Background: The consequences of a low prostate-specific antigen (PSA) in high-grade (Gleason 8-10) prostate cancer are unknown. We sought to evaluate the clinical implications and genomic features of this entity. Methods: Clinical and transcriptomic data from 626,057 patients with N0M0 prostate cancer were collected from two national cohorts and a large transcriptome database. Multivariable Fine-Gray and Cox regressions analyzed prostate-cancer specific mortality (PCSM) and all-cause mortality, respectively. GRID data were used to analyze transcriptomic features. Results: For Gleason 8-10 disease, the distribution of PCSM was U-shaped by PSA (PSA 4.1-10.0 ng/mL = referent), with adjusted hazard ratio (AHR) 2.70 for PSA ≤2.5 ng/mL (P < 0.001) versus 1.97, 1.36, and 2.56 for PSA 2.6-4.0, 10.1-20.0, and > 20.0 ng/mL, respectively. In contrast, distribution of PCSM by PSA was linear for Gleason ≤7 with AHR 0.41 for PSA ≤2.5 ng/mL (P = 0.127) versus 1.38, 2.28, and 4.61 for PSA 2.6-4.0, 10.1-20.0, and > 20.0 ng/mL, respectively (P_{Gleason*PSA interaction}< 0.001). Gleason 8-10, PSA ≤2.5 ng/mL disease had a significantly higher PCSM than standard high and very high-risk disease with PSA > 2.5 ng/mL (AHR 2.15, P = 0.009; 47-month PCSM 13.8% versus 4.9%). Among Gleason 8-10 patients treated with definitive radiotherapy, androgen deprivation therapy (ADT) was associated with a survival benefit for PSA > 2.5 ng/mL (AHR 0.87, P < 0.001) but not for ≤2.5ng/mL (AHR 1.36, P = 0.084; P_{ADT*PSA interaction}= 0.021). For Gleason 8-10 tumors, PSA ≤2.5 ng/mL was associated with a higher expression of neuroendocrine markers compared to > 2.5 ng/mL (P = 0.046), with no such relationship for Gleason ≤7. Conclusions: Low-PSA, high-grade prostate cancer appears to be a unique entity that has a very high risk for PCSM, potentially responds poorly to ADT, and is associated with neuroendocrine genomic features.