Purpose: Investigational cancer therapies may be available outside trials as "off-protocol therapy" (OPRx), with implications for patient safety, trial accrual, and access to care. We conducted a literature-based analysis of recent randomized trials to evaluate the potential scope and impact of OPRx in the United States.
Methods: A MEDLINE search identified all English-language phase III medical oncology randomized clinical trials (RCTs) published over a 2-year period ending April 17, 2008. Determination of OPRx availability was based on US Food and Drug Administration approval for any indication. We limited assessment of accrual to studies with US sites. Data from articles were extracted independently by two investigators.
Results: Among 172 eligible RCTs, the majority (108; 63%) evaluated drugs that were available OPRx in the United States at trial initiation, while an additional 19 (11%) evaluated interventions that became available during the trial. Among trials with US sites, time to accrual was slower (41 vs 22 months; P = .002) and less efficient (8.8 v 22.7 patients per month; P = .001) when OPRx was available. Sixty-six percent of RCTs reported at least one increased grade 3 to 4 toxicity in the experimental arm, 47% reported superior efficacy for at least one major clinical outcome in the experimental arm, and 27% reported improvement in overall survival. These outcomes did not vary on the basis of OPRx availability.
Conclusion: The majority of recent oncology RCTs involve experimental interventions that are available outside trials in the United States with potential impact on trial accrual. The safety and efficacy of novel interventions must be determined by clinical trials.