Different stability of free and complexed prostate-specific antigen in serum in relation to specimen handling and storage conditions

Tracking the Stability of Clinically Relevant Blood Plasma Proteins with Delta-S-Cys-Albumin-A Dilute-and-Shoot LC/MS-Based Marker of Specimen Exposure to Thawed Conditions

Biomolecular integrity can be compromised when blood plasma/serum (P/S) specimens are improperly handled. Compromised analytes can subsequently produce erroneous results-without any indication of having done so. We recently introduced an LC/MS-based marker of P/S exposure to thawed conditions called ΔS-Cys-Albumin which, aided by an established rate law, quantitatively tracks exposure of P/S to temperatures greater than their freezing point of -30 °C. The purposes of this study were to (1) evaluate ΔS-Cys-Albumin baseline values in gastrointestinal cancer patients and cancer-free control donors, (2) empirically assess the kinetic profiles of ΔS-Cys-Albumin at 23 °C, 4 °C, and -20 °C, and (3) empirically link ΔS-Cys-Albumin to the stability of clinically relevant proteins. ΔS-Cys-Albumin was measured at ≥ 9 different time points per exposure temperature in serum and K₂EDTA plasma samples from 24 separate donors in aliquots kept separately at 23 °C, 4 °C, and -20 °C. Twenty-one clinically relevant plasma proteins were measured at four time points per temperature via a multiplexed immunoassay on the Luminex platform. Protein stability was assessed by mixed effects models. Coordinated shifts in stability between ΔS-Cys-Albumin and the unstable proteins were documented by repeated measures and Pearson correlations. Plasma ΔS-Cys-Albumin dropped from approximately 20% to under 5% within 96 h at 23 °C, 28 days at 4 °C, and 65 days at -20 °C. On average, 22% of the 21 proteins significantly changed in apparent concentration at each exposure temperature (p < 0.0008 with >10% shift). A linear inverse relationship was found between the percentage of proteins destabilized and ΔS-Cys-Albumin (r = -0.61; p < 0.0001)-regardless of the specific time/temperature of exposure. ΔS-Cys-Albumin tracks cumulative thawed-state exposure. These results now enable ΔS-Cys-Albumin to approximate the percentage of clinically relevant proteins that have been compromised by incidental plasma exposure to thawed-state conditions.

Preanalytical stability of [-2]proPSA in whole blood stored at room temperature before separation of serum and plasma: implications to Phi determination

Background [-2]proPSA seems to outperform free/total prostate-specific antigen (PSA) ratio in prostate cancer diagnosis. However, [-2]proPSA stability remains an underestimated issue. We examined [-2]proPSA stability over time in whole blood before separation of serum and plasma and its implications for prostate health index (Phi) determination. Total PSA (tPSA) and free PSA (fPSA) stabilities were also assessed. Methods Blood was drawn from 26 patients and separated in two tubes for plasma (K2EDTA and K2EDTA plus protease inhibitors - P100) and one for serum (clot activator plus gel separator). Tubes were stored at room temperature before centrifugation 1, 3 and 5 h for serum and EDTA plasma or 1 and 5 h for P100 plasma. To investigate the influence of gel separator on markers' stability, blood was collected from 10 patients in three types of tubes to obtain serum: tubes with clot activator plus gel separator, with silica particles or glass tubes. Biomarkers were assayed with chemiluminescent immunoassays. Results [-2]proPSA and Phi levels significantly and progressively increased over time in serum (+4.81% and +8.2% at 3 h; +12.03% and +14.91% at 5 h, respectively, vs. 1 h; p<0.001). Conversely, [-2]proPSA levels did not change in plasma (EDTA or P100). tPSA levels did not change over time in serum or plasma, whereas fPSA decreased in serum. All markers were higher in plasma than in serum at any time point. This difference did not seem to be attributable to the use of gel for serum preparation. Conclusions EDTA prevented spurious in vitro modifications in PSA-related isoforms, confirming that a stabilized blood sample is a prerequisite for [-2]proPSA measurement and Phi determination.

Five-Year Stability Study of Free and Total Prostate-Specific Antigen Concentrations in Serum Specimens Collected and Stored at – 70°C or Less

The stability of total (t) and free (f) prostate-specific antigen (PSA) in male serum specimens stored at –70°C or lower temperature for 4.7 to 4.9 years was studied. Until now, the stability of these analytes in serum has not been evaluated systematically beyond 2 years of storage at –70°C. Aliquots of frozen serum were thawed in 2001 and 2006 and assayed for tPSA and fPSA using a Dade Behring Dimension(r) RxL analyzer and reagents. tPSA values ranged from 0.07 to 69.94 and 0.00 to 69.83 ng/mL in 2001 and 2006, respectively, whereas fPSA values for the tested specimens ranged from 0.02 to 5.72 and 0.00 to 5.92, respectively. Deming regression analyses showed agreement in assay values over time as tPSA values yielded a slope of 1.0112 and a y-intercept of 0.0195; fPSA values produced a slope 1.0538 and a y-intercept of –0.0442; f/tPSA values yielded a slope of 0.9631 and a y-intercept of 0.1195. A Bland-Altman analysis of the data demonstrated analyte and ratio stability over this time period. We conclude that serum, when collected properly and stored at –70°C or lower temperature, may be used for tPSA and fPSA clinical studies for at least 5 years after collection.

Release of urinary extracellular vesicles in prostate cancer is associated with altered urinary N-glycosylation profile

AIM

Nowadays, extracellular vesicles are of great interest in prostate cancer (PCa) research. Asparagine (N)-linked glycosylation could play a significant role in the pathological mechanism of these vesicles. We investigated if prostatic protein N-glycosylation profiles were related to urinary vesicle-associated prostate-specific antigen (PSA) extractability and if this parameter showed diagnostic potential for PCa.

METHODS

Urinary extracellular vesicles were visualised using transmission electron microscopy. Urinary extracellular vesicles extraction by means of n-butanol allowed determination of urinary vesicle-associated PSA extractability. Diagnostic value was assessed between benign prostate hyperplasia (BPH; n=122) and patients with PCa (n=85). Additionally, correlation with urine N-glycosylation was assessed.

RESULTS

Urinary extracellular vesicles with a diameter of approximately 100 nm were more abundantly present in urine of patients with PCa versus patients with BPH resulting in a higher vesicle-associated PSA extraction ratio (p<0.001). Next, vesicle-associated PSA extraction ratio was correlated to biantennary core-fucosylation (p=0.003). Finally, vesicle-associated PSA extraction ratio proved beneficial in PCa diagnosis, next to serum PSA and the urinary glycosylation marker (p=0.021).

CONCLUSIONS

The urinary vesicle-associated PSA extraction ratio is increased in PCa which is a direct result of the abundant presence of extracellular vesicles in urine of patients with PCa. The urinary vesicle-associated PSA extraction ratio was associated with changes in N-glycoforms and showed diagnostic potential. Further research is warranted to unravel the pathological link between N-glycosylation and extracellular vesicles in cancer, as well as to assess the prognostic value of this biomarker.

Stability and accuracy of total and free PSA values in samples stored at room temperature

INTRODUCTION

In 2010, an estimated 476,076 total PSA tests were performed in Ireland, at a cost of €3.6 million with the majority ordered by general practitioners. We aimed to replicate storage conditions at room temperature and see if prolonged storage affected total and free PSA values.

METHODS

Blood samples were taken from 20 male patients in four VACUETTE^® Serum Separator tubes (Greiner-Bio-One, Austria) and stored at room temperature (22 °C) for different time intervals (4, 8, 24, 48 h) before being centrifuged and analyzed. Total PSA (tPSA) and free PSA (fPSA) values were determined using the Tosoh AIA 1800 assay (Tokyo, Japan).

RESULTS

Mean tPSA values were measured at 4, 8, 24 and 48 h with values of 7.9, 8.1, 7.8 and 8.0 μg/L, respectively. Values ranged from -1.26 to +2.53 % compared to the initial 4 h interval reading, indicating tPSA remained consistent at room temperature. The tPSA showed no significance between groups (ANOVA, p = 0.283). Mean fPSA values at 4, 8, 24 and 48 h were 2.05, 2.04, 1.83, 1.82 μg/L, respectively. At 24 and 48 h there was 10.73 and 11.22 % reduction, respectively, in fPSA compared to the 4-h time interval, indicating prolonged storage resulted in reduced fPSA values. After 24 h, there was an 8.8 % reduction in the free/total PSA %. The fPSA showed significant differences between groups (ANOVA, p = 0.024).

CONCLUSIONS

Our recommendation is that samples that have been stored for prolonged amounts of time (greater than 24 h) should not be used for free PSA testing.

Objective non-intrusive markers of sperm production and sexual activity

Objective studies of men's reproductive function are hindered by their reliance on: (i) self-reporting to quantify sexual activity and (ii) masturbation to quantify sperm output rendering both types of estimate vulnerable to unverifiable subjective factors. We therefore examined whether detection of spermatozoa and measurement of prostate-specific antigen (PSA) in urine could provide objective semiquantitative estimates of sperm output and recent ejaculation, respectively, using widely available laboratory techniques. Of 11 healthy volunteers who provided urine samples before and at intervals for 5 days after ejaculation, sperm was present in 2/11 men before, and in all 11/11 samples immediately after ejaculation, but by the second and subsequent void, spermatozoa were present in ∼10%. PSA was detectable at high levels in all urine samples, peaking at the first post-ejaculatory sample but returning to baseline levels by the second post-ejaculatory void. We conclude that urinary spermatozoa and PSA are objective biomarkers for sperm production and sexual activity, but only for a short-time window until the first post-ejaculatory urine void. Hence, for a single urine specimen, the presence of spermatozoa and PSA are valid biomarkers, reflecting sperm production and recent ejaculation only until the next micturition, so their measurement should be restricted to the first morning urine void.

Short-term stability of the molecular forms of prostate-specific antigen and effect on percent complexed prostate-specific antigen and percent free prostate-specific antigen

Differences in stability of the free and complexed molecular forms of prostate-specific antigen (PSA) may influence the clinical utility of assays for these forms, as well as the calculated ratios to total PSA (tPSA), such as percent free PSA (fPSA) and percent complexed PSA (cPSA). The objective of this study was to directly compare the short-term stability of fPSA and cPSA under different storage conditions. Specimens (3 with prostate cancer, 3 biopsy-negative without cancer, 2 normal) from 8 men were analyzed at baseline within 2 hours of collection, and at 4 hours, 8 hours, 24 hours, 48 hours, and 1 week after storage at room temperature, 4 degrees C, or -20 degrees C. Serum specimens were analyzed in duplicate on the Bayer Immuno 1 analyzer (tPSA, cPSA) and on the Beckman Coulter Access analyzer (tPSA, fPSA Tandem assays). Baseline tPSA values ranged from 0.7 to 62.0 ng/mL, with a median of 7.9 ng/mL (Immuno 1). Overall, all forms of PSA were stable up to 24 hours at the 3 temperatures, with the exception of fPSA and percent fPSA, which decreased when stored at 4 degrees C. After 1 week, tPSA levels decreased when stored at room temperature and at 4 degrees C, as did cPSA stored at room temperature. Over the 7 days, percent cPSA was stable at room temperature, but increased at 4 degrees C. There were no significant changes in any PSA form or calculated ratio with storage at -20 degrees C for up to 1 week. In summary, in the short term (<1 week), fPSA is less stable with storage than tPSA or cPSA in a time- and temperature-dependent fashion. Thus, specimen handling should be considered when interpreting PSA results. It is recommended that specimens not analyzed the same day (within 8 hours of collection) be stored frozen at -20 degrees C.

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

The East Texas Medical Center Cancer Institute conducted a regional prostate cancer screening campaign over a 3-year period from 1998 through 2000. Total prostate-specific antigen (tPSA), complexed PSA (cPSA), and cPSA/tPSA ratio (c/tPSA) values were determined. To better define prostate cancer in the population, we chose to determine age-based reference ranges for these PSA isoforms in apparently healthy men. Participants (N = 12,902) between the ages of 20 and 94 were screened and demographic information and serum tPSA and cPSA values were collected across 41 centers throughout East Texas. Men with an abnormal digital rectal examination, a follow-up biopsy indicating prostatic disease, or any clinical signs and symptoms of prostatic disease were excluded. Sera from 7541 evaluable men were tested with the Bayer Immuno 1 PSA and cPSA methods at East Texas Medical Center. The resulting PSA data were then stratified by decade of age to determine age-related reference ranges for each PSA species. The cPSA values increased across all age decades: 40 to 49 years, 1.45 ng/mL; 50 to 59, 1.92 ng/mL; 60 to 69, 2.49 ng/mL; and 70 to 79, 2.77 ng/mL (95th percentile). tPSA levels also increased with age: 1.81 ng/mL, 2.45 ng/mL, 3.17 ng/mL, and 3.57, respectively. Comparatively, the c/tPSA levels remained constant (0.87), regardless of age. The upper limits of tPSA and cPSA values reported here suggest that men should be screened using lower cutoff values than are currently in use. These limits may more accurately identify prostate cancer among otherwise healthy men.

Complexed prostate-specific antigen for early detection of prostate cancer in men with serum prostate-specific antigen levels of 2 to 4 nanograms per milliliter

Complexed PSA (cPSA) has been shown to improve specificity in the detection of prostate cancer over that of total PSA (tPSA) testing in men with tPSA values greater than the cutoff value of 4.0 ng/mL. However, recent studies have reported a 25% incidence of prostate cancer in men with tPSA values in the 2.5- to 4.0-ng/mL range. We performed a multicenter study of cPSA in a population of men who underwent prostate biopsies because of elevated PSA levels or abnormal digital rectal examination (DRE). As part of this study, we sought to assess the clinical value of cPSA in comparison to tPSA, the free/tPSA ratio (f/tPSA) and the complexed/tPSA ratio (c/tPSA) in early detection of prostate cancer in men with tPSA values in the range of 2 to 4 ng/mL. The study was performed at 7 centers. Sera were drawn from men who underwent biopsy procedures consisting of >10 prostate tissue cores. Receiver operating characteristic (ROC) analysis was performed from the results of patients with tPSA values in the range of 2 to 4 ng/mL, including men with suspicious as well as unremarkable findings on DRE. Sera were collected and tested with the Bayer tPSA and cPSA assay and the Beckman free PSA and tPSA assays. ROC analysis was performed for all samples in the 2- to 4-ng/mL PSA range. At biopsy, 158 men had no evidence of malignancy and 57 (26.5%) were diagnosed with prostate cancer. ROC analysis indicated that the area under the curve (AUC) for cPSA was 0.64, which was statistically significantly greater than that achieved for tPSA (AUC, 0.57; P <0.0001). The AUC for f/tPSA and c/tPSA were 0.60 and 0.63, respectively, which was not statistically significantly different from that of tPSA or cPSA (P >or=0.252). A cutpoint of 2.5 ng/mL for tPSA and 2.1 ng/mL for cPSA provided a specificity of 20.3% and 34.2%, respectively, and sensitivity levels of 86%. Using cutpoints of 25% for f/tPSA and 74% for c/tPSA provided a specificity of 11.0% and 21.5%, respectively, and sensitivity levels of 97%. In all, >92% of the cancers treated with radical prostatectomy were organ confined, and the histologic grading of the tumors ranged from moderately to poorly differentiated with Gleason scores from 5 to 9. These data confirm that there is a high incidence of clinically significant prostate cancer in men with tPSA levels <4.0 ng/mL. Measurement of cPSA proved useful in stratifying men with tPSA values in the 2- to 4-ng/mL range into high- and low-risk groups for prostate cancer. The use of cPSA as a single test was found to enhance detection of prostate cancer over that of testing with tPSA and PSA ratios in men with tPSA values in the range of 2 to 4 ng/mL.

Intraindividual variation of PSA, free PSA and complexed PSA in a cohort of patients with prostate cancer managed with watchful observation

OBJECTIVE

To determine the intraindividual variation of prostate-specific antigen (PSA) isoforms in prostate cancer patients managed conservatively with watchful observation.

METHODS

Patients with favorable clinical parameters (stage T1b-T2b N0 M0, Gleason score </= 7, and PSA </= 15) were recruited to participate in a watchful observation program. Specimens were drawn for measurement of total (tPSA), free (fPSA) and complexed (cPSA) prostate-specific antigen isoforms. Total biologic variation and between-day analytical variation were used to calculate intraindividual variation.

RESULTS

Total variation for each isoform and two ratios were not greatly affected by the time window for measurements in the interval 6 months to 2.7 yr. Analytical variation made only a small contribution to the total biologic variation. Intraindividual variation for a 1-yr time interval for tPSA, fPSA, cPSA and the ratios of fPSA and cPSA to tPSA was, respectively, 21.6, 19.3, 25.4, 20.0 and 13.1%. The amount of change required for a significant difference between two readings (with 95% confidence) was, respectively, 59.8, 53.4, 70.4, 55.3 and 36.2%.

CONCLUSIONS

There is a significantly higher intraindividual variation of cPSA (25.4%), and a significantly lower intraindividual variation of the ratio cPSA to tPSA (13.1%) compared to the other individual PSA isoforms and to the ratio of fPSA to tPSA. The amount of change required for a significant difference between two concentrations is large for all variables studied, but the lowest is the ratio of cPSA to tPSA (36.2%). These results have significance for diagnosis and monitoring of patients with prostate cancer.