Background: The serum marker prostate-specific antigen (PSA) has been shown to be proportional to the volume of prostate cancer (Yang polyclonal assay). One gram of cancer on average produces 3.5 ng/ml of PSA elevation. Thus, changes in PSA in untreated patients should reflect tumor growth rate and the shape of the growth curve.
Methods: Forty-three patients with previously untreated prostate cancer had serial PSA determinations during a period of at least 12 months (mean, 30 months; range, 12-63 months). There was no treatment between the initial and final PSA. PSA was measured five or more times in half of these patients. Log-PSA values were plotted versus time, tested for linearity, and compared between different clinical stages and histologic grades.
Results: PSA values increased with time in 86% of patients but remained stable in 14%. The increase of PSA was exponential (log-linear) throughout the measured interval. Doubling times were faster in patients with higher disease stages and grades. Seventy-nine percent of all patients had a doubling time of more than 24 months. Twenty of 28 cancers thought to be clinically organ confined doubled at rates exceeding 48 months. Tumor doubling times were overestimated in patients with large-volume benign prostatic hyperplasia because hyperplasia increases serum PSA at an average rate of 0.3 ng/ml.
Conclusions: The authors conclude that prostate cancer has a constant (log-linear) growth rate that is very slow. This extraordinarily slow doubling time has substantial consequences for therapeutic decisions and for screening programs.