Background: nab-P + G showed promising efficacy in a phase I/II study in MPC, and decreases in CA19-9 correlated with OS. In MPACT, patients (pts) who received nab-P + G vs G had improved median OS (8.5 vs 6.7 mo; HR 0.72; p = 0.000015), PFS (5.5 vs 3.7 mo; HR 0.69; p = 0.000024) and ORR (23% vs 7%; p = 1.1 × 10⁻¹⁰). Here we present a prespecified exploratory analysis of CA19-9 from the MPACT trial. Methods: 861 previously untreated pts with MPC were randomized 1:1 to receive nab-P 125 mg/m² + G 1000 mg/m² days 1, 8, and 15 every 4 weeks or G alone 1000 mg/m² weekly for 7 weeks followed by a week of rest (cycle 1) and then days 1, 8, and 15 every 4 weeks (cycle ≥ 2). CA19-9 was evaluated at baseline and then every 8 weeks. OS comparisons at different CA19-9 criteria were performed by stratified Cox proportional hazards model (P by stratified log-rank test using randomization criteria). Results: 750 pts had an evaluable CA19-9 at baseline. More pts in the nab-P + G arm vs the G arm demonstrated a best CA19-9 decrease from baseline of ≥ 20% and ≥ 90% (61% vs 44% and 31% vs 14%, respectively; Table). At the first postbaseline assessment (week 8), greater proportions of pts in the nab-P + G arm vs the G arm had CA19-9 decreases of ≥ 20% and ≥ 90% (Table). At that time point, for pts with a decrease of ≥ 20% in CA19-9, nab-P + G demonstrated a significantly longer OS vs G. The risk reduction for pts with a ≥ 90% decrease was greater than in pts with a ≥ 20% decrease. In pts with an 8-week CA19-9 decrease < 20%, median OS for nab-P + G vs G was 8.3 vs 8.0 mo (HR 0.92; p = 0.705). The relationship of CA19-9 kinetics with OS will also be examined. Conclusions: Higher proportions of pts in the nab-P + G arm had CA 19-9 responses of ≥ 20% and ≥ 90% vs the G arm. Pts who achieved a CA19-9 decrease at 8 weeks of ≥ 20% or ≥ 90% had significantly longer OS with nab-P + G than with G. Clinical trial information: NCT00844649.