Clinical Toxicology (2011), 49, 452–456 Copyright © 2011 Informa Healthcare USA, Inc. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2011.594054
COMMENTARY
Clinical toxicology and drug regulation: A United Kingdom perspective W. STEPHEN WARING1 and PATRICIA MCGETTIGAN2 1Acute
Medical Unit, York Hospital, York, United Kingdom Medical School, University of Hull, Hull, United Kingdom
2Hull York
The Medicines and Healthcare products Regulatory Authority (MHRA) is the government body with responsibility for regulating new and existing medicines and medical devices in the United Kingdom. The Yellow Card scheme is a well-established pharmacovigilance system that collects voluntary reports of adverse effects associated with therapeutic drug use. In contrast, data concerning clinical toxicological effects are more poorly characterised. No comparable surveillance processes exist in the United Kingdom or elsewhere in Europe that might allow systematic collection of clinical data and outcomes after drug overdose. Toxicological effects are normally ascertained from individual patient reports or small case series from a few specialised poisons units, so that these data are generally under-represented in post-marketing consideration of risks and benefits. Safety concerns may lead to withdrawal of the Marketing Authorisation or restricted prescribing conditions, which are conveyed to health care professionals by means of safety warnings. These may have a variable impact, and three selected examples are presented to illustrate the complex interaction between drug regulation and clinical toxicology. First, the effects of the withdrawal of rofecoxib in 2004 shows that regulatory responses may reduce the prescribing of drugs across a particular class, and this has resulted in fewer enquiries to Poisons Control Centres regarding all cyclooxygenase-2 selective inhibitors. Secondly, data concerning the impact of safety warnings about antipsychotic medications illustrate that regulatory decisions may have a variable impact due to other factors that influence prescribing, including clinical guidelines, marketing pressures, and lack of alternative safe medications. Finally, the recent withdrawal of co-proxamol serves as an example of how clinical toxicology data can inform the drug regulation process and improve safety by minimising the risk of death associated with overdose. Greater reliance on clinical toxicology data could better inform the drug regulation process, perhaps through coordinated data collection systems that already exist in certain national poisons centres. Routine collection of high quality data concerning the effects of drug overdose could allow a more comprehensive review of risk and benefit by the regulatory authorities. Keywords
Drug regulation; Paracetamol; Safety; Pharmacovigilance; Adverse effects
The manuscript by Matlock et al. published in this issue calls for a more judicious use of the boxed warning system by the Food and Drug Administration (FDA).1 Historically, boxed warnings in the United States have applied only to severe, lifethreatening adverse effects or where mandatory prescribing restrictions are enforced. However, boxed warnings are now applied so widely that these include commonly prescribed medications such as atenolol, furosemide, lisinopril, iron supplements, and oral contraceptive pill. Concern has been raised that more inclusive criteria may diminish the overall effectiveness of safety warnings or that these might impede appropriate prescribing in certain situations.1 Data concerning the outcomes after overdose and the toxicological profile of a drug might allow a clearer understanding of potential hazards and a more judicious use of boxed warnings.
Different regulatory bodies and processes exist in Europe. In the United Kingdom, the Medicines and Healthcare Products Regulatory Authority (MHRA) is the government body responsible for regulatory decisions concerning both medicines and medical devices. This includes consideration of whether restrictions should be applied to new or existing medications based on data concerning their adverse effects and other safety issues. The European Medicines Agency (EMEA) is responsible for considering applications for a Marketing Authorisation that is valid across the European Union, and relies on an evaluation by the Committee for Medicinal Products for Human Use (CHMP), and all European Union members to this process.2 For example, the Commission on Human Medicines is an expert advisory committee that advises the United Kingdom CHMP members on the position that they should adopt. When a Marketing Authorisation is granted, this includes approval of the Summary of Product Characteristics (SPC) document. The SPC contains a variety of information for health care professionals, including details of the adverse effects, and Section 4.9 specifically addresses the effects of overdose and optimal management.
Received 25 April 2011; accepted 31 May 2011. Address correspondence to William Stephen Waring, PhD, FRCP, Acute Medical Unit, York Hospital, Wigginton Road, York, YO31 8HE, United Kingdom. Tel: ⫹44-0-1904-726276. E-mail:
[email protected]
452
Clinical toxicology and drug regulation 453 At the time when the SPC is approved, few data are normally available concerning the effects of overdose and, even when such data might become available later, it is rare that they would result in an SPC amendment.3,4 In the United Kingdom, the major source of post-approval data is the Yellow Card scheme, a centralised pharmacovigilance system that has for more than 40 years been receiving voluntary reports of adverse effects associated with therapeutic use of new and existing medications.5 Certain drugs are subject to intensive monitoring and these are indicated to prescribers by a black triangle in the British National Formulary (BNF), which is the standard prescribing reference source for health care professionals in the United Kingdom.6 The black triangle is typically applied to newly licensed medicines, medicines licensed for a new route or indication, or for a new combination of active substances. In contrast to the boxed warning system, the presence of a black triangle indicates a higher intensity of monitoring but does not indicate that there is a known hazard.7 The MHRA fulfils its regulatory role by considering the benefits and adverse effects of a medicine, and this is based on systematic evidence from observational databases and clinical trials, data from the Yellow Card scheme, and new information arising from the Pharmaceutical Industry. The MHRA or the Marketing Authorisation holder communicates information about adverse effects and the overall risk-benefit profile to health care professionals. The means of communication depends on the degree of urgency and potential seriousness of the problem, and may include safety warnings on the MHRA website, a ‘Dear healthcare professional’ postal letter, or communication via other professional bodies such as the National Patient Safety Agency or National Health Service alerting systems.8,9 Adverse effects are typically considered as those arising from use in accordance with the Marketing Authorisation and also those arising after inadvertent or intentional offlabel prescribing. Outcomes after drug overdose are often excluded from pharmacovigilance programmes and are more difficult to characterise. Therefore, this aspect of drug safety has been comparatively overlooked in consideration of actual benefit versus risk. Health care systems in the United Kingdom and Europe do not contain any systematic means for collecting clinical data in patients that seek help after drug overdose. Such data normally arise from individual patient reports and small case series arising from a few specialised poisons wards. National Poisons Centres across Europe independently collect clinical data from poisoned patients, and there are different systems and types of data collected between countries.10 Greater integration of data collection systems of the poisons centres might provide data from sufficiently large numbers of patients to allow earlier signal detection and better inform drug toxicity.11 Moreover, this would provide a unique and valuable means of capturing post-marketing toxicological data that could be considered by regulatory authorities in determining the overall benefit and risk. There have been a number of recent examples of safety concerns that have resulted in withdrawal of the Marketing Copyright © Informa Healthcare USA, Inc. 2011
Authorisation or led to prescribing restrictions. However, surprisingly few data exist to demonstrate what effect, if any, these changes have had on overall drug safety, particularly with respect to outcomes after drug overdose. Three selected examples are presented to show different aspects of the effect of safety warnings on subsequent patterns of drug overdose and clinical toxicology outcomes.
Withdrawal of rofecoxib: Less cyclooxygenase-2 selective inhibitor prescribing and fewer poisons centre enquiries Rofecoxib was withdrawn in September 2004 due to an apparent increase in cardiovascular risk, and additional data raised concerns that other non-steroidal anti-inflammatory drugs (NSAIDs) might also increase risk. The MHRA issued safety advice for prescribers and the public, and sent postal alerts to prescribers to advise that patients with established ischemic heart disease or cerebrovascular disease should switch from a cyclooxygenase-2 (Cox-2) selective agent to an alternative analgesic, although no specific advice was given about an appropriate alternative.12–14 In 2006, the CHMP concluded that while NSAIDs confer an increased risk of cardiovascular events, the overall balance between benefit and risk was favourable.15 The impact of MHRA guidance on NSAID prescribing is difficult to judge. At the time, national and international media coverage of cardiovascular events associated with NSAIDs was extensive, alerting both public and prescribers to potential risks. Community prescribing in England showed a marked increase in Cox-2 selective agents (rofecoxib, celecoxib, etoricoxib, and valdecoxib) between 2002 and 2004, such that these comprised 25% of all NSAID prescriptions.26 When rofecoxib was withdrawn, prescribing of Cox-2 selective agents fell dramatically, so that this group accounted for only 7% of NSAID prescriptions in 2005 and 3% in 2007. Since 2005, overall NSAID prescribing has continued to decline.16 The MHRA maintains surveillance of NSAID-associated cardiovascular risks and issues periodic update when new information becomes available, for example, that issued in January 2010 in response to publication of two observational studies.17 The latter studies indicate increased cardiovascular risk in patients receiving Cox-2 selected and non-selective NSAIDs, even after short-term use. The absolute risk increase in ‘healthy’ adults was considered low, so that there was no change to the regulatory position. The MHRA report from January 2010 noted an increased cardiovascular risk associated with high doses of two widely used non-selective NSAIDS, ibuprofen (in doses ⬎ 1200 mg/day) and diclofenac (⬎ 150 mg/day, though the studies reported the risk increased at ⬎ 100 mg/day, not 150 mg). While NSAID prescribing is falling in England, these two drugs account for over two thirds of prescribed NSAIDs, and are also available in the United Kingdom without prescription. Ibuprofen is available as a pharmacy-supervised product (200 mg and 400 mg in packs of up to 96 tablets) or under general sales list status (200 mg in packs of upto 16
454 W. S. Waring & P. McGettigan tablets); diclofenac is available in pharmacies only (12.5 mg in packs of 18 tablets). Non-prescription sales are rising but accurate usage data are not readily available, and there are no safeguards to ensure that safety warnings are being conveyed adequately to end-users. Therefore, while community prescribing data provides some evidence of the impact of safety warnings, it is difficult to ascertain the extent to which overall drug use might have changed. Few data are available concerning the impact of MHRA safety warnings on the rate of occurrence of overdose involving Cox-2 selective agents, or clinical outcomes. Data from the American Association of Poison Control Centers indicate that between 2004 and 2009 there was a year-on-year fall in the number of enquiries concerning Cox-2 selective inhibitors, whereas there had been an increased number of enquiries regarding other NSAIDS.18
Overdose involving antipsychotic medications: Contrasting responses to safety warnings concerning thioridazine and second-generation (atypical) antipsychotics In cases where the Marketing Authorisation is amended rather than withdrawn, the impact on prescribing may be more difficult to ascertain. For example, in 2000, the Committee for the Safety of Medicines issued a warning that thioridazine might be cardiotoxic especially in patients with underlying heart disease.19 These safety warnings were followed by substantial changes in antipsychotic prescribing patterns, and data from Scotland show a substantial decline in the number of patients that attended hospital due to thioridazine overdose after 2000, which had largely been substituted by overdoses involving second-generation antipsychotic agents.20 In September 2005, the MHRA issued a safety warning that the second-generation antipsychotics risperidone and olanzapine should not be used to treat behavioural problems in older patients with dementia.21 This followed a review by the Committee for the Safety of Medicines, which found that risperidone conferred a three-fold increased risk of stroke in older patients with dementia, and similar concerns were raised for olanzapine. At the time, they were licenced only for treatment of acute psychosis and schizophrenia, but it was estimated that in the previous year alone risperidone and olanzapine had been used off-label to treat around 30000 and 9000 elderly patients with dementia respectively. The drug safety warning was communicated to health care professionals by the Chief Medical Officer via the Public Health Link (later replaced by the National Health Service Central Alerting System), and the MHRA issued advice to prescribers and patients. Despite these warnings, there was little discernible impact on prescribing practice in England with respect to risperidone, olanzapine, or second-generation antipsychotic agents in general.22 A study between 2004 and 2007 in Scotland found that antipsychotic medications were implicated in 5.9% of overdoses involving adults less than 61 years age, and 6.1% of those more than or equal to 61 years of age.23 These proportions were constant throughout the study period, suggesting that the prescribing restrictions had done little to limit accessibility of antipsychotic agents among
older adults in Scotland, which is perhaps also attributable to the lack of a safe alternative treatment. The substantial apparent influence of the safety warning in 2000 is in stark contrast to the safety warning of 2005 that had little impact, indicating that other factors are important. These include the more widespread acceptance of secondgeneration antipsychotics by the time the second warning was issued, and perhaps the influence of active marketing by the Pharmaceutical Industry at that time. The earlier safety warning coincided with changing guidelines that emphasised the use of newer antipsychotic agents in favour of older ones such as thioridazine. Therefore, changes in prescribing practice cannot be attributed solely to the effects of regulatory authority intervention.
Clinical toxicology data that informs drug regulation to reduce death associated with overdose: Withdrawal of dextropropoxyphene and co-proxamol Co-proxamol is a fixed dose combination analgesic combining paracetamol and dextropropoxyphene, the latter having been recognised for many years to possess cardiotoxic properties.24 More than 20 years of clinical experience elapsed before evidence emerged to show that co-proxamol overdose is associated with excess mortality compared to paracetamol alone or to other fixed-dose combination analgesics, namely co-codamol and co-dydramol.25 The very long interval between pre-clinical safety concerns and clinical detection of increased cardiovascular risk highlights the inadequacy of the existing surveillance systems to detect adverse outcomes of drug overdose, especially when one considers that thousands of patients had sought medical attention due to co-proxamol overdose each year. In December 2007, the MHRA withdrew the Marketing Authorisation for co-proxamol, which was the only dextropropoxyphene-containing medicine available in the United Kingdom. A subsequent review by CHMP concluded that the risks associated with dextropropoxyphene, particularly of potentially fatal overdose, outweighed its potential benefits and, therefore, in June 2009 EMEA withdrew the Marketing Authorisation for dextropropoxyphene-containing medicines across the European Union. Some unlicensed use continues for patients that have had difficulty switching to alternative analgesia, but the volume of prescribed and non-prescribed ‘over-thecounter’ co-proxamol items in England has declined substantially (Fig. 1).26 Moreover, this intervention has led to a steep decline in the occurrence of co-proxamol overdose and may have prevented around 300–400 deaths annually in the United Kingdom alone. The change in prescribing has been accompanied by a significant increase in prescriptions for other analgesics, including co-codamol, paracetamol, co-dydramol, and codeine, but there has been no discernible increase in overdose deaths involving these agents.27,28 Co-proxamol provides a somewhat unusual example of the impact of a safety warning insofar as the withdrawal of the Marketing Authorisation was primarily due to the effects of overdose, and that there has been a demonstrable reduction in overdose deaths as a result of the intervention. Clinical Toxicology vol. 49 no. 6 2011
Clinical toxicology and drug regulation 455
Fig. 1. Volumes of pharmacy-dispensed medications for prescribed and non-prescribed ‘over-the-counter’ co-proxamol and paracetamol in England between 2004 and 2008.26
Summary Prescribing practice is influenced by a number of different factors such as adherence to clinical guidelines, marketing pressures, and public health interventions so that in some cases it may be difficult to ascertain the extent to which safety warnings have had an effect. Perhaps unsurprisingly, drug withdrawal is often accompanied by a dramatic fall in use of the specific agent, whereas prescribing restrictions have a less marked impact. In some cases, safety warnings about a particular agent can lead to an overall reduction in prescribing of agents of that class, as illustrated by the changes in COX-2 selective agent prescribing. In other situations, a reduction of a particular agent may be accompanied by substitution of other agents from within the same group so that overall prescribing of the drug class might increase, as exhibited by antipsychotic medications. Regulatory decisions can have a major impact by minimising access to potentially hazardous medications, and the recent withdrawal of co-proxamol is an example of how such an intervention can lead to improved health outcomes in patients that present to hospital after overdose. The traditional regulatory approach is to examine only the adverse effects that occur when a drug is used under the terms of its Summary of Product Characteristics, whereas data from Poisons Centers is not normally considered. Data concerning drug overdose effects could be collected by Poisons Centers and if included in pharmacovigilance planning, then this would better inform evaluation of drug safety. Several existing systems collect national data from Poisons Center enquiries, for example, the American Association of Poison Control Centers’ National Poison Data System (NPDS) and the United Kingdom Poisons Information Database (UKPID).10,18 A limitation is that these rely on spontaneous reporting and are subject to bias, and there is no reliable means of comparing the Copyright © Informa Healthcare USA, Inc. 2011
number of enquiries to drug exposures. Greater integration of international data collection systems might allow earlier signal detection, and this has been explored in syndrome surveillance for chemical threats.11,29 However, there are major barriers to implementing and maintaining such a data system, including substantial financial cost. Similarly, more robust collection of routine clinical toxicological data within health care facilities might provide additional data to link drug overdose to specific adverse outcomes. Greater reliance on clinical toxicology data might improve the quality of data concerning benefit versus risk and assist regulatory review of the relative toxicity of particular agents.
Declaration of interest The authors report no declaration of interest.
References 1. Matlock A, Allan N, Wills B, Kang C, Leikin JB. A continuing black hole? The FDA boxed warning: An appeal to improve its clinical utility. Clin Toxicol 2011; 49:443–447. 2. Borg JJ, Aislaitner G, Pirozynski M, Mifsud S. Strengthening and rationalizing pharmacovigilance in the EU: where is Europe heading to? A review of the new EU legislation on pharmacovigilance. Drug Saf 2011; 34:187–197. 3. Wall AJ, Bateman DN, Waring WS. Variability in the quality of overdose advice in summary of product characteristics (SPC) documents: gut decontamination recommendations for CNS drugs. Br J Clin Pharmacol 2009; 67:83–87. 4. Mallows J, Chan B, Graudins A. Quality of poisoning management advice in the Monthly Index of Medical Specialties Annual. Emerg Med Australas 2005; 17:511–519. 5. McLernon DJ, Bond CM, Hannaford PC, Watson MC, Lee AJ, Hazell L, Avery A. Yellow Card Collaboration. Adverse drug reaction reporting in the United Kingdom: a retrospective observational comparison of yellow card reports submitted by patients and healthcare professionals. Drug Saf 2010; 33:775–788.
456 W. S. Waring & P. McGettigan 6. Kendall M. Chairman of the Joint Formulary Committee, British National Formulary. Clin Med 2009; 9:349–352. 7. Heeley E, Riley J, Layton D, Wilton LV, Shakir SA. Prescriptionevent monitoring and reporting of adverse drug reactions. Lancet 2001; 358:1872–1873. 8. Swain E, Morgan S, Brewster W, Kauser S. Early communication of drug safety concerns: a feasibility study on enhancing interaction between the pharmaceutical industry and regulators. Pharmacoepidemiol Drug Saf 2010; 19:232–237. 9. Breckenridge A. The future of drug safety. Clin Pharmacol Ther 2007; 81:161–163. 10. Thanacoody HK, Good AM, Waring WS, Bateman DN. Survey of cases of paracetamol overdose in the United Kingdom referred to National Poisons Information Service (NPIS) consultants. Emerg Med J 2008; 25:140–143. 11. Waring WS, Palmer SR, Bateman DN. Alerting and surveillance using poisons information systems (ASPIS): outcomes from an international working group. Clin Toxicol 2007; 45:543–548. 12. MHRA. Immediate withdrawal of rofecoxib (Vioxx/Vioxxacutate). 30 September 2004. http://www.mhra.gov.uk/Safetyinformation/Safety warningsalertsandrecalls/Safetywarningsandmessagesformedicines/ CON1004263. Accessed 20 April 2011. 13. MHRA. Advice for patients following the withdrawal of a widelyused painkiller/anti-inflammatory: Vioxx. 30 September 2004. http://www.mhra.gov.uk/NewsCentre/Pressreleases/CON002056. Accessed 20 April 2011. 14. MHRA. MHRA issues updated advice on the safety of selective Cox-2 inhibitors. 17 Feb 2005. http://www.mhra.gov.uk/Safety information/Safetywarningsalertsandrecalls/Safetywarnings andmessagesformedicines/CON1004250. Accessed 20 April 2011. 15. European Medicines Agency. Public CHMP assessment report for medicinal products containing non-selective nonsteroidal antiinflammatory drugs (NSAIDs). 7 November 2006. http://www. ema.europa.eu/docs/en_GB/document_library/Report/2010/01/ WC500054344.pdf. Accessed 20 April 2011. 16. NHS Prescription Service. NSAIDs and analgesic national charts. June 2010. http://www.nhsbsa.nhs.uk/PrescriptionServices/Documents/PPDPrescribingAnalysisCharts/NSAIDs_National_June_10. pdf. Accessed 20 April 2011. 17. MHRA. Cardiovascular safety of Cox-2 inhibitors and non-selective NSAIDs. Public Assessment Report. http://www.mhra.gov.uk/Safety information/Generalsafetyinformationandadvice/Product-specific informationandadvice/Product-specificinformationandadvice-A-F/ CardiovascularsafetyofCOX-2inhibitorsandnon-selectiveNSAIDs/ index.htm. Accessed 20 April 2011.
18. Bronstein AC, Spyker DA, Cantilena LR Jr, Green JL, Rumack BH, Giffin SL. 2009 Annual Report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 27th Annual Report. Clin Toxicol (Phila) 2010; 48:979–1178. 19. Buckley NA, Whyte IM, Dawson AH. Cardiotoxicity more common in thioridazine overdose than with other neuroleptics. J Toxicol Clin Toxicol 1995; 33:199–204. 20. Poon H, Elliot V, Bateman DN, Waring WS. Impact of legislative changes on patterns of antipsychotic prescribing and self-poisoning in Scotland: 2000–06. J Toxicol Sci 2007; 32:1–7. 21. MHRA. New advice issued on risperidone and olanzapine. September 2005. http://www.mhra.gov.uk/Safetyinformation/Safetywarningsalertsandrecalls/Safetywarningsandmessagesformedicines/ CON002047. Accessed 20 April 2011. 22. NHS Prescription Service. Central nervous system national charts. April 2010. http://www.nhsbsa.nhs.uk/PrescriptionServices/ Documents/PPDPrescribingAnalysisCharts/NPC_Maps_CNS_ April_2010.pdf - 208KB. Accessed 20 April 2011. 23. Doak MW, Nixon AC, Lupton DJ, Waring WS. Self-poisoning in older adults: patterns of drug ingestion and clinical outcomes. Age and Ageing 2009; 38:407–411. 24. Nickander RC, Emmerson JL, Hynes MD, Steinberg MI, Sullivan HR. Pharmacologic and toxic effects in animals of dextropropoxyphene and its major metabolite norpropoxyphene: a review. Hum Toxicol 1984; 3(Suppl):13S–36S. 25. Hawton K, Simkin S, Deeks J. Co-proxamol and suicide: a study of national mortality statistics and local non-fatal self poisonings. BMJ 2003; 326:1006–1008. 26. MHRA. Prescription trends in analgesics 2003–2008: before, during and after the withdrawal of co-proxamol. http://www.mhra.gov. uk/home/groups/es-foi/documents/foidisclosure/con100023.pdf. Accessed 20 April 2011. 27. Corcoran P, Reulbach U, Keeley HS, Perry IJ, Hawton K, Arensman E. Use of analgesics in intentional drug overdose presentations to hospital before and after the withdrawal of distalgesic from the Irish market. BMC Clin Pharmacol 2010; 10:6. 28. Sandlands EA, Bateman DN. Co-proxamol withdrawal has reduced suicide from drugs in Scotland. Br J Clin Pharmacol 2008; 66: 290–293. 29. Watson WA, Litovitz TL, Belson MG, Wolkin AB, Patel M, Schier JG, et al. The toxic exposure surveillance system (TESS): risk assessment and real-time toxicovigilance across United States poison centers. Toxicol Appl Pharmacol 2005; 207(2 Suppl): 604–610.
Clinical Toxicology vol. 49 no. 6 2011
