Influence of initial presentation on treatment outcome of clinically localized prostate cancer treated by definitive radiation therapy.

Abstract

The increasing proportion of early stage prostate cancer diagnosed by various early detection methods together with reports espousing watchful waiting as a management option raise the possibility that patients may be selected for surveillance according to their initial presentation. The outcome for 427 men with clinical stages T1 to T4 localized prostate cancer treated with radiation therapy was evaluated according to their presentation: elevated prostate-specific antigen (PSA) level; abnormal digital rectal examination; or, urologic symptomatology. With a median follow-up of 30 months, there were no significant differences in disease outcome according to initial presentation. The actuarial incidence of relapse at 5 years was: PSA-detected (54 patients), 24%; digital rectal-detected (173 patients) 29%; and, symptom-detected (200 patients) 31% (p = 0.79). Likewise, there were no significant differences in the incidence of postradiation rising PSA profiles among the three groups. The actuarial incidence of relapse and/or rising PSA at 5 years was: PSA-detected 35%; digital rectal-detected 42%; symptom-detected, 48% (p = 0.72). On the other hand, T-stage, Gleason grade, pretreatment PSA, pretreatment acid phosphatase, and transurethral resection in T3/T4 disease were each highly correlated with outcome. In multivariate proportional hazards regression pretreatment PSA (p = 0.0003), Gleason grade (p = 0.045), and transurethral resection in T3/T4 disease (p = 0.0562) correlated with outcome, but initial presentation did not (p = 0.25). The absence of a prognostic gradient, good to bad, from PSA-detected through digital rectal-detected to symptom-detected cancer suggests that the initial presentation of patients with localized prostate cancer is not a valid basis for selecting watchful waiting vs. initial treatment.