One of the most important steps in developing a cocaine dependence treatment medication is conducting human clinical studies to establish a proposed medication's safety and efficacy. Three years ago, when NIDA and leading cocaine researchers first met to look at how clinical trials of potential cocaine treatment medications were being conducted, they found that differences in methodology and outcome measures made it difficult to assess and compare the results of different studies.
Clinical trials of cocaine treatment medications have come a long way since that initial workshop, Dr. Betty Tai of NIDA's Medications Development Division (MDD) told the most recent meeting of NIDA-funded cocaine researchers. NIDA held the workshop last fall to present an overview and update of issues critical to the success of clinical trials of potential cocaine treatment medications. The meeting was the culmination of a series of workshops NIDA's MDD has been holding with its treatment researchers, the Food and Drug Administration (FDA), and members of the pharmaceutical industry since 1992 to identify and resolve practical problems researchers have been confronting in conducting such clinical trials.
Through these meetings, workshop participants have identified issues that are critical to the design, implementation, analysis, and interpretation of the results of clinical trials of medications for cocaine abuse and dependence. The participants have also established and upgraded standards for conducting and evaluating these clinical trials and have developed clinical guidelines for deciding whether or not to take a proposed cocaine treatment medication further down the rigorous and costly path of safety and efficacy testing required for approval of a treatment medication by the FDA.
"Probably the most critical elements in clinical trials of cocaine treatment medications are outcome measures," says Dr. Tai. When MDD scientists reexamined data from early clinical trials, they found that individual investigators had used a potpourri of outcome measures to determine whether or not a medication worked, such as self-reported reductions in drug use or improvements in a patient's mental state or well-being. These researchers may have used many of these measures simply because an instrument was available to assess them, but they do not meet current standards for categorizing a drug as therapeutically effective, says Dr. Tai.
Through the clinical workshops, NIDA's MDD and the FDA have established four primary outcomes for assessing the safety and efficacy of a medication to treat cocaine abuse. Clinical trials of proposed cocaine treatment medications now measure patients' drug use by urinalysis or self-report, preferably both; how long patients remain in treatment; the physician's global assessment - a clinician's subjective assessment of how well a patient is progressing to a drug-free state compared to similar patients; and the patients' self-assessment of their own progress to a drug-free state.
The establishment of four core measures does not preclude individual researchers from looking at other outcome measures, such as the medication's effect on depression, that may be related to cocaine abuse, says Dr. Charles Grudzinskas, who directs NIDA's medications development program. However, the use of these four basic measures "will permit us to compare apples to apples, so that when we commission a meta-analysis a few years from now, people will be able to compare results using the same measures," he says. A meta-analysis is a statistical method used to summarize and describe the results of a number of studies.
Many of the workshops have also addressed the difficulties in assessing one of the core outcome measures - cocaine use. Early clinical trials often measured cocaine use with qualitative urinalysis. However, this method, which is used in workplace drug testing, only gives a positive or negative reading based on whether or not a preset amount of benzoylecgonine, a cocaine metabolite, is present in the urine when the test is performed. Because this method does not tell researchers the actual concentrations of benzoylecgonine in the urine, they cannot use it to determine if a medication is having an effect on the amount of cocaine use.
"Methods for conducting clinical trials of proposed cocaine medications have been getting much more sophisticated," says Dr. Tai. Now, almost every NIDA-funded trial is double-blind and placebo controlled. Researchers have also started using quantitative urinalysis to measure precise levels of cocaine in the urine. This method can give researchers a clearer picture of the effect of a medication on the extent of drug use, she says. At the workshop last fall, Dr. Kenzie Preston of NIDA's Division of Intramural Research showed how researchers can take that assessment one step further by using both quantitative urinalysis and patient self-reports of cocaine use to obtain a more accurate assessment of the timing, episodes, and amount of actual cocaine use than could be determined by either measure alone.
What NIDA has tried to do with this series of meetings, concludes Dr. Peter Bridge, who directs MDD's Clinical Trials Branch, is standardize clinical studies so that there is consistency across study sites.