I Introduction
Pharmaceuticals for human use have astounding health benefits.Footnote 1 They allow us to live longer and more productive and higher quality lives. At the same time, residues from the active pharmaceutical ingredient (API)Footnote 2 found in the natural environment (surface and ground water, soil and aquatic and in-land species) have been linked to negative and unintended effects on ecosystems, human health and animal health. Some pharmaceutical waste products have the ability to accumulate up the food chain, and waste products from pharmaceuticals, such as anti-cancer drugs, can directly disrupt our DNA.Footnote 3 Some prominent and widely-publicised examples of negative effects include: the presence of chronic low-levels of hormones in rivers and streams that have caused sex change in fish;Footnote 4 the decline of Indian vulture populations due to feeding off cattle treated with non-steroidal painkillers;Footnote 5 and the rise of anti-microbial resistance due to the chronic presence of antibiotics in effluent close to pharmaceutical manufacturing plants.Footnote 6 Recently, the US Geological Survey supported a study which found that pharmaceutical residues in waste water effluent can alter fish behaviour in ways that are central to sustaining their populations.Footnote 7
Pharmaceutical companies acknowledge the risks associated with pharmaceutical residues in the environment, but consider them insignificant compared to the benefits derived from their therapeutic use.Footnote 8 Because of their significant value to public health, the regulatory decision to place a new pharmaceutical drug on the market requires a comprehensive assessment of all the risks and benefits of the medicine in question. During this assessment, the risks to patient health are rigorously tested and balanced against the potential therapeutic benefits of the drug. However, the environmental risks are in practice not part of this balancing act, and there is debate as to whether environmental effects should be included in the risk-benefit analysis.Footnote 9
The pharmaceuticals in the environment problem (PiE) has been gaining the attention of public authorities, practitioners and academics. Within the legal field, the main focus has so far been on identifying legal challenges and gaps in the general regulatory framework. In the EU context, a major contribution can be found in the BIOIS report, summarising the legal areas needing more attention, including the marketing authorisation procedures, Good Manufacturing Practices, legislation regarding chemicals, regulation of the disposal of pharmaceutical waste and water legislation. This article does not cover all the identified legal challenges, but provides a focused analysis within one of the areas. The purpose is to compare the approaches taken in the EU and the US to managing the risks posed by pharmaceuticals to the environment within the marketing authorisation procedure (MA) in the EU and or, as it is known in the US, the new drug application (NDA) procedure.Footnote 10 The discussion about the risks posed by pharmaceuticals in the environment is, in essence, a discussion about the love/hate relationship society has with hazardous substances.Footnote 11 Some critics argue that medicines are given more leeway or regulated in a more relaxed manner than other hazardous substances, due to the significant health benefits they provide.Footnote 12 A debate about the risks and benefits of pharmaceuticals will always be controversial, as risks tend to lie with the more vulnerable sectors of society and the benefits with the more informed and powerful sectors of society.Footnote 13 It also requires a balancing act of different values, namely public health versus the environment and the priority placed upon these values can differ from place to place.Footnote 14
This article undertakes a comparative analysis of the EU and US pharmaceutical regulatory procedures in the context of the PiE problem, in order to detect the differences and similarities and to ask ourselves whether there is a better way forward.Footnote 15 The reason to focus this research on EU and US law is because these two regions are responsible for about two thirds of the global consumption of pharmaceutical drugs.Footnote 16 In addition, pharmaceutical residue predominantly enters the environment via patient use.Footnote 17 There is also a rich academic debate discussing the differences and similarities of the US and EU approaches to risk assessment of products in general. Some of this debate has taken place under the guise of some well-known trade disputes, for example theWTO Hormones decision.Footnote 18 Another reason is that the EU and the US are both founding members of the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH). The purpose of this organisation is to achieve greater global harmonisation of laws concerning human medicines. The European Commission and the US Food and Drug Authority (FDA) have committed to implementing the guidelines produced by the ICH.Footnote 19 Therefore, comparing the differences and similarities of the US and EU will provide an insight into the progress made into harmonising the law in this area.
Part II of the article begins by outlining the main known and unknown risks pharmaceuticals pose to the environment. It also explains the main gaps in our scientific knowledge about the environmental effects of pharmaceuticals. Part III of the paper gives a critical analysis of the US and EU’s approaches to assessing the risks posed by pharmaceuticals to the environment. There are many different ways to approach risk management, and to address all aspects of the law in this area is beyond the scope of the paper. Therefore, this article focuses on pharmaceutical regulatory procedures, and directly compares EU and US provisions which deal with the assessment of the environmental effects under this procedure, also known as the environmental risk assessment (ERA) in the EU and environmental assessment (EA) in the US.Footnote 20 It then compares the legal requirements to assessing the risk to the environment. The article concludes with a discussion of the findings and provides a different perspective on how to manage the risk of PiE. This is highly topical, as the European Commission is currently preparing a strategic approach to PiE.Footnote 21
II What do we (not) know
As briefly outlined in the Introduction, the environmental risk is just one of the weights on the scale when balancing the risks and benefits of a new drug entering the market. It is a weight that has so far been treated as a marginal one, as demonstrated in the following discussion.
In terms of their direct impact on human beings through their safety, quality and efficacy characteristics, pharmaceuticals are strictly evaluated before being placed on the market.Footnote 22 However, when it comes to assessing their impact on the environment – including their indirect impact on human health – there are obvious gaps in the applicable EU and US law, which arguably need addressing. Before the law can strike an appropriate balance between protecting patient’s access to good quality medicines and protecting the environment, gaps in our scientific knowledge must be addressed.Footnote 23 These gaps are caused by both regulatory practices and scientific limitations.
Firstly, the discussion about the effects of pharmaceutical residues on the natural environment is often treated separately from the discussion of the effects of pharmaceutical residue on human health. The World Health Organisation (WHO) concluded that pharmaceuticals in drinking water do not pose a threat to human health.Footnote 24 Pharmaceutical companies second this position. For example, Roche contends that the levels at which pharmaceuticals have been found in the environment are in general too low to have an adverse effect on humans,Footnote 25 and thus too low to trigger a detailed environmental risk assessment.Footnote 26 Nevertheless, some scientific studies have established that the levels of pharmaceuticals found in discharge from waste-water treatment plants are high enough to be harmful to aquatic speciesFootnote 27 and through this also harmful to humans. Therefore, the simplistic view of seeing human health as separate from environmental health is not correct when you consider the effects upon humans from exposure to pharmaceutical residues in water and soils.Footnote 28 In fact, the effects of long-term chronic exposure to even low levels of pharmaceuticals is one of the main topics in both academic and practitioners’ discourse.Footnote 29
The second major issue is a lack of scientific data. Public authorities in the US and EU in general do not routinely test for the presence of pharmaceuticals in water and soils.Footnote 30 The WHO recognises that this lack of good scientific data makes it challenging to effectively assess the human health risks from exposure to pharmaceutical residues.Footnote 31 In addition, confidential data generated during the MA process in the EU and US is private, making it difficult for individuals and NGOs to inform themselves about the environmental risks posed by pharmaceuticals.Footnote 32 This in turn presents a significant barrier for informed decision making in the public sphere about the risks and benefits of pharmaceuticals.
Lastly, the current approach to assessing the risks of medicines in the EU and US is for a particular medical product, not for the API within the medicine.Footnote 33 This approach is not effective in assessing the cumulative impact of PiE. Once pharmaceuticals enter the waste water system, they are mixed with other pharmaceuticals and all the other chemicals we use on a daily basis, and there is very little knowledge about the long-term effect these mixtures have.Footnote 34 The WHO has also acknowledged that there is not enough knowledge about the effects of long-term chronic exposure to pharmaceuticals and the possible combined effects of mixtures upon organisms in water and soil.Footnote 35
The known impacts of pharmaceuticals on the environment together with the much less documented impacts of pharmaceutical residue on human health call for precautions.Footnote 36 While we cannot speak about scientific certainty regarding the detrimental effects on human health, there is growing scientific evidence of the same.Footnote 37 This calls for more attention to the environmental risks during the pharmaceutical regulatory procedures.
III Analysis
This part describes and compares the current use and practice of E(R)A in the EU and the US pharmaceutical registration procedures. It starts by exploring the relevance and weight of the E(R)A within these procedures, then moves onto the scope of the E(R)A, and lastly compares the requirements related to assessment panels and public access to information. Reflections on the comparison are summarised in the Conclusion.
1 Relevance and weight of the E(R)A
The main focus of this article is to compare how the EU and US balance the risks to the environment against the benefits of using pharmaceutical drugs. A crucial part of this analysis is the impact the E(R)A has upon the risk-benefit analysis within the regulatory procedures. At first glance, the US and EU take a very similar position. Under both jurisdictions, the drug may be approved, even if the E(R)A shows there will be an adverse impact on the environment. However, there are slight differences in approach when you take a closer look.
Directive 2001/83/EC contains the EU code for medicinal products of human use, covering issues from marketing authorisation procedure to the disposal of unused pharmaceuticals. Article 8 sets out what must be included in the application for an MA of a new medicinal product. Article 8(3)(ca) states that an evaluation of the potential environmental harm, or an ERA, must be included in the application.Footnote 38 Article 8(3)(ca) is a relatively new requirement and came into force in October 2005.Footnote 39 This amendment can be seen as an expression of the integration principle as required by Article 11 of the Treaty of the Functioning of the European Union (TFEU).Footnote 40 Annex I, section 1.6 contains further details which the applicant must provide in the E(R)A, details of this will be discussed later on in the article.
The US takes a slightly different approach to regulating the effects of pharmaceuticals in the environment. The law in this area is contained in several pieces of legislation. The passing of the National Environmental Policy Act (NEPA) in 1969 under President Nixon, brought in the era of America’s modern environmental laws.Footnote 41 NEPA “requires all Federal agencies to assess the environmental impact of their actions”, and the FDA must implement this requirement by assessing the environmental impact of drugs during the new drug application (NDA) process.Footnote 42 Title 21, Part 355(a) of the United States Code (USC),Footnote 43 creates the necessity to apply for the approval of any new medicine that wishes to enter interstate commerce.Footnote 44 Part 355(b)(1)(A) requires the applicant to include full reports of investigations showing “whether or not the drug is safe for use and whether such drug is effective in use”. In addition to this, NEPA requires the FDA to assess the environmental impact of new medicines which seek the FDA’s approval to market a new medicine (Marketing Approval, or MA).Footnote 45 Title 21, part 25.15 and 25.20 of the Code of Federal Regulations (CFR) sets out the requirement to submit an EA when an applicant applies for the MA of a new medicine.Footnote 46 The FDA also has the power to refuse to file or approve an application for a new medicine if the applicant fails to provide an adequate E(R)A, or a claim of categorical exclusion.Footnote 47 An adequate E(R)A is one which addresses “the relevant environmental issues”, and needs to provide the FDA with enough information to decide whether the new medicine could “significantly affect the quality of the human environment”.Footnote 48
EU law takes a lenient approach to the obligation to submit a satisfactory E(R)A during the MA procedure. Practice indicates that an incomplete or missing E(R)A will not prevent an MA being given for a new medicine.Footnote 49 A study in 2014 found that 37% of the E(R)As for products registered through the centralised procedure carried out between 2011–2012, were submitted after the deadline.Footnote 50 They also found that 83% of the E(R)As submitted were of an incomplete or unsatisfactory nature.Footnote 51 The same study noted that this lack of compliance with the time-frame to submit an E(R)A indicates that companies do not prioritise E(R)A-related issues during the MA procedure.Footnote 52 A major reason for an MA being given to a medicine despite a missing E(R)A is because the E(R)A is not part of the risk-benefit analysis.Footnote 53
Article 2 of the 2001 Directive clearly states the following:
28. Risks related to use of the medicinal product:
— any risk relating to the quality, safety or efficacy of the medicinal product as regards patient’s health or public health;
— any risk of undesirable effects on the environment.
28a. Risk-benefit balance:
An evaluation of the positive therapeutic effects of the medicinal product in relation to the risks as defined in point 28, first indent. (emphasis added)
Because environmental risks are not included in the risk-benefit analysis, the E(R)A results have no impact on the final approval of a medicine. When there is an undeniable impact on the environment, risk management measures “shall be envisioned”.Footnote 54 The pharmaceutical industry supports the approach of granting MAs with post-marketing commitments, such as the submission of an E(R)A and monitoring under the pharmacovigilance system,Footnote 55 or eco-pharmacovigilance,Footnote 56 when considering environmental endpoints.Footnote 57 Those who defend the exclusion of the E(R)A from the risk-benefit analysis argue that lawmakers tried to reflect society’s concern that if environmental impacts were relevant in the ultimate balancing act of risks and benefits, then a potentially lifesaving medicine could be refused on the basis of its environmental impact.Footnote 58 Those who argue for the consideration of the E(R)A results during the risk-benefit analysis, state that its inclusion would be very unlikely to result in the exclusion of new medicines on the market. Rather it should target the approval of known problematic APIs.Footnote 59 This would be done by balancing the expected therapeutic benefits against the environmental impact, with particular consideration of whether there are already alternative and equivalent treatments available with less environmental impact.Footnote 60
Risk assessment involves the balancing of different values and the Court of Justice of the European Union (CJEU) in the Pfizer decision discussed (amongst other things) how the EU institutions determine the acceptable level of risk, and thereby the level of protection given to each value.Footnote 61 The court explains this is done by the EU institutions defining the political objectives to be pursued through a certain legal measure, within the bounds of the powers given to them by the Treaties.Footnote 62 The legal basis of Directive 2001/83/EC is the internal market.Footnote 63 Article 114 of the TFEU governs the EU’s law-making power in furtherance of the internal market. Article 114(3) states that the Commission must assign a high level of protection to health and the environment when making their proposal. This “high level of protection” is reflected in the second paragraph of the preamble of the medicines Directive,Footnote 64 making the Directive’s objective a set of rules allowing for the free movement of high quality and safe medicinal products. The environment arguably has less value under EU pharmaceutical law when it concerns human medicines. This could be because the possible detrimental effects on aquatic and land species is considered less significant than the availability of medicines and the competitiveness of the European pharmaceutical industry on the global market.Footnote 65 However, as argued above, this is a simplistic view, ignoring any indirect effects of pharmaceutical residue in the environment on human health.
The US District Court for the Southern District of New York, in the case Natural Defence Council v United States Federal Drug Authority, also discussed the threat of AMR due to non-therapeutic use of antibiotics in animals for human consumption.Footnote 66 The court ruled that when the FDA decides to approve an NDA or withdraw an existing medicine from the market, “the inquiry focusses on whether the drug is safe and effective”.Footnote 67 This includes the requirement to submit “full reports of investigations which have been made to show whether or not such drug is safe and effective for use”.Footnote 68 In this case, the FDA was found to have not evaluated the science presented by the plaintiff describing the risk of AMR, nor did they assess the safety of the relevant drugs.Footnote 69 The court confirmed the statutory mandate for the FDA’s action comes from Congress,Footnote 70 and upheld the decision of Massachusetts v EPA, that an agency “must ground its reason for action or inaction in the statute”.Footnote 71
In the US, when preparing an E(R)A, the applicant is required to consider whether there were other possible solutions available which were less environmentally damaging, and if this course of action was not taken, they must explain why it was not taken.Footnote 72 This balancing act is a practical example of how to include the E(R)A results into the risk-benefit analysis, and should be picked up by critics of EU law, who seek for the inclusion of the E(R)A results in a similar fashion. If the FDA decides to approve a new medicine, they can require the applicant to implement any measures mitigating the medicine’s effect on the environment and make these mitigating measures part of the medicine’s approval.Footnote 73 The FDA can also withdraw approval of the medicine if new information becomes available showing a serious adverse effect upon the environment.Footnote 74 This is similar to the pharmacovigilance system under EU law, which will also review the MA if new information becomes available relating to the safety of the medicine.Footnote 75
Under EU law, Annex I section 1.6 of Directive 2001/83/EC contains a more detailed requirement of the E(R)A and is directed solely at the applicant. The applicant “shall include a risk assessment overview evaluating possible risks to the environment due to the use and/or disposal of the medicinal product”.Footnote 76 This is in contrast to the US approach, where first the applicant is required to provide an EA.Footnote 77 The FDA will then evaluate the EA to ensure it is objective and accurate, and determine if the medicine’s approval will “significantly affect the quality of the human environment”.Footnote 78 If the FDA finds this to be the case, they will produce an environmental impact statement (EIS).Footnote 79 These two features of the law, the legal requirement to ensure that the EA is objective and accurate and the possibility of the FDA producing their own assessment of the potential environmental impacts, leads to the conclusion that the US approach arguably has more checks in place to ensure that the EA is of a high standard. However, this might not be the case, as the application of the US law is largely limited by the categorical exclusions, as will be shown below.
2 Scope of the E(R)A
This section highlights the main differences and similarities of what the law in the EU and US includes within the scope of the E(R)A.
As both the EU and US, along with Japan, are founding members of the ICH, they must adopt the guidelines issued by this body. To date the ICH has issued guidelines for the testing of the following effects of medicines during the MA procedure: carcinogenicity;Footnote 80 genotoxicity;Footnote 81 chronic toxicity;Footnote 82 reproductive toxicity;Footnote 83 and immunotoxicity.Footnote 84 There are no guidelines which address the risk of medicinal residues in the environment, which could be one explanation for the lack of harmonisation between the EU and US in this area.
A common feature of the scope of the E(R)A under EU and US law is if the medicine is unlikely to significantly affect the environment, the applicant is not required to prepare an E(R)A.Footnote 85 Under US law, the FDA will consider the cumulative effect of the medicine upon the environment.Footnote 86 This is not the case under EU law, where the potential environmental concentration (PEC) of the medicine is calculated on an individual basis.Footnote 87 Interestingly, under EU law if the medicine has endocrine disrupting properties, the applicant will still be required to perform an E(R)A even if the medicine does not present a risk. Conversely, US law does not consider the risk posed by endocrine disrupters to aquatic species, even at very low levels.Footnote 88
A striking difference between the US and EU is the approach to considering the effects upon the environment from manufacturing effluent. In the US, the applicant must consider the risks from manufacturing effluent to endangered species and habitats critical for those species. Under section 25.21(b) of the CFR, the FDA will require an E(R)A if a species or critical habitat protected under federal law or under the Convention on the International Trade in Endangered Species of Wild Flora and Fauna (CITES)Footnote 89 could be adversely affected by the approval of an NDA.Footnote 90 Guidelines issued by various federal bodies, including the FDA, require applicants to consider the adverse effects from manufacturing emissions upon species or habitats protected under the abovementioned laws.Footnote 91 By considering species protected under CITES and the Endangered Species Act, section 25.21(b) requires applicants to consider the effects of manufacturing effluent upon species outside of the US.Footnote 92 This is significant because China and India account for up to 90% of the global production of antibiotics.Footnote 93 An investigation revealed that poor manufacturing practices in China led to the release of antibiotics into the local environment, making these factories an important contributor to the global problem of AMR.Footnote 94
Unlike US law, applicants under the EU MA procedure are not required to consider the effects upon the environment from the manufacture of medicines.Footnote 95 Presumably, the Industrial Emissions Directive (IED) would cover the release of pharmaceuticals from manufacturing processes.Footnote 96 Annex II of the IED lists the substances for which monitoring and emission limit values should be set. To date, Annex II does not list any APIs. Consequently, pharmaceutical manufacturers are not required to specifically control the emission of APIs into the air or water.Footnote 97 In addition, unlike the US, applicants under the MA procedure in the EU are not required to evaluate the impact of the medicine upon endangered or protected species in the EU.Footnote 98 Nor does it mention the need to consider the impact upon sites within the Natura 2000 network.Footnote 99 By comparing the scope of the E(R)A in this instance, it appears the US places more value on the protection of endangered species and their habitats than the EU.
Another commonality between the EU and US is the large number of exceptions to the requirement to submit an E(R)A during the MA procedure. Under US law, Part 25.31 of the CFR lists the numerable exemptions which relate to human medicines.Footnote 100 In general, if the proposed approval of the medicine does not result in an increase in the use of the relevant active ingredient (25.31(a)) and/or an increase of discharge into the environment (25.31(b)), it will be covered by the categorical exclusion. In 1997, the number of categorical exemptions were significantly increased, due to President Clinton’s reinventing government (REGO) initiatives.Footnote 101 According to critics of this reform, a result of these REGO initiatives was that only medicines whose sales exceeded 44 tonnes per year required an E(R)A.Footnote 102
Despite this cutting of green tape, the FDA still has the right to require an E(R)A in extraordinary circumstances. Under part 25.21 of the CFR, even if the applicant is entitled to a categorical exclusion, the FDA will still require an E(R)A in extraordinary circumstances, namely if the “action may significantly affect the quality of the human environment”.Footnote 103 There are two specific examples of an extraordinary circumstance given in the CFR:Footnote 104 firstly, situations where the data points to “the potential for serious harm to the environment” (this includes long-term effects on the functioning of ecosystems and toxic effects on organisms);Footnote 105 secondly, direct and indirect effects (eg effluent from manufacturing), which could adversely affect an endangered species or habitat critical for that species’ survival.Footnote 106
Similar to the US, under the EU MA procedure not all applicants are required to provide an E(R)A. Since 30 October 2005, all applicants seeking an MA for both original and generic pharmaceutical products,Footnote 107 must submit an E(R)A with their applications. There are, however, some exceptions that apply to medicines authorised prior to 30 October 2005. For applications that present a “major variation”Footnote 108 of a previously authorised medicine, the applicant must submit an E(R)A only if there is an expected increase in environmental exposure.Footnote 109 Similarly, applications to extend an MA are only required to include an E(R)A if there is a potential increase of environmental exposure.Footnote 110 The authors of the BIOIS report argue that in practice this means that medicines authorised prior to 30 October 2005 are not required to carry out an E(R)A.Footnote 111 This is significant, because the number of new APIs arriving onto the market is likely to be small, compared to what has been authorised in the past. Some of these older generic medicines are the main culprits in the contamination of surface water in the EU.Footnote 112 Moreover, according to the EU EMA Guidelines, for some pharmaceutical products the E(R)A may consist solely of an explanation as to why no risk to the environment is expected, but there is a lack of guidance as to what is considered an adequate justification.Footnote 113
Experts from the BIOIS Report furthermore argue that the requirement to submit an E(R)A has no retroactive effect.Footnote 114 This is significant from an environmental perspective because the ten most commonly found pharmaceuticals in drinking water were given an MA before 2004, therefore they were not required to submit an E(R)A during their MA application.Footnote 115 However, in 2004 the 2001 Directive was amended by inserting the condition that the MA is valid for only five years.Footnote 116 After this period an MA may be renewed after an evaluation of a risk-benefit analysis.Footnote 117 It could thus be reasonably expected that the potential environmental risks of the “old” medicinal product should be assessed when an MA is renewed.
Another point of difference from the US law is when preparing the E(R)A, applicants under EU law whose product contains Genetically Modified Organisms (GMOs), must comply with a special set of requirements set out in Annex I, section 1.6 of Directive 2001/18/EC.Footnote 118 This includes preparing a risk management strategy, designed to address the risk of GMOs to the environment. This is strikingly different from US law, which makes no specific mention of the need to address risks associated with GMOs. When you can pinpoint a real difference in how countries manage the same risk, Jasanoff argues this is a sign that different countries have different values and politicians within those countries have made certain commitments to their electorate to uphold those values.Footnote 119 It is well known that the EU is more adverse to GMOs than the US.Footnote 120 Hornsby argues the European’s scepticism of GMOs stems from well-publicised food scares.Footnote 121 Consequently, very few Europeans are willing to trust their governments when it comes to biotechnology, hence the negative public perception of GMOs and the science behind them.Footnote 122 In addition, the US biotech and agricultural industry has been very influential in shaping public opinion and GMO regulations, and more organised than their European counterparts.Footnote 123
In both the EU and US, the applicant is required to develop risk mitigation measures if there is an unavoidable risk to the environment. In the US, the FDA can impose the obligation to implement the risk mitigation measures as a part of the MA process,Footnote 124 whereas in the EU, it is the MA applicant who is expected to envisage risk mitigation measures within the E(R)A.Footnote 125 Moreover, such precautionary safety measures within the EU are limited to those taken for the storage of the product, its administration to patients and for its disposal.Footnote 126
Regarding the packaging requirements of medicines, in the EU the MA applicant is required to provide in their application a “mock up” of the outer packaging.Footnote 127 This must, where appropriate,Footnote 128 include instructions for the disposal of unused medicines or the waste thereof and a reference to the appropriate collection system available in that Member State.Footnote 129 In the US, the MA applicant is also required to provide a specimen of the proposed labelling, however there is no specific requirement to inform patients about the correct way to dispose of expired or unused medicines. Alternatively, this information is provided to patients via a website maintained by the FDA.Footnote 130
While the legal tools within the EU and US are designed differently, the scope of medicines exempted from the E(R)A requirement is almost the same. However, when weighing up the pros and cons of both the EU and US approach to PiE, there are some factors which favour the US solution: firstly, the authorities can refuse the authorisation of a medicine because of a missing E(R)A; secondly, despite the large number of categorical exclusions, the FDA can still require an E(R)A in exceptional circumstances.
3 Assessment panels and public access to information
When it comes to assessing the data and information gathered by the applicant within the pharmaceutical regulatory processes, the types of expertise available within the panels which consider the application become relevant. US law requires the panel members to come from a variety of technical backgrounds, including expertise in biological and physical sciences.Footnote 131 They must nominate members who have experience in the different stages of the medicine’s lifecycle,Footnote 132 and they require a panel member who represents the interests of consumers and the pharmaceutical manufacturing industry.Footnote 133 In addition, scientific organisations have the opportunity to nominate individuals for appointment to the assessment panel.Footnote 134
Under EU law, the EMA is responsible for forming the advisory panels which consider the MA under the centralised procedure.Footnote 135 The Pharmacovigilance Risk Assessment Committee (PRAC) must consist of members with expertise in pharmacology and pharmacoepidemiology, a representative of healthcare professionals and a representative of patient interests.Footnote 136 The PRAC is expected to contain members with the highest level of qualifications and from a broad spectrum of interests, however ultimately members are appointed on the basis of their expertise in pharmacovigilance and risk assessment of medicinal products for human use.Footnote 137 Furthermore, the Committee for Medicinal Products for Human Use (CHMP), which evaluates the MA application, does not require members to have expertise in the area of environmental risk assessment.Footnote 138 By comparison, the US takes a more inter-disciplinary approach and has a wider range of stakeholders, whereas the EU is more focused on members with experience concerning the effects of medicines upon humans through the use of medicines. In addition, the rules for nominating members of the PRAC do not require experience in the different stages of the medicine’s lifecycle.
In both the EU and US, the presence of expert advisory committees is a common feature in policy-making, with their mission being “to ensure the scientific reliability of policy choices”.Footnote 139 However, there have been several scares on both sides of the Atlantic which caused the public to question the expert assessments which support policy decisions.Footnote 140 In particular, Europeans were alarmed by the illegal use of growth hormones in meat for human consumption. The policy decisions, and trans-Atlantic trade disputeFootnote 141 which followed, revealed a difference in how EU and US policy-makers understood and engaged with science while setting risk regulations. According to Hornsby, the EU institutions were unable to agree about how to manage the risk of growth hormones used during beef production, so they took a no-risk approach and cherry-picked the scientific evidence to support their decision.Footnote 142 This risk management strategy was taken despite most scientists agreeing that growth hormones in meat do not pose a risk to human health.Footnote 143 By contrast, US policy makers were more willing to trust international scientific bodies, such as the Codex Alimentarius, to correctly assess and set the appropriate level of risk for hormone-fed meat.Footnote 144
The WTO panel, established to settle the trade dispute between the EU and US, focused upon the method by which the EU established the level of risk it considered appropriate to protect its citizens from hormones in meat.Footnote 145 The panel agreed with the US approach, in that the appropriate level of risk should be applied in a consistent manner in line with the provisions of the Sanitary and Phytosanitary (SPS) Agreement, not upon consumer’s perceptions of that risk.Footnote 146 It is interesting to note Norway’s submission before the panel. Norway argues it is a sovereign right to determine the level of risk deemed appropriate for the protection of its citizens’ health and the measures appropriate to achieve this risk.Footnote 147 They contend that nations do not have “to scientifically prove the extent of the risk”, only show that a risk existed. It was then for the nation “to define the level of probability of harm it wanted to assume”.Footnote 148 The difference in the EU and US approach to the role of science in setting risk management measures is interesting because it highlights the different views on what is sufficient science and the inherently political nature of E(R)As.Footnote 149
In relation to the disclosure of information to the public gathered for the purpose of the E(R)A during the pharmaceutical regulatory processes, the US and EU differ only slightly in their approach. Both jurisdictions ensure a data protection (and data exclusivity) period to MA applicants of original medicines.Footnote 150 This means that data submitted with the MA application cannot be relied on by another applicant, primarily by an applicant for a generic product’s MA, before the data protection period expires. The rationale behind providing this special type of intellectual property right to medicines is the high costs and risks associated with the development of a new product. The data protection thus has introduced a right parallel to patent protection, in order to support innovation in the pharmaceutical sector. Moreover, confidential environmental data can be protected as a trade secret.Footnote 151 In order to secure proper data protection within the US context, the law distinguishes between confidential and non-confidential parts of data submitted with the E(R)A.Footnote 152 The confidential data must be summarised in the E(R)A to the extent possible, but will not be disclosed in full to the public.Footnote 153 Non-confidential information shall be made available to public upon request.Footnote 154
Similarly to the US, in the EU context, the EMA does not publish results of E(R)As in full, but only the non-confidential parts, such as the fate and effects endpoints of human and veterinary pharmaceuticals, in a standardised format.Footnote 155 The results of the E(R)A are made available to public within so-called European Public Assessment Reports (EPAR), which describes them in language understandable to the public.Footnote 156 The EMA does not require an applicant for an MA to distinguish between confidential and non-confidential parts of the data. This leads to the situation where the format of EPARs is standardised, but the content is not. The result is a highly non-transparent situation, with the duplication of environmental test data (including animal testing) as each applicant has to generate their own data set, and frequent inconsistencies in E(R)As’ results.Footnote 157
Experts have been calling for the EU to amend the law and make scientific data gathered during the MA more readily available to the general public and professionals.Footnote 158 This would, firstly, allow greater access to information about the safety and efficacy of the drug and, thus, promote more informed public debate about the risks and benefits of pharmaceuticals. Secondly, it would be helpful for MA applicants when preparing their E(R)A when identifying knowledge about the respective API. This would support consistency between E(R)As, allow increased learning and avoid duplications in testing.Footnote 159 However, the protection of intellectual property rights stand in the way of implementing this increased transparency. Implementation of such data publication and sharing would therefore need legislative changes.Footnote 160 Yet, recent case law of the CJEU broadened up the possibility to request more MA-related information within the current legal framework.Footnote 161 This is a positive development towards securing public access also to E(R)As’ results. The three cases discussed, among other things, the scope of commercially confidential information. The court noted that in order for information to be considered commercially confidential, the respective company must provide concrete evidence how the disclosure of the information would specifically undermine their commercial interests.Footnote 162
In summary, public access to information gathered during the E(R)A process is low in both the US and the EU. The US gives the public access to the environmental data, by allowing the disclosure of all the non-confidential part of the E(R)A upon request. Meanwhile, within the EU, the EMA decides which data from the non-confidential information is included in published EPARs. Moreover, the possibility of access upon request within the EU context is limited, but developing in a positive direction, at least in respect to centrally registered products. In respect to products registered via other procedures, a better utilisation of Directive 2003/4/EC on public access to environmental information to access non-confidential parts of data gathered for E(R)A could also be a possibility.Footnote 163 For now, the possibility to use the directive and allow for public access to the E(R)A data upon request is limited due to the broad definition of “commercial confidential information”,Footnote 164 which allows for refusal to render the sought information.Footnote 165 However, as the recent case in the CJEU shows, this broad definition is getting tighter. Still, the main reason for the limited availability of environmental data remains the protection of intellectual property rights, in particular the trade secret and data protection as mentioned in the discussion above.
IV Conclusion
Scientific evidence regarding the possible harmful effects of medicines to the environment, and arguably also to human health, have led legislators in both the EU and the US to implement environmental concerns into the field of pharmaceutical law.
When balancing the risks and benefits of new medicines, both the US and EU require applicants to submit an E(R)A during the MA procedure. However, only the US takes into account the E(R)A results during the risk-benefit analysis. In the EU, it is considered necessary to exclude the E(R)A results from the risk-benefit analysis in order to protect patient health and the competitiveness of the EU pharmaceutical industry. Therefore, the US approach shows a way to weigh up the environmental risk of new medicines alongside the possible therapeutic benefits; it may thus provide inspiration to EU legislators.
However, regarding the scope of the E(R)A, although the legal tools within the EU and US are designed differently, the scope of medicines exempted from the E(R)A requirement is almost the same. The broad categorical exclusions within the US law question the effectiveness of the environmental protection and so do the above-discussed limitations of the EU law. Although the goals and the scope of covered products in the two jurisdictions are similar, there are notable differences in what is considered during the E(R)A. For example, the EU requires the special consideration of endocrine disrupting substances and GMOs, whereas the US considers the cumulative impact and the effect of manufacturing effluent upon endangered species and ecosystems, including outside of the US. This difference in scope suggests a difference in values and the level of risk deemed acceptable.
In addition, the US and EU both take the approach of managing the impact of pharmaceuticals in the environment by assessing the impact of a particular medical product, not for the API within the medicine. However, the US, unlike the EU, requires applicants to consider the cumulative impact of the medicine within the E(R)A.
In discussing the gaps in our scientific knowledge about the effects of pharmaceuticals in the environment, we highlighted several regulatory challenges which are shared by the US and EU to a certain extent. In both the EU and US, the public lack access to sufficient scientific data about the presence of pharmaceuticals in the environment, firstly because of a lack of testing for the presence of pharmaceuticals in the environment and secondly because data generated during the MA procedure is protected by intellectual property rights. However, the US does allow the release of non-confidential data to the public upon request. In the EU, non-confidential parts of the data are reported within EMA’s European Public Assessment Reports. Each approach carries certain problems with it. The US solution relies on an active effort of the public to demand access to the data, since they have to file a request. In the EU, it is only preselected data that are presented in a standardised form by the EMA and, thus, there is no certainty that the data is comprehensive or even representative.
Both the US and EU seek to include more environmental interests into the regulatory processes through the formation of assessment panels with members with a wide field of experience. By including members from different stages of the medicine’s lifecycle and different fields of science, the US arguably comes closer to this objective than the EU.
Another underlying difference between the US and EU approach to risk assessment is the use of scientific knowledge during decision-making. The WTO hormones panel decision put on display the difference in this approach, with the US arguably taking a more objective, science-based approach to assessing the risk. Obiter dictum from the CJEU’s Pfizer decision shows risk assessment to be a more political exercise.Footnote 166 However, within the risk-benefit analysis of the MA procedure, both the EU and US decision-makers require a high level of scientific data, which plays a deciding role in the approval of a new medicine. Unfortunately, in the EU, this scientific data only concerns the efficacy, safety and therapeutic benefit of the medicine, not the potential environmental impact, as in the US case. Therefore, in the case of the risk assessment of PiE, values play a role in determining the type of scientific data presented before decision-makers in both the EU and US, which can partially explain the different approaches to risk assessment.
This article aimed to compare the letter of the law in the EU and the US concerning the inclusion of environmental concerns into the pharmaceutical regulatory processes. Further work needs to be done to understand all the underlying values in the two jurisdictions driving the legislative work in this area in order to be able to design appropriate and effective laws. The balance between public health concerns and environmental protection, especially, might need to be reconsidered in the light of the growing evidence of the detrimental effects of medicines’ residue in the environment on human health. While the main focus on human health might be difficult to change, environmental protection in pharmaceutical legislation might be strengthened through reflecting on the indirect effect of medicines’ residue on human health through the environment. We hope that our work may serve as evidence and provoke discussion around the conceptual differences between the US and EU regarding the relationship between public human health and environmental protection.
