Background: Soft tissue sarcoma (STS) is dependent on extracellular arginine as it often lacks expression of argininosuccinate synthase 1 (ASS1), the urea cycle enzyme needed to produce intracellular arginine. PEGylated arginine deiminase (ADI-PEG 20) is an extracellular arginine-degrading enzyme that causes ASS1 deficient tumors to enter the starvation state. Preclinical data demonstrated that addition of docetaxel (D) with ADI-PEG20 upregulates expression of the transporter for gemcitabine (G), overcoming transporter level resistance, and causing increased cell death. In vivo modeling demonstrated that the combination of ADI-PEG20 with G+D was superior to G+D alone. Therefore, we performed a phase 2 trial testing the addition of ADI-PEG20 to G+D. Methods: We performed an investigator-initiated, phase 2, multicenter, multi-arm clinical trial of ADI-PEG20 with G (90minute infusion)+D in STS, Ewing’s, osteosarcoma and small cell lung cancer. We are reporting Arm A, the STS arm. Eligible patients had STS that would be standardly treated with G+D that had progressed on at least one prior line of therapy with measurable disease by RECIST1.1 and had adequate organ function Based on a historic median PFS of 6.2 months for G+D, we targeted to enroll N = 75 patients in cohort A to detect a 2.8 month improvement with 80% power at a 5% alpha level. Primary endpoint was progression-free survival (PFS). Secondary endpoints included overall survival (OS), clinical benefit rate (CBR), safety, tolerability, cancer related mortality, and correlation with ASS1 expression by IHC. We evaluated PFS by Kaplan-Meier method and estimated overall response rate (ORR). Results: 75 patients were treated and deemed evaluable. The trial underwent two dose reductions by the data safety monitoring board due to prolonged neutropenia and thrombocytopenia preventing the use of day 8 G+D, consistent with preclinical mechanism of action data showing that ADI-PEG 20+D enhanced G uptake. Originally, the G dose was 900mg/m2 reduced first to 750mg/m2 then to 600mg/m2. D was dose reduced at the time of the second dose reduction from 75mg/m2 to 60mg/m2. ADI-PEG20 was given at a fixed intramuscular dose (36 mg/m2) weekly. The need for two dose reductions affected the PFS. The PFS/OS (months) were for the 600mg/m2 group (n = 31) was 6.0/N.D., leiomyosarcoma (LMS) (N = 33) 7.2/22.5, liposarcoma 5.1/17.4, and other (N = 36) 2.8/15.0. Responses were 8% complete (6/75) (3 LMS, 1 synovial and 2 angiosarcoma), 17% partial (13/75), and 43% stable disease (32/75), for an ORR of 25% (19/75) and CBR of 68% (51/75). There was a trend for ASS1 negative tumors to benefit more than ASS1 positive tumors. Conclusions: The combination of ADI-PEG20 with G+D can be safely and effectively given at a dose of 600mg/m2 G and 60mg/m2 D. Future randomized trials of ADI-PEG20 with G+D are planned. Clinical trial information: NCT03449901.