ABSTRACT
PURPOSE: To evaluate the hypothesis that telomere DNA content (TC) in breast tumor tissue correlates with TNM staging and prognosis. EXPERIMENTAL DESIGN: Slot blot assay was used to quantitate TC in 70 disease-free normal tissues from multiple organ sites, and two independent sets of breast tumors containing a total of 140 samples. Non-parametric Rank-Sums tests, logistic regression and Cox proportional hazards models were used to evaluate the relationships between TC and tumor size, nodal involvement, TNM stage, 5-year survival and disease-free interval. RESULTS: TC in 95% of normal tissues was 75-143% of that in the placental DNA standard, whereas only 50% of tumors had TC values in this range. TC was associated with tumor size (p=0.02), nodal involvement (p<0.0001), TNM stage (p=0.004), 5-year overall survival (p=0.0001) and 5-year disease-free survival (p=0.0004). A multivariable Cox model was developed using age at diagnosis, TNM stage and TC as independent predictors of breast cancer-free survival. Relative to the high TC group (>123% of standard), low TC (<101% of standard) conferred an adjusted relative hazard of 4.43 (95% CI 1.4-13.6, p=0.009). Receiver operating characteristic curves using thresholds defined by the TC distribution in normal tissues predicted 5-year breast cancer-free survival with 50% sensitivity and 95% specificity, and predicted death due to breast cancer with 75% sensitivity and 70% specificity. CONCLUSIONS: TC in breast cancer tissue is an independent predictor of clinical outcome and survival interval, and may discriminate by stage.
