Volume 58, Issue 1

Background Shortcomings in study design have been hinted at as one of the possible causes of failures in the translation of discovered biomarkers into the care of ovarian cancer patients, but systematic assessments of biomarker studies are scarce. We aimed to document study design features of recently reported evaluations of biomarkers in ovarian cancer. Methods We performed a systematic search in PubMed (MEDLINE) for reports of studies evaluating the clinical performance of putative biomarkers in ovarian cancer. We extracted data on study designs and characteristics. Results Our search resulted in 1026 studies; 329 (32%) were found eligible after screening, of which we evaluated the first 200. Of these, 93 (47%) were single center studies. Few studies reported eligibility criteria (17%), sampling methods (10%) or a sample size justification or power calculation (3%). Studies often used disjoint groups of patients, sometimes with extreme phenotypic contrasts; 46 studies included healthy controls (23%), but only five (3%) had exclusively included advanced stage cases. Conclusions Our findings confirm the presence of suboptimal features in clinical evaluations of ovarian cancer biomarkers. This may lead to premature claims about the clinical value of these markers or, alternatively, the risk of discarding potential biomarkers that are urgently needed.

The appropriate identification of bacterial infection is the basis for effective treatment and control of infective disease. Among this context, an emerging biomarker of infection is presepsin (PSP), recently described as early marker of different infections. PSP secretion has been shown to be associated with monocyte phagocytosis and plasmatic levels of PSP increase in response to bacterial infection and decrease after antibiotic treatment, therefore it can be considered a marker of activation of immune cell response towards an invading pathogen. Different methods have been developed to measure PSP and this review will briefly describe the different clinical fields of application of PSP, ranging from intensive care to neonatal infection, to orthopedic and pulmonary infection as well as fungal infections and cardiovascular infections.

As the largest immune organ, human gut microbiome could influence the efficacy of immune checkpoint inhibitor therapy (ICI). However, identifying contributory microbes from over 35,000 species is virtually impossible and the identified microbes are not consistent among studies. The reason for the disparity may be that the microbes found in feces are markers of other factors that link immune response and microbiotas. Notably, gut microbiome is influenced by stool consistency, diet and other lifestyle factors. Therefore, the ICI and microbiotas relationship must be adjusted for potential confounders and analyzed longitudinally. Moreover, a recent study where 11 low-abundance commensal bacteria induced interferon-γ-producing CD8 T cells, challenges the validity of the abundance-oriented microbiotas investigations. This study also confirmed the hierarchy in immunogenic roles among microbiotas. Fecal transplantation trials in germ-free mice provided “the proof of principle” that germ-free mice reproduce the donor’s microbiome and corresponding ICI efficacy. However, species-specific biological differences prevent direct extrapolation between the results in murine and human models. Fecal transplantation or supplementation with microbes found in ICI responders requires caution due to potential adverse events.

Background Interpretation of the complete blood count (CBC) parameters requires reliable biological variation (BV) data. The aims of this study were to appraise the quality of publications reporting BV data for CBC parameters by applying the BV Data Critical Appraisal Checklist (BIVAC) and to deliver global BV estimates based on BIVAC compliant studies. Methods Relevant publications were identified by a systematic literature search and evaluated for their compliance with the 14 BIVAC criteria, scored as A, B, C or D, indicating decreasing compliance. Global CV I and CV G estimates with 95% CI were delivered by a meta-analysis approach using data from BIVAC compliant papers (grades A–C). Results In total, 32 studies were identified; four received a BIVAC grade A, 2 B, 20 C and 6 D. Meta-analysis derived CV I and CV G estimates were generally lower or in line with those published in a historical BV database available online. Except for reticulocytes, CV I estimates of erythrocyte related parameters were below 3%, whereas platelet (except MPV and PDW) and leukocyte related parameters ranged from 5% to 15%. Conclusions A systematic review of CBC parameters has provided updated, global estimates of CV I and CV G that will be included in the newly published European Federation of Clinical Chemistry and Laboratory Medicine BV Database.

Background Measuring 24 h-urine calcium concentration is essential to evaluate calcium metabolism and excretion. Manufacturers recommend acidifying the urine before a measurement to ensure calcium solubility, but the literature offers controversial information on this pre-analytical treatment. The objectives of the study were (1) to compare pre-acidification (during urine collection) versus post-acidification (in the laboratory), and (2) to evaluate the impact of acidification on urinary calcium measurements in a large cohort. Methods We evaluated the effects of pre- and post-acidification on 24-h urine samples collected from 10 healthy volunteers. We further studied the impact of acidification on the calcium results for 567 urine samples from routine laboratory practice, including 46 hypercalciuria (≥7.5 mmol/24 h) samples. Results Calciuria values in healthy volunteers ranged from 0.6 to 12.5 mmol/24 h, and no statistical significance was found between non-acidified, pre-acidified and post-acidified conditions. A comparison of the values (ranging from 0.21 to 29.32 mmol/L) for 567 urine samples before and after acidification indicated 25 samples (4.4%) with analytical differences outside limits of acceptance. The bias observed for these deviant values ranged from −3.07 to 1.32 mmol/L; no patient was re-classified as hypercalciuric after acidification, and three patients with hypercalciuria were classified as normocalciuric after acidification. These three deviant patients represent 6.5% of hypercalciuric patients. Conclusions Our results indicate that pre- and post-acidification of urine is not necessary prior to routine calcium analysis.

Background There is a need for additional biomarkers to assist in the diagnosis and prognosis of rheumatoid arthritis (RA). The aim of our study was to evaluate the (pre-analytical, analytical and clinical) performance of serum calprotectin as a marker of inflammation in RA. Methods The study population included 463 rheumatologic patients (including 111 RA patients and 352 controls) who for the first time consulted a rheumatologist, 20 healthy controls and 27 patients with an infectious disease. Calprotectin was measured (using four different assays) in serum or in serum and EDTA plasma (healthy controls and infectious disease group). For rheumatologic patients, results for C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), rheumatoid factor (RF) and anti-cyclic citrullinated peptide antibody (ACPA) were available. Results Results for blood calprotectin were assay- and matrix-dependent, with higher values found in serum than in plasma. Serum calprotectin was higher in RA patients than in rheumatologic diseased controls and in healthy controls. Serum calprotectin was lower in RA patients than in patients with an infectious disease. Serum calprotectin was associated with disease activity (DAS score). The area under the curve (AUC) to discriminate RA from controls was 0.756 for CRP, 0.714 for ESR and 0.726–0.783 for calprotectin. Conclusions Our data document that calprotectin measurement is assay- and matrix-dependent. Serum calprotectin is associated with disease activity. Additional (prospective) studies are warranted to further evaluate the prognostic and diagnostic value of blood calprotectin measurements.

Background We investigated the dynamics and the predictive value of soluble syndecan-1 (Sdc-1), a biomarker of endothelial dysfunction, in uneventful pregnancies and pregnancies complicated by preeclampsia (PE). Methods Serum levels of Sdc-1 were measured at sequential time points during and after uneventful pregnancies (control, n = 95) and pregnancies developing PE (PE_long, n = 12). Levels were further measured in women with symptomatic PE (PE_state, n = 46) at a single time point. Results Sdc-1 levels increased consistently throughout pregnancy. In the PE_long group Sdc-1 levels were lower at all visits throughout pregnancy, and reached significance in weeks 18–22 (p = 0.019), 23–27 (p = 0.009), 28–32 (p = 0.006) and 33–36 (p = 0.008). After delivery, Sdc-1 levels dropped sharply in all pregnancies but were significantly elevated in the PE_long group. The predictive power of Sdc-1 was evaluated analyzing receiver operating characteristic (ROC) curves. A significant power was reached at weeks 14–17 (area under the curve [AUC] 0.65, p = 0.025), 23–27 (AUC 0.73, p = 0.004) and 33–36 (AUC 0.75, p = 0.013). Conclusions In summary, Sdc-1 levels were lower in women developing PE compared to uneventful pregnancies and Sdc-1 might be useful to predict PE. After delivery, Sdc-1 levels remained higher in women with PE. Additional studies investigating the link between glycocalyx degradation, Sdc-1 levels and placental and endothelial dysfunction in pregnancies affected by PE are warranted.

Background The Atellica Solution comprises chemistry (CH) and immunoassay (IM) analyzers. Recently, six early adopter clinical laboratories across Europe evaluated the analytical performance of 20 CH and IM assays. To measure analytical performance quality, Sigma metrics were calculated for individual-site and pooled-site results. Methods Precision, detection capability, linearity, and method comparison studies were performed according to Clinical Laboratory Standards Institute protocols. Global Sigma metrics across sites were calculated from pooled data at the medical decision level using total allowable error (TEa) goals from CLIA for CH assays, and TEa goals from RiliBÄK for IM assays; and, the equation: Sigma metrics=%TEa–%bias/%CV. A pooled %CV was calculated by combining the imprecision obtained from individual sites. Bias calculations were performed against the ADVIA Chemistry system or ADVIA Centaur system using Deming regression analysis (Passing-Bablok regression for electrolytes) on the pooled-site data. The 103 individual-site Sigma metric calculations used individual-site imprecision and pooled-bias. Results The limits of blank and detection results agreed with the manufacturer’s claims. Most assays were linear across the assay range tested. Pooled Sigma metrics were good or better (>4 Sigma) for 18 of 20 assays; and, acceptable for urea nitrogen (3.1) and sodium (3.9), the latter values attributable to higher imprecision at one of five sites. Conclusions Sigma metrics for data generated across multiple real-world sites evaluating the Atellica Solution demonstrated good or better performance of greater than 4 Sigma for 18 of 20 assays tested. Overall, results verified the manufacturer’s claims that methods were fit for use in clinical laboratories.

Background Monthly medians of patient results are useful in assessment of analytical quality in medical laboratories. Separate medians by gender makes it possible to generate two independent estimates of contemporaneous errors. However, for plasma creatinine, reference intervals (RIs) are different by gender and also higher over 70 years of age. Methods Daily, weekly and monthly patient medians were calculated from the raw data of plasma creatinine concentrations for males between 18 and 70 years, males >70 years, females between 18 and 70 years and females >70 years. Results The medians of the four groups were all closely associated, with similar patterns. The mean of percentage bias from each group defined the best estimate of bias. The maximum half-range (%) of the bias evaluations provided an estimate of the uncertainty comparable to the analytical performance specifications: thus, bias estimates could be classified as optimum, desirable or minimum quality. Conclusions Medians by gender and age are useful in assessment of analytical stability for plasma creatinine concentration ranging from 60 to 90 μmol/L. The daily medians are valuable in rapid detection of large systematic errors, the weekly medians in detecting minor systematic errors and monthly medians in assessment of long-term analytical stability.

Background This two-center study was designed to verify comparability of procalcitonin (PCT) values among 10 different commercial immunoassays. Methods A total number of 176 routine lithium-heparin plasma samples were divided in identical aliquots and simultaneously analyzed with 10 different PCT immunoassays, including Kryptor BRAHMS PCT sensitive, Abbott Architect BRAHMS PCT, Beckman Coulter Access PCT (on Access and DXI), BioMérieux Vidas BRAHMS PCT, Diasorin Liaison BRAHMS PCT, Fujirebio Lumipulse G BRAHMS PCT, Roche BRAHMS PCT (on Cobas E801), Diazyme PCT (on Roche Cobas C702) and SNIBE Maglumi PCT. Results Highly significant correlation was always found across multiple comparisons, with correlation coefficients comprised between 0.918 and 0.997 (all p < 0.001). Bland and Altman plots analysis revealed highly variable bias among immunoassays, ranging between ±0.2% and ±38.6%. Diazyme PCT on Roche Cobas C702 and SNIBE Maglumi PCT displayed the larger overestimation, whilst PCT values were underestimated by Cobas BRAHAMS PCT. The agreement was always >80% (all p < 0.001), but varied largely across multiple comparisons, ranging between 90%–99% at 0.1 μg/L, 81%–99% at 0.25 μg/L, 83%–100% at 0.5 μg/L, 94%–100% at 2.0 μg/L and 90%–99% at 10 μg/L, respectively. The larger disagreement was observed comparing Diazyme PCT and Maglumi PCT with the other methods. Conclusions Although we found acceptable correlation among 10 commercial PCT immunoassays, the limited agreement at clinical decision thresholds remains a major issue, especially at lower end of PCT concentration, thus potentially contributing to jeopardize the clinical value of this biomarker.

Background Serum free light chain (sFLC) measurements are increasingly important in the context of screening for monoclonal gammopathies, prognostic stratification and monitoring of therapy responses. In this study we have performed a method comparison of four sFLC assays that are currently available for routine clinical use. Methods In a retrospective study, sFLC analyses were performed on a cohort that included 139 patients with various monoclonal gammopathies and 54 control sera without an M-protein. Method comparisons of the following four FLC assays were performed: Freelite (Binding Site), N-Latex FLC (Siemens), Seralite (Abingdon Health) and Sebia FLC (Sebia). Results Bland-Altman agreement analysis showed biases varying between −0.1 and 16.2 mg/L for κFLC, −6.0 and 6.8 mg/L for λFLC and −0.04 and 0.38 for the ratio of the involved to uninvolved FLC. Strong agreements were observed for FLC-concentrations below 100 mg/L. The clinical concordance of the κ/λFLC-ratio of the four methods varied between 86% and 92%. Significant quantitative differences were observed between the different methods, mainly in sera with high FLC concentrations. Most assays consistently overestimated FLC concentrations compared to SPE. Conclusions Good overall clinical concordances were observed between the four sFLC assays that were compared in this study. Although good agreements were observed between the FLC assays, significant absolute differences in FLC concentrations in individual patients can be seen, particularly at higher FLC concentrations. Because of inequivalent absolute sFLC values between the methods in individual patients, none of the four sFLC assays can be used interchangeably.

Background This study aimed to quantify and compare serum aldosterone (sALD) levels through three different chemiluminescence immunoassays (CLIAs) and liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis. Methods Serum samples from 221 patients with suspected primary aldosteronism (PA) were retrospectively included in this study conducted at the Peking Union Medical College Hospital from June to August in 2017. sALD levels were determined using the LC-MS/MS method and three different CLIA systems, viz. , DiaSorin ® XL, iSYS and Auto Lumo A2000. Pooled fresh serum samples were used for recalibration. Passing-Bablok regression analysis, correlation matrix, and Bland-Altman plots were used to evaluate the concurrence among ALD levels determined using the three CLIAs. Results Within-laboratory precision of the four assays ranged from 2.1% to 9.4%, except the coefficient variation (CV) of one of the CLIAs, which exceeded 20.0% for samples with low sALD levels. sALD levels determined using LC-MS/MS were significantly lower than those determined using the other three CLIAs (p < 0.0001). Spearman’s correlation coefficient of the four assays ranged from 0.745 to 0.950 (p < 0.0001). The Bland-Altman plot showed that the average bias (%) for the three CLIAs and LC-MS/MS ranged from −69.3 to −49.2. After recalibration, this correlation did not improve among the assays. However, the bias and bias percentage at the medical decision level improved between LC-MS/MS and DiaSorin ® XL/iSYS. Conclusions Significant inconsistencies between the results of CLIAs and LC-MS/MS indicate that different sALD measures cannot be used interchangeably.

Background Characterization of lipoprotein particle profiles (LPPs) (including main classes and subclasses) by means of ultracentrifugation (UC) is highly requested given its clinical potential. However, rapid methods are required to replace the very labor-intensive UC method and one solution is to calibrate rapid nuclear magnetic resonance (NMR)-based prediction models, but the reliability of the UC-response method required for the NMR calibration has been largely overlooked. Methods This study provides a comprehensive repeatability and reproducibility study of various UC-based lipid measurements (cholesterol, triglycerides [TGs], free cholesterol, phospholipids, apolipoprotein [apo]A1 and apoB) in different main classes and subclasses of 25 duplicated fresh plasma samples and of 42 quality control (QC) frozen pooled plasma samples of healthy individuals. Results Cholesterol, apoA1 and apoB measurements were very repeatable in all classes (intraclass correlation coefficient [ICC]: 92.93%–99.54%). Free cholesterol and phospholipid concentrations in main classes and subclasses and TG concentrations in high-density lipoproteins (HDL), HDL subclasses and low-density lipoproteins (LDL) subclasses, showed worse repeatability (ICC: 19.21%–99.08%) attributable to low concentrations, variability introduced during UC and assay limitations. On frozen QC samples, the reproducibility of cholesterol, apoA1 and apoB concentrations was found to be better than for the free cholesterol, phospholipids and TGs concentrations. Conclusions This study shows that for LPPs measurements near or below the limit of detection (LOD) in some of the subclasses, as well as the use of frozen samples, results in worsened repeatability and reproducibility. Furthermore, we show that the analytical assay coupled to UC for free cholesterol and phospholipids have different repeatability and reproducibility. All of this needs to be taken into account when calibrating future NMR-based models.

Background Reference intervals (RIs) transference can expand the applicability of established RIs. However, the study on transference methodology is insufficient, and RIs validation based on small samples cannot adequately identify transferred risk under complex situations. This study aimed to find appropriate conditions to ensure the effect of transference. Methods We established the RIs of Roche and Beckman systems for 27 analytes based on 681 healthy individuals. Roche RIs were converted into the Beckman RIs using linear regression (least squares method) which is divided into two methods – Method ref (500 test numbers with relatively narrow data range) and Method ep (80 test numbers with relatively wide data range). Taking the RIs established by Beckman results as standard, we assessed the accuracy, precision and trueness of transferred results under various conditions. Results A total of 29.6% and 48.1% of analytes were consistent between the two systems for the lower and upper reference limits, respectively. The concordance rates between transferred and measured RIs for Method ref were up to 74.1% and 92.6%, which were better than Method ep (44.4% and 59.3%). The CV of transferred reference limits decreased gradually with increasing test number under the same data range. For most analytes, excluding some electrolyte tests, we could obtain accurate results when r > 0.800 and the test number was sufficient regardless of the regression equation types. Conclusions Transferability of RIs is affected by many factors, such as correlation, test number, regression equation type, and quality requirement. To reduce the risk of transference, it is very important to select right method with reasonable conditions.

Background Differentiated thyroid carcinoma (DTC) has an excellent prognosis and the role of high-sensitive thyroglobulin measured during levothyroxine (ON-T4 hs-Tg) testing to discriminate patients with structural from not-evidence of disease (SED and NED, respectively) has been investigated. We conducted a systematic review and meta-analysis evaluating the performance of ON-T4 hs-Tg in two clinical scenarios considering its negative predictive value (NPV) as primary outcome: (1) diagnostic performance of Tg when undetectable value and NED status are simultaneously demonstrated; (2) prognostic performance of undetectable Tg in predicting NED in the subsequent follow-up. Methods This systematic review and meta-analysis were registered on PROSPERO (CRD42019125092). PubMed, Scopus, CENTRAL and Web of Science were searched until February 12, 2019. Original articles reporting data on SED/NED in patients with detectable/undetectable ON-T4 hs-Tg were selected. Summary operating points were estimated using a random-effects model. Results Out of 1413 retrieved papers, 10 studies evaluating 1796 patients were included. Participants were outpatients diagnosed with DTC, treated with near-total (NTx) or total thyroidectomy (TTx) with or without radioactive iodine (RAI). The NPV of ON-T4 hs-Tg for diagnostic and prognostic performance was 99.4% (95% CI 98.9–99.9; I 2  = 13%) and 99.4% (95% CI 98.8–100; I 2  = 0%), respectively. Conclusions Our findings show that ON-T4 hs-Tg is an excellent diagnostic tool and prognostic factor to rule-out SED. A high level of evidence is provided to decrease the intensity and frequency of follow-up in those DTC patients having undetectable high-sensitive Tg.

Background Highly sensitive cardiac troponin assays (hs-cTn) are not available as point-of-care (POC) measurements. As rapid testing cannot be achieved at the expense of clinical performance, there is an urgent need to develop and rigorously validate POC hs-cTn. Konica Minolta (KM) has recently developed a surface plasmon-field enhanced fluorescence spectroscopy-based POC hs-cTn I system. Methods We validated the analytical characteristics of the KM POC system according to the international guidelines. Results Limit of blank (LoB) and limit of detection (LoD) were 0.35 and 0.62 ng/L, respectively, hs-cTn I concentrations corresponding to a total CV of 20%, 10% and 5% were 1.5, 3.9 and 11.0 ng/L, respectively. Method comparison studies showed that KM calibration was successfully traced to higher-order references. Limit of quantitation (LoQ), i.e. the hs-cTn I concentration having a total error of measurement of ≤34%, was 10.0 ng/L. The upper reference limit (URL) for 600 healthy blood donors was calculated at 12.2 ng/L (90% confidence interval [CI]: 9.2–39.2), while sex-partitioned URLs were 20.6 (males) and 10.7 ng/L (females), respectively (p < 0.0001). KM assay measured hs-cTn I concentrations >LoD in 65.7% of all reference individuals, in 76.7% of males and in 54.7% of females, respectively. Conclusions The KM system joins the characteristics of POC systems to the analytical performance of hs-cTn.