Study of cohort-specific consent and patient control in phase I cancer trials.

PURPOSE: To address the challenging ethical dilemmas created from the participation of advanced cancer patients in phase I trials, we assessed the feasibility of a clinical trial design that uses an interactive informed consent process in which patient-subjects can choose to become directly involved in decisions of dose escalation. PATIENTS AND METHODS: Subjects were advanced cancer patients in the Hematology/Oncology Clinics at the University of Chicago who were eligible to participate in a phase I trial in which they underwent a three-step informed consent process that used cohort-specific consent and allowed them the option to choose their own doses of the chemotherapeutic agents under study, vinorelbine (NVB) and paclitaxel (TAX), within predetermined limits. NVB and TAX were administered in conventional 21- to 28-day cycles for two cycles while on study. Dose escalation occurred when a patient-subject chose a higher untested dose after they received information on all previously assessable patient-subjects. In addition to the phase I trial itself, a survey that consisted of structured interviews, which sought to evaluate patients' experiences with the interactive subject-choice phase I trial design and consent process, was conducted with participating subjects. The phase I trial itself sought to determine the associated toxicities of the agents under study. The survey results were compared with a similar survey of a matched control population of subjects who participated in other concurrently active conventional phase I trials at our institution.
RESULTS: Twenty-nine patient-subjects participated in the phase I trial, with 24 who agreed to and completed the survey interviews. Seventy-six percent of patient-subjects opted to choose their dose of the agents under study, and 28% chose the highest available doses. More than half of the patient-subjects (56%) felt some degree of comfort in being asked to choose their dose of chemotherapy, with 53% stating that being asked to choose their dose made them feel in control, fully informed, or content. However, there were no statistically significant improvements in objective measures of the informed consent process, which included surveyed subjects' stated understanding of either provided information about phase I trials and alternatives to trial participation or of the research purpose of phase I trials. By making choices, the group of patients in the interactive subject choice trial changed the size of the dose cohorts and modified the process of dose escalation in this phase I study.
CONCLUSION: Although complex, our innovative phase I trial design is feasible. In addition to the use of cohort-specific consent, the trial design may reduce the magnitude of many of the commonly recognized ethical dilemmas associated with this form of clinical research, which include difficulties with information provision and the understanding of possible risks and benefits of phase I trial participation, through direct subject involvement in research decision making by otherwise potentially vulnerable cancer patients.