How to choose a time zero for patients in external control arms

Abstract

When a sponsor carries out a single-arm trial of a novel oncology compound, it may wish to assess the efficacy of the compound via comparison of overall survival to an external control arm, constructed using patients included in some retrospective registry. If efficacy of the novel compound is compared to efficacy of physician’s choice of chemotherapy, patients in the retrospective registry might qualify for inclusion in the external control arm at multiple different points in time, when they receive different chemotherapy treatments. For example, a patient might qualify at the start of their second, third and fourth lines of therapy. From the start of which line of therapy should this patient’s survival be compared to survival of participants in the single-arm trial?Some sponsors have elected to include patients in the external control arm from the last available line of therapy in the retrospective database. Another possibility is to randomly select a line of therapy for each external control arm patient from among those available. In this paper, we show, via probabilistic arguments and also via simulation based on real data, that both of these methods give rise to a bias in favor of the single-arm trial. We further show that this bias can be avoided by instead including external control arm patients multiple times in the external control arm, once for each time they receive qualifying treatment.

Comparative effectiveness of ciltacabtagene autoleucel in CARTITUDE‐1 versus physician's choice of therapy in the Flatiron Health multiple myeloma cohort registry for the treatment of patients with relapsed or refractory multiple myeloma

“…To account for this difference, patients in the RW cohort who received multiple subsequent therapies after meeting eligibility criteria contributed multiple observations (corresponding to all eligible LOTs) to the current analysis, provided they met eligibility criteria at the beginning of each LOT. The use of all eligible LOTs is supported by previous literature, which found that this approach provides the most statistical efficiency compared to including only the first or last eligible LOT [10, 11]. An exploratory analysis using only the first eligible LOT for patients in the RW cohort was conducted to assess potential differences in results from the two approaches.…”

Comparative Efficacy of Ciltacabtagene Autoleucel in CARTITUDE-1 vs Physician’s Choice of Therapy in the Long-Term Follow-Up of POLLUX, CASTOR, and EQUULEUS Clinical Trials for the Treatment of Patients with Relapsed or Refractory Multiple Myeloma

“…To account for this difference, if a patient in the physician's choice cohort received multiple LOTs after becoming eligible, that patient contributed multiple observations (corresponding to all eligible LOTs) to the present analysis, provided that the patient was eligible at the start of each LOT. Prior studies have shown this approach to be associated with improved statistical efficiency compared with restricting inclusion to only a single eligible LOT (e.g., first or last) [17,18]. An exploratory analysis using only the first eligible LOT received by patients in the physician's choice cohort was conducted to assess potential differences in the two approaches.…”

Natural History of Human Epidermal Growth Factor Receptor 2–Amplified and Human Epidermal Growth Factor Receptor 2 Wild-Type Refractory Metastatic Colorectal Cancer in US Clinical Practice

“…This study evaluated records of initiation of a treatment-refractory LOT as the primary unit of observation to minimize bias. 14 As such, a patient could contribute multiple records with corresponding index dates (thus, multiple index dates aligned with multiple LOTs). Where required, standard errors and confidence intervals were calculated using bootstrapping to account for correlations within subject observations.…”

“…This study evaluated records of initiation of a treatment-refractory LOT as the primary unit of observation to minimize bias. 14 As such, a patient could contribute multiple records with corresponding index dates (thus, multiple index dates aligned with multiple LOTs).…”