Background: Overcoming ARSI resistance is an unmet need for mCRPC patients. Known resistance mechanisms include genetic alterations augmenting androgen receptor (AR) function (e.g., amplifications of the AR gene and its enhancer and AR mutations) and treatment-emergent evolution from an AR-driven to an AR-indifferent phenotype such as small cell neuroendocrine prostate cancer. Other mechanisms remain unexplored. Methods: To phenotype ARSI-resistant prostate cancer, we compared the RNA-seq profiles of paired metastatic biopsies from 31 mCRPC patients obtained before the initiation of first-line abiraterone or enzalutamide, and upon radiographic progression. The patient samples were obtained from two independent cohorts, the Stand Up To Cancer/Prostate Cancer Foundation (SU2C/PCF) West Coast Prostate Cancer Dream Team (WCDT) and a multi-institutional cohort from the Netherlands sequenced through the Hartwig Medical Foundation. We first evaluated RNA-seq data to identify the most differentially expressed genes and then performed survival analysis in the retrospective WCDT cohort (n=115, including 54 ARSI-naïve and 61 ARSI-experienced tumors) to assess the association between gene expression and overall survival, defined as the time interval from biopsy to death from any cause. Results: In the analysis of paired RNA-seq data from 31 mCRPC patients, SSTR1 was identified as the most significantly altered gene; SSTR1 expression was consistently decreased in post-treatment, ARSI-resistant tumors (log2[fold change]=-2.5, p=1.08×10^-16). SSTR1 encodes somatostatin receptor 1, a G-protein coupled receptor that, when activated by the endogenous peptide hormone somatostatin or exogenous somatostatin analogues, mediates anti-proliferative, anti-migratory, and anti-secretory effects on target cells. Retrospective analysis in the 115 WCDT patients demonstrated that low SSTR1 expression was associated with worse overall survival (p=0.047); ARSI treatment improved survival for patients with SSTR1-high (≥ median) tumors (HR=0.29, p=0.0002) more significantly than those with SSTR1-low (< median) tumors (HR=0.67, p=0.17), with an interaction (SSTR1 expression * ARSI treatment) towards statistical significance (p=0.0598). A differential survival benefit of ARSI therapy with respect to SSTR1 expression was observed only among the 61 ARSI-experienced patients. Conclusions:SSTR1 expression is a potential predictive biomarker for response to ARSI in mCRPC, with higher SSTR1 expression predicting better response. This observation could be explained by an anti-tumor effect of SSTR1 signaling. Further work is warranted to prospectively validate these findings and to investigate SSTR1 as an actionable drug target to overcome ARSI resistance.