Prostate Specific Antigen (PSA) growth curves: A method to improve prostate cancer screening

Prostate cancer (PrCA) screening aimed at detecting aggressive disease represents a significant public health issue. Developing tools to predict aggressive PrCA represents a major public health challenge. In this study we retrospectively analyzed repeated measures of prostatic specific antigen (PSA) to develop and validate a tool to improve detecting high-risk PrCA.

Using 6 annual PSA measurements from the Prostate Lung Colorectal and Ovarian trial (PLCO), we established PSA growth curves for four groups of men; those who developed high-risk PrCA, those who developed low-risk PrCA, those who developed benign prostatic hyperplasia (BPH) and those who were not diagnosed with either PrCA or BPH. We used these curves to estimate individual PSA annual rate of change at one and two years before diagnosis. We then examined the area under the curve (AUC) to estimate specificity and sensitivity of PSA annual rate thresholds.

Our results show that PSA annual change rates varied significantly by cancer status. The difference between the means and medians of PSA rate values across the four groups of men were high and robust. Annual individual PSA rates among most men who developed high-risk PrCA were significantly higher than all others. This resulted in a high (AUC); 99.50 (99.34-99.66), in a logistic regression model that used these individual PSA annual rates to predict high-risk PrCA. A threshold of 0.37ng/ml/year had a sensitivity and specificity of 97.2%, and 97.3%, respectively. 

Our results suggest that it is feasible to improve screening specificity for high-risk PrCA by investigating PSA trends over time.