Aggressiveness features and outcomes of true interval cancers: comparison between screen-detected and symptom-detected cancers.

ABSTRACT

The question of whether screen detection confers an additional survival benefit in breast cancer is unclear and subject to several biases. Our aim was to examine the role of the diagnostic method (screen-detected, symptom-detected, and true interval cancers) and the clinical-pathological features in relapse-free survival and overall survival in breast cancer patients. We included 228 invasive breast cancers diagnosed in Barcelona from 1996 to 2008 among women aged 50-69 years. Ninety-seven patients were screen detected within the screening, 34 truly arose between 2-year screening mammograms (true interval cancers), and 97 were symptom detected outside the screening. The clinical-pathological features at diagnosis were compared. The overall and disease-free survival probabilities were computed using the Kaplan-Meier method. Cox proportional hazard models were applied, with adjustment by clinical-pathological variables. At diagnosis, symptom-detected and true interval cancers were in more advanced stages and were less differentiated. The highest proportion of triple-negative cancers was detected among true interval cancers (P=0.002). At 5 years of follow-up, the disease-free survival rates for screen-detected, true interval, and symptom-detected cancers were 87.5% (95% confidence interval, 80.5-95.2%), 64.1% (46.4-88.5%), and 79.4% (71.0-88.8%), respectively, and the overall survival rates were 94.5% (89.3-99.9%), 65.5% (47.1-91.2%), and 85.6% (78.3-93.6%), respectively. True interval cancers had the highest hazard ratio for relapse prediction (1.89; 0.67-5.31) and a hazard ratio of death of 5.55 (1.61-19.15) after adjustment for tumor-node-metastasis stage and phenotype. Clinically detected tumors, especially true interval cancers, more frequently showed biological features related to worse prognosis and were associated with poorer survival even after adjustment for clinical-pathological characteristics.