The 21-gene Oncotype DX Recurrence Score (RS) is widely used to guide adjuvant chemotherapy decisions in hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) breast cancer. However, the clinical interpretation of RS remains challenging in premenopausal women with node-positive disease, where discordance between genomic and clinicopathological risk is frequently encountered. This study aimed to elucidate the determinants of clinical–genomic risk discordance and to clarify its impact on real-world treatment selection, with a specific focus on premenopausal node-positive patients.

We retrospectively analyzed 290 patients with HR+/HER2- breast cancer who underwent curative surgery and Oncotype DX testing between September 2023 and November 2025. Clinical risk was defined according to the St. Gallen criteria, and concordance or discordance was determined based on agreement or disagreement between clinical risk and RS. Logistic regression analyses were performed to identify predictors of discordance in the overall cohort and in node-positive patients. Treatment allocation was further examined in a predefined subgroup of 28 premenopausal patients with node-positive disease. Treatment strategies were categorized as chemotherapy plus endocrine therapy, endocrine therapy combined with systemic intensification (abemaciclib or S-1), or endocrine therapy alone.

Clinical–genomic risk discordance was observed in 123 patients (42.4%). In multivariable analysis of the overall cohort, larger tumor size (OR 8.33) and nodal positivity (OR 6.43) were independently associated with discordance, indicating a strong association with tumor burden. In contrast, among node-positive patients, premenopausal status (OR 0.18, p = 0.045) and high proliferative activity (Ki-67 > 20%; OR 4.81, p = 0.048) were independently associated with discordance, indicating a distinct biological pattern associated with proliferative activity in premenopausal disease. In the premenopausal node-positive subgroup, all concordant cases received intravenous chemotherapy, whereas discordant cases showed marked heterogeneity in treatment selection: chemotherapy in 2 patients (8.0%), endocrine therapy combined with targeted or fluoropyrimidine-based intensification strategies (including abemaciclib or S-1) in 11 patients (44.0%), and endocrine therapy alone in 12 patients (p < 0.001).

Clinical–genomic risk discordance is frequent in premenopausal women with node-positive hormone receptor-positive, HER2-negative breast cancer and is associated with distinct biological characteristics, including high proliferative activity. In real-world clinical practice, discordance was linked to substantial variation in adjuvant treatment selection, with many patients managed using endocrine-based strategies. These findings highlight the clinical relevance of integrating genomic and clinicopathological information in treatment decision-making and support the role of genomic assays as complementary tools for individualized therapy in this challenging population.