Background: We previously reported the results of a Phase 2 study showing that a high proportion (59%) of men with PC being followed on AS will have a negative post-treatment biopsy after 90 days of apalutamide (Schweizer, et al. SUO Annual Meeting 2020). In order to identify candidate biomarkers for response, we conducted transcriptional profiling of tumor tissue obtained from men enrolled to the aforementioned trial. Methods: We analyzed FFPE tissue obtained from men enrolled to a Phase II study testing 90-days of apalutamide. Transcript profiles were assessed using Affymetrix Microarrays (Decipher Biosciences, Inc). Differences in signaling pathways were assessed between samples at baseline, day (D) 91 (post-treatment) and at D365. We also assessed differences in signaling pathways between patients that did vs. did not have a response (i.e. negative vs. persistent cancer on surveillance biopsy) at D91, which was the primary endpoint of the study. All comparisons were made using a Wilcoxon signed rank test unless otherwise indicated. Results: Samples from 22 subjects who completed 3-months of apalutamide and subsequently underwent post-treatment biopsy were available for analysis. From 19 Baseline and 15 post-treatment tissue samples, 25 passed pathology quality control (N = 12 at baseline, N = 8 at D91 and N = 5 at D365). Principal component analysis revealed distinct transcriptional profiles between tumor samples analyzed at baseline vs. D91. Surprisingly, D365 specimens still demonstrated a distinct profile compared to both baseline and D91 samples. Pathway analysis revealed up-regulation of angiogenesis signaling at D91 (P < 0.01) and D365 (P = 0.03) compared to baseline. As expected, estrogen (P < 0.01) and androgen receptor (P = 0.02) signaling were significantly lower at D91; however, only estrogen signaling was persistently suppressed at D365 (P = 0.03). Basal pathway signatures and markers associated with inflammatory response were also significantly upregulated at D91. There were no significant differences in Gleason grade group (GG) between responders and non-responders: 8/15 (53%) with GG1 vs. 5/7 (71%) with GG2 (Fisher’s exact P = 0.648). Decipher (P = 0.01) and Cuzick (P = 0.03) risk classifiers revealed an enrichment for responses in those with higher risk disease at baseline. There was also an enrichment for responses in those with higher nucleotide excision repair signature (P = 0.03) and those with signatures associated with TP53 mutations (P = 0.02). Conclusions: We observed significant transcriptional changes following 90 days of apalutamide, with evidence of persistent differences up to one year after enrollment. Higher baseline risk score was associated with improved responses to apalutamide treatment. Prospective studies evaluating the benefit of apalutamide in men on AS with higher risk transcriptional profiles are warranted. Clinical trial information: NCT02721979.