An artificial intelligence-driven support tool for prediction of urine culture test results.

BACKGROUND AND AIMS: We aimed to develop an easily deployable artificial intelligence (AI)-driven model for rapid prediction of urine culture test results. MATERIAL AND METHODS: We utilized a training dataset (n = 34,584 urine samples) and two separate, unseen test sets (n = 10,083 and 9,289 samples). Various machine learning models were compared for diagnostic performance. Predictive parameters included urinalysis results (dipstick and flow cytometry), patient demographics (age and gender), and sample collection method. RESULTS: Although more complex models achieved the highest AUCs for predicting positive cultures (highest: multilayer perceptron (MLP) with AUC of 0.884, 95% CI 0.878-0.89), multiple logistic regression (MLR) using only flow cytometry parameters achieved a very good AUC (0.858, 95% CI 0.852-0.865). To aid interpretation, prediction results of the MLP and MLR models were categorized based on likelihood ratio (LR) for positivity: highly unlikely (LR 0.1), unlikely (LR 0.3), grey zone (LR 0.9), likely (LR 5.0), and very likely (LR 40). This resulted in 17%, 28%, 32%, 10%, and 13% of samples falling into each respective category for the MLR model and 20%, 27%, 31%, 7%, and 16% for the MLP model. CONCLUSIONS: In conclusion, this robust model has the potential to assist clinicians in their decision-making process by providing insights prior to the availability of urine culture results in a significant portion of samples (∼2/3rd).

DeltaMSI: artificial intelligence-based modeling of microsatellite instability scoring on next-generation sequencing data.

BACKGROUND: DNA mismatch repair deficiency (dMMR) testing is crucial for detection of microsatellite unstable (MSI) tumors. MSI is detected by aberrant indel length distributions of microsatellite markers, either by visual inspection of PCR-fragment length profiles or by automated bioinformatic scoring on next-generation sequencing (NGS) data. The former is time-consuming and low-throughput while the latter typically relies on simplified binary scoring of a single parameter of the indel distribution. The purpose of this study was to use machine learning to process the full complexity of indel distributions and integrate it into a robust script for screening of dMMR on small gene panel-based NGS data of clinical tumor samples without paired normal tissue. METHODS: Scikit-learn was used to train 7 models on normalized read depth data of 36 microsatellite loci in a cohort of 133 MMR proficient (pMMR) and 46 dMMR tumor samples, taking loss of MLH1/MSH2/PMS2/MSH6 protein expression as reference method. After selection of the optimal model and microsatellite panel the two top-performing models per locus (logistic regression and support vector machine) were integrated into a novel script (DeltaMSI) for combined prediction of MSI status on 28 marker loci at sample level. Diagnostic performance of DeltaMSI was compared to that of mSINGS, a widely used script for MSI detection on unpaired tumor samples. The robustness of DeltaMSI was evaluated on 1072 unselected, consecutive solid tumor samples in a real-world setting sequenced using capture chemistry, and 116 solid tumor samples sequenced by amplicon chemistry. Likelihood ratios were used to select result intervals with clinical validity. RESULTS: DeltaMSI achieved higher robustness at equal diagnostic power (AUC = 0.950; 95% CI 0.910-0.975) as compared to mSINGS (AUC = 0.876; 95% CI 0.823-0.918). Its sensitivity of 90% at 100% specificity indicated its clinical potential for high-throughput MSI screening in all tumor types. Clinical Trial Number/IRB B1172020000040, Ethical Committee, AZ Delta General Hospital.

Serum free light chain analysis: persisting limitations with new kids on the block.

OBJECTIVES: Serum free light chain (sFLC) measurements have inherent analytical limitations impacting sFLC clinical interpretation. We evaluated analytical and diagnostic performance of three polyclonal sFLC assays on four analytical platforms. METHODS: sFLC concentration was measured using Diazyme FLC assays (Diazyme) on cobas c501/c503 analyzer (Roche); Freelite assays (The Binding Site) on Optilite analyzer (The Binding Site) and cobas c501 analyzer and Sebia FLC ELISA assays (Sebia) on AP22 ELITE analyzer (DAS). Imprecision, linearity, method comparison vs. Freelite/Optilite, antigen excess detection and reference value verification were assessed. Diagnostic performance was compared on 120 serum samples and on follow-up samples of five patients with κ and λ monoclonal gammopathy. RESULTS: Method comparison showed excellent correlation with Freelite/Optilite method for all assays. A large proportional negative bias was shown for both Sebia κ and λ ELISA and a significant positive proportional bias for λ in the low (<10 mg/L) Freelite/cobas c501 method. Clinically relevant underestimation of κ sFLC levels due to antigen excess was shown for 7% of each Diazyme/cobas application and for 11 and 32.1% of λ sFLC assay of respectively Diazyme/cobas and Sebia/AP22. sFLC reference values revealed application specific. Cohen's κ values were (very) good for κ sFLC but only moderate to good for λ sFLC. In 4/10 follow-up patients, significant differences in clinical interpretation between sFLC assays were noticed. CONCLUSIONS: Important analytical limitations remain for all sFLC applications. Differences in reference values and diagnostic performance hamper interchangeability of sFLC assays. Assay specific sFLC decision guidelines are warranted.

Pitfalls of SARS-CoV-2 antigen testing at emergency department.

BACKGROUND: Current method for diagnosis of SARS-CoV-2 infection is an RT-PCR test on the nasopharyngeal or oropharyngeal swab. Rapid diagnosis is essential for containing viral spread and triage of symptomatic patients presenting to hospital ER departments. As a faster alternative to RT-PCR, we evaluated a SARS-Cov-2 Rapid Antigen test in symptomatic patients presenting to hospital ER departments. METHODS: We evaluated the diagnostic performance of the Roche SARS-CoV-2 Rapid Antigen test (SD Biosensor) for detection of SARS-CoV-2 compared to RT-PCR. RESULTS: Our study showed inferior performance of the SARS-CoV-2 Rapid Antigen test for detection of SARS-CoV-2. Firstly, because of the lack of specificity, which is potentially life-threatening due to the association of nosocomial-acquired SARS-CoV-2 infection. Secondly, with a sensitivity of 45.5%, it is impossible to rule out SARS-CoV-2 infection, resulting in reflex PCR-testing. Comparison of viral load in RT-PCR positive samples with corresponding antigen results showed a significant difference between antigen positive and negative samples. COVID-19 infection will not be detected in patients admitted to the hospital in an early or late phase, typically associated with low viral loads. Sensitivity increases when testing within 5-7 symptomatic days, but the implementation of this cut-off is impractical in ER settings. However, diagnostic performance is better to detect high viral load (> = 5 log10 copies/mL) linked with contagiousness. CONCLUSION: Our study showed inferior performance of the Roche SARS-CoV-2 Rapid Antigen test (SD Biosensor) for detection of SARS-CoV-2 which limits its use as a diagnostic gatekeeper in ER departments, but is able to differentiate contagious individuals.

Cycle Threshold Probability Score for Immediate and Sensitive Detection of B.1.351 SARS-CoV-2 Lineage.

BACKGROUND: Detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants of concern associated with immune escape is important to safeguard vaccination efficacy. We describe the potential of delayed N gene amplification in the Allplex SARS-CoV-2 Assay (Seegene) for screening of the B.1.351 (20H/501.V2, variant of concern 2 [VOC.V2], South African SARS-CoV-2 variant) lineage. METHODS: In a study cohort of 397 consecutive polymerase chain reaction-positive samples genotyped by whole-genome sequencing, amplification curves of E/N/S-RdRP targets indicated delayedN vs E gene amplification characteristic of B.1.351. Logistic regression was used to calculate a VOC.V2 probability score that was evaluated as a separate screening test in an independent validation cohort vs sequencing. RESULTS: B.1.351 showed a proportionally delayed amplification of the  N vs E gene. In logistic regression, only N and E gene cycle thresholds independently contributed to B.1.351 prediction, allowing calculation of a VOC.V2 probability score with an area under the curve of 0.94. At an optimal dichotomous cutoff point of 0.12, the VOC.V2 probability score achieved 98.7% sensitivity at 79.9% specificity, resulting in a negative predictive value (NPV) of 99.6% and a positive predictive value of 54.6%. The probability of B.1.351 increased with an increasing VOC.V2 probability score, achieving a likelihood ratio of 12.01 above 0.5. A near-maximal NPV was confirmed in 153 consecutive validation samples. CONCLUSIONS: Delayed N vs E gene amplification in the Allplex SARS-CoV-2 Assay can be used for fast and highly sensitive screening of B.1.351.

The MicroRNA Landscape of Acute Beta Cell Destruction in Type 1 Diabetic Recipients of Intraportal Islet Grafts.

Ongoing beta cell death in type 1 diabetes (T1D) can be detected using biomarkers selectively discharged by dying beta cells into plasma. microRNA-375 (miR-375) ranks among the top biomarkers based on studies in animal models and human islet transplantation. Our objective was to identify additional microRNAs that are co-released with miR-375 proportionate to the amount of beta cell destruction. RT-PCR profiling of 733 microRNAs in a discovery cohort of T1D patients 1 h before/after islet transplantation indicated increased plasma levels of 22 microRNAs. Sub-selection for beta cell selectivity resulted in 15 microRNAs that were subjected to double-blinded multicenter analysis. This led to the identification of eight microRNAs that were consistently increased during early graft destruction: besides miR-375, these included miR-132/204/410/200a/429/125b, microRNAs with known function and enrichment in beta cells. Their potential clinical translation was investigated in a third independent cohort of 46 transplant patients by correlating post-transplant microRNA levels to C-peptide levels 2 months later. Only miR-375 and miR-132 had prognostic potential for graft outcome, and none of the newly identified microRNAs outperformed miR-375 in multiple regression. In conclusion, this study reveals multiple beta cell-enriched microRNAs that are co-released with miR-375 and can be used as complementary biomarkers of beta cell death.

Serum 25(OH)D Level on Hospital Admission Associated With COVID-19 Stage and Mortality.

OBJECTIVES: Vitamin D deficiency was previously correlated with incidence and severity of coronavirus disease 2019 (COVID-19). We investigated the association between serum 25-hydroxyvitamin D (25(OH)D) level on admission and radiologic stage and outcome of COVID-19 pneumonia. METHODS: A retrospective observational trial was done on 186 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected individuals hospitalized from March 1, 2020, to April 7, 2020, with combined chest computed tomography (CT) and 25(OH)D measurement on admission. Multivariate regression analysis was performed to study if vitamin D deficiency (25(OH)D <20 ng/mL) correlates with survival independently of confounding comorbidities. RESULTS: Of the patients with COVID-19, 59% were vitamin D deficient on admission: 47% of females and 67% of males. In particular, male patients with COVID-19 showed progressively lower 25(OH)D with advancing radiologic stage, with deficiency rates increasing from 55% in stage 1 to 74% in stage 3. Vitamin D deficiency on admission was not confounded by age, ethnicity, chronic lung disease, coronary artery disease/hypertension, or diabetes and was associated with mortality (odds ratio [OR], 3.87; 95% confidence interval [CI], 1.30-11.55), independent of age (OR, 1.09; 95% CI, 1.03-1.14), chronic lung disease (OR, 3.61; 95% CI, 1.18-11.09), and extent of lung damage expressed by chest CT severity score (OR, 1.12; 95% CI, 1.01-1.25). CONCLUSIONS: Low 25(OH)D levels on admission are associated with COVID-19 disease stage and mortality.

Diagnostic Performance of Chest CT for SARS-CoV-2 Infection in Individuals with or without COVID-19 Symptoms.

Background The use of chest CT for coronavirus disease 2019 (COVID-19) diagnosis or triage in health care settings with limited severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) polymerase chain reaction (PCR) capacity is controversial. COVID-19 Reporting and Data System (CO-RADS) categorization of the level of COVID-19 suspicion might improve diagnostic performance. Purpose To investigate the value of chest CT with CO-RADS classification to screen for asymptomatic SARS-CoV-2 infections and to determine its diagnostic performance in individuals with COVID-19 symptoms during the exponential phase of viral spread. Materials and Methods In this secondary analysis of a prospective trial, from March 2020 to April 2020, parallel SARS-CoV-2 PCR and CT with categorization of COVID-19 suspicion was performed with CO-RADS for individuals with COVID-19 symptoms and control participants without COVID-19 symptoms admitted to the hospital for medical emergencies unrelated to COVID-19. CT with CO-RADS was categorized on a five-point scale from 1 (very low suspicion) to 5 (very high suspicion). Area under the receiver operating curve (AUC) was calculated in symptomatic versus asymptomatic individuals to predict positive SARS-CoV-2 PCR, and likelihood ratios for each CO-RADS score were used for rational selection of diagnostic thresholds. Results A total of 859 individuals (median age, 70 years; interquartile range, 52-81 years; 443 men) with COVID-19 symptoms and 1138 control participants (median age, 68 years; interquartile range, 52-81 years; 588 men) were evaluated. CT with CO-RADS had good diagnostic performance (P < .001) in both symptomatic (AUC, 0.89) and asymptomatic (AUC, 0.70) individuals. In symptomatic individuals (42% PCR positive), CO-RADS 3 or greater detected positive PCR with high sensitivity (89%, 319 of 358) and specificity of 73%. In asymptomatic individuals (5% PCR positive), a CO-RADS score of 3 or greater detected SARS-CoV-2 infection with low sensitivity (45%, 27 of 60) but high specificity (89%). Conclusion CT with Coronavirus Disease 2019 Reporting and Data System (CO-RADS) had good diagnostic performance in symptomatic individuals, supporting its application for triage. Sensitivity in asymptomatic individuals was insufficient to justify its use as a first-line screening approach. Incidental detection of CO-RADS 3 or greater in asymptomatic individuals should trigger testing for respiratory pathogens. © RSNA, 2020 Online supplemental material is available for this article.

Humoral Immune Response to SARS-CoV-2.

OBJECTIVES: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) serology tests are clinically useful to document prior SARS-CoV-2 infections. Data are urgently needed to select assays with optimal sensitivity at acceptable specificity for antibody detection. METHODS: A comparative evaluation was performed of 7 commercial SARS-CoV-2 serology assays on 171 sera from 135 subjects with polymerase chain reaction-confirmed SARS-CoV-2 infection (71 hospitalized patients and 64 paucisymptomatic individuals). Kinetics of IgA/IgM/IgG seroconversion to viral N and S protein epitopes were studied from 0 to 54 days after onset of symptoms. Cross-reactivity was verified on 57 prepandemic samples. RESULTS: Wantai SARS-COV-2 Ab ELISA and Orient Gene COVID-19 IgG/IgM Rapid Test showed superior overall sensitivity for detection of SARS-CoV-2 antibodies. Elecsys Anti-SARS-CoV-2 assay and EUROIMMUN Anti-SARS-CoV-2 combined IgG/IgA showed acceptable sensitivity (>95%) vs the consensus result of all assays from 10 days post onset of symptoms. Wantai SARS-COV-2 Ab ELISA, Elecsys Anti-SARS-CoV-2 assay, and Innovita 2019-nCoV Ab rapid test showed least cross-reactivity, resulting in an optimal analytical specificity greater than 98%. CONCLUSIONS: Wantai SARS-COV-2 Ab ELISA and Elecsys Anti-SARS-CoV-2 assays are suitable for sensitive and specific detection of SARS-CoV-2 antibodies from 10 days after onset of symptoms.

Analytical and clinical performance evaluation of two POC tests for therapeutic drug monitoring of infliximab.

Background Infliximab (IFX) is an effective therapy in patients with inflammatory bowel disease. Serum IFX trough concentrations correlate well with clinical, biological and endoscopic outcomes. Therefore, therapeutic drug monitoring (TDM) of infliximab is useful for dose optimization and prevention of secondary treatment failure. In the present study, analytical and clinical performance of two point-of-care (POC) tests, RIDA®QUICK IFX Monitoring assay (R-biopharm) and Quantum Blue® Infliximab assay (Bühlmann), have been evaluated and compared to our established enzyme-linked immunosorbent assay (ELISA) (apDia IFX ELISA). Methods Analytical performance was assessed according to the CLSI EP5-A2 protocol using the manufacturer's kit controls and different serial dilution series. Method comparison with our established ELISA was done using a wide range of consecutive patient samples (n=180). Clinical concordance was evaluated by categorization based on well-known therapeutic cut-off points (3-7 µg/mL). Results The analytical performance of both POC tests was inferior to the established ELISA, but acceptable based on the manufacturer's quality claims. Eight-point serial dilution confirmed the analytical performance data in the low-level measuring range. Eleven-point serial dilution demonstrated linearity for both POC tests over the studied concentration range. Method comparison with the ELISA showed significant negative proportional bias for the RIDA®QUICK IFX Monitoring assay. However, good correlation and clinical concordance were shown. Quantum Blue® Infliximab assay showed a significant positive proportional and a negative systematic bias in comparison with the ELISA, resulting in overestimation of IFX levels with impact on clinical concordance data. Conclusions Both POC tests have their own specific benefits and drawbacks but are suitable for therapeutic drug monitoring of IFX. However, long-term monitoring of IFX trough levels requires measurement of IFX concentrations with the same assay.

Combined Analysis of GAD65, miR-375, and Unmethylated Insulin DNA Following Islet Transplantation in Patients With T1D.

Aim: Several biomarkers have been proposed to detect pancreatic ß cell destruction in vivo but so far have not been compared for sensitivity and significance. Methods: We used islet transplantation as a model to compare plasma concentrations of miR-375, 65-kDa subunit of glutamate decarboxylase (GAD65), and unmethylated insulin DNA, measured at subpicomolar sensitivity, and study their discharge kinetics, power for outcome prediction, and detection of graft loss during follow-up. Results: At 60 minutes after transplantation, GAD65 and miR-375 consistently showed near-equimolar and correlated increases proportional to the number of implanted ß cells. GAD65 and miR-375 showed comparable power to predict poor graft outcome at 2 months, with areas under the curve of 0.833 and 0.771, respectively (P = 0.53). Using receiver operating characteristic analysis, we defined likelihood ratios (LRs) for rationally selected result intervals. In GADA-negative recipients (n = 28), GAD65 <4.5 pmol/L (LR = 0.15) and >12.2 pmol/L (LR = ∞) predicted good and poor outcomes, respectively. miR-375 could be used in all recipients irrespective of GAD65 autoantibody status (n = 46), with levels <1.4 pmol/L (LR = 0.14) or >7.6 pmol/L (LR = 9.53) as dual thresholds. The posttransplant surge of unmethylated insulin DNA was inconsistent and unrelated to outcome. Combined measurement of these three biomarkers was also tested as liquid biopsy for ß cell death during 2-month follow-up; incidental surges of GAD65, miR-375, and (un)methylated insulin DNA, alone or combined, were confidently detected but could not be related to outcome. Conclusions: GAD65 and miR-375 performed equally well in quantifying early graft destruction and predicting graft outcome, outperforming unmethylated insulin DNA.

Performance Evaluation of Serum Free Light Chain Analysis: Nephelometry vs Turbidimetry, Monoclonal vs Polyclonal Reagents.

OBJECTIVES: Free light chain (FLC) measurement gained a lot of interest for diagnostic workup of monoclonal gammopathy. METHODS: We evaluated the performance of turbidimetric polyclonal Freelite (The Binding Site, Birmingham, UK) assays on Cobas 6000 (Roche Diagnostics, Rotkreuz, Switzerland) and nephelometric monoclonal N Latex (Siemens Healthcare Diagnostics, Marburg, Germany) assays on BN ProSpec (Dade Behring, Deerfield, IL) vs established nephelometric Freelite assays on BN ProSpec. RESULTS: Analytical performance was acceptable. Method comparison (n = 118) showed significant proportional FLC differences for N Latex assays. However, good correlation and clinical concordance were shown. Recovery study in the low concentration range demonstrated consistent over- and underrecovery for Freelite reagents, hampering future research on prognostic value of suppressed noninvolved FLC. Antigen excess detection was successful for κ FLC in three-fourths of cases with Freelite reagents and in all cases with N Latex reagents. However, the latter resulted in underestimated κ FLC concentrations. CONCLUSIONS: FLC analysis requires continuous awareness of analytical limitations. Monitoring of disease response requires FLC analysis on the same platform using the same reagents.

Analytical and diagnostic performance of two automated fecal calprotectin immunoassays for detection of inflammatory bowel disease.

BACKGROUND: We evaluated the (pre-)analytical and diagnostic performance of two automated fecal calprotectin (FC) immunoassays, Liaison® Calprotectin (Diasorin) on Liaison® XL and fCAL™ turbo (Bühlmann laboratories AG) on Cobas C501 (Roche Diagnostics), and compared it with our established Bühlmann ELISA method. METHODS: Our study comprised 229 consecutive patients with clinical suspicion of inflammatory bowel disease (IBD). RESULTS: All assay related stool extraction procedures showed excellent correlation with the established method, but the new stool extraction devices tend to give higher results as compared with stool weight methods. Both automated assays demonstrated good performance in terms of precision (CVt≤8.1%), accuracy (bias≤6.7%) and total error (≤16.4%). Method comparison with established enzyme linked immunosorbent assay (ELISA) showed good correlation (rs>0.925), but regression analysis showed significant proportional differences. Diagnostic performance characteristics with regard to diagnosis of IBD were good and in line with other reports. In addition, we were able to show that optimization of manufacturer's cut-off and moreover, the introduction of a gray zone resulted in a significant increase of post-test probability. CONCLUSIONS: In conclusion, the newly developed stool extraction device protocols showed acceptable and comparable performance to the stool weight method. Overall, the automated Liaison® Calprotectin and fCAL™ turbo assay showed good analytical and diagnostic performance for detection of IBD.

Comparison of two immunoassays for measurement of faecal calprotectin in detection of inflammatory bowel disease: (pre)-analytical and diagnostic performance characteristics.

BACKGROUND: Symptoms of inflammatory bowel disease (IBD) and irritable bowel syndrome (IBS) can overlap. Faecal calprotectin has recently been established to be a non-invasive marker for neutrophilic intestinal inflammation. We compared two devices for extraction of faecal calprotectin. Based on these results, two immunoassays for measurement of faecal calprotectin were evaluated. METHODS: Samples were extracted using the Thermo Fisher extraction device (Thermo Fisher Scientific) and Smart Pep extraction device (Roche Diagnostics) and measured with the EliA Calprotectin immunoassay (Thermo Fisher Scientific) on ImmunoCAP 250. The performance of both assays was investigated by enrolling 183 consecutive patients (79 males, 104 females; median age 32 years) with clinical suspicion of IBD. Faecal calprotectin was measured using a recently launched immunoassay, EliA Calprotectin in comparison with an established immunochomatographic point-of-care-test (POCT, Quantum Blue Calprotectin; Bühlmann). Results were compared with endoscopic and histological findings. RESULTS: The use of the Thermo Fisher extraction device resulted in an underestimation of faecal calprotectin concentrations, especially in liquid stool samples. IBD was diagnosed in 51/183 patients (27.9%) [Crohn's disease (CD, n=37), ulcerative colitis (UC, n=14)]. After adjusting the optimal cut-off for detection of IBD using receiver operating curve analysis, a sensitivity of 94.1% and 90.2% and specificity of 87.9% and 90.9% for the EliA and POCT assay, respectively, were obtained. CONCLUSIONS: The Thermo Fisher device is not reliable for extraction of faecal calprotectin. The performance characteristics of the EliA Calprotectin assay are statistically equivalent to the Bühlmann POCT.

Enzymatic pyruvate measurement by Cobas 6000 open channel assay.

BACKGROUND: Blood pyruvate measurement in conjunction with lactic acid is useful for differentiating pyruvate dehydrogenase deficiencies from primary or secondary disorders of mitochondrial electron transport. METHODS: We evaluated the analytical performance of pyruvate measurement by an enzymatic open channel assay on a Roche Cobas 6000. RESULTS: The assay was linear from 0.07 to 0.50 mmol/L pyruvate. Total imprecision ranged from 15.7% to 7.1% at pyruvate levels of 0.08 to 0.31 mmol/L, respectively. Functional sensitivity was 0.07 mmol/L. The assay showed no interference by lipids or bilirubin, whereas haemolysis influenced pyruvate concentrations in a hemoglobin concentration-independent manner. Method comparison with patient samples (n = 41) showed that the Cobas 6000 enzymatic method correlated well (r2 = 0.930) with a similar enzymatic assay on a Cobas Mira platform and showed better accuracy in external control schemes. CONCLUSIONS: Enzymatic pyruvate measurement by a Cobas 6000 open channel shows satisfactory analytical performance. The assay can be integrated in the automated laboratory workflow and is always ready for use thanks to its on-board reagents.

Diagnostic potential of CD34+ cell antigen expression in myelodysplastic syndromes.

The World Health Organization introduced flow cytometry as an additional criterion for diagnosis of myelodysplastic syndromes (MDS). Aberrant antigen expression on bone marrow (BM) blasts may identify "low-grade MDS." This study aimed to examine differences in antigen expression on CD34+ BM cells between patients with MDS and those with secondary cytopenia. BM aspirates of 175 patients with cytopenia were classified as MDS or secondary cytopenia. Expression of stem cell antigens (CD34, CD133), myeloid antigens (CD13, CD33), B-cell antigens (CD19, CD10), growth factor receptors (CD117, CD123), and chemokine receptor (CD184) was examined. Thirty-two normal adults and 49 patients with CD34+ acute myeloid leukemia (AML) were also examined. High percentage of CD34+ cells, CD117 and CD123 overexpression, and abnormal CD45 expression on these cells are the best markers for MDS. These phenotypic aberrancies correlate with number of blasts and degree of dysplasia, and were similar to those in CD34+ AML, thus reflecting the relationship between these disorders.

Use of likelihood ratios improves interpretation of laboratory testing for pulmonary sarcoidosis.

Laboratory tests for pulmonary sarcoidosis (percentage lymphocytes and CD4/CD8 ratio in bronchoalveolar lavage fluid and serum angiotensin-converting enzyme activity) lack sensitivity and specificity. In a retrospective study of 153 subjects under suspicion of pulmonary sarcoidosis (36 cases and 117 patients with other diseases [control patients]), we defined likelihood ratios (LRs) for rationally selected result intervals of these tests, which improve clinical interpretation as compared with dichotomous interpretation based on a single cutoff value. By using logistic regression analysis, we further integrated the 3 individual tests into a unified algorithm that could rule out diagnosis in 57 (48.7%) of the 177 control subjects and confirm diagnosis in 12 (33%) of the 36 pulmonary sarcoidosis cases. We conclude that use of LRs improves interpretation of laboratory tests for pulmonary sarcoidosis. In addition, we present a prediction algorithm based on the combination of laboratory tests that helps clinicians confirm or exclude diagnosis in almost half of the study population.