Key words: Drug safety * Europe * Pharmacovigilance * Post- marketing studies * UK
ABSTRACT
The granting of regulatory approval allows medical practitioners to prescribe a drug in a controlled way to a group of patients defined within the licence. Prior to this, the new product may have been evaluated often in less than 5000 patients and usually in a selected environment in which many patients have been excluded, including for example, women of childbearing potential, the elderly and children. Co-existent disease and the concomitant use of a number of common drug treatments also frequently exclude patients from pre-licensing trials. It is hardly surprising, therefore, that many adverse drug reactions are only detected once the product has been prescribed to the general population. National and international regulatory bodies, therefore, provide systems for post- marketing pharmacosurveillance, although participation in these by clinicians is generally voluntary and under-reporting is widespread. Post-marketing surveillance (PMS) studies are not generally an integral component to launching a new drug and many clinicians are sceptical over data generated in trials which do not conform to the ‘gold standard’ randomised control trial (RCT) design. However, in dismissing such studies, a great opportunity to obtain information, often from many thousands of subjects, is being missed. This article discusses post-marketing pharmacovigilance and the role of PMS studies in the context of current UK and European legislation.
Introduction
Estimates suggest that bringing a new drug to market is an expensive exercise costing somewhere in the region of 500 million ($800 million) and taking on average 10 years-12 years for the successful completion of clinical trials, specifically designed to gain approval by the regulatory bodies’2. Very few therapeutic molecules complete the successful journey from, laboratory discovery to the pharmacist1 and only three out of ten products launched ever recoup their development costs. A 20-year patent usually protects all new drugs, although most of this time is taken up during the development phase, such that the ‘in-use’ factual patent term can be less than 8 years. This places enormous pressures on pharmaceutical companies to achieve a substantial market share as soon as possible.
In 2004 two high profile pharmaceutical companies arguably marketing two well-known drugs experienced significant financial embarrassment and media attention. Following consultation with the European regulatory authorities, AstraZeneca issued a ‘Dear Doctor’ letter, in June 2004, in which attention was drawn to the risk of rhabdomyolysis with rosuvastatin (Crestor) and in the UK, the Medicines and Healthcare Products Regulatory Agency (MHRA) recommended a starting dose of 1Og in all patients3. A few months later a press release from Merck announced the withdrawal of rofecoxib (Vioxx) because of an increased cardiovascular risk in patients taking the drug for more than 18 months4. These are, however, only the latest in a series of actions needed to be taken by a number of companies marketing a whole range of pharmaceutical products. While some companies have been criticised for aggressive marketing of new drugs, the most important question which now needs asking is how do we bring new therapies into clinical practice quickly enough to benefit patients while at the same time provide maximum protection.
Acquiring regulatory approval
The process involved in obtaining regulatory approval, which allows new therapeutic products into the market place, is complex, although detailed guidance and advice can be found on the website of the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH] (www.ich.org) and the European Medicines Agency EMEA (www.emea.eu.int).
Most health-care professionals are familiar, however, with some of the regulatory requirements, including Phase I, II and III clinical trials. Phase I trials, usually addressing pharmacokinetic and pharmacodynamic properties of the drug, define pharmacological safety in a relatively small number of healthy volunteers. Phase II, exploratory therapeutic trials, are the first attempts to show proof of concept in a relatively small number of patients with disease, usually a few hundred, in very controlled circumstances, typically over periods of around 3 months. Finally, Phase III trials are therapeutic confirmatory studies, performed on much larger groups of subjects, often in excess of 1500 subjects. The inclusion and exclusion criteria while strict, more closely match the general ‘target’ disease population; the studies may be conducted over 6- and 12-month periods and are usually of the random control trial (RCT) ‘gold standard’ design.
On average, approximately 4500 individuals will be exposed to a new drug during the development programme. This is clearly insufficient to reliably detect an adverse drug reaction (ADR) with, for example, an incidence of 1 in 10000, which for a commonly prescribed drug may be considered to be an unacceptably high safety/ efficacy ratio.
While contributing to the acquisition of regulatory approval and a product licence, these early studies, therefore, rarely provide answers to many questions posed by health-care professionals. They are usually performed on highly selected groups of patients often with an under-representation of large groups including women of childbearing age, the elderly and also children5-7. Co-existent disease and concomitant use of a number of common drug treatments often leads to study exclusion. While demonstrating therapeutic benefits, comparative data with existing therapies are usually not available nor are long term end-points and safety profiles. For these reasons regulatory agencies such as the Committee on Safety of Medicines (CSM) and the MHRA, the EMEA and the Food and Drug Administration (FDA) all have Systems in place for monitoring drug therapies and in particular close surveillance of new drugs.
Monitoring new drugs after general release
In 1964 following the well-known problems with thalidomide, the Yellow Card Scheme was introduced in the UK8. This provided a mechanism of postmarketing pharmacosurveillance of suspected ADRs. In 2002 an electronic card system was introduced but as with the original yellow card, its success is dependent upon the reporting of ADRs to the CSM by doctors on a voluntary basis.
The CSM also encourages the reporting of all suspected ADRs to new drugs and vaccines marked with an ‘inverted black triangle’ in the British National Formulary (BNF), Nurse Preservers’ Formulary (NPF), MIMS, the ABPI Compendium of Datasheets and Summaries of Product Characteristics, and advertising material9.
A black triangle will be assigned to a product if the drug is a new active substance. However, a product containing previously licensed active substances may also be monitored if it is a new combination of active substances. Black triangles can also be added if administration of a product is via a novel route or drug delivery system, or if there is a significant new indication which could alter the established risk/benefit profile of that drug. As with the yellow card scheme, the reporting of ADRs for black triangle drugs is voluntary.
Information received by the CSM and MHRA is evaluated and where appropriate feedback is provided. Product information is updated ‘routinely’ as necessary when there are safety concerns. Three to four times each year serious drug safety issues are communicated via the Current Problems in Pharmacovigilance bulletin mailed to all doctors, dentists, pharmacists and coroners in the UK. Not infrequently urgent safety issues including a newly identified ADR or drug withdrawal are communicated via letter, fax or via the Chief Medical Officer’s EPINET. Along with other governmental pharmacovigilance centres around the world reports are also submitted to the ICH, the EMEA and also the WHO programme for International Drug Monitoring.
Despite these procedures, concerns remain over drug safety, particularly when new products are released into general use following the limited information gained from trials designed for regulatory purposes. Systems dependent upon voluntary monitoring of ADRs lead to an under-reporting, which is common and widespread10- 14. Although serious ADRs are more likely to be reported10, overall rates of spontaneous reporting are often well below 10% of all ADRs11-13.
There are, however, other methods by which ADRs can be identified, evaluated and monitored. These include observational safety studies, usually sponsored by the market authorisation holder, prescription event monitoring (PEM) as undertaken by the Drug Safety Research Unit at the University of Southampton and follow-up via large computerised databases such as the General Practice Research Database (GPRD).
Post-marketing surveillance (PMS) studies
Company sponsored PMS studies are regarded by many as studies generating poor quality data that do not match that from clinical RCTs and which cannot stand up to the rigors of high quality external peer-review. Much of the criticism arises from the traditional view that such studies are really ‘seeding’/marketing trials designed by a company aimed at promoting sales of new drugs. However, to dismiss such studies as having no \role deprives us of a potentially valuable source of information obtained from real-life patients that have not been subject to, nor excluded by, standardised clinical trial circumstances.
In an attempt to provide studies of greater scientific credibility that could withstand peer-review scrutiny, the Safety Assessment Marketed Medicines (SAMM) guidelines were developed in the UK in 1994 by a working group that comprised the Medicines Control Agency (now MHRA), CSM, the Association of the British Pharmaceutical Industry (ABPI), the British Medical Association (BMA) and the Royal College of General Practitioners (RCGP)15. Guidelines are also available in other European countries including the Netherlands and Germany and in the UK there are also clear guidelines for Phase IV clinical trials, where medication is provided by the sponsoring pharmaceutical company. Multinational studies including pan-European multicentre studies have the great advantage of allowing many thousands of patients to be recruited. In this respect pan-European guidance is provided by the PostAuthorisation Safety Study (PASS) guidelines16 which are very similar to the SAMM guidelines.
In the UK, a SAMM study has been defined as ‘a formal investigation conducted for the purpose of assessing clinical safety of marketed medicines in clinical practice’ and the guidelines are intended to cover all company-sponsored studies carried out to evaluate the safety of marketed medicines. Such studies may be hypothesis-generating, to identify previously unrecognised safety issues or hypothesis testing, to investigate possible hazards. The design of the study will depend upon the objectives but may include observational cohort studies, case-control studies, case surveillance and clinical trials.
Company sponsored observational PMS studies are those subject to the greatest scepticism but are also those capable of producing some of the most useful data to health-care professionals. Guidance here is very clear and includes:
(i) Patients included in the study should be representative of the general population of users.
(ii) Comparator groups should normally be included (i.e. identical patients on an alternative therapy).
(iii) Only once a drug has been prescribed as a result of normal clinical practice can the patient be entered into the study. Recruitment into the study must then be in accordance with study protocol.
For all studies the general advice is also clear:
(i) Companies are advised to discuss the draft study plan with the MCA (now MHRA) and then submit a finalised plan to the MHRA before starting the study. Standard regulatory requirements for reporting ADRs must be fulfilled and both interim and final reports should be submitted to the MHRA.
(ii) Appropriate ethics committee approval is required.
(iii) The companies’ medical department is responsibility for the conduct and quality of the study under the supervision of a named UK registered medical practitioner.
(iv) Where an agent for the company performs the study, the agent should identify a UK registered medical practitioner to supervise the study and liaise with the company.
(v) An independent advisory board should be appointed to oversee the study.
(vi) Studies should not be conducted for the purposes of promotion and company representatives should not be involved in such a way that it could be seen as a promotional exercise.
(vii) Study payment to doctors to recompense for time and expenses should be in accordance with BMA guidelines.
Pharmaceutical studies that adhere to these guidelines should, therefore, be scientifically credible and contribute in a meaningful way to our understanding of the drug being evaluated. Probably the most important feature of a PMS study, which is often overlooked, is that the decision is made to prescribe the drug as part of routine clinical practice before the patient is entered into the study. This lack of randomisation is a critical feature of a PMS study as it leads to the monitoring of real-life clinical use of the new drug. The decision by the clinician to ‘prescribe before entering’ based on clinical circumstances should also reduce pharmaceutical company intervention and bias.
Combined with other post-marketing schemes and Phase IV clinical trials, PMS studies registered with the MHRA, that have appropriate Multicentre/Local Research Ethics Committee MREC/LREC approval, an independent steering committee for data monitoring and that adhere to all other PASS or SAMM regulatory guidelines, will increase our confidence in the validity and importance of the data generated.
Most recently in November 2004, the ICH issued a new tripartite guideline on pharmacovigilance planning (designated E2E) for recommendation to the regulatory bodies of the European Union, Japan and USA17. The guideline is intended to assist with the planning of pharmacovigilance, especially in the post-marketing period. It will make it mandatory that the applicant pharmaceutical company of a new drug, or of an existing product undergoing significant change in its usage or where new safety concerns have arisen, formulate a Pharmacovigilance Plan with advice from the regulator, which will be based upon a safety specification detailing the important identified risks of a drug, important potential risks and important missing information. The ICE E2E document, also endorsed by the WHO Advisory Committee on Safety of Medicinal Products (ACSoMP) and which will be implemented during 2005 in the European legislation for pharmacovigilance (volume 9) by the EMEA, will lead to the involvement in the design, and the monitoring of follow-up and the completion of pharmacoepidemiological studies by the EMEA. A further contribution to the development of a European Risk Management Strategy will be the inclusion of detailed regulation for the implementation of PMS studies as a follow-up measure that may be a condition for approval of a drug.
What are the benefits of PMS studies?
Searches of electronic databases reveal an increasing number of PMS studies, often on many thousands of patients. In some instances data confirm that of the original early phase clinical trials, although in many others, novel hypothesis-generating data provide invaluable ADR information. In a recent study which looked at the differences between the characteristics of treated populations and treatment patterns in clinical trials and post-marketing settings for tacrine, simvastatin and celecoxib, the authors concluded that ‘their results plead for systematic ad hoc observational post- marketing studies for any novel and/or expensive medicine…’5. This view has recently been endorsed by clinicians in the USA with a request for a more integrated and comprehensive approach to post- market approval surveillance, following recent concerns over restenosis rates with the novel Cypher stent in patients with coronary artery disease18.
The potential benefits of a PMS study are that reallife patients are being evaluated in real-life situations and because of a lack of tight inclusion and exclusion criteria, large numbers may be recruited. The consequences, therefore, are that infrequent (1 in 1000) yet important adverse reactions can be documented in studies enrolling 10000-20000 patients. In Germany, PMS studies are already considered to be an integral component to the launching of a new drug, with some studies exceeding 50 000 patients.
Although such studies in the UK are not commonplace, we are now starting to see high quality PMS studies with the potential to play an increasing role in patient safety and our understanding of how new drugs behave in the general population. Following the recent launch of a new basal insulin analogue, insulin detemir, Novo Nordisk and Kendle International (Novo Nordisk’s partner contract research organisation) initiated a large international observational PMS study, PREDICTIVE (Predictable Results and Experience in Diabetes through Intensification and Control to Target: an International Variability Evaluation). The primary endpoint is to evaluate the incidence of serious ADRs, including major hypoglycaemic events, during treatment with insulin detemir. The secondary endpoints include weight change, fasting blood glucose variability and HbA1 . The study will involve 25000 patients and 5000 clinicians world-wide. Importantly, prior to starting the study, advice was sought from the appropriate regulatory body (MHRA), ethics committee approval was granted, an independent steering Committee was established to monitor the data and remuneration to clinicians is in line with the BMA fee schedule for PMS studies19.
While there are obvious safety benefits for the patient there are advantages too for clinicians and pharmaceutical companies. Following the launch and almost immediate withdrawal of the thiazolidinedione, troglitazone in the UK, the subsequent release of rosiglitazone and pioglitazone has been accompanied by guidelines from the National Institute of Clinical Excellence (NICE), recommending 2-monthly liver function tests for the first 12 months for all patients prescribed either of these agents20. These drugs have now been prescribed to many thousands of patients, and unlike troglitazone, hepatic toxicity does not appear to be a side effect. A properly conducted PMS study could by now have produced evidence for consideration by NICE to reduce what many diabetologists believe to be a waste of valuable time and expense in performing unnecessary tests. The respective pharmaceutical companies would also benefit, not only by having additional safety data on their products but probably by increased sales which have almost certainly been held back by the burdens associated with prescribing these drugs.
The way forward
High quality, scientifically credible, efficacy and safety data are needed for all drugs so that we can prescribe to our patients with confidence. The Department \of Health, on behalf of the UK Clinical Research Collaboration, is leading the work to establish a new UK clinical research network to facilitate the conduct of randomised prospective trials of interventions and other well designed studies in the broad area of clinical research21. This should include good PMS studies. As clinicians we need to appreciate the benefits of well-conducted and properly regulated PMS studies so that we can encourage pharmaceutical companies to invest in them and develop mutually beneficial collaborations between clinicians, regulatory bodies and pharmaceutical companies. This appreciation could begin with the incorporation and teaching of all types of clinical studies into our undergraduate curriculum training programmes at university. These activities could be extended to post- graduate education and even continuing medical education. Ultimately, data obtained from well-conducted PMS studies, analysed by independent advisory groups, should be written up for peer- review publication and should be viewed positively by editors of good quality journals.
When a licence for a new product is submitted a pharmaceutical company must provide all ‘positive’ and ‘negative’ data and a summary of this is available to the public through European Public Assessment Reports (EPARs), produced by the EMEA on the granting of a licence. In addition the ABPI actively encourages the publication of all clinical trial data including those that show both ‘positive’ and ‘negative’ results to minimise publication bias. Although the ABPI have produced a website for companies to publish information from clinical trials (www.cminteract.com/clintrial/), this remains voluntary. As clinicians we should encourage companies to commit to this or other independent and transparent electronic databases.
Finally, the European Risk Management Strategy, conducted by the EMEA, including the E2E ICH guideline, due to be implemented this year22, should have a significant impact on pharmaceutical companies seeking regulatory approval for new products including the need for more widespread implementation of post-marketing pharmacoepidemiological safety studies.
Conclusions
The safety of medicines we give to our patients is of paramount importance.
The costs and time involved in the development of new products are great and place enormous pressure on companies to recoup some of the costs as quickly as possible. Acquiring regulatory approval for a new drug allows us to begin to understand how it might behave in an unselected, real-life, population. Post-marketing pharmacovigilance employing a variety of complementary approaches supported by new European legislation provides us with the opportunity of providing early access to the latest forms of treatment combined with the highest standards of medical care and safety.
Acknowledgements and declaration of interests
Novo Nordisk Ltd have provided support for the publication of this manuscript, but I have received no financial support or sponsorship from Novo Nordisk Ltd or Kendle International as reimbursement for writing this article, which expresses my own views without influence from outside sources.
I am grateful to Alan Davies (Kendle International) and Chris Martin (Novo Nordisk) for information and helpful comments on the manuscript, including the provision of clinical data searches.
References
1. The Association of the British Pharmaceutical Industry. http// www. abpi.org.uk/statistics/intro.asp [last accessed 4 March 2005]
2. DiMassi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ2003;22:151- 85
3. New prescribing advice for the 40 mg dose of Crestor (rosuvastatin) http://medicines.mhra.gov.uk/ [last accessed 4 March 2005]
4. Merck. Merckannounces voluntary worldwide withdrawal of VIOXX”. http://www.vioxx.com/vioxx/documents/english/ vioxx_press_release.pdf [last accessed 4 March 2005]
5. Martin K, Bgaud B, Latry P, Miremont-Salam G, Fourier A, Moore N. Differences between clinical trials and post-marketing use. Br J Clin Pharmacol 2003;57:86-92
6. Schmucker DL, Vesell ES. Underrepresentation of women in clinical drug trials. Clin Pharmacol Ther 1993;42(Suppl):l 1-5
7. Bugeja G, Kumar A, Banerjee AK. Exclusion of elderly people from clinical research: a descriptive study of published reports. Br Med J 1997;315:1059
8. The yellow card scheme: how the yellow card scheme protects patients. https://www.mhra.gov.uk/medicines/safety/ yellowcard/ howprotects.htm [last accessed 4 March 2005]
9. Monitoring the safety and quality of medicines; new drugs under intense surveillance, http://medicines.mhra.gov.uk/ourwork/ monitorsafequalmed/newdrugs/newdrugs.htm [last accessed 4 March 2005]
10. Heeley E, Riley J, Layton D, Wilton LV, Shakir SAW. Prescription-event monitoring and reporting of adverse drug reactions. Lancet 2001;358:1872-73
11. Smith CC, Bennett PM, Pearce HM, et al. Adverse reactions in a hospital general medical unit meriting notification to the Committee on Safety of Medicines. Br J Pharmacol 1996;42:423-9.
12. Mittmann N, Knowles SR, Gomez M, Fish JS, Cartotto R, Shear NH. Evaluation of the extent of under-reporting of serious adverse drug reactions: the case of toxic epidermal necrolysis. 18. Drug Safety 2004;27:477-87
13. Begaud B, Martin K, Haramburu F, Moore N. Rates of reporting of adverse drug reactions in France. J Am Med Assoc 2002;288:1588
14. Backstrom M, Mjorndal T, Dahlqvist R. Under-reporting of serious adverse drug reactions in Sweden. Pharmacoepidemiol Drug Saf 2004; 13:483-7
15. Guidelines for company-sponsored Safety Assessment of
Marketed Medicines (SAMM) guidelines. Medicines Control Agency, Committee on Safety of Medicines, Royal College of General Practitioners, British Medical Association and association of British Pharmaceutical Industry (November 1993). Br J Pharmacol 1994;2:95-7 22.
16. Pharmacovigilance medicinal products for human and veterinary use, vol. 9: section 1.5. Company-sponsored post-authorisation studies. http://www.eudravigilance.org/human/docs/Vol9en.pdf [last accessed 4 March 2005]
ICH harmonised tripartite guideline. Pharmacovigilance planning E2E. http//www.fda.gov/cder/guidance/6002dft.pdf [last accessed 4 March 2005]
Mehran R, Leon MB, Feigal DA, et al. Post-market approval surveillance: a call for a more integrated and comprehensive approach. Circulation 2004; 109:3073-7
BMA fees guidance schedule, section 6, April 2003 http://www/ bma.org.uk [last accessed 4 March 2005]
Guidance on the use of glitazones for the treatment of type 2 diabetes. http://www.nice.org.uk/pdf/TA63_Glitazones_ Review_Guidance.pdf [last accessed 4 March 2005]
Department of Health. Research for Patient Benefit Working Party. Final report 2004. http://www.dh.gov.uk/ assetRoot/04/08/26/75/ 04082675.pdf [last accessed 4 March 2005]
Work programme for the European Medicines Agency 2005. http:// www.emea.eu.int/pdfs/general/direct/emeawp/ 13835104en.pdf [last accessed 4 March 2005.
CrossRef links are available in the online published version of this paper: http://www.cmrojournal.com
Paper CMRO-2937_2, Accepted for publication: 09 March 2005
Published Online: 18 March 2005
doi:10.1185/030079905X41426
Stephen Gough
Division of Medical Sciences, University of Birmingham, Birmingham, UK
Address for correspondence: Professor Stephen C. L. Gough, Division of Medical Sciences, Institute of Biomedical Research (2nd floor), The Medical School, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. email: [email protected]
Copyright Librapharm Apr 2005
Comments