Breast cancer subtypes are routinely described as molecularly distinct but the unifying organising principle — why these specific six subtypes exist, what single biological axis orders them, and what specific molecular lock maintains each subtype — has not been stated in any published framework. This document presents the OrganismCore six lock type classification of breast cancer, derived from Waddington landscape attractor geometry applied to 19,542 single cancer cells across six subtypes (GSE176078, Wu et al. Nature Genetics 2021) before literature review. Six distinct lock types are identified: (1) Luminal A — CDK4/6 cycle lock (Type 1, slope arrest). Therapeutic first step: CDK4/6 inhibitor + endocrine therapy. Novel: CDKN1A (p21) as quantitative CDK4/6i benefit predictor. (2) Luminal B — DNMT3A/HDAC2 chromatin lock suppressing ER output (Type 1, epigenetic brake). DNMT3A/HDAC2 co-expression coupling r=+0.267 vs r=+0.071 in LumA (p=5.68e-56). Therapeutic first step: entinostat (HDACi) then endocrine therapy. Novel: TFF1/ESR1 decoupling as patient selector for entinostat benefit (NCT07235618 active 2026). (3) HER2-enriched — ERBB2 amplicon kinase override (Type 1, kinase arrest). Therapeutic first step: anti-HER2. Novel: EZH2i + anti-HER2 for CDH3-high, AR-low, EZH2+118% deep fraction. (4) TNBC/Basal-like — EZH2/PRC2 H3K27me3 epigenetic lock silencing FOXA1/GATA3/ESR1 (composite Type 1→2). EZH2 elevated +189% above normal luminal reference. Therapeutic sequence: tazemetostat → fulvestrant (CS-LIT-16, Zenodo 2026-03-06) or tazemetostat maintenance post-chemotherapy (CS-LIT-17, Zenodo 2026-03-06). (5) Claudin-low — stem root lock with no residual luminal programme (Type 4). Not the same as TNBC: luminal identity was never written, not silenced. Tazemetostat does not apply. Therapeutic approach: anti-TIGIT + anti-PD-1 in memory-low subgroup. (6) ILC — CDH1 structural lock, the geometric inverse of TNBC (Type 3). FOXA1 hyperactivated. Fulvestrant superior to AI by FOXA1 IHC stratification. All six subtypes are ordered by the FOXA1/EZH2 ratio, validated across ~7,500 patients in seven independent datasets on four platforms (companion Zenodo deposit 2026-03-06). The classification provides a mechanistic explanation for known clinical paradoxes: why entinostat fails in unselected ER+ but works in LumB; why TNBC relapses late despite pCR; why anti-TIGIT fails in unselected TNBC; why AI is inferior to fulvestrant in FOXA1-high ILC. No competing attractor geometry classification of breast cancer currently exists in the published literature. Full computational record: https://github.com/Eric-Robert-Lawson/attractor-oncology (Repository ID: 1172048282)
Eric Robert Lawson (Fri,) studied this question.