Drug Safety and Shortages Challenge Manufacturers

A prime objective of FDA policies is to ensure the safety of the nation's drug supply, a task that entails calculating risks and preventing harm. The recent emergence of serious side effects associated with Roche/Genentech's Avastin (bevacizumab) prompted FDA to re-evaluate the risks and benefits associated with this cancer therapy and ultimately to recommend rescinding the drug's indication to treat breast cancer. That decision generated an outcry from patients who believe the therapy is saving lives.

Jill Wechsler

Similar situations for patients and physicians arise from inadequate supplies of crucial drugs due to manufacturing difficulties, supply-chain problems, and regulatory actions. More effective adverse-event reports and databases may help patients avoid unsafe therapies in the future. Methods for accessing and evaluating this type of information are advancing rapidly, but require considerable research on appropriate assessment techniques and communications strategies. In the future, such data analysis may help healthcare providers and payers assess comparative treatments and quality of care, and also support clinical trials for new medical therapies.

Early warnings

Faster information about drug-safety problems is the goal of FDA's Sentinel surveillance system for detecting adverse-drug events. As required by the FDA Amendments Act of 2007 (FDAAA), Sentinel augments FDA's passive adverse-event reporting system by accessing electronic drug-use information held independently by private health plans and government health programs [see the March 2010 Washington Report in Pharmaceutical Technology]. While this idea sounds fairly straightforward, it's a complex undertaking that involves linking diverse databases and establishing standard analytical methods that can detect and assess resulting safety signals.

As a prelude to a fully operational real-time safety monitoring system, FDA has established a Mini-Sentinel pilot program to meet the FDAAA requirement to access data on at least 100 million individuals by July 1, 2012. The agency exceeded its first mandated goal for Sentinel in July 2010 by tapping into drug-use information on more than 25 million patients. These accomplishments and how leaders of the Sentinel Initiative have laid the groundwork for the system during the past two years were discussed at a January workshop organized by the Brookings Institute and at the Drug Information Association's January 2011 pharmacovigilance conference.

Now the program is moving forward with plans to test scientific methods for further investigating specific drug use issues, explained Judy Racoosin, scientific lead for Sentinel in FDA's Center for Drug Evaluation and Research (CDER). A Coordinating Center operated by Richard Platt of the Harvard Pilgrim Health Care Institute has been established to send queries to databases operated by Kaiser Permanente, Aetna, Partners Healthcare, and Humana, among others, and then evaluate resulting responses. FDA also has formed a drug surveillance collaboration with other federal agencies, including the Centers for Medicare and Medicaid Services, the Veterans Administration, and the Department of Defense, to query these health databanks on safety issues of mutual interest.

The project has established basic principles and policies, including a Common Data Model and core safety surveillance methods. Researchers have examined which regression methods are applicable for sequential surveillance programs and have identified Health Outcomes of Interest that deserve evaluation. Because the Sentinel Initiative involves public health activities, as opposed to research, the query and analysis process does not require institutional review board approval to proceed.

A key feature of this distributed data system is that patient health information remains with the data holders, an approach designed to ensure the privacy and security of personal health information. While the system currently taps into administrative and claims data, a goal is to add clinical data in the future, at least for some patient cohorts. For now, data holders may be asked to provide certain patient-level information if needed to followup on unclear findings, but that data would be stripped of direct patient identifiers before transmission.

This year, the Mini-Sentinel pilot is testing several protocols to see how well the system can evaluate emerging safety concerns. One initiative will assess whether certain postmarketing regulatory initiatives, such as risk evaluation and mitigation strategies (REMS) can enhance the safe use of medicines. Another project is monitoring adverse events associated with four relatively new, but routinely administered vaccines. An extensive study examining adverse cardiological events in users of diabetes drugs will compare the incidence of myocardial infarction in patients prescribed saxagliptin (Bristol-Myers Squibb's Onglyza) to those taking other diabetes treatments to demonstrate how well Sentinel can monitor drug safety in large populations.

These studies can use some of the findings of the Observational Medical Outcomes Partnership (OMOP), a collaboration between industry, FDA, and the Foundation for the NIH. OMOP is conducting research on how well different methods and data sources can identify valid safety signals and whether meaningful information can be gleaned from disparate observational data sources that use a wide variety of terms. The program has identified important technical, organizational, and resource capabilities of a successful active surveillance site and ways to assess the strength of associations between drug exposure and specific adverse events, such as liver failure or bleeding. OMOP's success in identifying methods and tools through an open, collaborative process has prompted the participants to extend the partnership beyond its initial two-year timeframe and to expand from drug safety issues to methods that could apply to medical devices and biologics, plus approaches for comparative effectiveness resarch (CER) and healthcare quality measures.

Communications challenges

As Sentinel moves forward, manufacturers and other parties are examining how to issue early drug-safety signals without leading to public misinformation or alarm. The pilot study on diabetes treatments and cardiac events, for example, raises questions about when and how FDA will make any findings public, noted GlaxoSmithKline general counsel Daniel Troy at the Sentinel workshop. Even if FDA says it won't release early results prior to full analysis, outside lawyers and policymakers may try to compel disclosure of such information.

In Washington this month

CDER Associate Director for Medical Policy Rachel Behrman acknowledged that liability is on the agency's mind because it could drive companies away from the Sentinel project. Unrealistic expectations about what an active drug surveillance system can do raises liability concerns for pharmaceutical manufacturers as well as for data providers that could be accused of failing to warn the public about emerging safety issues, Troy added. FDA needs to make it clear to the public that Sentinel signals are based on observational data that often is not sufficient to make informed judgments about medical safety issues.

National resource

FDA is not the only government agency seeking access to health-plan data banks for secondary purposes, but its collaborative experience in establishing Sentinel may influence the shape of similar initiatives, Behrman suggested. It was not clear initially that Sentinel should be established as a public–private partnership, she said, but if "FDA had done it alone, in a silo, we would not be where we are today."

A new initiative for the US Department of Health and Human Services (HHS) is to build a multipayer claims database (MPCD) to support CER, as authorized and funded by the economic stimulus legislation of 2009. To meet this requirement, HHS Assistant Secretary for Policy and Evaluation Sherry Glied is establishing a centralized component that will hold de-identified claims and administrative data on some 100 million individuals in private health plans and Medicare. In addition, the project will build a distributed network component, similar to Sentinel.

Glied also is working with several states that have established their own MPCDs, primarily as cost-control strategies. About a dozen state initiatives from Maine to Utah are analyzing eligibility and claims data to track use and quality of healthcare services and products, while also looking for fraud, waste, and trends relevant to continuity of care and case management.

Ultimately, a national interoperable electronic medical-records system will feed patient-level information into all these data systems. With appropriate privacy protections, this arrangement will facilitate assessments of healthcare delivery and medical care for many purposes. The relatively early experience of FDA and its partners in establishing a system that can answer common drug-safety questions with rigor and credibility provides a "national resource for evidence development," said Mark McClellan, former FDA commissioner and current director of the Engelberg Center for Health Care Reform at the Brookings Institute. Demonstrating ways to link and analyze different data sources to develop evidence on medical product performance should be useful in drug development, comparative research, and assessing the quality of care delivered by healthcare organizations and providers, McClellan explained.

At the Sentinel workshop, Glied emphasized the need to coordinate the HHS MCPD project with Sentinel, and CDER Director Janet Woodcock similarly suggested that it's important to avoid "reinventing the wheel" with each initiative. Woodcock acknowledged that Sentinel is part of a broader effort to better use data to improve healthcare, and FDA, she said, "is willing to be a node" on these larger e-health information undertakings.

UnitedHealth Group Medical Affairs Chief Reed Tuckson echoed these remarks, pointing out that health plans are receiving hundreds of requests for data from federal, state, intrastate and demonstration projects, and that standards and efficient governance systems are needed for these projects to work. Tuckson agreed that these data initiatives can help improve healthcare and product safety, but "data is expensive," he said, and meeting all the requests costs a lot of money.

At a symposium sponsored by OMOP, Woodcock emphasized the importance of building a solid methodological foundation for interpreting results. The hypothesis is that one can identify characteristics in drugs that provide causal processes for adverse events, Woodcock observed. Better information on drug characteristics and health outcomes and a solid methodological foundation for interpreting results, she said, is critical for reimbursement purposes, for medical practice, and for regulatory decision-making.

Scarce supplies

Although early information about adverse drug events can curb prescribing of unsafe drugs, early warnings also can help prevent serious shortages of important medicines, which have been on the rise in recent years. The University of Utah Drug Information Service recorded 211 drug shortages in 2010, up from 166 in 2009 and 70 in 2006. The most dire shortages are affecting anesthetics and treatments for cancer, pain, and serious infections. FDA notes that sterile injectable drugs accounted for almost half of all drug shortages in 2009 because fewer companies are able to make these more complex products, and any interruptions in production lines can affect multiple products and cause lengthy production delays.

The situation has led to complaints from hospitals and physicians about having to use less familiar and sometimes inappropriate alternative products that can cause dosing errors and compromise patient safety. The Institute for Safe Medication Practices (ISMP) reported an increase in medication mishaps related to switches to alternate medications to replace unavailable products. Changes in opioid pain killers, for example, generated confusion about correct dosage, as has the need to use sedation agents with various strengths and patient monitoring protocols.

The issue has made front-page news as state correction departments seek out substitutes for sodium thiopental, the drug commonly used for lethal injections. It no longer is made in the US, following the exit of Hospira (Lake Forest, IL) from the market in January. States have run into trouble trying to import the drug from the United Kingdom and other countries that oppose capital punishment, and Ohio recently faced demands by Lundbeck, the maker of the barbiturate pentobarbital, not to use its product for lethal injection. Further complicating the issue, lawyers for a group of prisoners on death row have filed suit against FDA for permitting the import of thiopental supplies and similar products that are registered overseas, but not by FDA safe and effective.

The shortages stem from a number of factors. Many of these products are low-margin generic drugs that are not sufficiently profitable to support a production upgrade when FDA inspectors uncover manufacturing violations. Quality problems with the anesthetic propofol in 2009 prompted recalls by Hospira and Teva Pharmaceuticals that created severe shortages. And after receiving multiple warning letters from FDA, Teva closed its Irvine, California, manufacturing facility for injectibles, and it seems far from being able to reopen it.

Pharmacists and physicians discussed remedies to the shortage crisis at a meeting last November organized by ISMP and the American Society of Health-System Pharmacists. The group's report acknowledges that manufacturers generally run production lines at full capacity, making it difficult to respond quickly to increased market demand. Limited sources of active ingredients can cause disruptions, while just-in-time inventory practices make it hard for manufacturers to deal with sudden shortages.

Some established sterile products are not as profitable to companies as new therapies and may be dropped, especially when continued production requires an overhaul of facilities. This observation appears particularly true for those decades-old products that have been marketed without formal FDA approval but have been the target of an enforcement crackdown in recent years. While a rise in FDA inspections of injectable drug manufacturing processes has created problems for several companies, no one wants contaminated products on the market.

Although many of these situations are unavoidable, health professionals believe that more early warnings from manufacturers about emerging shortages will help them deal with serious short supply situations. Pharmacists and physicians want Congress to require companies to notify FDA earlier and more quickly about looming supply problems for a broader range of "medically necessary" products, including drugs with a single active ingredient source. Such a proposal has been introduced into the Senate, and further remedies may be offered.

In addition to waving the stick, FDA could encourage the production of short-supply drugs by speeding through new applications and offering tax credits to manufacturers that expand production or upgrade manufacturing facilities for needed products. A process for extending stability for products in short supply also might help. Without some way to lessen shortages, manufacturers will appear to be putting profits above patients and leaving the public to cope with ineffective, and even unsafe, alternative therapies.

Additional information on the Sentinel Initiative is available at www.fda.gov/Safety/FDAsSentinelIntiative/ucm2007250.htm.

Jill Wechsler is Pharmaceutical Technology's Washington editor, 7715 Rocton Ave., Chevy Chase, MD 20815, tel. 301.656.4634, jwechsler@advanstar.com.