were again instructive on this matter. An expert scientist at the CMR commented:
There is a [ICH] guideline that says that you do not necessarily have to do two-rodent species so you can go down to one, but there has to be an alternative, and that alternative, at the moment, has not been validated. It's not clear at the moment whether companies are still carrying out their two-rodent tests, using both the rat and the mouse, until the alternatives are validated. Now there are still question marks on the validation and they are currently being validated and the criteria under which they are being validated are being challenged. The other aspect is whether or not these new tests are actually going to help us in terms of - are they going to give us added value - whether or not these new tests are only going to detect genotoxic carcinogens - because if they were genotoxic they would be knocked out by the genotoxicity battery anyway. So what we are looking for is how do we detect carcinogenic risk of non genotoxic compounds (Interview with CMRI).
Indeed, there is so much scientific uncertainty in this field that one industrial scientist suggested that there was no scientific basis for such validation:
There's a lot of work being done on short-term models but none of them are validated to pick up non genotoxic carcinogens. So there is, again, no scientific basis to choose an appropriate model - that background data does not exist - it's being collected but, as we speak, it does not exist.... And I suspect that when it's collected it will show that it isn't particularly useful...and I think also that there is starting to become some evidence that the mouse study probably does have more value than they [ICH] gave it credit for (Interview with UK company3)
It is not being suggested that efforts to find alternative and effective short and/or medium term carcinogenicity tests should be abandoned or discouraged, but the purpose of ICH was supposed to be to harmonise existing technical guidelines, rather than to encourage the reduction of safety checks by reference to non-existent scientific knowledge. This Expert Working Group would have been far better engaged if it had put its energies into harmonising carcinogenicity testing with the clinical guidelines defining long-term exposure of patients during clinical trials. This would require the completion of at least one carcinogenicity study prior to long-term clinical trial exposure for drugs intended for long-term use and not for life-threatening diseases. This would be a harmonisation that would increase regulatory requirements, but provide added and important protection of health for patients on clinical trial. Sadly, the ICH committee appears not even to have considered this possibility. This failure is unacceptable. Given such failure, this should now be a priority in any future ICH work.
5c. Chronic (Repeat Dose) Toxicity Testing in Non-Rodents
As mentioned above, regarding the field of chronic toxicity testing in non-rodents, a consensus was reached at the ICH in July 1997 that 9-month tests were adequate, whereas previously in the US, the FDA had required these tests to continue for 12 months. The European regulatory authorities had previously accepted a repeat dose toxicity test of merely 6 months duration and continue to do so. Hence, this is another case of the ICH recommending reduced testing requirements (in the US at any rate).
As I have already mentioned, the scientific basis for this recommendation is puzzling. Here again, interviews with representative of EFPIA and IFPMA were instructive. The respondent at EFPIA commented:
I think it wasn't based on science - I think that the most telling thing I could say - it wasn't based on science - there probably isn't enough science to have anything based on it. One could even question the value of any of these things ...and it's, it's, in some sense, at least, let's say, we don't believe it was based on science - I think there were some people in other regions who do think it was based on science (Interview with EFPIA).
Even a representative of IFPMA acknowledged the science was heavily influenced by political considerations in reaching agreement on this guideline:
It isn't pure science. There you are in the US where drugs have always been tested with a year's toxicity and suddenly because of some negotiating with Europe, you're now reducing the safety margin on drugs being tested. ... I think the EU had to be very careful about the public reaction which says "hey wait a minute, all these years we've had drugs on the market which were only tested for 6 months and now you're telling us they should have been tested for 9 months" - you know - and that's why the EU absolutely have not said that drugs have to be tested for 9 months - they will accept 6 months. They will continue to accept 6 months.
There were arguments and when you heard them [the FDA] arguing they looked at these one or two cases which they would have picked up at 9 months. You know, they'd [the FDA] have seen something happening at 9 months which they wouldn't have seen at 6 months - that really was the telling factor. Now the extent to which that was put within a frame because they didn't want to go back to the 6 months - they didn't feel it was going to be politically acceptable (Interview with IFPMA).
In other words, the social viability or political credibility of the regulatory authorities in the various regions significantly influenced the changes to the "scientific" design of this toxicity testing in non-rodents. Other interviewees from the UK Committee on Safety of Medicines (CSM) and the European Commission tended to confirm these perspectives on the ICH negotiation of this guideline (Interviews with CSM and European Commission DGIII).
It is highly likely that IFPMA and other enthusiasts for ICH will want this guideline to remain in the future and to be consolidated. However, it is equally clear that the construction and content of this guideline are entirely unacceptable, and that the whole issue needs to be reconsidered in a much more thorough scientific review in the full view of public scrutiny. In the meantime, in the interests of public health, and of women of non-childbearing potential in particular, the ICH should recommend the adoption of the FDA's original regulatory requirement on one-year chronic toxicity testing in non-rodents.
5d. Efficacy (and Clinical Safety)
As indicated above, many of these ICH guidelines represent good clinical practice and may well make a contribution to public health so long as they are adhered to by pharmaceutical companies. The Expert Working Groups on some of these guidelines are to be congratulated on this. However, there are areas that need to be tightened up, especially those concerning industry reporting of ADRs and industry discretion in determining exemptions. Recommendations need to be much more precise and prescriptive on these matters, where flexibility is a threat to public health, rather than a help.
6. Relationship to WTO, International Trade
The ICH "technical" guidelines are not necessarily linked to the issues raised by "harmonisation" of international trade and intellectual property rights the focus of the World Trade Organisation (WTO). On the other hand, insofar as ICH guidelines raise the standards of clinical testing in developing countries (and perhaps other countries), then they may present barriers of entry to the market for smaller pharmaceutical companies who find it difficult to reach higher regulatory standards. Without doubt the raising of such standards is directly in the interests of patients. However, this also places a burden of responsibility on transnational pharmaceutical manufacturers who can meet the higher regulatory standards to make the drugs developed to those high standards available to people who need them in developing countries. Unfortunately reports concerning the conduct of some pharmaceutical companies and the European Commission suggest the possibility of a bizarre paradox, namely, that raising regulatory standards might make it impossible for poor people in developing countries to access the drugs that they need because they are unaffordable.
In 1998, the European Commission challenged Canada's right to utilise the Bolar provisions, alleging that these provisions are not compatible with Canada's trade obligations under the WTO agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS). Bolar provisions provide specific exemptions from patent infringement and allow generic manufacturers to research and develop a product, and conduct the tests required for approval prior to the expiry of the patent. Indeed generic drugs may even be stockpiled up to 6 months before such expiry so that the drugs are ready to ship as soon as patents expire. The US has adopted similar provisions. The EU, operating on behalf of the transnational brand-name European pharmaceutical industry, asked the WTO to outlaw the Bolar provisions. The consequence of doing so would almost certainly have meant higher prices for consumers. This would make even more drugs very costly and/or inaccessible to poor people in developing countries and their national health systems, whose revenues are needed for many other pressing public health problems besides drug purchasing.
On March 17, 2000, the WTO Panel's report on EU challenge upheld Canada's Bolar exception. That finding was not appealed by the EU and the report has now been adopted by the Dispute Settlement Body of the WTO.
Moreover, it is reported that pharmaceutical companies in alliance with the US Government have attempted to prevent some developing countries, most notably South Africa and Thailand from legally utilising Article 31 of the WTO's guidelines on TRIPS. This permits countries to engage in the domestic manufacture and provision of cheaper versions of brand name drugs prior to patent expiry in cases of emergency ("compulsory licensing") or shop around for cheaper versions prior to patent expiry in cases of emergency ("parallel importing"). Both South African and Thailand consider that their AIDS epidemics represent an emergency and they need anti-AIDS drugs. The brand name version costs US $10,000 per patient per year, while the average income in South Africa is US $1,000 per year. Hence cheaper access to anti-AIDS drugs must be found in the interests of public health.
Allegedly, the US Government has been involved in trade wars with South Africa and Thailand on behalf of US-based transnational pharmaceutical companies in order to intimidate these developing countries into refraining from compulsory licensing or parallel importing. Given this situation, the ICH and the WTO should condemn these intimidatory and litigious challenges by the US Government, the European Commission and transnational pharmaceutical companies. Furthermore, the Canadian Government should continue to defend the Bolar provision as it did with the European Union.
Bangemann, M. (1992) "Welcome and Introduction: Welcome address by Mr martin Bangemann, Vice-President of the Commission of the European Communities" in P.F. D'Arcy and D.W.G. Harron (eds.) Proceedings of the First International Conference on Harmonisation. Belfast: IFPMA.
Cecchini, P. (1988) The European Challenge, 1992, Aldershot, England: Wildwood House.
ICH (1993) Guideline on Studies in Support of Special Populations: Geriatrics, 24 June.
ICH (1994a) Guideline on The Extent of Population Exposure to Assess Clinical Safety for Drugs Intended for Long-Term Treatment of Non-Life-Threatening Conditions, 27 October.
ICH (1994b) Clinical Safety Data Management: Definitions & Standards for Expedited Reporting, 27 October.
ICH (1995a) Guideline on the Need for Carcinogenicity Studies of Pharmaceuticals, 29 November.
ICH (1997a) Testing for Carcinogenicity of Pharmaceuticals, 16 July.
ICH (1997b) Duration of Chronic Toxicity Testing in Animals, 16 July.
ICH (1997c) Guideline on Ethnic Factors in the Acceptability of Foreign Clinical Data, 6 March.
IFPMA (1998) "A brief history of ICH" in ICH: Past and Future at www.ifpma.org/ich1.html
IFPMA (2000) The Value and Benefits of ICH to Industry. IFPMA.
TPP (1996) "Dear client" letter, re guideline on the need for carcinogenicity studies of pharmaceutials, April.
Copyright ©2006-2007 Women and Health Protection