Background: ASCOLT is an international Phase III randomized control trial (NCT00565708) investigating the impact of adjuvant aspirin on recurrence and survival after curative resection of colorectal cancer (CRC). Retrospective data suggest that COX2 and HLA Class 1 antigen expression may predict for aspirin benefit (Reimers MS, JAMA internal medicine, 2014;174(5):732-739; and Chan AT, JAMA, 2009;302(6):649-658.) Translational studies aim to confirm putative biomarkers of aspirin sensitivity in this prospective trial. This study reports the IHC assessment of COX2 and HLA in representative samples. Methods: FFPE slides from consenting Australian patients were stained with COX2 and HLA Class 1 antibodies (Chan AT, JAMA, 2009;302(6):649-658.). Digital images were captured using Aperio software. IHC expression was scored by a pathologist (DW) blinded to sample identification. Intensity was assessed semi-quantitatively for each antibody in one area of dominant (primary) intensity and one of secondary intensity in each section. A scale was applied for staining intensity: 0=absent, 1=weak, 2=moderate and 3=strong; and for amount of tumor epithelium stained: 1= 5%-24%, 2= 25%-49%, 3=50%-74% 4= >75%. Normal epithelium, which typically has strong COX2 and HLA staining, was used as internal control. Only tumor epithelium was scored (not inflammatory cells). Results: 90 tumors were stained for both COX2 and HLA Class 1. All tumors showed diffuse staining for both markers, mostly strong, but with many tumors having regions with lesser degrees of expression (Table). Conclusions: COX2 expression was generally strong and HLA Class 1 was more likely to show regions of weak to moderate expression. No significant correlation was detected between COX2 and HLA expression levels. Further pathological analysis is underway and correlations will be made with outcome once trial data has matured in 2020. Clinical trial information: NCT00565708