Early indicators of accrual success: Time-to-first-patient and accrual performance at an anticipated enrollment milestone-A study of NCI-CTEP-sponsored clinical trials.

Abstract

6001 Background: This research investigates whether indicators of time from trial opening to first enrollment and accrual performance may predict ultimate successful trial accrual. Methods: National Cancer Institute Cancer Therapy Evaluation Program (NCI-CTEP) Sponsored trials for all therapeutic, nonpediatric phase I, I/II, II, and III trials activated and closed to accrual from 2000 to 2007 were investigated (n = 764). Two indicators were evaluated. The first indicator was months from opening of the trial to first patient enrolled. The second indicator was the accrual performance, computed as the percentage of actual accrual compared to the minimum accrual goal, evaluated at the anticipated time to achieve the minimum goal. For example, if a trial had a minimum accrual goal of 150 and was anticipated to accrue 15 pts/month, the anticipated time is 10 months. Multivariable logistic regression analysis was used to calculate unadjusted and adjusted odds ratios, controlling for study size, cancer incidence, and cancer mortality. Results: Trials that enrolled the first patient beyond the first two months of study activation were statistically significantly less likely to achieve the minimum accrual goal regardless of the time they remain open (odds ratio:0.616; 95% CI:0.464-0.875). Almost half of all trials (n = 379, 49.6%) enroll the first patient outside this window. Overall, 18.5% (n = 141) of CTEP-approved oncology trials achieve their minimum accrual goal within their anticipated time. Of the trials that remained open after this time (n = 603, 78.9%), trials that did not enroll a least 60% of their minimum goal within the anticipated period were statistically significantly less likely to achieve this minimum accrual goal by study closure, regardless of the time to closure (odds ratio:0.190; 95% CI: 0.055- 0.652). Conclusions: Early indicators in accrual performance can be used to identify trials that are highly likely to close without achieving their intended accrual goal. Time to first patient and accrual performance at anticipated trial enrollment time appear to be two such predictors. No significant financial relationships to disclose.