BACKGROUND: This study determines the prognostic value of the luminal-like subtype in patients with hormone receptor-positive breast cancer and explores whether the efficacy of extended anastrozole therapy differs between patients with luminal A-like versus luminal B-like tumours. MATERIALS AND METHODS: The phase III DATA study (NCT00301457) examined the efficacy of 6 versus 3 years of anastrozole in postmenopausal women with early-stage hormone receptor-positive breast cancer who had received 2-3 years of tamoxifen. Patients with available formalin-fixed paraffin-embedded tissue blocks were identified and classified by immunohistochemical luminal-like subtype. Distant recurrence (DR) and breast cancer-specific mortality (BCSM) were compared by luminal-like subtype and treatment arm using competing risk methods. RESULTS: This study included 788 patients: 491 had a luminal A-like tumour and 297 had a luminal B-like tumour. The median follow-up time was 13.1 years. Patients with luminal B-like tumours experienced a higher risk of DR [subdistribution hazard ratio (sHR) 1.44, 95% confidence interval (CI) 1.03-2.01, P = 0.03] and BCSM (sHR 1.68, 95% CI 1.15-2.45, P = 0.008) than patients with luminal A-like tumours. The efficacy of extended anastrozole therapy differed between patients with luminal A-like tumours (DR: sHR 0.51, 95% CI 0.30-0.88, P = 0.02
BCSM: sHR 0.39, 95% CI 0.19-0.82, P = 0.01) and patients with luminal B-like tumours (DR: sHR 2.09, 95% CI 0.96-4.53, P = 0.06
BCSM: sHR 2.36, 95% CI 0.80-7.00, P = 0.12) (P-interaction = 0.03 and P-interaction = 0.06, respectively). CONCLUSION: In patients with hormone receptor-positive breast cancer, the luminal B-like subtype was associated with a significantly worse prognosis when compared with the luminal A-like subtype. Extended anastrozole therapy halved the risk of DR and BCSM in patients with luminal A-like tumours, whereas no effect was seen in patients with luminal B-like tumours.