Aims Oncotype DX recurrence score (RS) is a clinically validated assay, which predicts the likelihood of disease recurrence in oestrogen receptor‐positive/HER2‐negative (ER+/HER2−) breast cancer (BC). In this study we aimed to compare the performance of Oncotype DX against the conventional clinicopathological parameters using a large BC cohort diagnosed in a single institution. Methods and results A cohort (n = 430) of ER+/HER2− BC patients who were diagnosed at the Nottingham University Hospitals NHS Trust and had Oncotype DX testing was included. Correlation with the clinicopathological and other biomarkers, including the proliferation index, was analysed. The median Oncotype DX RS was 17.5 (range = 0–69). There was a significant association between high RS and grade 3 tumours. No grade 1 BC or grade 2 tumours with mitosis score 1 showed high RS. Low RS was significantly associated with special tumour types where none of the patients with classical lobular or tubular carcinomas had a high RS. There was an inverse association between RS and levels of ER and progesterone receptor (PR) expression and a positive linear correlation with Ki67 labelling index. Notably, six patients who developed recurrence had an intermediate RS; however, four of these six cases (67%) were identified as high‐risk disease when the conventional clinical and molecular parameters were considered. Conclusion Oncotype DX RS is correlated strongly with the conventional clinicopathological parameters in BC. Some tumour features such as tumour grade, type, PR status and Ki67 index can be used as surrogate markers in certain scenarios. Based on our findings, there are some patients who are low risk, such as patients with low tumour grade, special tumour types (classical lobular carcinoma, tubular, or invasive cribriform), low Ki67 <10% with high progesterone receptor (PR) >10%. Those patients should be treated as low‐risk patients.