Objective To determine whether patients with Gleason score 5 + 3 = 8 prostate cancer have outcomes more similar to other patients with Gleason score 8 disease or to patients with Gleason score 9 disease. Patients and Methods The Surveillance, Epidemiology and End Results (SEER) database was used to study 40 533 men diagnosed with N0M0 Gleason score 8 or 9 prostate cancer from 2004 to 2011. Using Gleason score 4 + 4 = 8 as the referent, Fine and Gray competing risks regression analyses modelled the association between Gleason score and prostate cancer‐specific mortality (PCSM). Results The 5‐year PCSM rates for patients with Gleason score 4 + 4 = 8, 3 + 5 = 8, 5 + 3 = 8, and 9 disease were 6.3%, 6.6%, 13.5%, and 13.9%, respectively (P < 0.001). Patients with Gleason score 5 + 3 = 8 or 9 disease had up to a two‐fold increased risk of PCSM (adjusted hazard ratio AHR 1.89, 95% confidence interval CI 1.50–2.38, P < 0.001; and AHR 2.17, 95% CI 1.99–2.36, P < 0.001, respectively) compared with the referent group of patients (Gleason score 4 + 4 = 8). There was no difference in PCSM between patients with Gleason score 5 + 3 = 8 vs 9 disease (P = 0.25). Conclusions Gleason score 8 disease represents a heterogeneous entity with PCSM outcomes distinguishable by the primary Gleason pattern. The PCSM of Gleason score 3 + 5 = 8 and Gleason 4 + 4 = 8 disease are similar, but patients with Gleason score 5 + 3 = 8 have a risk of PCSM that is twice as high as other patients with Gleason score 8 disease and should be considered to have a similar poor prognosis as patients with Gleason score 9 disease. Such patients should be allowed onto trials seeking the highest‐risk patients in which to test novel aggressive treatment strategies.