Impact on Survival of Early Versus Late Initiation of Adjuvant Chemotherapy After Pancreatic Adenocarcinoma Surgery: A Target Trial Emulation.

Abstract

We examined the impact of early (0-4weeks after discharge) versus late (> 4-8weeks after discharge) initiation of adjuvant chemotherapy on pancreatic adenocarcinoma survival. We used Danish population-based healthcare registries to emulate a hypothetical target trial using the clone-censor-weight approach. All eligible patients were cloned with one clone assigned to 'early initiation' and one clone assigned to 'late initiation'. Clones were censored when the assigned treatment was no longer compatible with the actual treatment. Informative censoring was addressed using inverse probability of censoring weighting. We included 1491 patients in a hypothetical target trial, of whom 32.3% initiated chemotherapy within 0-4weeks and 38.3% between > 4 and 8weeks after discharge for pancreatic adenocarcinoma surgery; 206 (13.8%) initiated chemotherapy after > 8weeks, and 232 (15.6%) did not initiate chemotherapy. Median overall survival was 30.4 and 29.9months in late and early initiators, respectively. The absolute differences in OS, comparing late with early initiators, were 3.2% (95% confidence interval [CI] - 1.5%, 7.9%), - 0.7% (95% CI - 7.2%, 5.8%), and 3.2% (95% CI - 2.8%, 9.3%) at 1, 3, and 5years, respectively. Late initiators had a higher increase in albumin levels as well as higher pretreatment albumin values. Postponement of adjuvant chemotherapy up to 8weeks after discharge from pancreatic adenocarcinoma surgery is safe and may allow more patients to receive adjuvant therapy due to better recovery.