Abstract The application of positron emission tomography (PET)‐computed tomography (CT) in treatment response evaluation has increased in diffuse large B‐cell lymphoma (DLBCL), although its predictive value is controversial. We retrospectively analyzed the rate of false‐positive PET‐CTs performed as interim (n = 94) and end‐of‐treatment (n = 8) assessments among 102 DLBCL patients treated during 2010–2017 at Oulu University Hospital. In PET‐CT Deauville score ≥4 was regarded as positive. A biopsy was performed on 35 patients, and vital lymphoma tissue was detected from nine patients. Positive biopsy findings were associated with poor disease outcomes in this study. This difference was statistically significant: 2‐year failure‐free survival (FFS) was 44% in patients with a positive biopsy versus 83% for those with a negative biopsy (p = 0.003). The corresponding overall survival (OS) rates were 53% versus 95% (p = 0.010). In the multivariate analyses, a negative biopsy was an independent protective factor in FFS (Hazard Ratio (HR) 0.093 (95% confidence interval [CI] 0.017–0.511); p = 0.006) unrelated to the International Prognostic Index (IPI) (HR 1.139 [95% CI 0.237–5.474] p = 0.871) or stage (HR 1.365 [95% CI 0.138–13.470]; p = 0.790). There was no statistically significant difference in OS according to the PET results, but the FFS rate was significantly higher in patients with a negative PET. The value of PET‐CT as an evaluation method suffers from a high false‐positive rate, and it is inadequate alone for the justification of treatment decisions. Biopsy results provide more reliable prognostic information for the evaluation of treatment response and outcome and should be used to assess patients with positive PET‐CT scans.