Abstract Background: We are conducting an ongoing project using archival tissue samples from primary breast cancer cases treated with surgery alone at our institution, without any adjuvant systemic therapy. Tissue microarrays (TMAs) have been constructed, and various protein expressions are being re-evaluated using contemporary immunohistochemical techniques. This treatment-naïve cohort provides a unique opportunity to assess the intrinsic prognostic significance of biomarker expression, free from the confounding effects of systemic therapy. Androgen receptor (AR) is a steroid hormone receptor expressed in a subset of breast cancers and has attracted growing interest as a potential prognostic and therapeutic marker. Forkhead-box protein A1 (FOXA1) is a pioneer transcription factor known to act as a super-enhancer of the estrogen receptor (ER), playing a crucial role in mammary gland development. It is well established as a favorable prognostic marker in ER-positive, HER2-negative breast cancer. In recent years, however, FOXA1 expression has also been reported in other subtypes, including ER-negative tumors, although its clinical significance in these contexts remains uncertain. Whether FOXA1 functions primarily as a true prognostic marker or rather as a surrogate for endocrine responsiveness remains an open question. In this study, we report the results of immunohistochemical analyses for AR and FOXA1 in a treatment-naïve historical cohort, aiming to clarify their intrinsic prognostic relevance across breast cancer subtypes. Methods: We retrospectively analyzed primary breast cancer cases operated on at our institution between 1979 and 1982, from which TMAs were constructed by sampling three representative areas of the invasive tumor component. All pathological features, including immunohistochemical markers, were re-evaluated using current standardized protocols. Clinical outcomes were obtained from institutional records. Results: Among 935 eligible cases, 758 were available for AR and FOXA1 immunohistochemical evaluation. Subtype distribution based on ER and HER2 status was as follows: ER+/HER2− (n = 538, 71.3%), ER+/HER2+ (n = 47, 6.2%), ER−/HER2+ (n = 74, 9.8%), and ER−/HER2− (n = 96, 12.7%), comparable to current subtype frequencies. Using a 10% cutoff for positivity, AR expression was observed in 69.1% of ER+/HER2−, 57.5% of ER+/HER2+, 44.6% of ER−/HER2+, and 16.7% of ER−/HER2− tumors. In survival analysis, strong AR expression was significantly associated with improved distant recurrence-free survival (DRFS) (p = 0.0239). FOXA1 expression, using the same cutoff, was detected in 96.4% of ER+/HER2−, 100% of ER+/HER2+, 92.9% of ER−/HER2+, and 23.2% of ER−/HER2− tumors. Notably, FOXA1 was expressed in a subset of ER-negative tumors. FOXA1 expression was not significantly associated with DRFS, including in ER-positive tumors. In the ER−/HER2− subtype, although no significant difference in DRFS was observed based on Kaplan-Meier analysis, among patients who experienced recurrence, those with FOXA1-positive tumors had a significantly longer disease-free interval (DFI) (p = 0.0005). Conclusion: In this historical cohort of breast cancer patients untreated with adjuvant therapy, strong AR expression was significantly associated with improved DRFS. FOXA1 was expressed across all intrinsic subtypes and, contrary to previous reports, did not demonstrate prognostic significance in ER-positive tumors in the absence of systemic therapy. These findings suggest that FOXA1 may be more reflective of endocrine responsiveness rather than intrinsic tumor aggressiveness. The observation that FOXA1-positive tumors had a longer DFI suggests a potential role in late recurrence, particularly in ER-negative subtypes, and warrants further investigation. Citation Format: Y. Inoue, T. Osako, M. Akiya, T. Chiba, Y. Haruyama, U. Nakadaira, J. Masuda, Y. Kimura, A. Iesato, M. Nishimura, Y. Ozaki, T. Maeda, N. Uehiro, N. Yamashita, T. Kobayashi, T. Sakai, T. Takano, T. Ueno. Androgen receptor and FOXA1 in breast cancer: Prognostic insights from a treatment-naïve historical cohort abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-01-12.
Inoue et al. (Tue,) studied this question.