The aim of this study was to determine whether the excess risk of axillary lymph node metastases (N+) differs between interval breast cancers arising shortly after a negative mammography and those presenting later. In a registry-based series of pT1a–pT3 breast carcinoma patients aged 50–74 years from the Italian screening programmes, the odds ratio (OR) for interval cancers (n = 791) versus the screen-detected (SD) cancers (n = 1211) having N+ was modelled using forward stepwise logistic regression analysis. The interscreening interval was divided into 1–12, 13–18, and 19–24 months. The prevalence of N+ was 28% among SD cancers. With a prevalence of 38%, 42%, and 44%, the adjusted (demographics and N staging technique) OR of N+ for cancers diagnosed between 1–12, 13–18, and 19–24 months of interval was 1.41 (95% confidence interval 1.06–1.87), 1.74 (1.31–2.31), and 1.91 (1.43–2.54), respectively. Histologic type, tumour grade, and tumour size were entered in turn into the model. Histologic type had modest effects. With adjustment for tumour grade, the ORs decreased to 1.23 (0.92–1.65), 1.58 (1.18–2.12), and 1.73 (1.29–2.32). Adjusting for tumour size decreased the ORs to 0.95 (0.70–1.29), 1.34 (0.99–1.81), and 1.37 (1.01–1.85). The strength of confounding by tumour size suggested that the excess risk of N+ for first-year interval cancers reflected only their higher chronological age, whereas the increased aggressiveness of second-year interval cancers was partly accounted for by intrinsic biological attributes.
Breast screening: axillary lymph node status of interval cancers by interval year
FERRETTI, Stefano;
2008
Abstract
The aim of this study was to determine whether the excess risk of axillary lymph node metastases (N+) differs between interval breast cancers arising shortly after a negative mammography and those presenting later. In a registry-based series of pT1a–pT3 breast carcinoma patients aged 50–74 years from the Italian screening programmes, the odds ratio (OR) for interval cancers (n = 791) versus the screen-detected (SD) cancers (n = 1211) having N+ was modelled using forward stepwise logistic regression analysis. The interscreening interval was divided into 1–12, 13–18, and 19–24 months. The prevalence of N+ was 28% among SD cancers. With a prevalence of 38%, 42%, and 44%, the adjusted (demographics and N staging technique) OR of N+ for cancers diagnosed between 1–12, 13–18, and 19–24 months of interval was 1.41 (95% confidence interval 1.06–1.87), 1.74 (1.31–2.31), and 1.91 (1.43–2.54), respectively. Histologic type, tumour grade, and tumour size were entered in turn into the model. Histologic type had modest effects. With adjustment for tumour grade, the ORs decreased to 1.23 (0.92–1.65), 1.58 (1.18–2.12), and 1.73 (1.29–2.32). Adjusting for tumour size decreased the ORs to 0.95 (0.70–1.29), 1.34 (0.99–1.81), and 1.37 (1.01–1.85). The strength of confounding by tumour size suggested that the excess risk of N+ for first-year interval cancers reflected only their higher chronological age, whereas the increased aggressiveness of second-year interval cancers was partly accounted for by intrinsic biological attributes.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
