Data from a series of > 1,200 Danish patients, all referred by their primary primary care physicians to a single, specialized urology practice, seems to confirm what has long been suspected about the relative values of total and complex PSA levels in the diagnosis of prostate cancer.
Hillig et al. set out to assess the relative diagnostic performances of total PSA (tPSA), complex PSA (cPSA), and percent free PSA (%fPSA). As suggested above, they collected data from a total of > 1,200 among patients referred — between June 2005 and August 2010 — by general practice doctors in Denmark to the urology department at the Frederikssund Hospital in Cøpenhagen. They then evaluated prognostic factors and survival data for this cohort of men.
Here is what they found:
- 1988 individual patients were referred to the Department of Urology, Frederikssund Hospital.
- Relevant and appropriate PSA data were collected from 1,422 men.
- 1,261 men who had no prior history of prostate cancer were eligible for inclusion in the study database.
- On evaluation by specialists at the urology department
- 299/1,261 men (23.7 percent) were diagnosed with prostate cancer within 1 year of the referral.
- All 299 men found to have prostate cancer were biopsied., whereas 209 (22%) of the 962 non-PCa patients were biopsied.
- 962/1,261 men (76.3 percent) were found to have no prostate cancer within 1 year of referral.
- 209/962 men (21.7 percent) who were found not to have prostate cancer were biopsied.
- Among the men found to have prostate cancer
- Average (median) age was 70.8 years (range, 50.1 to 95.3 years)
- Average (median) tPSA was 13.4 ng/ml (range, 0.1 to 5,920.5 ng/ml)
- Average (median) cPSA was 10.8 ng/ml (range, 0.1 to 4,908.7 ng/ml)
- Average (median) %fPSA was 12.6 percent (range, 0.0 to 79.3 percent)
- Among the men who showed no evidence of prostate cancer
- Average (median) age was 67.5 years (range, 19.1 to 95.4 years)
- Average (median) tPSA was 2.5 ng/ml (range, 0.1 to 118.1 ng/ml)
- Average (median) cPSA was 1.9 ng/ml (range, 0.1 to 83.8 ng/ml)
- Average (median) %fPSA was 24.9 percent (range, 0.0 to 78.8 percent)
- The sensitivity, specificity, positive predictive value, negative predictive value, and efficiency of the tPSA and cPSA values overlapped.
- Among men with a tPSA value between >4 and ≤ 20 ng/ml, the %fPSA was able to exclude prostate cancer with a negative predictive value of 72.4 percent.
- At 10 years of follow-up,
- 412/1,261 patients included in the cohort (32,7 percent) had died of all causes.
- The median follow-up time of the remaining 849 patients was 110 months (range, 102 to 118 months).
- 115/299 men with prostate cancer (38.5 percent) had
- A tPSA concentration >20 ng/ml at time of initial referral
- A reduced 10-year survival as compared to patients with tPSA concentrations ≤20 ng/ml.
- Among all 1261 patients in the cohort
- Age was a significant prognosticator for 10-year overall and prostate cancer-specific survival.
- tPSA was a significant prognosticator for 10-year prostate cancer-specific survival. tPSA and Gleason score were stronger prognosticators for PCa-specific survival than age among PCa patients.
- Among the 299 patients with prostate cancer
- Age was not a prognosticator for 10-year prostate cancer-specific survival.
- tPSA level at diagnosis was a significant prognosticator for 10-year overall survival.
- Gleason score was a prognosticator for 10-year survival.
- tPSA and Gleason score were stronger prognosticators for prostate cancer-specific survival than age.
- Overall 10-year survival was
- 72 percent among men in the tPSA concentration interval 0 to ≤ 4 ng/ml
- 73 percent among men in the tPSA concentration interval > 4 to ≤ 10 ng/ml
- 64 percent among men in the tPSA concentration interval > 10 to ≤20 ng/ml
- 37 percent among men with a tPSA concentration > 20 ng/ml
- Prostate cancer-specific 10-year survival was significantly reduced among men with tPSA concentrations >20 ng/ml.
Hillig et al. conclude that:
tPSA and cPSA showed similar diagnostic performances. %fPSA provided additional diagnostic information at tPSA concentrations > 4 to ≤ 20 ng/ml]. The high percentage of patients with tPSA concentrations >20 ng/ml indicate delayed use of tPSA resulting in advanced disease at presentation and reduced patient survival.
These data would seem to confirm the widely held beliefs that cPSA and tPSA levels provide similar information about initial risk for prostate cancer but that lower %fPSA values are indeed helpful in discriminating between risk for prostate cancer and risk for other prostate disorders among men whose initial tPSA value is between about 4 and 20 ng/ml.
Filed under: Diagnosis, Risk | Tagged: complex, Diagnosis, free, mortality, PSA, risk, total |
What has happened to the updates on research articles? They were most useful.
Marcia Falconer
On Wed, Oct 26, 2016 at 2:59 PM, THE “NEW” PROSTATE CANCER INFOLINK wrote:
> Sitemaster posted: “Data from a series of > 1,200 Danish patients, all > referred by their primary primary care physicians to a single, specialized > urology practice, seems to confirm what has long been suspected about the > relative values of total and complex PSA levels in t” >
Dear Marcia:
I am not sure what you mean by updates on “research articles”, but since October 26 last year we have published > 60 commentaries on a whole range of issues … and the majority of them have been based on data from research reports in the clinical and medical scientific media.