P2-34 Association of anthropometric and lifestyle factors with prostate specific antigen (PSA) trajectories in men with localised prostate cancer undergoing active monitoring

Introduction Widespread use of Prostate-specific Antigen (PSA) testing has lead to a rapid increase in the identification of low risk prostate cancer, with PSA trajectories often used to monitor tumour progression after diagnosis. Here we develop novel age-specific multi-level growth curves to assess the effects of lifestyle and anthropometric measures on PSA trajectories. Methods Serial PSA measures from 513 men aged 50–70 years with localised prostate cancer undergoing active monitoring were used to develop models. Gleason score, height, weight, body mass index, waist, waist to hip ratio, smoking and alcohol consumption were each added to the basic age-specific growth curve model as explanatory variables. Results The basic growth curve gave an average PSA at age 50 of 2.10 ng/ml (95% CI 1.85 to 2.38) and a yearly increase in PSA of 1.07 ng/ml (95% CI 1.06 to 1.08) (or 7%; 95% CI 6% to 8%). In these preliminary findings, Gleason score at baseline was strongly associated with PSA growth: the yearly increase in PSA was 3.4% greater (95% CI 1% to 6%) for men with Gleason of 7 or greater vs those with Gleason 6 or less. Current smoking was positively associated with PSA change: the yearly increase in PSA was 3.2% greater (95% CI 0% to 7%) for current vs never smokers. No other factors were strongly associated with either initial PSA or yearly increase in PSA. Conclusions Smoking status may be associated with PSA trajectory in men being followed up by active monitoring but other lifestyle and anthropometric factors have little association.