Demographic and clinical factors as determinants of serum levels of prostate specific antigen and its derivatives.

BACKGROUND To characterise the association between demographic and clinical factors and levels of total prostate specific antigen (tPSA) and its molecular derivatives complexed PSA (cPSA), free PSA (fPSA) and the ratio of free to total PSA (%fPSA)] in New Zealand Maori, Pacific Islanders and Europeans, in order to determine whether reported ethnic differences in PSA can be explained by lifestyle and social factors. MATERIALS AND METHODS Demographic and clinical factors were examined in relation to tPSA, fPSA and cPSA levels, in 1405 Maori, Pacific Island and New Zealand European men with no clinical evidence of prostate cancer, in the Wellington region of New Zealand. Any associations between levels of PSA and PSA derivatives and body mass index, smoking status, family cancer history, non-steroidal anti-inflammatory/vitamin supplement usage, number of sexual partners, age at first intercourse, previous vasectomy, marital/partnership status, educational level and socioeconomic status were investigated by backwards stepwise regression analysis, correcting for age, ethnicity and urinary symptoms. RESULTS Not being married/partnered was associated with increased tPSA, fPSA and cPSA. tPSA and cPSA decreased with regular non-steroidal anti-inflammatory use. cPSA was decreased in subjects with a first degree relative with any form of cancer. tPSA and fPSA were decreased if the body mass index was > 34. fPSA and %fPSA were decreased in current and former smokers. CONCLUSION Demographic and clinical factors appear to have a significant effect on levels of PSA and its various derivatives and may account for previously observed ethnic differences. It is important that these associations are taken into account when comparing individual PSA results with standard reference ranges.