Cargando…
Estrogen Receptor Status Predicts Late-Onset Skeletal Recurrence in Breast Cancer Patients
Estrogen receptor-positive (ER+) breast cancer (BCa) often recurs after long latency, and is known to favor bone as a metastatic site. We hypothesized that skeletal recurrence of ER+ BCa follows a different chronological pattern from that of nonskeletal recurrence. We retrospectively evaluated 434 m...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Wolters Kluwer Health
2016
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4779030/ https://www.ncbi.nlm.nih.gov/pubmed/26937933 http://dx.doi.org/10.1097/MD.0000000000002909 |
Sumario: | Estrogen receptor-positive (ER+) breast cancer (BCa) often recurs after long latency, and is known to favor bone as a metastatic site. We hypothesized that skeletal recurrence of ER+ BCa follows a different chronological pattern from that of nonskeletal recurrence. We retrospectively evaluated 434 matched pairs of ER+ and ER− female patients who underwent surgery for clinically localized BCa between 2005 and 2009. Patient age, tumor size, lymph node involvement, and adjuvant treatment biases were adjusted by the propensity score method. We conducted competing risk analysis to determine the prognostic significance of ER expression status on the risk of overall recurrence and late recurrence (after 3 years). We also compared chronological patterns of ER+ and ER− tumor recurrence, stratified by the first metastatic site (skeletal vs nonskeletal). After 3 postoperative years, ER+ tumor had a significantly higher risk of overall distant recurrence than ER− tumor (P = 0.02). When further stratified by first site of metastasis, only late skeletal recurrence was significantly associated with ER status (P = 0.029). In multivariate analysis, ER and lymph node involvement status were significant prognostic factors for late skeletal recurrence, with adjusted hazard ratios of 5.2 (95% CI = 1.2–22.4, P = 0.025) and 5.2 (1.7–16.3, P = 0.005), respectively. For nonskeletal distant recurrence, tumor size (>2 cm) was the only significant risk factor with adjusted hazard ratio of 2.8 (1.4–5.7, P = 0.005). Annual hazard of skeletal recurrence events of ER+ tumors continued to exist up to 10 years, while annual hazard of nonskeletal recurrences decreased after peaking at 5 years. ER− tumor recurrences exhibited similar annual hazard patterns across skeletal and nonskeletal sites. ER expression and lymph node involvement status were strong predictors of BCa late-onset (>3 years) recurrences, especially in skeletal sites. Therefore, skeletal system surveillance is mandatory for long-term follow-up of this subpopulation. |
---|