Purpose <p>Based on SOFT and TEXT trials data, a composite recurrence risk score (CR-score) model was developed for early premenopausal women with hormone receptor -positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) breast cancer using the subgroup treatment effect map (STEPP) method to guide ovarian function suppression (OFS) application. However, the CR-score model has yet to be validate in real-world settings.</p> Methods <p>Our study included patients diagnosed between January 1, 2013, and December 31, 2021, from 42 breast centers in China. We utilized restricted cubic splines (RCS) to visualize continuous CR-score and hazard ratios for breast cancer recurrence. After adjusting for confounding factors via propensity score matching (PSM), Kaplan-Meier curves were used to compare disease-free survival (DFS) among premenopausal patients between the OFS and non-OFS groups.</p> Results <p>The hazard ratio of recurrence consistently increased with higher CR-scores. Notably, 87.68% of patients who received OFS had a CR-score above 1.42. Following PSM, adjuvant OFS significantly improved DFS in the high CR-score group (CR-score above 1.42)(HR 0.571; 95% CI 0.403–0.809; <i>p</i> = 0.001). Among patients younger than 35 years old, those receiving OFS had significantly better DFS compared to those without OFS. After matching for age, grade, ER, PR, and lymph node status, OFS can significantly improve the DFS of those chemotherapy-treated patients with CR-score above 1.42 (<i>p</i> = 0.006). Furthermore, the group with high CR-score but ER expression below 50% did not benefit from OFS.</p> Conclusion <p>The CR-score model can effectively guide clinicians in making decisions regarding OFS for premenopausal patients with HR+/HER2- breast cancer.</p>