Breast cancer tumour growth modelling for studying the association of body size with tumour growth rate and symptomatic detection using case-control data

Abstract

Introduction: A large body size is associated with larger breast cancer tumours at diagnosis. Standard regression models for tumour size at diagnosis are not sufficient for unravelling the mechanisms behind the association. Methods: Using Swedish case-control data, we identified 1352 postmenopausal women with incident invasive breast cancer diagnosed between 1993 and 1995. We used a novel continuous tumour growth model, which models tumour sizes at diagnosis through three submodels: for tumour growth, time to symptomatic detection, and screening sensitivity. Tumour size at other time points is thought of as a latent variable. Results: We quantified the relationship between body size with tumour growth and time to symptomatic detection. High body mass index and large breast size are, respectively, significantly associated with fast tumour growth rate and delayed time to symptomatic detection (combined P value=5.0×10 -5 and individual P values=0.089 and 0.022). We also quantified the role of mammographic density in screening sensitivity. Conclusions: The times at which tumours will be symptomatically detected may vary substantially between women with different breast sizes. The proposed tumour growth model represents a novel and useful approach for quantifying the effects of breast cancer risk factors on tumour growth and detection.