Potential for Abuse

With diseases of the central nervous system affecting an estimated 50 million Americans each year, the pharma industry is harnessing an explosion of new compounds in development in this area. However, the search for new and efficacious medications for conditions such as anxiety, depression, pain, and ADHD is tempered by the potential public health impacts of abuse liability and dependence potential.

Dr. Edward M. Sellers

Indeed, in 2002, more than six million people in the United States reported using psychotherapeutic drugs for "non-medical" purposes at some time in their lives, according to the Substance Abuse and Mental Health Services Administration's national survey on drug use and health. Forms of inappropriate use include taking a larger or more frequent dose than what was prescribed, while abuse involves non-therapeutic use with a risk of addiction, such as crushing the drug and inhaling or snorting it to get high.

Drug abuse risk has implications for clinical research studies, and for the post-marketing stage. At the study stage, regulatory authorities require that clinical research assess the potential for drug abuse or dependency. They seek a scientific basis on which to schedule drugs into different legal and medical classes. For example, earlier this year, federal regulators recommended that Pfizer's Lyrica (pregabalin), for persistent nerve pain, be scheduled as a controlled substance under the Controlled Substances Act (CSA).

Abuse liability assessment Many executives in the industry are not fully aware of FDA's regulations requiring abuse liability assessments, which evaluate the likelihood that a drug will be abused. FDA has made these assessments a mandatory part of the NDA submission process.

Detailed assessment of a drug's abuse liability includes information on the chemical similarity to known drugs of abuse, as well as information on the solubility, ease of synthesis, extractability, pharmaceutical properties, and dosage form of the brand. Companies must examine receptor-binding profiles, the impact of the drug on neurotransmitter release, and animal behavior. In addition, FDA requires analysis of behavioral pharmacology experiments, such as self-administration, drug discrimination, and drug withdrawal in rats and non-human primates.

Companies must also analyze all clinical trial data to interpret the potential for abuse. This includes adverse events (like euphoria) and reports of abuse, tampering, loss, misuse, discontinuation, and overdose. In specific human abuse liability studies, measures of subjective, performance, pharmacologic, and observer effects, and physical dependence, are also analyzed.

This information is used as part of a NDA to formulate a recommendation for scheduled or unscheduled status, and if scheduled, at what level. The schedules of the CSA affect controls on manufacturing, distribution, and prescription restrictions, and may affect the ease of patient access to a medication.

Risk management When there is potential for abuse, companies should prepare a Risk Management Action Plan (RiskMAP). In that plan, sponsors and regulators must be sensitive to balancing the legitimate needs of the patients who require the medication against the potential for abuse. In other words, companies should ensure that drugs are available to patients that require them, but minimize inadvertent exposure. A comprehensive RiskMAP includes action items for both the intended population (patients) and the unintended population (abusers). For example, in response to reports of abuse of OxyContin (oxycodone), Purdue Pharma implemented a RiskMAP that strengthened the warning label, implemented a physician education program, and developed a tracking system to identify and monitor abuse.

Preparation is key The best defense is good offense. Often, abuse and dependence potential is an afterthought in the development process. By instituting a company-coordinated effort—with a team mandated to oversee and direct this work—companies can avoid costly delays and surprises at the regulatory level.

No single test offers a complete characterization of a drug. Investigators must remain flexible in their designs, and reach out to consultants who can help them perform studies with the appropriate methods, data collection, and interpretation.

Dr. Edward M. Sellers, president and CEO, Ventana Clinical Research