Prostate-specific antigen. An important marker for prostate cancer treated by external beam radiation therapy.

BACKGROUND

Prostate-specific antigen (PSA) is a valuable serum marker for prostate cancer. However, the prognostic importance of baseline PSA values in relation to other prognostic factors has not been elucidated. The incidence of postradiation rising PSA values has not been documented, and the extent to which PSA influences the assessment of radiation therapy is unclear. This study was designed to address these issues.

METHODS

Three hundred and fourteen consecutive patients with baseline PSA values who were treated between 1987 and 1991 with external beam radiation alone were reviewed for disease outcome and posttreatment PSA levels.

RESULTS

Clinical stages at diagnosis were: Stage A2, 87 (28%); Stage B, 108 (34%); and Stage C, 119 (38%). At a mean follow-up of 21 months, 25 patients had relapsed, 53 had developed rising PSA profiles, and 58 had either relapsed or had rising PSA profiles. The actuarial relapse rate was 20% at 4 years, the incidence of rising PSA profiles was 38% at 4 years, and the incidence of either relapse or rising PSA was 40% at 4 years. In multivariate analysis, baseline PSA value was the single most important factor predicting for local relapse, metastatic relapse, any disease relapse, and posttreatment rising PSA values. Using relapse or rising PSA as endpoints, the following four prognostic groupings based on baseline PSA and M.D. Anderson (MDA) grade were delineated: I, PSA less than or equal to 4 ng/ml, any grade; II, PSA greater than 4 but less than or equal to 10 ng/ml, Grades 1 and 2; III, PSA greater than 4 but less than or equal to 10 ng/ml, Grades 3 and 4 or PSA greater than 10 but less than or equal to 30 ng/ml, Grades 1 and 2; and IV, PSA greater than 10 but less than or equal to 30 ng/ml, Grades 3 and 4 or PSA greater than 30 ng/ml, any grade. The actuarial incidence of relapse or rising PSA in these groups was: I, less than 10% at 3 years; II, 20% at 3 years; III, 55% at 3 years; and IV, 90% at 30 months. When using traditional endpoints of disease outcome, the patients in this series had an outcome equivalent to that in 799 patients treated in our institution in the pre-PSA era; when using rising PSA profiles as endpoints, treatment was significantly less effective.

CONCLUSIONS

Pretreatment serum PSA level is the single most significant predictor of disease outcome after radiation therapy for local-regional prostate cancer. Moreover, postirradiation PSA values may potentially serve as an early endpoint to evaluate treatment efficacy. Using a rising posttreatment PSA profile as an index of treatment failure reveals that total and permanent eradication of prostate cancer with radiation therapy alone is not achieved as often as previously believed and that multimodal treatment approaches to prognostically unfavorable early stage disease need investigation.