Assessing the Clinical Impact of Prostate-Specific Antigen Assay Variability and Nonequimolarity: A Simulation Study Based on the Population of the United Kingdom

Verification of Harmonization of Serum Total and Free Prostate-Specific Antigen (PSA) Measurements and Implications for Medical Decisions

BACKGROUND

Previous studies have shown that the harmonization of prostate-specific antigen (PSA) assays remained limited even after the introduction of WHO International Standards. This information needs updating for current measuring systems (MS) and reevaluation according to established analytical performance specifications (APS) and the characteristics of antibodies used.

METHODS

Total (tPSA) and free (fPSA) PSA were measured in 135 and 137 native serum samples, respectively, by Abbott Alinity i, Beckman Access Dxl, Roche Cobas e801, and Siemens Atellica IM MSs. Passing-Bablok regression and difference plots were used to compare results from each MS to the all-method median values. Agreement among methods was evaluated against APS for bias derived from biological variation of the 2 measurands.

RESULTS

The median interassay CV for tPSA MSs (11.5%; 25-75th percentiles, 9.2-13.4) fulfilled the minimum APS goal for intermethod bias (15.9%), while the interassay CV for fPSA did not [20.4% (25-75th percentiles, 18.4-22.7) vs goal 17.6%]. Considering the all-method median value of each sample as reference, all tPSA MSs exhibited a mean percentage bias within the minimum goal. On the other hand, Alinity (+21.3%) and Access (-24.2%) were out of the minimum bias goal for fPSA, the disagreement explained only in minimal part by the heterogeneity of employed antibodies.

CONCLUSIONS

The harmonization among tPSA MSs is acceptable only when minimum APS are applied and necessitates further improvement. The marked disagreement among fPSA MSs questions the use of fPSA as a second-level test for biopsy referral.

Serum Prostate-Specific Antigen Testing for Early Detection of Prostate Cancer: Managing the Gap between Clinical and Laboratory Practice

BACKGROUND

Current clinical practice guidelines (CPGs) for early detection of prostate cancer recommend for clinical decision-making a personalized prostate-specific antigen (PSA)-based management to improve the risk-benefit ratio of the screening strategy. Some important critical issues regarding the PSA determination in the clinical framework are, however, still neglected in current guidelines and a major focus of recommendations on those aspects would be needed to improve their effectiveness.

CONTENT

Evidence sources in the available literature concerning the interchangeability of total PSA results measured with different commercial methods were critically appraised. We discuss how the heterogeneity of the measurand, the intermethod bias, and the design and selectivity of immunoassays may affect the diagnostic accuracy of selected PSA thresholds, and how knowledge of the analytical characteristics of assays in service, such as the recognized PSA circulating forms and the cross-reactivity with PSA homologs, is basic for improving both clinical decision-making in cancer screening and the reliability of the clinical interpretation of results at the individual level.

SUMMARY

Current CPGs ignore the poor interchangeability of PSA results obtained from different assays and the substantial role of laboratory issues in clinical performance of PSA testing. Involved stakeholders should contribute to fill the existing gap by: (a) preparing commutable reference materials for immunoassay calibration; (b) providing analytical characteristics that may explain the different performance of assays; (c) deriving outcome-based analytical performance specifications for PSA measurement; and (d) giving more focus on laboratory items when CPGs are prepared.

Interchangeability and Diagnostic Accuracy of Two Assays for Total and Free Prostate-Specific Antigen: Two not Always Related Items

The variation between different PSA assays seems to influence the interpretation of individual PSA values and the clinical decisions about prostate cancer. One reason for this variability could be the different reactivity of antibodies for the various molecular forms of serum PSA; as a result, samples containing the same amount of tPSA but different proportions of fPSA can produce very different values. In this study, serum samples were collected prospectively from 152 consecutive patients referred to 2 institutions (Regional Hospital, Venice, 90 subjects; San Bortolo Hospital, Vicenza, 62 subjects) for PSA elevation and/or symptoms. Serum samples were assessed according to the manufacturers’ instructions on the following 2 analyzers: the Immulite 2000 assay (Diagnostic Products Corporation, Los Angeles, USA), which measures tPSA and fPSA, and the ADVIA Centaur (Bayer Diagnostics, Tarrytown, USA), which assays tPSA and cPSA. cPSA values were transformed into fPSA by the equation fPSA=tPSA-cPSA. When taking Immulite tPSA and f/tPSA values as 100%, ADVIA Centaur values were 92.6% and 122%, respectively, which means that 20% of patients would be classified differently according to the traditional biopsy cutoff. In conclusion, there are considerable differences between the 2 methods, which could affect clinical decisions.

Clinical impact of prostate specific antigen (PSA) inter-assay variability on management of prostate cancer

PURPOSE

To evaluate the inter-assay variability of six commercially available prostate specific antigen (PSA) assays, its clinical impact in prostate cancer (PCa) and comparison of automated versus manual assays.

PATIENTS AND METHODS

Sera from 495 patients (425 with PCa and 70 men with Benign Prostatic Hyperplasia (BPH), were measured with six different assays [three automated assays (a-PSA) and three manual ELISA based assay (m-PSA)]. Variability, agreement and bias were measured and compared among assays using Bland Altman plots and Passing and Bablok regression analysis. The possible impact of inter-assay variability on important clinical scenarios was also studied.

RESULTS

All the assays were well correlated (r: 0.88-0.98); however there was significant disagreement and bias between the systems, which were more pronounced among the a-PSA assays. The Bland Altman plot showed that the variability was high between the m-PSA assays and the standard Abbott system with mean difference of 3.8-5.8ng/ml. In contrast, the a-PSA had better agreement with mean difference of 0.8-2.3ng/ml. Beckman Coulter showed the best agreement to the institutional reference (slope-1.097; 95% CI: 1.06-1.14; p<0.05, and intercept-0.20; 95% CI-0.38-0.58; p<0.05, Passing Bablok). It led to significant variability in PCa risk stratification and failure to detect biochemical failure in more than 50% cases.

CONCLUSIONS

The discrepancies between the assays lead to significant clinical misinterpretation with risk group migration and detection of biochemical failure post radiotherapy. There are significant discordances between automated and ELISA based assays.

Variability of assay methods for total and free PSA after WHO standardization

The variability of total PSA (tPSA) and free PSA (fPSA) results among commercial assays has been suggested to be decreased by calibration to World Health Organization (WHO) reference materials. To characterize the current situation, it is necessary to know its impact in the critical cutoffs used in clinical practice. In the present study, we tested 167 samples with tPSA concentrations of 0 to 20 μg/L using seven PSA and six fPSA commercial assays, including Access, ARCHITECT i2000, ADVIA Centaur XP, IMMULITE 2000, Elecsys, and Lumipulse G1200, in which we only measured tPSA. tPSA and fPSA were measured in Access using the Hybritech and WHO calibrators. Passing-Bablok analysis was performed for PSA, and percentage of fPSA with the Hybritech-calibrated access comparison assay. For tPSA, relative differences were more than 10 % at 0.2 μg/L for ARCHITECT i2000, and at a critical concentration of 3, 4, and 10 μg/L, the relative difference was exceeded by ADVIA Centaur XP and WHO-calibrated Access. For percent fPSA, at a critical concentration of 10 %, the 10 % relative difference limit was exceeded by IMMULITE 2000 assay. At a critical concentration of 20 and 25 %, ADVIA Centaur XP, ARCHITECT i2000, and IMMULITE 2000 assays exceeded the 10 % relative difference limit. We have shown significant discordances between assays included in this study despite advances in standardization conducted in the last years. Further harmonization efforts are required in order to obtain a complete clinical concordance.

Roadmap for harmonization of clinical laboratory measurement procedures

Results between different clinical laboratory measurement procedures (CLMP) should be equivalent, within clinically meaningful limits, to enable optimal use of clinical guidelines for disease diagnosis and patient management. When laboratory test results are neither standardized nor harmonized, a different numeric result may be obtained for the same clinical sample. Unfortunately, some guidelines are based on test results from a specific laboratory measurement procedure without consideration of the possibility or likelihood of differences between various procedures. When this happens, aggregation of data from different clinical research investigations and development of appropriate clinical practice guidelines will be flawed. A lack of recognition that results are neither standardized nor harmonized may lead to erroneous clinical, financial, regulatory, or technical decisions. Standardization of CLMPs has been accomplished for several measurands for which primary (pure substance) reference materials exist and/or reference measurement procedures (RMPs) have been developed. However, the harmonization of clinical laboratory procedures for measurands that do not have RMPs has been problematic owing to inadequate definition of the measurand, inadequate analytical specificity for the measurand, inadequate attention to the commutability of reference materials, and lack of a systematic approach for harmonization. To address these problems, an infrastructure must be developed to enable a systematic approach for identification and prioritization of measurands to be harmonized on the basis of clinical importance and technical feasibility, and for management of the technical implementation of a harmonization process for a specific measurand.

Taking a new biomarker into routine use--a perspective from the routine clinical biochemistry laboratory

There is increasing pressure to provide cost-effective healthcare based on “best practice.” Consequently, new biomarkers are only likely to be introduced into routine clinical biochemistry departments if they are supported by a strong evidence base and if the results will improve patient management and outcome. This requires convincing evidence of the benefits of introducing the new test, ideally reflected in fewer hospital admissions, fewer additional investigations and/or fewer clinic visits. Carefully designed audit and cost-benefit studies in relevant patient groups must demonstrate that introducing the biomarker delivers an improved and more effective clinical pathway. From the laboratory perspective, pre-analytical requirements must be thoroughly investigated at an early stage. Good stability of the biomarker in relevant physiological matrices is essential to avoid the need for special processing. Absence of specific timing requirements for sampling and knowledge of the effect of medications that might be used to treat the patients in whom the biomarker will be measured is also highly desirable. Analytically, automation is essential in modern high-throughput clinical laboratories. Assays must therefore be robust, fulfilling standard requirements for linearity on dilution, precision and reproducibility, both within- and between-run. Provision of measurements by a limited number of specialized reference laboratories may be most appropriate, especially when a new biomarker is first introduced into routine practice.

Clinical Impact of New Prostate-Specific Antigen WHO Standardization on Biopsy Rates and Cancer Detection

Background: Clinicians may be unaware that replacement of the historical total prostate-specific antigen (tPSA) standard with the WHO 96/670 international standard leads to difficulties in interpreting tPSA results. Our aim was to investigate the relationship between the Hybritech and WHO calibrations of the Beckman Coulter tPSA assay, and to assess the impact on prostate cancer (PCa) detection.

Methods: tPSA concentrations were measured in 106 serum samples with both Hybritech and WHO calibrations. The established relationships were used for an in silico experiment with a cohort of 5865 men. Differences in prostate biopsy rates, PCa detection, and characteristics of missed cancers were calculated at biopsy thresholds of 3.0 and 4.0 μg/L.

Results: A linear relationship was observed between the 2 calibrations, with a 20.3% decrease in tPSA values with the WHO standard compared with the Hybritech calibration. Applying the WHO calibration to the cohort of 5865 men yielded a 20% or 19% decrease in prostate biopsies and a 19% or 20% decrease in detected cancers compared with the Hybritech calibration, at a cutoff for biopsy of 3.0 or 4.0 μg/L, respectively. The decrease in detected cancers declined to 9% or 11% if an abnormal result in a digital rectal examination or a transrectal ultrasound evaluation was used as trigger for prostate biopsy (cutoff of 3.0 or 4.0 μg/L, respectively).

Conclusions: Application of the WHO standard for tPSA assays with commonly used tPSA thresholds leads to a significant decrease in PCa detection. Careful assessment of the relationship between the WHO standard and the thresholds used for prostate biopsy is hence necessary.

Toward metrological traceability in the determination of prostate-specific antigen (PSA): calibrating Beckman Coulter Hybritech Access PSA assays to WHO standards compared with the traditional Hybritech standards

BACKGROUND

The metrological traceability of prostate-specific antigen (PSA) assay calibration to WHO standards is desirable to potentially improve the comparability between PSA assays. A method comparison was performed between the traditionally standardized Beckman Coulter Hybritech Access PSA and free PSA (fPSA) assays and a new alternate calibration of assays aligned to the WHO standards 96/670 and 96/668, respectively.

METHODS

Sera from 641 men with and without prostate cancer, various control materials and mixtures of different proportions of the WHO standards were measured with both assay calibrations.

RESULTS

Excellent comparability between the corresponding assay calibrations was observed, with correlation coefficients of at least 0.996. The Passing-Bablok slopes were 0.747 for total PSA (tPSA), 0.776 for fPSA and 1.02 for the percentage ratio of fPSA to tPSA (%fPSA), while the corresponding percentages of the new WHO-aligned assay results related to the traditional assays were 76.2%, 77% and 102.2%. Receiver operating characteristics revealed no differences between the two PSA assay calibrations.

CONCLUSIONS

The WHO calibration yields results approximately 25% lower for tPSA and fPSA values when compared with the conventional Hybritech calibration. Using the WHO-aligned PSA assay, a tPSA cut-off of 3 microg/L should be considered in clinical practice, while %fPSA cut-offs could be retained.

Major inter-laboratory variations in PSA testing practices: results from national surveys in Ireland in 2006 and 2007

BACKGROUND

Ireland had the highest prostate cancer incidence in Europe in 2006. In that year, the National Cancer Forum (NCF) recommended against prostate specific antigen (PSA) testing for population-based screening.

AIMS

To investigate (1) PSA services and (2) impact of the NCF recommendation.

METHODS

Questionnaires were dispatched to biochemistry laboratories nationwide in 2006 and 2007.

RESULTS

All 55 laboratories responded in 2006; 33/36 (89%) responded in 2007. 36 laboratories measured total PSA (tPSA); 14 measured free PSA (fPSA). Laboratories with higher tPSA workload were more likely to measure fPSA (P = 0.024). A total of 15 laboratories used age-specific PSA ranges. In 2006, there were [382,000 tPSA and [48,000 fPSA tests costing an estimated euro 4,900,000. During 2006-2007 tPSA tests increased by 11%; fPSA tests decreased by 36%.

CONCLUSIONS

There is considerable inter-laboratory variation in PSA testing practices. Because of the potential clinical consequences, standardisation should be considered. Testing practice was unaffected by the NCF recommendation.

National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines for use of tumor markers in clinical practice: quality requirements

BACKGROUND

This report presents updated National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines summarizing quality requirements for the use of tumor markers.

METHODS

One subcommittee developed guidelines for analytical quality relevant to serum and tissue-based tumor markers in current clinical practice. Two other subcommittees formulated recommendations particularly relevant to the developing technologies of microarrays and mass spectrometry.

RESULTS

Prerequisites for optimal use of tumor markers in routine practice include formulation of the correct clinical questions to ensure selection of the appropriate test, adherence to good clinical and laboratory practices (e.g., minimization of the risk of incorrect patient and/or specimen identification, tube type, or timing), use of internationally standardized and well-characterized methods, careful adherence to manufacturer instructions, and proactive and timely reactions to information derived from both internal QC and proficiency-testing specimens. Highly desirable procedures include those designed to minimize the risk of the reporting of erroneous results attributable to interferences such as heterophilic antibodies or hook effects, to facilitate the provision of informative clinical reports (e.g., cumulative and/or graphical reports, appropriately derived reference intervals, and interpretative comments), and when possible to integrate these reports with other patient information through electronic health records. Also mandatory is extensive validation encompassing all stages of analysis before introduction of new technologies such as microarrays and mass spectrometry. Provision of high-quality tumor marker services is facilitated by dialogue involving researchers, diagnostic companies, clinical and laboratory users, and regulatory agencies.

CONCLUSIONS

Implementation of these recommendations, adapted to local practice, should encourage optimization of the clinical use of tumor markers.

The Clinical Impact of who Standardization of PSA Assays

The determination of serum level of the prostate specific antigen (PSA) is widely used for detection and management of prostate cancer. Analytical variability between the various PSA assays on the market has been reported. This discrepancy in the PSA results was shown to be related to non-equimolar detection of total PSA (tPSA) but also to a lack of assay standardization and could have serious clinical repercussions on the diagnostic performance of PSA testing. The recalibration of equimolar assays to common reference preparations (tPSA WHO 96/670 and fPSA 96/668) was thought to promote standardization of PSA assays and limit the clinical implication of assay variability. Comparison studies have demonstrated that PSA assay calibration to the WHO standard certainly improves the harmonisation of PSA testing, but differences between assays remain. Recent evaluations of the clinical impact of analytical variations induced by a calibration to the WHO standard reported that 15% to 30% of prostate cancer could be missed if the historical tPSA cut-off was used. In order to avoid unacceptable erosion of the clinical diagnostic performance of PSA determination for the detection of prostate cancer with WHO calibrated assays, it is critical to define new specific clinical decision points.

Improving the comparability of immunoassays for prostate-specific antigen (PSA): Progress and problems

BACKGROUND

Commutability of immunoassay test results is an important objective for laboratory medicine.

METHODS

PSA is a clinically important analyte for which, as a consequence of a number of national and international initiatives over the last decade, considerable progress has been made towards improving method comparability. However, results from different assays are still not interchangeable, a situation that is only likely to improve once broad recommendations can be made about the most clinically relevant antibody combinations.

CONCLUSIONS

Universal implementation of such recommendations would almost certainly improve between-method agreement substantially, provided careful attention were paid to assay design and use of appropriately pure secondary standards ensured.

Interchangeability of measurements of total and free prostate-specific antigen in serum with 5 frequently used assay combinations: an update

BACKGROUND

The comparability of total and free prostate-specific antigen (tPSA and fPSA) results among commercial PSA assays has been suggested to be improved by calibration to WHO PSA reference materials and the development of equimolar-response assays. To characterize the current situation, we assessed 5 frequently used commercial assay combinations for tPSA and fPSA regarding the interchangeability of the PSA values and the ratio of fPSA to tPSA (%fPSA), equimolar characteristics, and diagnostic accuracy.

METHODS

Sera from 314 patients with prostate cancer (PCa) and 282 men with no evidence of prostate cancer (NPCa) were measured with tPSA and fPSA assays from Abbott (AxSYM), Beckman Coulter (Access), Diagnostic Products Corporation (Immulite 2000), and Roche (Elecsys 2010) and with tPSA and complexed PSA (cPSA) assays from Bayer (ADVIA Centaur).

RESULTS

Method comparisons (Passing and Bablok regressions; Bland-Altman plots) showed assay-dependent results for tPSA, fPSA, and %fPSA. With the Access tPSA values taken as 100%, tPSA concentrations varied from 87% (AxSYM and ADVIA Centaur) to 115% (Immulite), leading to different numbers of patients classified according to the commonly recommended tPSA cutoffs for performing a biopsy. Different %fPSA values also led to assay-dependent ROC analysis results, a finding that shows the importance for the diagnostic accuracy.

CONCLUSION

Interchangeability of tPSA, fPSA, and %fPSA values obtained by commercial PSA assays remains inadequate, but attention to this issue may minimize the misinterpretation of PSA results obtained by different assays.