When the drug trial fails: an approach to clinical drug studies.

Drug development involves the chemical identification and characterization of a compound to determine stability and define the drug's preliminary actions. Preclinical research follows in animals to develop a pharmacokinetic profile to determine dose range, biotransformation, elimination, and toxicology. The 4 phases of clinical research, phase 1 to phase 4, encompass a progressive investigation of healthy subjects, otherwise healthy patients, to patients with a target disease to obtain US Food and Drug Administration (FDA) approval. Clinical studies include open-label noncomparative studies during phases 1 and 2, and double-blind, comparative, and placebo-controlled studies during phases 2 and 3. Approval from the FDA follows the successful evaluation of the drug. After drug marketing, phase 4 clinical trials continue to collect safety and efficacy information. Many drugs that undergo this drug development process succeed in obtaining FDA approval and are marketed for clinical use. There are several circumstances, however, that preclude the successful completion of drug development, FDA approval, and marketing. This study describes a clinical trial of a new benzodiazepine, Ro 48-6791. Ro 48-6791 was being developed as an ultra-short-acting benzodiazepine with clinical effects of shorter duration than midazolam. The purpose of this study was to define a safe dose range for the induction and maintenance of conscious sedation of patients in an outpatient gastroenterology laboratory. Efficacy criteria to be evaluated included time to onset of action, duration of action, and psychomotor fitness upon recovery. Patients were assessed by using the Observer's Assessment of Alertness/Sedation score (OAA/S) and a 5-m heel-toe-line-walk test (HTLW). The patients were divided into 2 groups. Group 1 patients received Ro 48-6791. Group 2 patients were premedicated with meperidine before administration of Ro 48-6791. Ro 48-6791 was titrated over 30 seconds, and patients were observed for 90 seconds before the next dose was given. The OAA/S score, oxygen saturation, and vital signs were charted every minute through induction and every 5 minutes during the procedure. Patients received Ro 48-6791 until they reached on OAA/S score of 3, corresponding to slowed patient response to name calling. Group 1 (Ro 48-6791 alone) required greater induction doses and increased time to induction. Maintenance doses were the same for both groups. The duration of action of Ro 48-6791 as measured by the OAA/S score and HTLW test did not differ between groups. Ro 48-6791 seemed to be a safe and effective agent to achieve conscious sedation in outpatients undergoing short invasive procedures. However, clinical drug development of Ro 48-6791 was stopped because it did not meet the efficacy criteria of an ultra-short-acting benzodiazepine.