The association between patient age and prostate cancer stage and grade at diagnosis.

Abstract

To investigate the association of age with prostate cancer stage and grade, as the latter factors at the time of diagnosis influence management and prognosis, with some studies suggesting that they may change as a function of patient age. The charts were reviewed of an age-stratified (<60, 60-69, 70-79 and > or = 80 years) random sample of men with newly diagnosed, histologically confirmed prostate cancer in 1995/96 from three geographical areas in Ontario, Canada. Patients were identified using a comprehensive cancer registry, and the chi-square analysis used to examine the relationship between age and stage, logistic regression for the effect of age on clinically localized disease, and linear regression to assess the age and grade relationships. In all, 347 charts were reviewed; more men in the oldest group had T1 and metastatic disease than had younger men (P = 0.034). The proportion of patients with clinically localized disease (T1 and T2) did not change with age (P > 0.10). Tumour grade, as assessed by Gleason score, increased slightly with age (R(2) = 0.017, P = 0.011). Excluding those patients diagnosed by transurethral prostatectomy did not influence either the age/stage or age/grade relationship. Adjusting for prostate-specific antigen level attenuated the age/grade relationship. The stage and grade of prostate cancer at diagnosis changes only slightly with age, probably because of a lower intensity of screening and later diagnosis in older men, rather than any change in prostate cancer biology with age.