First-Line Ribociclib Plus Endocrine Therapy Versus Chemotherapy With or Without Subsequent Endocrine Therapy in Patients With Rapidly Progressive HR-Positive/HER2-Negative Advanced Breast Cancer

Inclusion criteria for the prospective cohort

1. Female patient aged 18 years or older.
2. ECOG PS of 0-2.
3. Histologically or cytologically confirmed recurrent, metastatic, or unresectable locally advanced breast cancer not amenable to curative surgery or radiotherapy.
4. HR-positive/HER2-negative disease: ER expression in at least 10% of tumor-cell nuclei; HER2 IHC 0 or 1+, or IHC 2+ with negative FISH/ISH. When metastatic-tissue pathology is available, the metastatic result is preferred.
5. At least one feature of rapid disease progression, as determined by the investigator: symptomatic visceral metastasis; rapidly progressive disease or impending visceral compromise; or markedly symptomatic nonvisceral disease.
6. No prior systemic anticancer therapy for recurrent or metastatic disease. Prior neoadjuvant or adjuvant therapy is permitted.
7. At least one measurable lesion according to RECIST 1.1.
8. Postmenopausal, premenopausal, or perimenopausal status. Premenopausal or perimenopausal patients must agree to receive OFS.
9. Adequate baseline organ function: hemoglobin at least 90 g/L; white blood cell count at least 3.5 × 10\^9/L; absolute neutrophil count at least 1.5 × 10\^9/L; platelet count at least 100 × 10\^9/L; serum creatinine not above the institutional ULN; and clinically acceptable hepatic function.
10. Negative pregnancy test for patients of childbearing potential and agreement to use effective contraception according to applicable product information and institutional requirements.
11. Written informed consent and ability to comply with treatment and follow-up procedures.

Exclusion criteria for the prospective cohort

1. Prior systemic anticancer therapy for recurrent or metastatic disease.
2. Prior CDK4/6 inhibitor therapy in the neoadjuvant or adjuvant setting.
3. Symptomatic central nervous system metastasis requiring urgent local intervention. Treated, clinically stable, asymptomatic CNS metastasis may be permitted at investigator discretion.
4. Known contraindication or serious hypersensitivity to ribociclib or the selected endocrine agent.
5. Clinically significant uncontrolled cardiac disease, arrhythmia, congenital long-QT syndrome, uncorrected electrolyte abnormality, or baseline QTcF at or above 450 ms.
6. Severe or active cardiovascular, hepatic, respiratory, renal, hematologic, infectious, or psychiatric disease that may increase risk or interfere with efficacy assessment.
7. Inability to swallow oral medication or clinically significant gastrointestinal disease that may impair drug absorption.
8. Pregnancy or breastfeeding.
9. Clinical condition judged by the investigator to preclude safe systemic treatment.
10. Any other condition that, in the investigator's opinion, makes participation inappropriate.

Eligibility for the historical external control cohort

The external control cohort must satisfy the same core disease, treatment-line, biomarker, rapid-progression, and measurable-disease criteria as the prospective cohort. The following retrospective adaptations are permitted:

* ECOG PS may be taken from an explicit medical-record entry. If functional descriptions are mapped to ECOG PS according to a prespecified rule, the inferred status will be flagged and excluded in sensitivity analyses.
* Rapid disease progression will be determined from contemporaneous symptoms, laboratory findings, imaging descriptions, physician notes, and the treatment-decision context.
* Baseline laboratory values will be the closest available results within the prespecified window before the index chemotherapy date.
* At least one measurable lesion must be retrospectively assessable according to RECIST 1.1. Patients with unavailable or inadequate imaging will be excluded.
* Patients with critically missing baseline covariates, treatment details, tumor-response information, or follow-up outcomes will be excluded according to prespecified data-quality rules.