All Times EDT
Keywords: Interim analysis, multiple endpoints, oncology, non-inferiority trials, sample size
In oncology phase 3 trials, non-inferiority concepts (e.g., shortening duration of chemotherapy) have sometimes been evaluated without preliminary knowledge of an experimental treatment being non-inferior to the standard treatment, and thus, a prompt consideration for futility stop is vital. Such concepts are usually examined in a patient population with a good prognosis, in a phase 3 trial setting single primary endpoint (e.g., overall survival or disease-free survival), and thus, it is often the case that the number of events is quite small at earlier interim analyses, which would miss an opportunity for a harmful treatment to be recommended a futility stop. To address these issues, the incorporation of an aggressive futility monitoring rule using an earlier-available intermediate endpoint is attractive. When the primary endpoint is overall survival, typical intermediate endpoints include progression-free survival, disease-free survival, and so on. Past studies have shown from simulation and case studies that the monitoring rule accelerated the timing of futility recommendation (Goldman et al., 2008;Wang et al., 2012). However, they missed a crucial issue of using an early aggressive rule in that the study power is sometimes dropped to a nonnegligible level due to its aggressiveness.
In the present study, we numerically assess the power reduction of when an aggressive inefficacy monitoring based on an earlier-available intermediate endpoint is used. We then propose a novel sample size determination method to achieve power at a predetermined level (e.g., 80-90%). Numerical and simulation studies showed that our proposed method successfully attain power with an appropriate recommendation of futility at earlier interim analyses. This paper motivates an ongoing phase 3 non-inferiority trial evaluating the value of breaking immune-checkpoint inhibitor agents (e.g., pembrolizumab and nivolumab) for non-small cell lung cancer patients.