Objective: To evaluate the efficacy on physical functioning and safety of tailored exercise therapy in patients with knee osteoarthritis (OA) and comorbidities. Methods: In a randomized controlled trial, 126 participants were included with a clinical diagnosis of knee OA and at least 1 of the following target comorbidities: coronary disease, heart failure, type 2 diabetes mellitus, chronic obstructive pulmonary disease, or obesity (body mass index ≥30 kg/m2), with severity score ≥2 on the Cumulative Illness Rating Scale. The intervention group received a 20-week, individualized, comorbidity-adapted exercise program consisting of aerobic and strength training and training of daily activities. The control group received their current medical care for knee OA and were placed on a waiting list for exercise therapy. Primary outcome measures were the Western Ontario and McMaster Universities Osteoarthritis Index, subscale physical functioning (WOMAC-pf), and the 6-minute walk test (6MWT). Measurements were performed at baseline, after 20 weeks (directly posttreatment), and at 3 months posttreatment. Results: Statistically significant physical functioning differences over time were found between the intervention and control group (WOMAC: B = −7.43 [95% confidence interval (95% CI) −9.99, −4.87], P < 0.001; and 6MWT: B = 34.16 [95% CI 17.68, 50.64], P < 0.001) in favor of the intervention group. At 3 months followup, the mean improvements in the intervention group were 33% on the WOMAC scale and 15% on the 6MWT. These improvements are of clinical relevance. No serious adverse events occurred during the intervention. Conclusion: This is the first study showing that tailored exercise therapy is efficacious in improving physical functioning and safe in patients with knee OA and severe comorbidities.