Background: Poor outcome in pancreatic ductal adenocarcinoma (PDA) is associated partly with stromal hyaluronan (HA) accumulation, which compromises chemotherapy perfusion. PEGPH20 (PEGylated recombinant human hyaluronidase) potentiates chemotherapy by degrading HA. An ongoing, phase II, randomized study is comparing PEGPH20 + nab-paclitaxel (Nab) + Gemcitabine (Gem) (PAG) vs Nab+Gem (AG) in untreated Stage IV PDA. Interim data were presented at ASCO 2015 using a prototype assay to determine the HA-high subpopulation. Final efficacy data are now available with a newly developed Ventana HA companion diagnostic assay and scoring algorithm, which identifies HA-high (staining in the extracellular matrix [ECM] ≥ 50% of tumor surface) pts. Methods: Pts received PEGPH20 3 µg/kg twice weekly (C1), then weekly (C2+) with standard AG dosing. The protocol was amended after a temporary clinical hold (Apr-Jul 2014) for an imbalance in thromboembolic (TE) events to exclude high-TE-risk pts and add enoxaparin (LMWH) prophylaxis to both treatment arms. Primary endpoints are PFS and TE events. PFS and ORR data through Dec 2014 are for pts enrolled up to clinical hold (Stage 1); TE data through Dec 2015 (Stage 2) are outlined below. Results: 135 pts were treated (74 PAG, 61 AG). PFS results are shown below (median follow-up 7 mo). In HA-high pts receiving PAG vs AG, ORR was 55% (1 CR) vs 33%. TE events (PAG vs AG) were: Stage 1 (no LMWH) 42% vs 25%; Stage 2 (+LMWH 1 mg/kg/d) 7% vs 4%; overall (+LMWH 40 mg/d or 1 mg/kg/d) 12% vs 9%. No Grade 5 events have occurred with 1 mg/kg/day LMWH. Conclusions: Pts with HA-high tumors receiving PAG vs AG showed clinically meaningful improvements in PFS and ORR. TE events reduced with LMWH. A global phase III trial of PAG is scheduled to initiate in Q1 2016. Clinical Trial Information: NCT01839487.