Trophoblast cell-surface antigen 2 (TROP2), a transmembrane receptor expressed in many carcinomas, is a target for novel antibody-drug conjugates such as sacituzumab govitecan. TROP2-targeted therapy is used for unresectable locally advanced or metastatic triple-negative and hormone receptor-positive, HER2-negative breast cancers. The role of TROP2 as a predictive marker is yet unclear. Standardized interpretation criteria for TROP2 immunohistochemistry (IHC) are lacking. Here, we compared three antibody clones and two methods for semi-quantitative assessment, aiming to establish reproducible evaluation criteria. First, TROP2 IHC was performed on normal tissues and nine breast cancers, using the BSB-148, EPR20043 and SP293 clones. EPR20043 was selected for subsequent evaluation in 69 breast cancers without pathological complete response to neoadjuvant chemotherapy (NAC). Four pathologists applied the ASCO/CAP guidelines for HER2 IHC testing (designated as the 'membrane score') and the H-score. All H-scores were categorized as low (0-100), intermediate (101-200) and high (201-300). Although the membrane scores strongly correlated with the categorized H-scores, the latter showed higher interobserver variability. Next, TROP2 IHC was performed on 94 breast cancer metastases and evaluated by six pathologists, confirming the strong correlation between the membrane scores and H-scores. In metastases, the interobserver variability was similar for both methods. Our observations support the application of the HER2 ASCO/CAP guidelines for semi-quantitative evaluation of membranous TROP2 protein expression, as this method strongly correlates with the H-score and is less prone to interobserver variability in post-NAC breast resections. Future studies should investigate the association between the TROP2 membrane score and response to TROP2-targeted therapy.