The New Era of Risk Management

FDA has issued the industry a new charge-pay closer attention to risk management. Now that prescription drug user fees have helped the agency approve candidates more rapidly, FDA has returned to its basic mandate: assuring that marketed pharmaceuticals are safe. In the past, that meant clear labeling with adequate directions and warnings based on clinical trials. The agency now believes that product safety extends beyond warning labels and wants to ensure that prescriptions are used safely as well. As a result, it is asking the pharma industry to demonstrate products' safety before approval and to further control their use after approval.

First, it is important to understand what "risk management" means today. Traditionally, pharma companies have mitigated risks by assessing safety throughout clinical trials and through post-marketing surveillance. Product labels have always specified the parameters for safe use, promotional communications have been limited to approved indications, and fair-balance statements have conveyed the meaningful risks. Those things have not changed. Now, however, FDA wants greater assurances that pharmaceuticals will be used as safely as possible. As Deputy FDA Commissioner Lester Crawford recently said, "There was a time when we put the drug out there and didn't worry much about it. That is changing."

For pharma companies intent on developing drugs as rapidly as possible and launching products on a foundation that uniquely promotes the product's benefits, contemporary risk management presents new challenges. It means that companies must better assess products' risks, more fully evaluate their meaning and implications, communicate the known risks more completely, and institute systems that minimize risks and maximize safe use.

That will require more careful thought and new forms of evidence, backed by research, about how products will be used when they reach the market. It will also require coordinated planning by companies' regulatory, clinical development, and marketing teams. It will require increased reliance on epidemiologists to understand user populations' prescribing and utilization proclivities, and it will call for pharma companies to modify questionable patient behavior through targeted communications and specialized distribution systems.

This article reviews the background and recent history of the new era in risk management and how it has affected drug development and marketing. It also outlines how pharma companies must adjust their planning to take into account each of the four essential risk management functions: identification, evaluation, communication, and systems development.

Tolerance Shift

Culture can change slowly as new ideas replace old concepts, or it can change rapidly as dramatic events shift public views in an instant. The 1999 release of the Institute of Medicine's report, "To Err is Human: Building a Safer Health System," is the instant that ushered in the new era of risk management. Based on research by Lucian Leape of Harvard University, the report stated that 44,000-98,000 Americans die each year from medical errors, more than from car accidents or breast cancer. Although those figures include unavoidable adverse drug reactions, they had an immediate and far-flung impact on the public.

In reaction, numerous public health agencies and private sector organizations initiated programs to reduce medical errors, and FDA responded with a series of initiatives to better manage medical product risk. Those initiatives and a new philosophical perspective on risk management were summarized in an internal FDA report to the commissioner in May 1999. It defined a safe medical product as "one that has reasonable risks, given the magnitude of the benefit expected and the other alternatives available." Unlike most federal publications, in which the report itself is the culmination of work on a project, that report established a blueprint for future FDA actions to better manage medical product risks.

The most dramatic evidence of changing tolerance for drug risks is the number of product withdrawals in the past few years. The General Accounting Office reported that in each half decade since the late 1970s, two to four drugs have been removed from the US market. Yet, in the past four years, at least ten therapies have been pulled. (See "Off the Market.") The relative safety of the product, compared with others on the market, was a key issue in several withdrawals. With the Rezulin (troglitazone) withdrawal, for instance, a Health and Human Services' press statement said,"FDA took this action after its review of recent safety data…showed that Rezulin is more toxic to the liver than the other two drugs." Similarly, with Baycol (cerivastatin), Dr. John Jenkins of FDA was quoted in the New York Times as saying, "Baycol really stood out as being different. Baycol did not offer any benefits beyond those of the other statins. But it carried a potential risk, and that leads to a conclusion that it is no longer safe to be marketed."

Off the Market

The failure of physicians and patients to comply with product directions and warnings was an additional factor in product withdrawals. CDER Director Janet Woodcock said in an April 4, 2000, presentation at Temple University, "We've had to withdraw drugs from the market that would have been safe if used according to label instructions." Thus, in this new era of risk management, safety means more than describing risks on the product label; now the expectation is that manufacturers must influence safe drug usage.

To enforce the new drug safety standards, FDA has reorganized its Center for Drug Evaluation and Research and created an Office of Drug Safety. Through increased user fees, it seeks to fund the hiring of 100 new employees during the next five years.

Industry Responds

Following FDA's lead, pharma companies have begun to incorporate the demands for risk reduction in both drug development and marketing practices. High-profile risk management systems, such as Celgene's Thalomid (thalidomide) STEPS program and Roche's Accutane (isotretinoin) START program, incorporating multifaceted communications and special distribution controls, were instituted to reduce pharmaceutical exposure during pregnancy. Other companies have also established programs to assess, evaluate, and communicate risks to health professionals and patients on a more routine basis.

Although the industry fully supports increased safety, concerns about the costs of risk management are also under debate. As companies seek to meet sales targets by marketing effective new drugs, FDA's increased demand for safety data can seriously dampen the industry's ability to introduce new compounds. Drug development may be delayed if the agency demands longer-term trials to more fully assess and evaluate risks.

FDA, for example, recently denied a new drug application (NDA) for Sepracor's allergy drug, Soltara (tecastemizole), because the trials conducted were of insufficient duration. There was no evidence that the therapy had any specific risks, but FDA wanted greater assurance that it was safe for longer use. Similarly, the filing of a supplemental NDA for Celgene's Thalomid to treat multiple myeloma has been delayed despite the company's completion of four clinical trials. The agency has requested a fifth study, postponing the filing by two years.

Furthermore, once a candidate is saddled with the "risky drug" label, requiring special forms to complete or information sheets to distribute, companies may question whether they can effectively market the product, especially if there are competing brands without such requirements. Indeed, that stigma, coupled with heavy risk management requirements, may inhibit sales.

On the other hand, risk management activities may increase public confidence that a product can be used safely, improve the relationship among prescribers, patients, and pharma companies, and stave off generic competition by making it difficult for new products to enter the market unless they duplicate the risk management activities. Although the rule of thumb is that 75 percent of sales are lost within two years of US patent expirations, analysts predict that sales of Roche's Accutane will erode by only 25 percent after generic introduction because of the stringent safety monitoring required.

Regardless of the industry's concerns, FDA has clearly established new rules for drug development and marketing, and the industry's decision to support FDA's risk management activities through user fees, at least symbolically, endorses the agency's plans. It is now up to individual companies to build their own risk management frameworks as they develop new products and discover new risks for existing therapies.

How do companies build such risk management programs? Following are strategies for establishing the four building blocks of a meaningful risk management planning.

Identification

The first critical step is to classify risks as predictable or unpredictable. Predictable risks are by far the most common and should be considered at the earliest states of drug development. Researchers can identify such risks by considering a broad range of potential information sources and answering a set of questions:

• Does the product have potential adverse reactions that can be predicted by its pharmacokinetics (patient absorption, distribution, metabolism, elimination)?

• What adverse reactions have been found for similar compounds already on the market?

• Do toxicology studies indicate potential risks?

• What side effects and adverse reactions were found during Phase I and Phase II trials and are they associated with more serious events such as liver failure?

• What risks are inherent in the patient population and are they related to the risks of the drug?

• What risks could be incurred during manufacturing, storage, packaging, and distribution?

• Are there potential problems with the method of delivery or with inactive ingredients?

• Is there any potential for abuse or withdrawal reactions?

• What concomitant medications are likely to be a problem? Are they key medications for the target population?

Researchers can also employ the following tactics to identify potential risks:

• data mining for signals-clusters of events associated temporally with use of a drug-that may signify an adverse reaction

• epidemiological studies of patient populations to profile baseline morbidities.

Understanding the common diseases or morbidities associated with a drug indication by researching the natural history of disease in the target patient population may also be useful. Such studies appear in literature or are available from literature reviews. They may also be conducted in retrospective databases that can map the progression of conditions over time to identify the other diseases and conditions that affect the target patients. Longitudinal database studies, in particular, can also identify the potential for drug interactions by developing profiles of drug utilization to target those patients likely to be using interacting drugs. Such studies can also determine which sub-populations may be at the most risk for serious problems such as arrhythmias or liver injuries that may be attributed to the new therapy.

Evaluation

Once identified, risks must be evaluated to determine their potential for hazard. Such evaluation helps identify and quantify an acceptable risk level. It puts the hazard of the individual therapy in the context of the magnitude of the benefit expected and the other alternatives available. Following are some tactics to evaluate the dangers:

Epidemiological studies. These studies can compare the risks in the target population with those of the general population. The prevalence and incidence of cardiovascular co-morbidities in the population of patients with schizophrenia, for instance, is much higher than in the general population. Studies can quantify those rates so they can be used to inform physicians and as a benchmark for future reported rates for the therapy under investigation.

Root cause analysis. Included in the responsibility to ensure that drugs are used safely is the need to identify and overcome problems in the healthcare delivery system. For instance, Janssen discontinued commercial marketing of cisapride because its interaction with other medications that inhibit the cytochrome P450 3A4 enzyme function could lead to serious cardiac arrhythmias. An analysis of prescribing patterns found that half of the patients who received concomitant prescriptions for cisapride and contraindicated drugs received them from two different doctors, yet most patients took them to the same pharmacy. In the past, pharma companies' interventions to prevent contraindicated prescribing have been targeted to physicians, but the analysis indicated that more can be done to bring pharmacies and patients into the problem-solving loop.

Communication

It is a dictum at FDA-and the law as well-that pharmaceuticals cannot be safely marketed unless they are properly labeled. Labeling not only constitutes the physical label that accompanies the commercial product, but it also includes the vast majority of company communications that describe the therapy, its use, and its effects. Under the new rules for risk management, communications must not only disclose known risks, they must also clearly communicate them. To achieve the goal of safe product use, risk communications must motivate readers to comply with advocated behaviors that minimize risks.

Pharma companies can communicate risk information to healthcare professionals and patients at every stage of drug development and marketing. (See "Risk Documentation.") Regulatory agencies must review each of those documents to ensure that they provide adequate risk disclosure. Increasingly, regulatory agencies seek evidence that important information is conveyed in an easy-to-understand way. Also, questions have emerged about the value of simple communications to influence behavioral change. Marketers are designing multifaceted communication programs and evaluating them to ensure adequate communication of pharmaceutical risks. Following are some tactics pharma companies can use to do so:

Risk Documentation

Communications planning. Communications strategies must be designed to send specific messages to specific audiences, and multifaceted programs may be essential to achieving the company's objectives. Multiple interventions that provide repeated messages through a variety of media are clearly superior to a single intervention strategy. However, it is important to develop an integrated system that avoids loading the audience with too much information in any one document.

Document design principles. Each risk management document directed to health professionals and patients bears the distinct responsibility of clearly communicating warning messages. Those documents may also need to convey a vast amount of additional information that companies or regulatory officials think would be helpful. Unfortunately, document readers can absorb only a limited amount of information. Having clearly delineated objectives can make it easier to design effective communications. Reducing the cognitive load and providing clear "signals" to readers about important information can improve the likelihood that risk communication goals are met. (See "The Pharmacokinetics of Patient Communications," Morris and Aikin, Drug Information Journal, 2001.)

Document testing. Regardless of how well participants think they have designed documents, testing is essential. Qualitative testing using individuals or groups to review the documents may provide some general reactions during the drafting process. FDA has asked for quantitative research to support risk communication designs. Target audiences should include both healthcare professionals and potential patients. Companies can use comprehension testing and label perception tests to determine whether risk and benefit messages clearly communicate how to use the product safely.

Evaluation. Regardless of how much testing is completed prior to use, pharma companies cannot predict with certainty that risk communications will have the intended impact. Therefore, an evaluation plan to accompany the implementation of a risk communication program can help assess its impact. It is important that such evaluations provide specific information about all of the program's elements and sub-goals. If overall physician and patient behavioral goals are not met, it is important that the company determine what aspects of the program failed. Diagnosing the reason for risk communication failure is imperative. Without such information, the program will be scrapped and regulators may require more drastic risk management interventions.

Management Systems

One contribution of the Institute of Medicine's report and a follow-up report, "Crossing the Quality Chasm," published in 2001, was to introduce healthcare professionals to the concept of error prevention analysis within medical systems. A system can be defined as the coming together of parts, interconnections, and purpose. In other words, by going through the process of building a system to minimize errors, the industry can better understand the elements necessary for a risk management plan. By delineating the steps necessary to use a therapy safely-starting with the patient's safe use and working backwards-pharma companies can learn what informational and environmental conditions must be present at each stage. Such "systems thinking" lays the foundation for designing a series of interventions to improve drug safety. To achieve that goal, the industry must understand and integrate all of the elements of the risk management system. Pharma companies can use the following tactics to manage risk:

Systems analysis and design. To assure safe medicine use, it is important to delineate each of the necessary steps and to determine how each is functioning. Without such delineation, medical errors may occur. For an example of a systems analysis of prescription writing and dispensing, see "Safety Steps."

Safety Steps

Test risk management systems. Although logical analysis of a system can be a useful exercise, a systematic review by field experts can provide insights into the actual working of a system or show how it must be designed to consider environmental constraints. Some systems are simple and unidirectional; others, such as those required in the treatment of hypertension, can be quite complex. Beta testing risk management systems with a group of volunteers in real or simulated environments can also provide important feedback for improved design.

Systems evaluation. Registries to track initial product users can provide a database for systems evaluation. Systematic interviews with patients and providers can be particularly useful in determining whether the system is working as designed, if there are unanticipated conditions undermining safe therapy use, and if the system is misunderstood. Again, evaluations must be specific enough to allow the design to correct and modify risk management programs. Using registry participants to test new interventions-such as new patient information materials-can also provide for positive modifications.

Risk-Benefit Ratio

As the industry seeks to develop more potent treatments, pharmaceutical benefits will continue to ascend. So will their risks. As companies develop new products, they will need to identify risk signals sooner, evaluate the risks, communicate known risks and benefits as thoroughly as possible, and develop systems to manage risks.

Those four fundamentals will serve pharma companies by making drug development more focused and by giving them more confidence when regulatory authorities ask hard questions about their products. The best products are those with minimal risks and maximum benefits. Risk management seeks to improve that ratio to the fullest extent possible.