Global Risk Management
The changing safety landscape in the EU and U.S.
By Peter Summerfield and David Shoemaker
The safety of medicines is always an important topic, but of late it has become even more critical. The availability of new medicines is also a focus for patients. In the past decade, there has been a shift to earlier (often conditional) approval of new medicines. The public’s perception is that some medicines are being approved too soon, leading to the discovery of significant side effects shortly after introduction into the marketplace. Regulatory agencies for human medicines have the critical role of safeguarding public health. There is an impression that pharmaceutical companies are not adhering consistently to the regulatory requirements for capturing, evaluating and reporting all notifications of adverse effects that occur with new medicines whether they are detected during a clinical trial or reported spontaneously by patients/consumers or healthcare professionals during routine use of the products after approval for sale. To tackle these perceptions, there has been a global shift to increase the consistency of the capture, analysis and reporting of patients’ safety during pharmaceutical product development and after approval. These initiatives are being led from Europe and the U.S.
History of Safety in the EU: How we got here
In 1995, the European Medicines Evaluation Agency (EMEA) was founded, providing central coordination of the regulation of pharmaceutical medicines throughout all Member States within Europe. Since 2001, the European Commission (EC) has moved to a more proactive approach to safety of medicines and created a pharmacovigilance working party (PhVWP) to implement this strategy.
Since October 2005, it has been mandatory for companies to do the following:
- Discuss and prepare as appropriate Risk Management Plans (RMP) for all new marketing authorization applications (MAA), or on request for products in development or already approved.
- Submit Periodic Safety Update Reports (PSUR) throughout the lifecycle of a marketed product; six monthly at least for the first two years of marketing, annually up to the fifth year of marketing and every three years beyond that.
- Submit all Individual Case Safety Reports (ICSR) and Suspected Unexpected Serious Adverse Reactions (SUSAR) electronically to the EMEA’s EudraVigilance safety system, either through EVWEB or the ESTRI Gateway.
- Submit to Pharmacovigilance Inspection by regulatory authorities to ensure the appropriate pharmacovigilance systems are in place to comply with all EU and national regulations.
And there is an overarching expectation that the Qualified Person for Pharmacovigilance (QPPV) — an EU resident with appropriate PVG experience — will oversee delivery of all aspects of pharmacovigilance within his/her company and that he/she is, in turn, fully supported by the company.
Superimposed on these risk management requirements, and in order to address the needs of special populations, there is an additional requirement within the EU since 2008 for pediatric investigation plans (PIP) to be developed for all products that may be used in any pediatric population. In the future, approval of a Marketing Application Authoriza-tion will be dependent on a valid PIP.
Current and Future EU Safety Landscape
As with any new legislation, it takes time for change to be implemented. Progress has been impressive, but there is still some way to go.
Risk Management Plans/Periodic Safety Update Reports
The requirement for RMPs is firmly established. The template is not particularly user friendly, differs from that in the U.S., and requires input from many different sources to compile properly. It takes time, so the RMP should be planned well in advance. Improvements to the template have been made. These improvements, along with a requirement to submit key elements electronically to EudraVigilance in parallel, demonstrate the legally binding aspect of the plans. Updates on implementation of the plan are essential elements of the PSUR.
About half the Member States in the EU have full EudraVigilance capabilities. In addition, some have additional user-testing nuances to those that need to be certified with EMEA. While it is taking longer than expected to implement, the progress to date is significant and the goals should be achieved shortly. Transmission of ICSRs is becoming easier. Submission of legacy cases for products approved since the creation of the EMEA in 1995 remains a huge challenge though, as reflected in a further extension to the deadline for compliance.
Routine and for-cause inspections are now part of routine risk management. Many companies have hosted more than one inspection. The more complex the development and marketing organization, the greater the challenges are to comply with all elements of EU and local legislation. Large and small pharmaceutical companies have had to invest significant efforts in compliance, whether it is in preparing the Summary of Pharmacovigilance Systems, or in response to inspection findings. This effort is never ending, and the role of the QPPV is central.
Europe has found its feet. The number of products referred to the Pharmaocovigilance Working Party for review is increasing. Establishing networks within the EMEA and externally with other regulatory agencies and academic groups is a sign that the complexity and quality of risk management is developing. Layered on this is greater involvement of patient groups. This will impact the content of RMPs, the design of studies in special populations and the commitments pharmaceutical companies will be expected to adhere to. While this is all proactive, it should not be forgotten that safety signals are often detected retrospectively. The evolving power of EudraVigilance to detect safety signals will greatly enhance regulators’ ability to avoid damaging product withdrawals. However, no system is perfect.
History of Safety in the U.S.
While protecting the U.S. public health is the primary mission of the FDA, the mechanism by which the agency accomplishes this task has undergone several organizational transformations. With the passage of the Food, Drug, and Cosmetic Act in 1938, the U.S. Government first required products to demonstrate their safety prior to marketing. A safe product was sufficient until 1962, when the Kefauver Harris Amendments required drug products to be efficacious as well as safe. Since that time, the FDA has evaluated drugs by assessing the benefit:risk ratio in both the pre-marketing and post-marketing arena. This concept of benefit:risk assessment has been the cornerstone of FDA’s critical evaluation of products and the ultimate determinant of whether a product will be approved for marketing and will be allowed to remain on the market.
Pre-marketing Risk Assessment
The assessment of investigational drugs has historically been the responsibility of the Office of New Drugs, where integrated teams of scientists and physicians evaluated both the safety and the efficacy of drugs during their clinical development. Risk is routinely evaluated from preclinical, manufacturing and clinical perspectives throughout the development of an investigational drug. With the advent of gene therapy clinical investigations in the late 1990s, the public at large and the U.S. Congress became much more aware of the potential safety risks with investigational products. This increased awareness coincided with the creation of the Office of Drug Safety at the FDA.
While there was some ambiguity of responsibility for safety assessment within the agency, the Office of Drug Safety in its current reincarnation, the Office of Surveillance and Epidemiology, appears to be coordinating much better with the Office of New Drugs in the evaluation of the risks associated with investigational products. As recently as June of this year, there appears to be agreement between the Office of New Drugs and the Office of Surveillance and Epidemiology regarding the responsibilities of the two offices in evaluating the safety of investigational drugs.
Post-marketing Risk Assessment
Throughout the FDA’s history, post-marketing drug safety evaluation in the U.S. has largely depended on the spontaneous reporting of adverse events to pharmaceutical companies and the agency by physicians and patients. Pharmaceutical companies are required to periodically report the safety results obtained with their products to the agency. This process has not proven to be especially robust.
In some cases, pharmaceutical companies are requested by the FDA to perform post-marketing safety studies to explore potential safety concerns identified by the FDA during development. Unfortunately the FDA lacked the authority to ensure these requests were carried out, and in the vast majority of cases, these trials were never conducted.
Current and Future U.S. Safety Landscape
In March of 2005, the FDA issued the three risk management guidances listed below:
- Pre-marketing Risk Assessment
- Development and Use of Risk Minimization Action Plans
- Good Pharmacovigilance Practices and Pharmacoepidemiologic Assessments
These three guidances form the foundation of the current thinking at the FDA regarding the subject of safety evaluation through risk management.
In September of 2007, the U.S. Congress passed the Food and Drug Administration Amendments Act (FDAAA), which had several provisions for improving safety monitoring of U.S. patients using approved products and for better public awareness of the investigational studies being conducted by the pharmaceutical industry.
First, and foremost, the Act provides the authority for the FDA to require post-approval studies for the purpose of identifying safety risks, assessing product-related safety signals or to identify product-related serious risks when the potential for serious risk has been identified during the investigational stage of development. The U.S. Congress conferred authority to the agency to level civic monetary penalties if these post-approval studies were not conducted to the agency’s satisfaction or were delayed. Finally, the FDA was given the authority to require the product sponsor to make any changes to the label the agency “deems appropriate to address the new safety information."
Additional provisions of the Act include the concept of a Risk Evaluation and Mitigation Strategy (REMS) that the FDA may request of product sponsors at any point during development, if safety data warrant it. The REMS requirement extends the concepts developed in the 2005 risk management guidances listed previously and encompasses the tools contained in the FDA’s accelerated approval regulations. One important distinction between the requirement for REMS by the FDA and a Risk Management Plan by the EMEA is that in the U.S., a REMS is not mandatory for all approved drugs, but rather requires a safety signal warranting the implementation of a REMS. The future success of the FDA in preventing product recalls will determine whether this framework is sufficient or if it evolves into a more compulsory requirement for REMS.
With regard to improving the safety monitoring of investigational products, FDAAA now requires product sponsors to complete a Form FDA 3674, “Certification of Compliance with Requirements of ClinicalTrials.gov Data Bank,” which is submitted along with a company’s protocol elements to a National Institutes of Health-sponsored website. Pharmaceutical product sponsors had previously been required to post their protocol elements to the clinicaltrials.gov website only if they were conducting a study in a life-threatening indication. The stipulation that all trials be entered into a national database will obviously increase the transparency of investigational trials and presumably the accountability for product sponsors regarding their safety results.
Recently, Dr. Robert Temple of the FDA categorized risk into four categories associated with pharmaceutical products that will assist with thinking about risk in the post-market environment. The first category of risk Dr. Temple defined is “Common and Severe, Obviously Drug-related.” This category of risk is characterized by those risks exemplified by the following:
- cytotoxic effects (oncologic drugs)
- infections, malignancies (anti-inflammatory tumor necrosis factor inhibitors for rheumatoid arthritis)
- agranulocytosis, seizures, myocarditis (clozapine for schizophrenia)
- bleeding (thrombolytics, anticoagulants)
These products produce relatively common serious risks that do not generally occur spontaneously. They are easily detected in pre-marketing trials and are deemed acceptable only for drugs that have no alternatives and produce significantly large therapeutic effects.
The second category of risk is characterized as “Rare and Severe, Obviously Drug-related.” These risks, by virtue of their lower frequency, are not always detected during investigational clinical trials. Examples of products in this category would be those products causing myocardial complications due to QTc prolongation or contributing to severe drug-induced liver injury by a mechanism potentially capable of being identified by evaluation using Hy’s Law as described in the recent 2008 FDA guidance. These are the type of risks that would be easily detected in post-approval safety studies and typify the risks associated with the products that have been most recently withdrawn from the U.S. market.
The third type of risk category is associated with “Spontaneous Events, Not Obviously Drug-related but Greatly Increased in Rate by Drug.” These risks are usually not able to be interpreted from a single or a small number of occurrences. However, the tight association with the administration of the drug, coupled with the low background rate in the general population, makes these risks also readily identifiable in post-approval safety studies. Examples of this third category of risk include the following:
- Deep vein thrombosis (oral contraceptives)
- Endometrial cancer (estrogens)
- Valvulopathy (fenfluramine, pergolide)
Finally, the fourth type of risk category defined by Dr. Temple is the “Spontaneous Events but only Modestly (< 2 fold) Increased by Drug.” These risks are obviously the most difficult to predict due to the fact that even epidemiological data are uncertain in many of these cases. They typically represent risks such as cardiovascular events, tumors and disease exacerbations.
Categorization of risks associated with pharmaceutical products is something that is long overdue if there is ever going to be a means to make the study of safety anything approaching a quantifiable effort. Dr. Temple has taken the first step, and while there may be others who refine these categories in the future, it is a necessary first step in the development of the science of safety. Once a better quantitative assessment is developed for the evaluation of risk, we can then begin to develop algorithms for benefit:risk assessment that the general public, healthcare providers, regulators and pharmaceutical companies will use to communicate with one another quantitatively.
Globalization of Your Risk Management Plan
The obvious trend in the industry is to begin thinking about your risk management plan during the development process. Inasmuch as the vocabulary defining risk management plans is not yet finalized or harmonized globally, it requires companies to commit to understanding the safety of their products by conducting studies investigating potential safety concerns prior to submitting marketing applications. Certainly, if any safety signal of concern is identified during investigational development, it makes sense to begin executing a coordinated risk management plan that satisfies both the EU and U.S. regulatory bodies.
Not only do companies have regional regulatory requirement differences to contend with, but also within a regulatory region various therapeutic disciplines are beginning to adopt safety requirements to assess risk. The June meeting of the Endocrinologic and Metabolic Drugs Advisory Committee of the FDA recently addressed the issue of cardiovascular safety of therapies for diabetes. The committee generally agreed with the approach advocated by panel member (and discussant) Steve Nissen (Medical Director, Cleveland Clinic) to mandate a pre-approval “screening” study followed by a “confirmatory” long-term study to be initiated pre-approval and completed post-approval. The long-term study should last at least three years, and preferably five.
Registries and observational studies are becoming increasing important research tools in the post-approval regulatory environment. These trials are traditionally the centerpiece of risk management programs and are likely to continue to be so. These focused studies are set up with minimal logistical commitments on the part of the investigator and consequently are attractive in today’s world of multi-tasking physicians. Focused regulatory document collection and the employment of centralized ethics committees/institutional review boards facilitate keeping effort to a minimum while still ensuring safety of the subjects. Investigator involvement in follow-up can also be kept to a minimum and in some cases no investigator input is required during this phase of data collection.
The acceptance of electronic data capture additionally facilitates the conduct and data analysis for these studies. Data requirements are such that minimal data query generation is the norm. Gaining the acceptance of regulators in the EU and U.S. for these studies at the pre-approval stage allows companies to proceed with global studies, which generate safety data bases of maximal utility. In addition to monitoring specific safety endpoints, these trials serve to garner additional data that assist in penetrating global markets.
Registrational and observational studies are not without their challenges with regard to implementation. The regulatory requirements vary from region to region requiring companies to have a familiarity with changing regulations throughout the world. While the requirement for investigators to participate actively in these studies may be reduced relative to pre-approval studies, their experience with Good Clinical Practices for human research is highly recommended.
The reporting of results from Risk Management Plans is also highly varied from region to region and requires familiarity with the specific region’s regulations. The recent FDAAA carries with it the requirement for companies to provide the FDA with an assessment of the Risk Evaluation and Mitigation Strategy within 15 days of a request from the regulatory authority. The Act requires companies to effectively be on alert at all times with regard to the safety profile of their products so that they may present an accurate evaluation of the risks associated with their products upon request. Periodic updates are required in both the EU and the U.S. to ensure regulatory authorities are apprised of the safety profiles of companies’ products on a reasonably frequent basis. It is probably not too far into the future when this safety data will be shared with the public at large.
As the safety environment continues to shift in the coming months and years, pharmaceutical companies will have to be flexible to address changing requirements. The relatively conservative nature of the current regulatory safety environment may eventually swing back to a less conservative nature. This shift has been the trend over the past several decades and will more than likely be the case in the future. However, it is impossible to predict what the next public health crisis will be (perhaps the onset of pandemic flu) that convinces the public and their government representatives that it is a higher priority to expedite the development of needed medications rather than ensuring the public health by preventing the marketing approval of medications that produce serious adverse events in a fraction of the treated population. No medication is without risk, but it is still a largely qualitative opinion what constitutes risk in peoples’ minds and what benefit justifies the corresponding level of risk.