Abstract
Premenopausal breast cancer patients have an increased risk of distant recurrence, but their long-term risk remains unclear. Notably, over 90% of estrogen receptor (ER) positive tumors in premenopausal patients are also progesterone receptor (PR) positive, compared to 70% in postmenopausal patients. We aimed to determine whether PR intra-tumor heterogeneity influences long-term risk of distant recurrence and endocrine therapy benefit in premenopausal breast cancer patients. We conducted a secondary analysis of the Stockholm tamoxifen (STO-5) randomized controlled trial (1990-1997) with 20-year complete follow-up, including 924 premenopausal women with operable breast cancer in Stockholm, Sweden. Patients were randomized to 2 years of adjuvant endocrine therapy or control, with lymph node-positive patients receiving standard chemotherapy (CMF). Tumor blocks were available for 731 patients. PR intra-tumor heterogeneity was assessed by measuring variation in PR immunohistochemical staining intensity in whole tumor slides and was categorized as high or low for 520 ER-positive/PR-positive patients. Distant recurrence-free interval (DRFI) by PR heterogeneity was analyzed using Kaplan-Meier, multivariable Cox proportional-hazards regression, and multivariable time-varying flexible parametric modeling. We found PR intra-tumor heterogeneity significantly impacted 20-year DRFI (log-rank p = .002). Patients with high intra-tumor heterogeneity had a significantly increased risk of distant recurrence, compared to patients with low heterogeneity (hazard ratio [HR] = 1.42; 95% CI, 1.02-1.96). Similar results were observed in HER2-negative patients. Patients with high PR heterogeneity significantly benefited from endocrine therapy (HR = 0.41; 95% CI, 0.24-0.71). These findings suggest that premenopausal patients with high PR heterogeneity have increased long-term risk but significantly benefit from endocrine therapy.