Age-related reference ranges for complexed prostate-specific antigen and complexed/total prostate-specific antigen ratio: results from East Texas Medical Center Cancer Institute screening campaign

Age-Adjusted Reference Values for Prostate Specific Antigen - A Systematic Review and Meta-Analysis

OBJECTIVES

To systematically evaluates the evidence on ethnic differences in age-adjusted reference values of PSA.

MATERIALS AND METHODS

In concordance with the Preferred Reporting Items for Systematic Review and Meta-analysis statement, a review of English articles using Medline, Embase and Cochrane databases, from inception to December 2019 was conducted. Studies that reported the PSA upper reference value as 95th percentile of the cohort distribution, in healthy men aged 40 to 79, were included. Methodological quality was assessed with a modified version of the Agency for Healthcare Research and Quality checklist for cross-sectional studies.

RESULTS

Forty-three studies examining 325,514 participants were included in the analysis. These were published between 1993 and 2018. Majority were prospective observational studies and reported the reference values in ten-year age intervals. Only five reports directly compared ethnic differences in PSA values. Due to missing data, six studies were not considered in the quantitative synthesis. For the remainder (37/43), heterogeneity in PSA reference values was considerable (Higgin's index = 99.2%), with age and ethnicity being the sole identified significant contributors. Accordingly, the pooled upper limits for PSA reference values were 2.1, 3.2, 4.9 and 6.5 ng/ml for men in their 40 s, 50 s, 60 s, and 70 s, respectively.

CONCLUSION

Moderate quality evidence suggest that upper PSA reference limits increased with age and significant ethnic differences were present.

The distribution of serum prostate-specific antigen levels among American men: implications for prostate cancer prevalence and screening

BACKGROUND

The purpose of this study was to describe the distribution of serum prostate-specific antigen (PSA) among American men and to estimate the number of prevalent cases of biopsy detectable prostate cancer among men with normal serum PSA.

METHODS

We analyzed data of the National Health and Nutrition Examination Survey 2001-2002 (NHANES 2001-2002) data and combined these results with published data from the Prostate Cancer Prevention Trial (PCPT).

RESULTS

Most men in the US have a serum PSA < or = 4.0 ng/ml, and mean and median serum PSA values rise steadily with age. There are an estimated 1,607,585 (95% CI 1,370,848-1,844,322) prevalent cases of biopsy detectable prostate cancer in men aged 62-85 years with a serum PSA < or = 4 ng/ml. Among men aged 62-75 years, there are an estimated 1,252,143 (95% CI 1,054,677-1,449,609) prevalent cases, including an estimated 195,499 (95% CI 140,234-250,764) high-grade tumors.

CONCLUSION

A large number of prevalent cases of biopsy detectable prostate cancer exist in American men with a normal PSA.

The measurement of complexed prostate-specific antigen has a better performance than total prostate-specific antigen

The aim of this study was to compare the diagnostic utility of complexed prostate-specific antigen (cPSA) with total PSA (tPSA) in screening for prostate cancer. Serum concentrations of tPSA and cPSA were measured in 4479 adult men during the prostate cancer screening program in the Saarland region (Germany). The percentage of men with c/tPSA ratio above the cut-off value of 0.75 increased with increasing tPSA intervals: tPSA 0-0.9 microg/l, 4.4%; 1.0-1.9 microg/l, 24.3%; 2.0-2.9 microg/l, 43.9%; 3.0-3.9 microg/l, 50.4%; and 4.0-20 microg/l, 60.2%. The commonly accepted tPSA cut-off value of 3.9 microg/l matched to the 93rd percentile of the overall population (corresponding cPSA value, 2.9 microg/l). A total of 202 men out of 313 with increased cPSA had increased c/tPSA ratio (cut-off > or = 0.75) vs. 186 out of 312 men with increased tPSA. Thus, an additional 16 men at high risk for prostate cancer were selected only if cPSA was utilised as a first line parameter. Our data show that, compared to tPSA, cPSA measurement will always detect more high-risk patients, independent of the cut-off levels utilised for cPSA, tPSA and c/tPSA ratio. cPSA is more effective than tPSA in selecting subjects with an elevated c/tPSA ratio who are at high risk of prostate cancer. Thus, cPSA might be seen as the superior first-line parameter in screening for prostate cancer. Using lower cut-off values for tPSA or cPSA than the commonly accepted values seems reasonable for screening purposes.

Racial influence on the prevalence of prostate carcinoma in Brazilian volunteers

PURPOSE

To investigate the prevalence of prostate carcinoma in a sample of volunteers known to have a large proportion of Bantu African ancestors, and the performance of total PSA (tPSA), PSA density (PSAD) and free-to-total PSA ratio (f/tPSA) on the diagnosis.

MATERIALS AND METHODS

A total of 473 volunteers (range: 40 - 79 years) were screened for prostate carcinoma. Those with tPSA >2 ng/ml and/or abnormal digital rectal examination were submitted to a transrectal ultrasound-directed biopsy (10 cores). The volunteers were classified as White, Mulatto or Black according to physical characteristics and to ancestors race reference. The following variable number of tandem repeats (VNTR) were analyzed in the blood of 120 volunteers without cancer and in 27 patients with prostate cancer: D4S43, PAH, F13A1, APOB and vW-1.

RESULTS

The biopsies performed in 121 volunteers revealed cancer in 27 (5.7% of 473). The proportions of cancer in White, Mulatto and Black were respectively: 0.6% (1/148), 6.7% (6/90) and 8.5% (20/235) (p = 0.006). The VNTRs analysis revealed heterogeneity in White, Mulatto and Black anthropologic phenotypes with the following admixture of Caucasian, African and Amerindian gene lineages: 67.5 +/- 8%, 20.8 +/- 8%, 11.7 +/- 7%; 54.8 +/- 9%, 36.3 +/- 5%, 8.9 +/- 7%; and, 45.3 +/- 3%, 45.9 +/- 4%, 8.8 +/-7%. Such a mixture was 50.5 +/- 9%, 49 +/- 8% and 0.5 +/- 4% in volunteers bearing cancer, and 59.1 +/- 7%, 31.7 +/- 8% and 9.2 +/- 5% in those without cancer. The sensitivity and specificity of tPSA at cut-off levels of 2, 2.5 and 4 ng/ml for volunteers with tPSA </= 10 ng/ml were respectively: 100% and 6,6%, 100% and 36,6%, 69,2% and 62,2%. PSAD at a cut-off level of 0.08 or 0.10, and f/tPSA at a cut-off level of 20% were able to increase significantly tPSA specificity without loss on sensitivity.

CONCLUSIONS

The tumor prevalence was higher in Non-White than in White phenotype. The association of tPSA at a cut-off level of 2.5 ng/ml with a PSAD of 0.08 or a f/tPSA of 20% for biopsy indication deserves further investigations as an alternative to tPSA cut-off level of 4 ng/ml.

Complexed prostate specific antigen (PSA) reduces unnecessary prostate biopsies in the 2.6-4.0 ng/mL range of total PSA

OBJECTIVE

To compare the performance of complexed prostate-specific antigen (cPSA) to total PSA (tPSA) and percentage free PSA (f/tPSA) in the diagnosis of prostate cancer for the tPSA range 2.6-4.0 ng/mL.

PATIENTS AND METHODS

Consecutive men scheduled for prostate biopsy were enrolled prospectively at 14 different sites in two multicentre studies in Europe and the USA. Serum obtained before biopsy was tested with the ACS:180 and Immuno 1 tPSA and cPSA assays (Bayer Diagnostics, Tarrytown, NY, USA) and the Access fPSA and tPSA assays (Beckman, Inc., San Diego, CA, USA). Receiver operating characteristics (ROC) curves were generated to compare the diagnostic performance of tPSA, cPSA and f/tPSA.

RESULTS

Of 316 men with a tPSA of 2.6-4.0 ng/mL, 82 (26%) were diagnosed with prostate cancer on biopsy. ROC analysis of all 316 men showed an area under the curve (AUC) for cPSA of 0.63, significantly greater than the AUC for tPSA of 0.56 (P = 0.008). At a sensitivity of 95%, threshold values of 2.3 ng/mL for cPSA and 2.73 ng/mL for tPSA provided specificities of 20.1% and 9.8%, respectively. f/tPSA was only available for 205 of the 316 (65%) men and the AUC for cPSA was 0.63, and did not significantly differ from the f/tPSA AUC of 0.64 (P = 0.58).

CONCLUSIONS

As a single test, cPSA provides improved specificity over tPSA and comparable specificity to f/tPSA for detecting prostate cancer, and may reduce the number of unnecessary prostate biopsies in the 2.6-4.0 ng/mL tPSA range.

Impact of age on complexed PSA levels in men with total PSA levels of up to 20 ng/mL

OBJECTIVES

To evaluate the relationship between age and complexed prostate-specific antigen (cPSA) in healthy white men with no evidence of prostatic disease and total PSA (tPSA) levels of 0 to 20 ng/mL.

METHODS

A total of 10,530 tests for cPSA and tPSA in men between 40 and 79 years old were performed at Innsbruck University, Austria, between 2001 and 2002. After excluding 263 patients because of follow-up biopsy findings indicating prostatic disease, 10,267 PSA tests were available for study. All serum samples were analyzed using the Bayer Immuno 1 cPSA and tPSA assays. cPSA and tPSA values were determined in subjects with tPSA levels between 0 and 20 ng/mL. The concentrations were calculated for the 50th, 90th, and 95th percentiles of four age groups (40 to 49, 50 to 59, 60 to 69, and 70 to 79 years). For statistical analysis, the Kruskal-Wallis test was used. The 95th percentile was defined as the upper limit of normal.

RESULTS

A total of 10,267 PSA tests were available for study. cPSA and tPSA levels correlated linearly and positively with age (P <0.001). Percent free PSA showed a significant increase with age (P <0.001), which may be attributed to the increase in prostate volume with advancing age.

CONCLUSIONS

Both cPSA and tPSA levels were found to increase with age. The low tPSA and cPSA levels observed at the 95th percentile in men aged 40 to 59 years suggest that lower cutoff values should be used in place of the currently accepted tPSA level of 4 ng/mL. Lower cutoff values may help improve the detection of prostate cancer among otherwise healthy men.

A neurocomputational model for prostate carcinoma detection

BACKGROUND

Current guidelines for prostate carcinoma screening rely primarily on the digital rectal examination (DRE) and prostate specific antigen (PSA). Well described patient risk factors for prostate carcinoma also include age, ethnicity, family history, and complexed PSA. However, due to the nonlinear relation of each of these variables with prostate carcinoma, it is difficult to predict reliably each patient's risk based on linear univariate analysis. The authors investigated a neural network to model the risk of prostate carcinoma by seven readily available clinical features.

METHODS

The database for the current study comprised 3268 men recently evaluated for the early detection of prostate carcinoma. The seven clinical features evaluated included age, race, family history, International Prostate Symptom Score (IPSS), DRE, and total and complexed PSA. Three hundred forty-eight subjects in the dataset included men with determined prostate biopsy outcomes and for whom at least 6 of 7 features were available. The dataset was divided randomly into a training set (60%) and a test set (40%), with n1/n2 cross-validation used to evaluate model accuracy, and was modeled with linear and quadratic discriminant function analysis and a neural computational system. After a model with acceptable goodness of fit was achieved, reverse regression analysis using Wilks's generalized likelihood ratio test was performed to evaluate the statistical significance of each input variable.

RESULTS

The receiving operating characteristic (ROC) area for the neural computational system in the test set was 0.825, whereas total PSA and complexed PSA alone had ROC areas of 0.678 and 0.697, respectively. The ROC area of logistic regression in the test set was 0.510, linear discriminant function analysis was 0.674, and quadratic discriminant function analysis was 0.011. All were significantly less than the ROC area of the neural computational model (all Ps < 0.002). Reverse regression based on Wilks's generalized likelihood ratio test demonstrated each input feature to be highly significant to the model (all Ps << 0.000001).

CONCLUSIONS

The authors modeled a combination of well described patient risk factors for prostate carcinoma using a neural computational system with acceptable goodness of fit. They demonstrated that each of the seven variates on which the model was based was critically significant to model performance. The authors presented this model for clinical use and suggested that clinicians use it in deciding to perform prostate biopsy.