Pre-treatment PSA velocity predicts recurrence risk

Boston-A preoperative rise in serum PSA of >2.0 ng/mL predicts an increased mortality risk in patients with localized prostate cancer undergoing radical prostatectomy and might identify men who require additional therapy, investigators have concluded after reviewing outcomes in more than 1,000 cases.

A PSA velocity exceeding 2 ng/mL in the year before prostate cancer diagnosis predicted a significantly shorter time to death from prostate cancer and death from all causes after radical prostatectomy. Notably, half of the patients with a preoperative PSA rise >2.0 ng/mL had low-risk characteristics.

"The results are striking because they show that half of a group of men with a PSA rise of more than 2 points in the year prior to diagnosis had low-risk prostate cancer, ie, PSA <10.0 ng/mL and Gleason 6 or less and T1c," said Anthony D'Amico, MD, professor of radiation oncology at Harvard Medical School and Brigham and Women's Hospital in Boston. "This redefines how we look at PSA and what it is telling us."

Post-prostatectomy screens

The study included data on 1,804 men who participated in a prostate cancer screening study at Northwestern University Feinberg School of Medicine, Chicago, and subsequently underwent radical prostatectomy between 1989 and 2002. All had stage T1c or T2 prostate cancer.

Dr. D'Amico and colleagues excluded of 689 men who had only a single PSA measurement in the year prior to diagnosis and 20 others who had adjuvant radiation therapy, leaving 1,095 who had at least two preoperative PSA evaluations and were treated solely by radical prostatectomy.

In all cases, the preoperative workup included a digital rectal examination, PSA measurement, and transrectal needle biopsy of the prostate. DRE and PSA evaluations occurred at 6- to 12-month intervals and just before diagnosis of prostate cancer. After surgery, most patients had PSA measurements every 6 months and an annual DRE. After recurrence-defined as two consecutive PSA values >2.0 ng/mL-the median interval between PSA measurements was 4 months. Follow-up ran for a median of 5.1 years.

Using linear-regression analysis, the investigators calculated PSA velocity on the basis of the measurement closest to diagnosis of prostate cancer and all other PSA values obtained within the year before diagnosis. Other factors evaluated for a relationship to prostate cancer-specific mortality and all-cause mortality were PSA value at diagnosis, Gleason score, and tumor stage.

Recurrence and death risks Men who had a PSA velocity >2.0 ng/mL in the year before prostate cancer diagnosis had a significantly greater risk of recurrence, death from prostate cancer, and death from any cause, compared with men who had lower PSA velocity values.

In multivariate analysis, the relative risk conferred by a faster rise in PSA was 1.5 for recurrence (p=.003); 9.8 for prostate cancer death (p<.001); and 1.9 for death from any cause (p=.01). None of the other factors evaluated (PSA at diagnosis, Gleason score, tumor stage) had a significant association with any of the outcomes included in the analysis.

Investigators emphasized that half of the patients with a PSA velocity >2.0 ng/mL had what might be considered classic low-risk characteristics: PSA <10.0 ng/mL, clinical stage T1c diesase, and Gleason score •6.