Gemcitabine and nab-paclitaxel in older adults with metastatic pancreatic cancer: Are two doses per cycle enough?
655 Background: The dosing of Gemcitabine and Nab-Paclitaxel (GA), a frontline regimen to treat metastatic pancreatic cancer (mPC), is frequently altered from the traditional dosing schedule (TDS) of day 1, 8, and 15 of a 28 day cycle to a modified dosing schedule (MDS) of 2 doses/cycle. Previous work showed that overall survival (OS) was similar between patients (pts) treated with the MDS vs the TDS. We sought to analyze a larger real-world database to assess these trends. Methods: We retrospectively analyzed de-identified pts with mPC ≥ 65 y/o treated with GA in the Flatiron Health nationwide EHR-derived database. Demographics, treatments (tx), and outcomes were collected. Pts were grouped as either starting with the TDS or MDS. Analysis included time on treatment (TOT) as well as OS. A Cox model was used to test non-inferiority of the MDS vs the TDS for both TOT and OS, adjusting performance status, age, race, gender, and line of therapy (LOT). The upper bound for non-inferiority was a Hazard Ratio (HR) = 1.2. Results: 1497 pts were treated between 1/1/14-5/31/19; 883 pts with the TDS and 614 with the MDS. Median TDS age was 72 (65-85) and MDS was 73 (65-84) (p<0.001). 1237 pts received first- line GA; 60% received the TDS, 40% the MDS. The use of the TDS vs MDS did not vary significantly by LOT, gender, or race, but more pts with a PS of ≥2 received the MDS (p=0.03). In the first-line, outcomes were better for the TDS vs the MDS (unadjusted median TOT 5.3 vs 3.2 mo, p<0.001, OS 9.2 vs 5.3 mo; p<0.001), with consistent results in the ≥ second-line. The MDS did not meet its non-inferiority boundary: first-line TOT HR=1.4 [95% CI 1.2-1.6]; second+ line TOT HR=1.3 [95% CI 1.0-1.7]; first-line OS HR=1.6 [95% CI 1.4-1.8]; second+ line OS HR=1.3 [95% CI 1.0-1.8]. Results were consistent when additionally stratified by PS 0-1 vs 2+. Conclusions: In this large real-world cohort, first-line GA tx with a MDS did not meet criteria for non-inferiority for TOT and OS vs a TDS in older adults with mPC. With the caveats of potential confounding that exist in a de-identified retrospective database, these results suggest that dose intensity may be important in pts with mPC. Further prospective studies are necessary to ensure we utilize effective tx strategies in older adults with mPC.