Prostate specific antigen and prostatic acid phosphatase for monitoring therapy of carcinoma of the prostate

Serial serum prostatic acid phosphatase (PAP) and prostate specific antigen (PSA) measurements were performed in 871 patients treated with hormonal combination therapy for stage C (95 patients) or stage D2 (776) prostate cancer for an average followup of 26 months. The relative efficacy of serum PAP and PSA for predicting recurrence of the disease was evaluated by 2 statistical methods at the time of progression as well as 6 and 12 months before clinical relapse of disease using optimized cut-off values of 2.0 and 4.0 micrograms/l. for serum PAP and PSA, respectively. At the time of progression the sensitivity (plus or minus standard deviation) of the 2 tests was estimated at 61.1 +/- 3.2% and 86.7 +/- 3.1% for PAP and PSA, respectively, while the specificity (plus or minus standard deviation) was calculated at respective values of 79.6 +/- 1.3% and 92.4 +/- 4.1%. Receiver operating characteristic analysis disclosed a greater accuracy for PSA at 89.2 +/- 1.7% versus 78.7 +/- 1.6% (plus or minus standard deviation) for PAP. The somewhat lower positive predictive value of the PSA test (81.4% versus 89.6%) is more than compensated by its superior negative predictive value (92.4% versus 79.6%). The present data also show that serum PSA measurements are superior to those of serum PAP for predicting disease recurrence in stages C and D prostate cancer patients treated by combination endocrine therapy and they indicate that measurement of serum PAP does not add significantly to single measurement of serum PSA alone.