Background: Understanding the impact of sarcopenia on clinical outcomes in patients with metastatic cancer will assist clinicians with risk stratification, treatment decision-making, and inform the need for targeted supportive care strategies. Our objective was to comprehensively assess sarcopenia using measures of muscle mass (radiographically) and function (i.e., muscle strength and walking speed) and examine its impact on severe treatment toxicity, time to first emergency room (ER) visit, prostate-specific antigen (PSA) progression, radiographic progression, and overall mortality in men initiating androgen receptor-axis targeted therapy (ARAT) or chemotherapy for mCRPC. Methods: This was a secondary analysis of a prospective observational study of older men with mCRPC at the Princess Margaret Cancer Centre. Sarcopenia was defined as the combination of low muscle strength (grip strength < 35.5kg), slowness (walking speed < 0.8m/s), and low muscle quantity or quality prior to treatment initiation. The skeletal muscle index and skeletal muscle density were assessed through computed tomography scans prior to ARAT or chemotherapy initiation to determine muscle quantity and quality, respectively, using published cut-offs. Severe treatment toxicity, unplanned healthcare use, and disease progression were assessed from treatment initiation until treatment termination or loss to follow up. The associations between sarcopenia and severe treatment toxicity (i.e., grade 3+ toxicity) were assessed using multivariable logistic regression. Survival analyses were used to assess the impact of sarcopenia on the time to first ER visit, PSA progression, radiographic progression, and overall mortality. An interaction term for sarcopenia by treatment was introduced in all multivariable models and when necessary, an analysis by treatment was performed. Results: In total, 110 men participated, of whom 30 (27.3%) had sarcopenia prior to initiating treatment. Sarcopenia was associated with severe treatment toxicity (adjusted odds ratio (aOR) = 6.26, 95%CI = 1.17-33.58, P = 0.032) and time to first ER visit (adjusted hazard ratio (aHR) = 4.41, 95%CI = 1.26-15.43, p = 0.020) in the group that was initiating ARAT but not chemotherapy. Sarcopenia was a significant predictor of radiographic progression (aHR = 2.39, 95%CI = 1.06-5.36, p = 0.035) and overall mortality (aHR = 2.44, 95%CI = 1.17-5.08, p = 0.018) in the entire cohort. Conclusions: Sarcopenia may predict severe treatment toxicity and emergency room visits in men starting ARAT for mCRPC. Additionally, sarcopenia predicts radiographic progression and overall mortality regardless of treatment type in men with mCRPC. Confirmation is needed from large-scale studies. Assessment of sarcopenia can assist clinicians in treatment decision making while identifying high-risk patients that require targeted supportive care strategies.