Patients' expectations and beliefs regarding the potential benefits and harms of medical interventions may induce placebo and nocebo effects, and affect the response to pain therapies. In a randomized clinical trial, we examined the effect of placebo and nocebo expectations on pain relief and adverse events (AEs) in association with a topical treatment among 65 cancer survivors experiencing chronic musculoskeletal pain. Participants received either a 1% camphor-based topical pain patch or a placebo treatment for 14 days. We measured pain severity with the worst pain item of the Brief Pain Inventory (BPI) at baseline and 14 days and treatment expectations at baseline with validated expectation questionnaires. We found that high vs. low nocebo expectations decreased pain severity improvements by 2.5 points (95% confidence interval [CI] −3.8 to −1.2; p<0.001) on a 0–10 numeric rating scale of the BPI and pain response rate by 42.7% (95% CI 0.2–0.6; p<0.001) at day 14, irrespective of placebo expectation status or treatment arms. Patients with high vs. low nocebo expectations in the true arm reported 22.4% more unwanted AEs. High nocebo expectations were associated with increased AEs by 39.5% (odds ratio: 12.0, 95% CI 1.2, 145.5; p=0.029) and decreased pain response in the true arm vs. placebo. Our study demonstrated that nocebo expectations, rather than placebo expectations, elevate the risk of AEs and compromise the effect of topical pain interventions. The findings raise the possibility that nocebo expectations may worsen somatic symptoms through heightening central pain amplification and should be further investigated.