Prolonged time to treatment initiation in advanced pancreatic cancer patients has no major effect on treatment outcome: a retrospective cohort study controlled for lead time bias and waiting time paradox.

Abstract

A prolonged time to treatment initiation (TTI) correlates with an adverse prognosis in different cancer types including resectable pancreatic cancer (PC). Only limited evidence on the correlation between TTI and prognosis in advanced PC exists. Consecutive PC patients (n = 368) who were diagnosed or treated at our high-volume comprehensive cancer center were included in a prospectively maintained database. We retrospectively analyzed time from first imaging showing advanced PC to initiation of palliative first-line chemotherapy. Lead time bias and waiting time paradox were addressed by landmark analysis and correlation of tumor burden with TTI. Two hundred and ninety-seven patients met the pre-specified in- and exclusion criteria of our study. Median TTI was 29days (range: 1-124days). Most common reasons for prolonged TTI (> 21days) were referral from an external treatment center (39%) and a second biopsy (31%). A TTI above the median-, 75th or 90th percentile (43 or 60days, respectively) had no impact on overall survival. Furthermore, no correlation between levels of carbohydrate antigen 19-9 (CA 19-9) at time of treatment initiation and TTI was observed. While a timely work-up of advanced PC patients remains important, delays in treatment initiation due to repeated biopsies, inclusion in a clinical study or transfer to a specialized cancer center appear to be justified in light of the absence of a strong adverse effect of prolonged TTI on prognosis in advanced PC patients.