The PSA Velocity

The change in PSA values over time is called the PSA velocity. This parameter was first introduced by The Johns Hopkins Medical Institutions in a collaborative effort with the Baltimore Longitudinal Study of Aging. This important study showed for the first time that the "rate of change" of serum PSA over time provides useful information and increases the specificity of PSA for cancer detection. These authors showed that a cutoff of 0.75ng/mL/y increased the sensitivity of PSA testing from 66% to 79% among men with normal PSA level (< 4 ng/mL) followed up for over 2 years. (If you go the Sloan Kettering Nomogram calculator it will calculate PSA velocity for you, or the doubling time, go here.)

Current recommendations for the use of PSA velocity include collection of PSA levels over a period of no less than 18 months and the use of multiple values (minimum of 3) to perform the calculation. Currently, PSA velocity has been best used in younger men who have elected to begin early detection programs before age 50. These men have predominantly lower (normal) serum PSA levels and face the prospect of having their PSA levels followed up over several decades. The PSA velocity test was designed to help avoid unnecessary, costly, and potentially morbid repeat biopsies in these men. Patients and physicians electing to monitor prostate disease by measuring PSA velocity should be cautioned that fluctuations between measurements can occur as a result of either laboratory variability related to inter-assay variability from the use of different commercially available sources or from individual biologic variability. Consequently, both of these causes may confuse the interpretation of PSA velocity measurements. The NCCN 2004 guidelines make the following statement:

"PSA Velocity: For men with PSA < 10.0 ng/mL, data suggest that a PSA velocity of 0.75 ng/mL/yr. is suspicious for the presence of cancer, and biopsy is recommended. Measurement should be made on three specimens drawn over at least an 18-mo interval."

The guidelines don't trigger concern unless the PSA rises 0.75ng/ml in a year, but some studies suggest that even a rise  in PSA of as little as 0.1 ng/ml per year is worrisome as to the long term chance of being diagnosed with prostate cancer as in the study below (Urology 2002; 59:889.)
PSA Velocity Prostate Cancer Risk by 10 Years
< 0.1 ng/ml per year 3% (0 - 8.6%)
> 0.1 ng/ml per year 65% (44 - 86%)
Any rapid rise in PSA is worrisome both for the likelihood of developing cancer but of having more serious (high grade prostate cancer.) A recent study (Preoperative PSA Velocity and the Risk of Death from Prostate Cancer after Radical Prostatectomy, D'Amico. NEJM 2004;351:125) found that men whose PSA jumped more than 2 points in the year prior to being diagnosed had much higher death rate when treated with surgery (radical prostatectomy.)

Also patients treated with radiation for prostate cancer whose PSA starts to rise do worse if the PSA doubles in less than 12 months (see here.)