Weighing the benefits and risks of diabetes drugs: a call for a unified global approach.

Abstract

Most people, and notably physicians and healthcare professionals, are familiar with the Chinese proverb that says, “Good medicine tastes bitter.” Drugs provide therapeutic benefits, from alleviating symptoms to slowing down and curing diseases, but they also carry risks for adverse side effects, which can range from minor gastrointestinal upset and headaches to potentially serious life-threatening catastrophies. Clinicians and pharmacists have ingrained in their daily clinical practices that they must weigh the benefits and risks of each drug before they prescribe and dispense them. Yet a majority of our patients would prefer natural or herbal remedies to taking prescribed medications. The rapid advances in science, medicine and technology and the easy access to a plethora of information through the Internet have enabled us to evaluate for ourselves the promised benefits and potential risks of drugs to harm human health and the environment. The withdrawal of high-profile drugs from the market in recent years has led to increased public concerns and skepticisms about the safety of medicinal products and devices. Like many physicians, I routinely have to take time to explain in great detail the benefits and risks of medications before prescribing them to my patients who are affected by diabetes. Increasingly, I have encountered patients who are extremely reluctant to consider even life-saving medications, like statins, because of their concerns with adverse side effects. This is not at all surprising and, in fact, is justifiable in light of frequent news in the media concerning medication-induced problems and the incredible proliferation of self-help websites and blogs that are readily available on the Internet, purporting to note the adverse side effects of various prescribed drugs. On the other hand, healthcare professionals and consumers alike are incessantly bombarded by ads in the mass media and on the Internet touting the benefits of alternative remedies for all kinds of diseases, including cancer. With respect to drugs that we prescribe to patients with diabetes, there is a long history of controversies surrounding the safety of many medications, including oral glucose-lowering drugs and insulins. A recent example is rosiglitazone, an effective glucose-lowering, insulin-sensitizer drug that was implicated in a meta-analysis as being associated with increased risk for myocardial infarction and other cardiovascular events (1Nissen S. Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes.New Engl J Med. 2007; 356: 2457-2471Crossref PubMed Scopus (4092) Google Scholar). This and other publications concerning observational studies led the US Food and Drug Administration (FDA) to add a black-box warning and place additional restrictions on the use of this drug. Recently, the FDA reviewed the updated published data and concluded that concerns for cardiovascular safety were not well substantiated and rescinded the prescribing and dispensing restrictions on rosiglitazone (2Hiatt W.R. Kaul S. Smith R.J. The cardiovascular safety of diabetes drugs: Insights from the rosiglitazone experience.New Engl J Med. 2013; 369: 1285-1287Crossref PubMed Scopus (152) Google Scholar). Nonetheless, the FDA issued a guidance in 2008 mandating that all new antidiabetes drugs that treat type 2 diabetes must demonstrate that the therapy will not result in an unacceptable increase in cardiovascular risk (3Food and Drug Administration. Guidance for industry: diabetes mellitus: Evaluating cardiovascular risk in new antidiabetic therapies to treat type 2 diabetes. Silver Spring, 2008.Google Scholar). A meta-analysis of important prespecified cardiovascular events across phase 2 and phase 3 clinical trials is required to exclude an 80% increase in the relative risk involved with the new agent (that the upper boundary of the 2-sided 95% confidence interval for the estimated risk ratio is less than 1.8) prior to approval. In addition, postmarketing cardiovascular outcome safety trials that rule out a 30% increase are also generally required (3Food and Drug Administration. Guidance for industry: diabetes mellitus: Evaluating cardiovascular risk in new antidiabetic therapies to treat type 2 diabetes. Silver Spring, 2008.Google Scholar). Another example of safety concerns is the link between glargine, a synthetic basal insulin, and increased cancer risk based on preclinical animal data. This controversy was finally refuted by cardiovascular and other outcome data from a 6-year, large–scale, randomized clinical trial, the Outcome Reduction With Initial Glargine Intervention (ORIGIN) trial (4The ORIGIN Trial Investigators Basal insulin and cardiovascular and other outcomes in dysglycemia.N Engl J Med. 2012; 367: 319-328Crossref PubMed Scopus (1259) Google Scholar). Quantitative estimates of benefit and risk assessments have been developed over the years to provide a more objective evaluation. The number needed to treat (NNT) and the number needed to harm (NNH) are simple methods that are useful for assessing benefits and risks in a single clinical trial. The NNT is the number of patients who needed to be treated with the drug in order to achieve one or more occurrence of efficacious treatment of the disease targeted by the drug. The NNH refers to the number of patients who needed to be treated before 1 patient experiences an adverse side effect. The NNH:NNT ratio is another tool to measure the increase in the number of therapeutic successes vs. adverse side effects. However, these numerical tools are often viewed as somewhat simplistic and even misleading because they do not take into account the degree of benefits or the severity of the harm caused by the adverse side effects. They also do not take into consideration the quality-of-life issues. Other measures include quality-adjusted time without symptoms or toxicity and quality-adjusted life years gained by drug treatment. Multicriteria decision analysis is a tool to support decision making in which several benefits and risks are taken into account. It should be noted that drug-regulatory authorities rely solely on qualitative rather than quantitative assessments and expert opinions in their decision-making processes. How can we help the baffled healthcare professionals and consumers by providing them with unbiased and evidenced-based information about the various medications so that they can make the appropriate informed decisions? This is particularly relevant to those people who suffer from chronic diseases, such as diabetes and heart disease, who have to take not a single but multiple medications. I began searching for answers to this somewhat simplistic question and started with how government agencies across the world approve drugs or biological products that improve the health of their citizens. In order for a drug to be approved for marketing in a country, it must be proven safe and effective for its intended use. Although the definition of safe is not explicitly stated in the regulations or statues, it is recognized that all drugs have the potential of causing adverse effects. The safety of a drug is assessed on the basis that its benefits outweigh its risks. That said, none of the drug-regulatory agencies have uniform approaches to evaluating the benefits and risks of each compound submitted for approval. The FDA and the European Medicines Agency (EMA) are the 2 largest regulatory agencies in the world, and they set the standards to which most other countries look for guidance. The drug-approval process is complicated and often tedious, and for good reasons. A drug is approved only after careful scrutiny of preclinical, clinical and pharmacological and toxicity data by a panel of experts. In general, the FDA requires the submission of at least 2 pivotal phase-3, randomized, double-blinded clinical trials with positive results compared with placebo controls before a drug is approved. Both the FDA and the EMA publish the various relevant documents on their websites and take great care to explain the reasons behind regulatory decisions, but the clinical analyses may not always be readily understood or accepted by a broad audience, including the drug manufacturers, the healthcare professionals and the public. These concerns about the lack of transparency and clarity have led some authorities to lobby drug-regulatory agencies to base their decisions on formalized and quantitative approaches to benefit-risk assessment, including the assignment of weights to benefits and risks. Both the FDA and the EMA have undertaken initiatives to address these concerns but each takes a slightly different approach. In 2009 the FDA began to develop a framework for a structured approach for drug-benefit risk assessments that could serve as a template for product reviews as well as explain the bases of its regulatory decisions concerning drug approvals. The framework includes the categorization of several key decision factors: analysis of condition, current treatment options, benefit, risk and risk management. Within each of these factors, there are 2 considerations that inform the regulatory decision. The first consideration includes evidence as well as uncertainties and any assumptions that need to be made to address what is not known. The second consideration provides the conclusions and reasons explaining the bases for the regulatory decisions. The framework was piloted in 2012, and the 5-year plan for fiscal years 2013 to 2017 will include refinement and implementation (5Food and Drug Administration. Draft PDUFA V implementation plan: Structured approach to benefit-risk assessment in drug regulatory decision-making. Silver Spring, 2013.Google Scholar). Of interest to clinicians are 2 areas of focus by the FDA that characterize uncertainties in benefits and risks. The first is how well the benefit-risk assessment, based on premarket clinical trial data, translates into the postmarket setting after the drug has been approved and used in a much broader patient population. The second area pertains to the level of uncertainty about an adverse event that becomes known in the postmarketing period, when the basis for the observation comes from sources of varying levels of rigor. The EMA initiated a benefit-risk methodology project in 2009 to enhance the transparency of the benefit-risk decision-making process and to facilitate the communication of the rationale for each decision, both within the regulatory system and to the public (6European Medicines Agency. Benefit-risk methodology project: Report on risk perception study module, 2012.Google Scholar). A key question that the EMA wanted to address was whether European drug regulators are risk averse, risk neutral or risk seeking. They were particularly interested in whether the medical assessors’ belief systems and values might contribute to divergent opinions and ultimately lead to inconsistencies in the approval of drugs in the European Union. The study involved 80 assessors (physicians, PhDs and pharmacists) from 9 different countries in the European Union who participated in a web-based questionnaire to gain insights into risk attitudes and perceptions. The assessors were given a list of 28 medicinal products and asked to rate their benefits and risks using a 7-point scale. The findings of this study showed divergent responses by the assessors. For example, insulin was rated as being extremely beneficial by 90% of assessors, whereas 80% of them rated it as being slightly to moderately risky. The main conclusion of this study was that the assessors’ perceptions were moderated by group discussions, and the final outcome was likely to be a group effort rather than an individual effort. The benefit-risk methodology project was divided into 5 work packages, with the first 4 consisting of research and the fifth intended as implementation and training. A recommendation arising from the research phase of work package 4 was the development of a tool to guide assessors in understanding the risk attitude with regard to medicinal products (6European Medicines Agency. Benefit-risk methodology project: Report on risk perception study module, 2012.Google Scholar). The effects table was developed as a tool to summarize the key benefits and risks based on clinical trial data and the uncertainties that form the bases of the decisions (7European Medicines Agency. Benefit-risk methodology project: Update on work package 5: Effects table pilot (phase I), 2014.Google Scholar). It bears remarkable resemblance to the framework developed by the FDA, even though the 2 regulatory agencies approached the benefit-risk assessment by totally different paths. Health Canada, unlike the FDA and EMA, has not initiated its own benefit-risk assessment. Instead, it issued a Guidance Document in 2013, the Periodic Benefit-Risk Evaluation Report, which adopted the International Conference on Harmonisation guidance developed jointly by the regulatory bodies of the European Union, the United States and Japan (8Health Canada Health Products and Food Branch. Guidance document. Periodic benefit-risk evaluation report (PBRER), 2013.Google Scholar). Taken together, it appears that the time is ripe to have a roundtable discussion among the drug-regulatory agencies across the globe to develop a unified approach to assess the benefits and risks of drugs used to treat chronic diseases. Diabetes is a complex chronic disease, and most people affected by it will require medical therapy. Having a simple and more uniform tool to guide clinicians in choosing various drugs to treat diabetes and related complications will go a long way in simplifying the lives of primary care practitioners while improving the health of people affected by diabetes.