Cocaine use in patients receiving methadone is associated with worse treatment outcomes. The association between cocaine use and office-based buprenorphine/naloxone treatment outcomes is not known. We evaluated the association between baseline and in-treatment cocaine use, treatment retention, and urine toxicology results in 162 patients enrolled in a 24-week trial of primary care office-based buprenorphine/naloxone maintenance. Patients with baseline cocaine metabolite-negative urine toxicology tests compared with those with cocaine metabolite-positive tests had more mean weeks of treatment retention (18.3 vs. 15.8, p = .04), a greater percentage completed 24 weeks of treatment (50% vs. 33%, p = .04) and had a greater percentage of opioid-negative urines (47% vs. 34%, p = .02). Patients with in-treatment cocaine metabolite-negative urine toxicology tests compared with cocaine metabolite-positive patients had more mean weeks of treatment retention (19.0 vs. 16.5, p = .003), a greater percentage completed 24 weeks of treatment (60% vs. 30%, p < .001), and had a greater percentage of opioid-negative urines (51% vs. 35%, p = .001). We conclude that both baseline and in-treatment cocaine use is associated with worse treatment outcomes in patients receiving office-based buprenorphine/naloxone and may benefit from targeted interventions.