Glioblastoma (GB) remains incurable despite aggressive chemotherapy, radiotherapy, and surgical interventions; immunotherapies remain experimental in clinical practice. Relevant preclinical models that can accurately predict tumor response to therapy are equally challenging. This study aimed to validate the effect of the naturally occurring agent diallyl trisulfide (DATS) in human GB in relevant pre-clinical models. Ex vivo slice culture, in vivo cell line derived orthotopic xenograft and patient-derived orthotopic xenograft (PDX) animal models of GB were utilized to assess efficacy of treatment with DATS. Our results showed 72-h treatments of 25µM DATS induced cell death in ex vivo human GB slice culture. We treated U87MG orthotopic xenograft models (U87MGOX) and patient-derived orthotopic xenograft models (PDX) with daily intraperitoneal injections of DATS for 14 days. Magnetic resonance (MR) imaging of mice treated with DATS (10mg/kg) demonstrated reduced tumor size at 5 weeks when compared with saline-treated U87MGOX and PDX controls. Hematoxylin (H&E) staining demonstrated dose-dependent reduction in gross tumor volume with decreased proliferation and decreased angiogenesis. Western blotting showed that DATS was associated with increases in histone acetylation (Ac-Histone H3/H4) and activated caspase-3 in this novel preclinical model. Histological assessment and enzyme assays showed that even the highest dose of DATS did not negatively impact hepatic function. DATS may be an effective and well-tolerated therapeutic agent in preventing tumor progression and inducing apoptosis in human GB.