Background: GA evaluates aging-related domains (e.g., function) known to be associated with cancer treatment toxicity. In this CRCT, we evaluated if providing a GA summary with management recommendations to oncologists can reduce toxicity in older patients (pts) with advanced cancer receiving chemotherapy and/or other agents with a high reported prevalence of grade 3-5 toxicity. Methods: Pts aged > 70 with incurable solid tumors or lymphoma and > 1 impaired GA domain starting a new treatment regimen were enrolled. Community oncology practices were randomized to intervention (oncologists received GA summary/recommendations for impairments) or usual care (none given). The primary outcome was proportion of pts who experienced any grade 3-5 toxicity (CTCAE v.4) within 3 months. Practice staff prospectively captured toxicities; blinded oncology clinicians reviewed medical records to verify. Secondary outcomes included 6 month overall survival (OS) and treatment intensity (standard vs reduced). Outcomes were analyzed using generalized linear mixed/Cox models with Arm as a fixed effect, controlling for practice. Results: From 2013-19, 718 pts were enrolled from 41 practices. Age (mean 77 yrs), sex (43% women), number of impaired GA domains (median 4/8), and treatment type (chemotherapy 88%) were not different by Arm. More pts in intervention were Black (12% vs 3%, p<0.01), had GI cancer (38% vs 31%, p<0.01), and had prior chemotherapy (31% vs 23%, p=0.02). Pts in intervention experienced a lower proportion of grade 3-5 toxicity (175/349; 50%) than pts in usual care (262/369; 71%). The relative risk (RR: intervention vs usual care) of grade 3-5 toxicity was 0.74 (95% CI: 0.63-0.87; p=0.0002); the difference was mostly driven by non-heme toxicities (RR 0.73; 95% CI: 0.53-1.0, p<0.05). OS was not significantly different (71% vs 74%, p=0.3). More pts in intervention received reduced intensity treatment at cycle 1 (49% vs 35%, RR 0.81, p=0.01). Dose modifications due to toxicity were lower in intervention (42% vs 58%, p<0.0001), but results were not significant after controlling for practice (RR 0.85; 95% CI: 0.67-1.08, p=0.2). Conclusions: Providing GA information to oncologists reduces the proportion of older pts who experience grade 3-5 toxicity from high-risk palliative cancer treatment, without compromising OS. Reduced treatment intensity at cycle 1 may explain these results. Funding: R01CA177592, U01CA233167, UG1CA189961. Clinical trial information: NCT02054741.