Microsoft word - innogen working paper 14 - final.doc

Therapeutic advance of biopharmaceuticals . 6 Table 1. Prescrire definitions of the evaluation categories . 7 Table 2. Prescrire evaluations of the therapeutic value of biopharmaceuticals and all other drugs (Jan 1986 – June 2004) . 9 Table 3. Prescrire evaluations of biopharmaceutical indications over time .10 Drug share and prevalence rates for biopharmaceuticals .11 The social benefits of biopharmaceuticals include both economic and health effects. Due to the poor correlation between drug sales and health outcomes, benefits are better assessed using non-economic indicators. We define biopharmaceuticals as large molecule drugs produced using rDNA technology and identify 59 biopharmaceuticals with market approval in the US or Europe by the end of June 2004. We then compare the therapeutic advance offered by biopharmaceuticals and small molecule drugs, using independent evaluations by Prescrire for 48 evaluated biopharmaceuticals with 85 indications and for 2,235 evaluated indications for other drugs. Biopharmaceuticals have a distinctive therapeutic advance, with 27.1% of biopharmaceuticals providing ‘some advance’ or higher, compared to only 10.2% of all other drugs. However, the therapeutic edge of biopharmaceuticals is declining over time due to an increase in ‘me too’ biopharmaceuticals. Over the short-term, the aggregate health benefits of biopharmaceuticals will be limited because the approved indications are mostly for rare diseases. In the late 1990s, approximately 3.4 billion US dollars (Purchasing Power Parity) of public funds were spent per year on biotechnology research in the OECD, excluding the United States (OECD, 2001), while the US accounted for an additional 8 billion. The majority of public spending was for research in areas with potential health benefits, particularly for the development of new prescription drugs and diagnostics. A recent US Department of Commerce (2003) survey of biotechnology firms estimates that the private sector in the United States invested 16.4 billion dollars in biotechnology R&D in 2001, of which between 70% and 90% was probably for health applications. An important question for science and technology policy is whether the large public and private investments in biotechnology are producing notable public benefits (Nightingale, forthcoming). One way of measuring public benefits is in terms of the economic impact of biotechnology, such as on sales revenues or employment, which is the focus of many reports by private consultants such as IMS Health, DataMonitor, and Ernst and Young, or on the competitiveness of pharmaceutical firms (Allansdottir et al, 2001). However, we would argue that economic measures of the application of biotechnology to health only tell one part of the story, and the less important one at that. Economic measures of impacts assume a strong positive correlation between indicators such as sales or value-added and social benefits. This assumption is acceptable for many complex and simple products such as computers or pencils. Economists assume that a consumer that decides to buy pencils and paper instead of a computer is making a rational decision based on his or her preferences. However, rational decision-making is constrained in pharmaceutical markets because of severe distortions from acute information asymmetries, extremely elastic consumer demand, and monopoly pricing. Furthermore, the public benefits of pharmaceuticals do not derive from the amount of consumption per se, measured in either monetary units or the number of prescriptions, but from their therapeutic benefits. One possible measure of the public benefits is the number of biopharmaceuticals that receive market approval. Drug regulatory agencies such as the Food and Drug Administration (FDA) in the United States or the European Agency for the Evaluation of Medicinal Products (EMEA) assess the safety and efficacy of new drug applications. Efficacy is mainly determined from comparisons with placebos in randomized double-blinded clinical trials. With the exception of a few categories such as antibiotics, a pharmaceutical firm does not need to include head-to-head trials with other drugs for the same indication in its submission for marketing approval. Therefore, many drugs that receive market approval are ‘me-too’ drugs that offer little or no therapeutic improvement over existing drugs already on the market, although theoretically they should create other social benefits through price competition. A better measure of public benefits is to determine whether the large investments in biopharmaceuticals have created new drugs that offer therapeutic advances over existing drugs and how many patients will benefit from these new drugs. In this paper, we first develop a working definition of a biopharmaceutical. We then use Prescrire data on the therapeutic advance provided by new drug indications to compare biopharmaceuticals against small molecule drugs. Prescrire is a non-profit French organization that provides an independent, expert evaluation of the therapeutic advance of all new drugs and new drug indications coming onto the French market since 1981. The evaluations are based on a complete review of published scientific literature and unpublished data from drug regulators, including the FDA and EMEA, and pharmaceutical firms. The results show that biopharmaceuticals have, so far, offered a larger therapeutic advance than other new drugs. We then briefly evaluate a few other indicators of the therapeutic benefit of new drugs that overlap with economic indicators, such as the share of all new drugs that are biopharmaceuticals and the prevalence rates for diseases targeted by biopharmaceuticals. At first sight, it should be a relatively easy task to identify all biopharmaceuticals, obtain the Prescrire evaluation of each drug, and compare the results against the evaluations for all other new drugs. In fact, the task is not easy because of the first step - how to identify a biopharmaceutical? This problem highlights the importance of developing neutral, unbiased statistics for biopharmaceuticals that are based on clear definitions. This requirement is necessary not only for evaluating the therapeutic value of biopharmaceuticals, but also for estimating their economic impacts. There are wide variations in estimates of the number of biopharmaceuticals on the global market and consequently the estimated share of pharmaceutical sales that are due to biopharmaceuticals. For example, the American Biotechnology Industry Organization (BIO) includes 197 drugs in its ‘Approved Biotechnology Drugs’ list, covering all FDA approvals up to the end of 2003 (BIO, 2004). IMS Health (2004) estimates that there were 119 biopharmaceuticals on the global market in 2004, accounting for 32 billion US in sales, or about seven percent of world-wide pharmaceutical sales. Our own estimate (see below) is that there are currently about 60 biopharmaceuticals on the US or European markets – or half the IMS Health estimate and one-third the BIO estimate. These differences are mostly due to how a biopharmaceutical is defined. Unfortunately, many sources fail to give a clear definition, which makes it a challenge to interpret the validity of different estimates of the economic or health impacts of biopharmaceuticals. Not surprisingly, given rapid advances in the health sciences, it is difficult to produce a clear, workable definition of a biopharmaceutical. One option is to include all drugs whose discovery, development or manufacture depends on one or more biotechnologies, such as proteomics, genomics, and rDNA technology. Unfortunately, this definition is unworkable because it is difficult to identify which drugs rely, to some extent, on one or more biotechnologies. Furthermore, the extent to which modern biotechnologies are used to develop a specific drug will vary widely, requiring complex subjective decisions on whether or not many drugs are biotech drugs or not. The alternative, which is what we use here, is to use a restricted but operational definition. We define a biopharmaceutical as a large molecular weight compound that is manufactured using rDNA technology, including monoclonal antibodies (MABs). These drugs are comparatively (although not always) easy to identify and are based on the main technical breakthrough for biotechnology. When the definition is undefined, we use the term ‘biotech drug’. Our definition of a biopharmaceutical excludes two groups of drugs that could reasonably be potential candidates. We do not include vaccines or large molecular weight drugs that are extracted from human, animal or plant products, such as non-rDNA immunoglobulins and insulin. Many of the technologies for biologics have been available for decades, although modern biotechnology has also been used to improve them, as when rDNA technology is used to produce vaccines. Second, we exclude the use of rDNA to produce small molecule drugs that can be manufactured through chemical synthesis1. Three well-known lists of biotech drugs are the US FDA’s (2004) list of Therapeutic Biologic Approvals (TBA), which are primarily biopharmaceuticals, the BIO list of Approved Biotechnology Drugs (BIO, 2004), and the monthly column of new product announcements in the journal Nature Biotechnology. Each source gives widely different numbers of biotech drugs. For example, for 2002, the TBA lists 7 new market introductions, Nature Biotechnology lists 19, and BIO lists 22. All drugs listed by the TBA as a biologic are included in the BIO and Nature Biotechnology lists, but BIO includes an additional 15 drugs and Nature Biotechnology an additional 12 that the FDA does not include in its TBA. Clearly, using the BIO or Nature Biotechnology lists would substantially increase the number of biopharmaceuticals compared to the definition used by the FDA. Why do BIO and Nature Biotechnology include many drugs that the TBA does not define as a biologic?2 The main reason is that both appear to define all drugs produced by ‘biotech firms’ as a biotech drug. As an example, BIO includes both tadalafil (Cialis) and fluoxetine hydrochloride, the active ingredient in Prozac. In addition, the definition of a ‘biotech’ firm is highly elastic. A logical definition of a biotech firm would be one that has ever received marketing approval for a biopharmaceutical that the firm developed in-house or which is actively developing a biopharmaceutical. However, we could find no evidence for marketing approval of a biopharmaceutical for several firms that have produced ‘biotechnology’ drugs listed by Nature Biotechnology or BIO. Some of these could have biotech drugs under development, but a large number appear to be active in drug delivery technology and show no evidence, based on their website, of active development of biopharmaceuticals. For example, of the 79 firms cited in Nature Biotechnology as having introduced new biotechnology products in the US or Europe up until November 2003, 44 (56%) developed and marketed one or more biopharmaceuticals and eight (10%) developed a non rDNA biologic. The remaining 29 firms (34%) had not marketed any drugs within these two categories. The main rationale for their inclusion in the BIO and Nature Biotechnology definitions of a biotech firm is that they are relatively new firms active in new fields such as light activation or liposome and pegylation drug delivery systems. 1 rDNA techniques can be used, along with other techniques, to synthesize single enantiomer version of stereoisomer molecules. One application of this technology appears to be to extend patent protection rather than to create additional therapeutic value, as with esomeprazole (Nexium), which is the single enantiomer version of someprazole (Losec or Prilosec). 2 We used the EMEA and FDA websites to find product information for drugs that the FDA did not classify as biologics. In most cases there is little doubt that the excluded drugs are not biopharmaceuticals drugs because of their small molecular weight and other chemical characteristics. BIO identifies key biopharmaceuticals based on ‘recombinant proteins and monoclonal antibodies’ in its annual lists of new biotech drug approvals and it notes that its lists include ‘selected small-molecule’ products, although it gives no details on how these are selected. Of note, only 7 (28%) of the 25 new drugs for 2003 were identified as recombinant proteins or MABs and of these six are included in the TBA. The seventh is pegvisomant (Somavert), which is not included on the FDA’s list even though it is produced using rDNA technology. This is possibly because approval for pegvisomant and many other biopharmaceuticals was moved from the Center for Biologics Evaluation and Research (CBER) to the Center for Drug Evaluation and Research (CDER) in October 2003. Teriparatide (Forteo) is another rDNA drug that does not appear in the TBA3. These and other examples, such as the exclusion of early biopharmaceuticals such as rDNA insulin and human growth hormone from the TBA, suggests that neither the FDA’s TBA nor industry lists are entirely reliable as a source of biopharmaceuticals, although the TBA is substantially more accurate than either the Nature Biotechnology or BIO lists. Additional sources need to be used to identify both biopharmaceuticals that have not received marketing approval in the US and biopharmacueticals that do not appear in the TBA list. We identified biopharmaceuticals from the BIO lists on product launches, from the FDA’s TBA (www.fda.gov) and from EMEA (www.emea.eu.int), which approves all biopharmaceuticals in Europe4. We used the FDA product label information or information from EMEA to verify that each drug was a large molecule produced using rDNA technology5. This process identified 59 biopharmaceuticals. We exclude small molecule drugs produced using rDNA technology, non-rDNA biologics, vaccines, and rDNA diagnostics or devices. This is partly because the coverage of vaccines in Prescrire appears to be incomplete, while diagnostics are not covered at all. Prescrire has evaluated 48 of the 59 biopharmaceuticals. Of the remaining eleven, four have not received market approval or a positive evaluation for Europe as of June 30, 2004 (barbepoetin alfa, bevacizumab, acefacept, and omalizumab) and three have received a positive evaluation and are therefore very likely to receive market approval for Europe in the near future (cetuximab, efalizumab and tositumomab). Four have received EMEA market approval but have not yet been evaluated by Prescrire (ibritumomab tiuxetan, laronidase, pegvisomant, and teriparatide). All of the latter were approved between 2002 and 2004. Prescrire’s evalulations are between one and three years behind the first drug marketing approval for Europe by the EMEA. However, the same problem applies to all drugs, so any potential bias in the comparison of biopharmaceuticals versus other drugs (or variation due to random effects) should be relatively small in comparison to the observed differences in the distribution of the level of therapeutic advance. Therapeutic advance of biopharmaceuticals Biotechnology provides firms with the ability to develop new drugs with new modes of action that were not achievable using chemical synthesis. This suggests that biopharmaceuticals should offer therapeutic advances. Ashton (2001) uses commercial data to calculate that 56% of the biopharmaceuticals (using a similar definition as ours: 3 The EMEA scientific discussion papers describe Forteo as a ‘recombinant 1-34 terminal fragment of endogenous human parathyroid hormone’ and Somavert as a ‘pegylated recombinant analogue of the human growth hormone (GH) which has been genetically engineered to function as a GH receptor antagonist”. 4 Other types of drugs can proceed through a national regulatory agency, if the manufacturer does not intend to market it another EU country (Senker and Zwanenberg, 2000). 5 It is still possible to make mistakes, either by including a drug that is not a biopharmaceutical, or excluding drugs that are. For this reason we give a full list of biopharmaceuticals for which we have a therapeutic evaluation in the Appendix. Non-evaluated biopharmaceuticals are listed below. We welcome additions, comments and criticisms of our list from readers. “recombinant proteins and monoclonal antibodies”) launched between 1982 and 2000 in the United States were for orphan diseases, compared to 14% of other new pharmaceuticals. He also found that 25% of biopharmaceuticals had a unique mode of action compared to 15% of small molecule drugs. This is good evidence that biopharmaceuticals provide an important therapeutic advance, but the evidence is incomplete, since Ashton is unable to compare biopharmaceuticals against all existing drugs and the comparison is not based on a measure of the degree of therapeutic advance. Prescrire is an especially useful source of drug evaluations because it is immune to the enormous pressure on the evaluation process from industry, industry-financed consumer groups, and some Government agencies. The organization does not accept advertising and receives no financial support from the French government, pharmaceutical firms, or other organizations. The cost of the drug evaluations, designed to guide the prescribing and dispensing decisions of French doctors and pharmacists, is entirely funded by subscriptions to its journal, La Revue Prescrire. Prescrire is the best available source for the relative therapeutic value of new drugs. The Drugs and Therapeutics Bul etin in the UK or the Medical Letter in the US provide similar prescribing advice as Prescrire but do not cover all drugs nor provide a ranking system. The FDA classifies drug assessments into a priority and standard review. The decision to assign a priority review is based on clinical evidence at the start of the review for a significant therapeutic improvement compared to products already on the market6. However, this classification system was not used for biologics, although the transfer of many biologic approvals to CDER in late 2003 may change the approval process in the future. Prescrire assesses the therapeutic value of all new drugs that receive market approval in France, plus approvals for existing drugs to be used for a new indication. An indication is a specific disease or health condition. For example, interferon-alpha is approved in France for the treatment of both melanoma and lymphoma. Based on the results of its drug evaluation, Prescrire assigns each drug to one of six therapeutic categories, ranging from a ‘major’ advance to ‘not acceptable’ (the drug offers no benefits over existing alternatives but has potential or real disadvantages). A full definition of each evaluation category is given in Table 1. When alternative drugs are available, the measure of therapeutic advance is based on comparisons with existing drugs for the same indication that are already on the market. In addition, a seventh category is used when the available data are insufficient for assessing the therapeutic value of the drug. Table 1. Prescrire definitions of the evaluation categories The drug is a major therapeutic innovation in an area where previously no treatment was available. The product has some value but does not fundamentally change the present therapeutic 6 There is some controversy over the criteria used to assign a drug to a priority or standard review. A report by the National Institute for Health Care Management (NIHCM, 2002, page 6) notes that ‘the criteria for obtaining a priority rating are broad enough to include drugs that some observers would not regard as providing important therapeutic advances”. Éventuel ement The product has minimal additional value and should not change prescription practices except in rare circumstances. The product may be a new molecule but is superfluous because it does not add to the clinical possibilities offered by previously available products. In most cases it concerns a me-too product. Product without evident benefit but with potential or real disadvantages. The editors postpone their judgement until better data and a more thorough evaluation of the drug are available. Source: English definitions are from Prescrire International. Prescrire evaluated 48 biopharmaceuticals between 1986 (the first evaluation of a biopharmaceutical) and June 2004. These biopharmaceuticals were evaluated for 85 indications. The distribution of evaluations is given in Table 2 for both the highest evaluation received by each biopharmaceutical for any indication and the distribution of the evaluations for all indications. There is little difference in the percent distribution of evaluations based on either the highest evaluation given to each biopharmaceutical or the distribution for all indications. For example, 10.4% of the highest evaluations were assigned to the ‘important advance’ category, compared to 9.4% of all indications. The Appendix gives the generic name for each of the 48 biopharmaceuticals, the indication that received the highest evaluation, and the number for the evaluation, as defined in Table 1 (ranging from 1 to 7). The five highest-ranked biopharmaceuticals are algucerase, Blood Factor VIII, the first recombinant insulin, interferon alfa and human protein C. Table 2. Prescrire evaluations of the therapeutic value of biopharmaceuticals and all other drugs (Jan 1986 – June 2004) Source: Prescrire issues between 1986 and June 2004. For biopharmaceuticals, analysis of Prescrire records by the authors. All other drugs: 1986 data on page 58, Prescrire Jan 2000, 1987 – 2001 data on page 56, Prescrire, Jan 2002; data for 2002 – 2004 from individual Prescrire issues. The evaluations for biopharmaceuticals were subtracted from the totals for all drugs. Prescrire updates evaluations when new information becomes available, so the highest rating for a biopharmaceutical could be due to either a revised rating for the same indication or to differences in the ratings among indications. The ‘all indications’ column includes all ratings, whether revised or not, in order to maintain comparability with the results for all other drugs, given in the last two columns of Table 2. Based on the results for indications, a higher percentage of biotechnology than all other drugs provide ‘some advance’ or higher: 27.1% versus 10.2% of all other drugs. In addition, less than a quarter (23.5%) of biopharmaceuticals are rated as offering no therapeutic advance over existing drugs on the market, versus almost two-thirds (65.6%) of all other drugs. These results show that biopharmaceuticals have so far offered greater therapeutic advances than other types of drugs. Although we know of no other consistent evaluations of the therapeutic advance of biopharmaceuticals, the Prescrire evaluations for all other drug indications can be benchmarked against the percentage of FDA new drug applications (NDAs) that received priority review status. The National Institute for Health Care Management (NIHCM, 2002) reports that 24% of 1,035 NDAs between 1989 and 2000 inclusive were given priority status. This is over double the 10.2% of Prescrire indications that received a rank of ‘some advance’ or higher and slightly less than the 26.9% of Prescrire indications with a rank of a ‘minimal advance’ or higher. This suggests that Prescrire’s criteria for a significant therapeutic advance are more demanding than that of the FDA. However, part of the difference will be due to the fact that the FDA determines priority status at the start of its review process, before it has access to all information, while Prescrire’s reviews are based on both the final FDA and EMEA reviews and clinical trials published after marketing approval. Of note, the results in Table 2 raise a few concerns over the rush to bring drugs to market. The two categories of ‘not acceptable’ and ‘judgment reserved’ refer to drugs that should not have received marketing approval, either because the drug is more harmful than alternatives or because the available data are insufficient for assessing drug safety and efficacy. Twenty percent of biopharmaceutical indications fall in this group, which is almost three times higher than the rate of 7.4% for all other drugs. The higher figure for biopharmaceuticals does not appear to be due to a general decline in regulatory standards because the percentage of all other drugs within these two categories declined from 10.9% between 1986 and 1995 to 6.1% after 19967. The greater therapeutic advance of biopharmaceuticals compared to all other drugs is partly expected, given that biopharmaceuticals, as noted by Ashton (2001), are based on a new technology with new modes of action. In this respect biopharmaceuticals display some of the characteristics of an emerging technology in contrast with the ‘mature’ technology characteristics of small molecule drugs. However, the biopharmaceutical sector is also showing signs of a maturing technology, with a decline in the level of therapeutic advance over time. Table 3 divides the Prescrire evaluations into three time periods with roughly equal numbers of evaluations. The share of biopharmaceutical indications offering some therapeutic advance or greater declined from 39.3% of 28 indications evaluated between 1986 and 1998 inclusive, to 29.6% of 27 indications evaluated between 1999 and 2001, and then to 13.4% of 30 indications evaluated after 20018. Over this time period, the percentage of ‘me too’ indications increased markedly from 7.1% to 40.0%. Table 3. Prescrire evaluations of biopharmaceutical indications over time 7 However, a more detailed breakdown for the period after January 1996 shows an increase in the percentage of drugs receiving a rank of 6 (not acceptable) or 7 (judgment reserved). Between 1996 and the end of 2001, 5.4% of all drug evaluations were ranked at a 6 or 7, compared to 11.2% of all drug evaluations between January 2002 and June 2004. 8 The therapeutic advance of all other drugs is also falling, from 14.3% before 1996 to 8.6% afterwards, but the decline was not as steep. The decline in therapeutic advance is partly due to the diffusion of the technology to an increasing number of firms, with competitors bringing comparable drugs onto the market. A good example is interferon, with eight different versions currently available. It is important to note that in absolute terms the number of biopharmaceuticals per year offering some therapeutic advance or greater doubled, from 0.85 per year between 1986 and 1998 to 1.9 per year after 1999. Drug share and prevalence rates for biopharmaceuticals We often hear that the percentage of all drugs that are biopharmaceuticals is increasing over time. We estimate, using FDA data for new molecular entities (NMEs) between 1998 and 2004 and the number of biologic license applications, that biopharmaceuticals accounted for 16.8% of approximately 200 new drugs over this time period. In comparison, IMS Health (2004) estimates that 27% of new drugs in clinical trials in 2003 were biopharmaceuticals. Even though IMS Health does not define a biopharmaceutical, it is likely that the share of biopharmaceuticals will increase in the future. However, simple counts of the number of biopharmaceuticals or the percentage of all drugs that are biopharmaceuticals will not provide a useful measure of health impacts. This is because health impacts depend both on the number of new drugs and the number of patients per year with a disease (the prevalence rate) that is an approved indication for each drug. Prevalence rates are also linked to economic indicators for sales, since total sales depend on both the annual number of prescriptions (determined by disease prevalence rates) and the cost per prescription. Most biopharmaceuticals target rare diseases compared to many small molecule drugs. Of the approximately 30 disease groups targeted by biopharmaceuticals, eight are very rare with prevalence rates of less than 1 in 10,000 in the United States and only four have prevalence rates close to or above 1 in 100: diabetes, stroke, heart attacks, and rheumatoid arthritis. With the exception of diabetes, biopharmaceuticals are only approved for the remaining three diseases under specific circumstances or as a second-line drug when other alternatives fail. This substantially reduces the number of patients for which the biopharmaceutical is appropriate. An estimate of the social benefits from the large public and private investments in the health applications of biotechnology partly depends on how we define a biopharmaceutical and on the therapeutic advance of these drugs. Our analysis, limited to large molecule drugs produced using rDNA technology, shows that a higher percentage of biopharmaceuticals than other drugs provide a significant therapeutic advance over existing drug therapies. Of concern, however, is the higher percentage of biopharmaceuticals that received market approval on the basis of inadequate evidence for safety and efficacy. A second concern is the decline over time in the percentage of biopharmaceuticals that offer a therapeutic advance, although the absolute number of biopharmaceuticals that provide a notable advance over alternative drugs has increased from slightly less than one per year before 1998 to almost two per year after 1998. Data on the prevalence rates of the disease targets for biopharmaceuticals suggest that biopharmaceuticals are unlikely to account for a substantial share of all drug prescriptions over the near future because the approved applications for most biopharmaceuticals are limited to rare diseases. There are other good reasons why biotechnology, at least over the short term, might remain a niche player in public health. The mode of delivery – via injections – reduces their public acceptance, and many biopharmaceuticals have severe side effects. This limits their use to cases where other drugs fail, such as in the treatment of rheumatoid arthritis, where biopharmacuetical TNF receptor binding proteins (Etanercept, Adalimumab Anakinra), had global sales equivalent to only 7% of the therapy class ‘antirheumatic non-steroidals’ in 2002 (IMS Health, 2004b). From a public policy perspective, it would be unrealistic to assign all of the benefits from public and private investments in biotechnology to the definition used in this paper of biopharmaceuticals. Research on the benefits, using a broader definition of biotechnology drugs (and including diagnostics) is also needed. As pointed out above, this could be very difficult, requiring a careful evaluation of the drug discovery and development process behind all new drugs. An alternative perspective with policy implications is to ignore the technology and examine the structure of the pharmaceutical industry, essentially by asking who develops new drugs with therapeutic advances. For example, are small pharmaceutical firms more likely to discover drugs with a high therapeutic advance than the large established pharmaceutical firms? And, what factors might influence the average level of therapeutic advance in a firm’s pipeline over time? We hope to be able to explore these issues in the near future. References Allansdottir A, Bonaccorsi A, Gambardella A, et al. Innovation and Competitiveness in European Biotechnology. Report for DG Enterprise, European Commission, Luxembourg, 2001. Ashton G. Growing pains for biopharmaceuticals. Nature Biotechnology 19:307-311, 2001. BIO, Kathy Stover, www.findarticles.com/cf_dls/mONKV/2_4/97592356/p1/article.jhtml. For a full list from 1982, see http://www.bio.org/speeches/pubser/approveddrugs.asp, 2004. Last accessed August 2, 2004. FDA, Center for Drug Evaluation and Research (CDER). www.fda.gov/cder/biologics/biologics_table.htm (last accessed August 1, 2004). IMS Health. Biopharmaceuticals – Moving to Center Stage, and Brighter Prospects Ahead for Biopharmaceuticals in Europe? www.imshealth.com/ims/portal/front, last accessed August 1, 2004. IMS Health. 2002 World pharma sales growth: slower but still healthy. www.ims-global.com/insight/news_story/0302, last accessed May 20, 2004b. Nightingale P. The myth of the biotech revolution. Trends in Biotechnology, forthcoming. NIHCM (National Institute for Health Care Management). Changing Patterns of Pharmaceutical Innovation. NIHCM, Washington DC, May 2002. OECD. A Statistical Framework for Biotechnology Statistics. DSTI/EAS/STP/NESTI (2001) 39, OECD, Paris, 2001. Senker J and Zwanenberg P. European Biotechnology Innovation System, EC Policy Overview. DG Research, European Commission, TSER Contract No. SOE1-CT98-1117, 2000. US Department of Commerce. A Survey of the Use of Biotechnology in US Industry. US Department of Commerce, Technology Administration, Bureau of Industry and Security, Washington DC, October 2003. Highest Prescrire therapeutic evaluation and indication for biopharmaceuticals reteplase recombinant plasminogen activator 1: Molecular weight of 7,122.27 Daltons (lower than most large molecule pharmaceuticals), but produced using rDNA technology. 2: This rating is unexpectedly low, possibly because we did not find a Prescrire evaluation of this drug shortly after its introduction onto the market in France in the 1980s. The highest evaluation here is from 1996.

Source: http://www.innogen.ac.uk/downloads/Innogen-WP-014-The-Benefits-of-Biopharmaceuticals.pdf