Machine Learning-Based Prediction of Hemoglobinopathies Using Complete Blood Count Data.

BACKGROUND: Hemoglobinopathies, the most common inherited blood disorder, are frequently underdiagnosed. Early identification of carriers is important for genetic counseling of couples at risk. The aim of this study was to develop and validate a novel machine learning model on a multicenter data set, covering a wide spectrum of hemoglobinopathies based on routine complete blood count (CBC) testing. METHODS: Hemoglobinopathy test results from 10 322 adults were extracted retrospectively from 8 Dutch laboratories. eXtreme Gradient Boosting (XGB) and logistic regression models were developed to differentiate negative from positive hemoglobinopathy cases, using 7 routine CBC parameters. External validation was conducted on a data set from an independent Dutch laboratory, with an additional external validation on a Spanish data set (n = 2629) specifically for differentiating thalassemia from iron deficiency anemia (IDA). RESULTS: The XGB and logistic regression models achieved an area under the receiver operating characteristic (AUROC) of 0.88 and 0.84, respectively, in distinguishing negative from positive hemoglobinopathy cases in the independent external validation set. Subclass analysis showed that the XGB model reached an AUROC of 0.97 for ß-thalassemia, 0.98 for α0-thalassemia, 0.95 for homozygous α+-thalassemia, 0.78 for heterozygous α+-thalassemia, and 0.94 for the structural hemoglobin variants Hemoglobin C, Hemoglobin D, Hemoglobin E. Both models attained AUROCs of 0.95 in differentiating IDA from thalassemia. CONCLUSIONS: Both the XGB and logistic regression model demonstrate high accuracy in predicting a broad range of hemoglobinopathies and are effective in differentiating hemoglobinopathies from IDA. Integration of these models into the laboratory information system facilitates automated hemoglobinopathy detection using routine CBC parameters.

Hemoglobinopathy screening in primary care in the Netherlands: exploring the problems and needs of patients and general practitioners.

The prevalence of hemoglobinopathies in The Netherlands is increasing due to migration. Hemoglobinopathies are severe hereditary diseases. An informed reproductive choice by at-risk couples, such as pre-implantation diagnosis or termination of affected pregnancies, can be made if carriers are detected prior to conception. Using a qualitative design, the needs and wishes of patients, carriers and general practitioners were evaluated regarding carrier detection of hemoglobinopathies in primary care practice. 30 semi-structured interviews were established with 10 general practitioners, 10 patients and 10 carriers. The interviews were audio-recorded, transcribed verbatim and analysed using content analysis to identify recurring themes. Three themes were generated regarding carrier detection of hemoglobinopathies: (1) a need for more information about hemoglobinopathy, (2) a need for indications when to refer for analysis (carrier diagnostics) and (3) insight concerning organization and roles in care for hemoglobinopathy carriers and patients. These themes reflected a need to increase awareness of hemoglobinopathy, improve competences among general practitioners through better education and improvement of communication with patients and their unidentified family members. This study shows the scope of the problem and the critical need for action to improve informed reproductive decision making for the at-risk population.

Evaluation of in silico predictors on short nucleotide variants in HBA1, HBA2, and HBB associated with haemoglobinopathies.

Haemoglobinopathies are the commonest monogenic diseases worldwide and are caused by variants in the globin gene clusters. With over 2400 variants detected to date, their interpretation using the American College of Medical Genetics and Genomics (ACMG)/Association for Molecular Pathology (AMP) guidelines is challenging and computational evidence can provide valuable input about their functional annotation. While many in silico predictors have already been developed, their performance varies for different genes and diseases. In this study, we evaluate 31 in silico predictors using a dataset of 1627 variants in HBA1, HBA2, and HBB. By varying the decision threshold for each tool, we analyse their performance (a) as binary classifiers of pathogenicity and (b) by using different non-overlapping pathogenic and benign thresholds for their optimal use in the ACMG/AMP framework. Our results show that CADD, Eigen-PC, and REVEL are the overall top performers, with the former reaching moderate strength level for pathogenic prediction. Eigen-PC and REVEL achieve the highest accuracies for missense variants, while CADD is also a reliable predictor of non-missense variants. Moreover, SpliceAI is the top performing splicing predictor, reaching strong level of evidence, while GERP++ and phyloP are the most accurate conservation tools. This study provides evidence about the optimal use of computational tools in globin gene clusters under the ACMG/AMP framework.

A Novel Tool for the Analysis and Detection of Copy Number Variants Associated with Haemoglobinopathies.

Several types of haemoglobinopathies are caused by copy number variants (CNVs). While diagnosis is often based on haematological and biochemical parameters, a definitive diagnosis requires molecular DNA analysis. In some cases, the molecular characterisation of large deletions/duplications is challenging and inconclusive and often requires the use of specific diagnostic procedures, such as multiplex ligation-dependent probe amplification (MLPA). Herein, we collected and comprehensively analysed all known CNVs associated with haemoglobinopathies. The dataset of 291 CNVs was retrieved from the IthaGenes database and was further manually annotated to specify genomic locations, breakpoints and MLPA probes relevant for each CNV. We developed IthaCNVs, a publicly available and easy-to-use online tool that can facilitate the diagnosis of rare and diagnostically challenging haemoglobinopathy cases attributed to CNVs. Importantly, it facilitates the filtering of available entries based on the type of breakpoint information, on specific chromosomal and locus positions, on MLPA probes, and on affected gene(s). IthaCNVs brings together manually curated information about CNV genomic locations, functional effects, and information that can facilitate CNV characterisation through MLPA. It can help laboratory staff and clinicians confirm suspected diagnosis of CNVs based on molecular DNA screening and analysis.

The hemoglobinopathies, molecular disease mechanisms and diagnostics.

Hemoglobinopathies are the most common monogenic disorders in the world with an ever increasing global disease burden each year. As most hemoglobinopathies show recessive inheritance carriers are usually clinically silent. Programmes for preconception and antenatal carrier screening, with the option of prenatal diagnosis are considered beneficial in many endemic countries. With the development of genetic tools such as Array analysis and Next Generation Sequencing in addition to state of the art screening at the hematologic, biochemic and genetic level, have contributed to the discovery of an increasing number of rare rearrangements and novel factors influencing the disease severity over the recent years. This review summarizes the basic requirements for adequate carrier screening analysis, the importance of genotype-phenotype correlation and how this may lead to the unrevealing exceptional interactions causing a clinically more severe phenotype in otherwise asymptomatic carriers. A special group of patients are ß-thalassemia carriers presenting with features of ß-thalassemia intermedia of various clinical severity. The disease mechanisms may involve duplicated α-globin genes, mosaic partial Uniparental Isodisomy of chromosome 11p15.4 where the HBB gene is located or haplo-insufficiency of a non-linked gene SUPT5H on chromosome 19q, first described in two Dutch families with ß-thalassemia trait without variants in the HBB gene.

Hemoglobinopathy prevention in primary care: a reflection of underdetection and difficulties with accessibility of medical care, a quantitative study.

As in most Northern European countries, the prevalence of hemoglobinopathies in The Netherlands is increasing due to migration. Although hemoglobinopathies are severe chronic diseases with few treatment options, timely detection of carriers allows at-risk couples to make informed reproductive choices such as pre-implantation diagnosis, prenatal diagnosis or termination of affected pregnancies. Using a quantitative design, we evaluated the prevalence of hemoglobinopathies in The Hague region, The Netherlands. Patient and carrier registries from hospital, laboratory and general practitioners allowed this quantitative analysis. The highest prevalence of hemoglobinopathies was seen in immigrant neighborhoods, and a large gap was noted between estimated carrier prevalence and the actual registration of carriers in electronic patient records. Carrier prevalence was estimated to be 13,704; however, the ELAN database contains only 1542 cases with ICPC codes for sickle cell disease or thalassemia. Although more research is needed to define the requirements of the healthcare system to address this challenge, this study clearly shows the gap between estimated carrier prevalence and registration and thereby the pressing need for action.

Adapting the ACMG/AMP variant classification framework: A perspective from the ClinGen Hemoglobinopathy Variant Curation Expert Panel.

Accurate and consistent interpretation of sequence variants is integral to the delivery of safe and reliable diagnostic genetic services. To standardize the interpretation process, in 2015, the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) published a joint guideline based on a set of shared standards for the classification of variants in Mendelian diseases. The generality of these standards and their subjective interpretation between laboratories has prompted efforts to reduce discordance of variant classifications, with a focus on the expert specification of the ACMG/AMP guidelines for individual genes or diseases. Herein, we describe our experience as a ClinGen Variant Curation Expert Panel to adapt the ACMG/AMP criteria for the classification of variants in three globin genes (HBB, HBA2, and HBA1) related to recessively inherited hemoglobinopathies, including five evidence categories, as use cases demonstrating the process of specification and the underlying rationale.

Cyanosis, hemolysis, decreased HbA1c and abnormal co-oximetry in a patient with hemoglobin M Saskatoon [HBB:c.190C > T p.His64Tyr].

We describe a first Dutch case of Hb M Saskatoon (HBB:c.190C > T p.His64Tyr) in a 47-year-old female Dutch patient who presented with cyanosis, hemolysis, and abnormal co-oximetry. A mean corpuscular volume (MCV) of 105 fL caused by reticulocytosis (160 × 109/L) and low red blood cell count (3.6 × 1012/L) suggested an increased erythrocyte turnover. An HPLC glyco-globin analysis revealed a decreased HbA1c fraction of 12.3 mmol/mmol, HbA0 of 93.3% and an additional unidentified fraction at 1.2 min. DNA sequencing revealed a missense mutation in the HBB gene, (HBB:c.190C > T p.His64Tyr), known as Hb M Saskatoon, a variant which has been previously identified as an unstable hemoglobin variant leading to methemoglobinemia and anemia. In this report, we describe the clinical and remarkable laboratory aspects of our patient with Hb M Saskatoon, and the consequences for treatment and drug use.

Recommendations for diagnosis and treatment of methemoglobinemia.

Methemoglobinemia is a rare disorder associated with oxidization of divalent ferro-iron of hemoglobin (Hb) to ferri-iron of methemoglobin (MetHb). Methemoglobinemia can result from either inherited or acquired processes. Acquired forms are the most common, mainly due to the exposure to substances that cause oxidation of the Hb both directly or indirectly. Inherited forms are due either to autosomal recessive variants in the CYB5R3 gene or to autosomal dominant variants in the globin genes, collectively known as HbM disease. Our recommendations are based on a systematic literature search. A series of questions regarding the key signs and symptoms, the methods for diagnosis, the clinical management in neonatal/childhood/adulthood period, and the therapeutic approach of methemoglobinemia were formulated and the relative recommendations were produced. An agreement was obtained using a Delphi-like approach and the experts panel reached a final consensus >75% of agreement for all the questions.

The Evolving Role of Next-Generation Sequencing in Screening and Diagnosis of Hemoglobinopathies.

During the last few years, next-generation sequencing (NGS) has undergone a rapid transition from a research setting to a clinical application, becoming the method of choice in many clinical genetics laboratories for the detection of disease-causing variants in a variety of genetic diseases involving multiple genes. The hemoglobinopathies are the most frequently found Mendelian inherited monogenic disease worldwide and are composed of a complex group of disorders frequently involving the inheritance of more than one abnormal gene. This review aims to present the role of NGS in both screening and pre- and post-natal diagnostics of the hemoglobinopathies, and the added value of NGS is discussed based on the results described in the literature. Overall, NGS has an added value in large-scale high throughput carrier screening and in the complex cases for which common molecular techniques have some inadequacies. It is proven that the majority of thalassemia cases and Hb variants can be diagnosed using routine analysis involving a combined approach of hematology, hemoglobin separation, and classical DNA methods; however, we conclude that NGS can be a useful addition to the existing methods in the diagnosis of these disorders.

Breakpoint characterization of a rare alpha0 -thalassemia deletion using targeted locus amplification on genomic DNA.

INTRODUCTION: The high-sequence homology of the α-globin-gene cluster is responsible for microhomology-mediated recombination events during meiosis, resulting in a high density of deletion breakpoints within a 10 kb region. Commonly used deletion detection methods, such as multiplex ligation-dependent probe amplification (MLPA) and Southern blot, cannot exactly define the breakpoints. This typically requires long-range PCR, which is not always successful. Targeted locus amplification (TLA) is a targeted enrichment method that can be used to sequence up to 70 kb of neighboring DNA sequences without prior knowledge about the target site. METHODS: Genomic DNA (gDNA) TLA is a technique that folds isolated DNA, ensuring that adjacent loci are in a close spatial proximity. Subsequent digestion and religation form DNA circles that are amplified using fragment-specific inverse primers, creating a library that is suitable for Illumina sequencing. RESULTS: Here, we describe the characterization of a rare 16 771 bp deletion, utilizing gDNA TLA with a single inverse PCR primer set on one end of the breakpoint. Primers for breakpoint PCR were designed to confirm the deletion breakpoints and were consequently used to characterize the same deletion in 10 additional carriers sharing comparable hematologic data and similar MLPA results. CONCLUSIONS: The gDNA TLA technology was successfully used to identify deletion breakpoints within the alpha-globin cluster. The deletion was described only once in an earlier study as the --gb , but as it was not registered correctly in the available databases, it was not initially recognized as such.

Hemoglobin Yamagata [ß132(H10)Lys→Asn; (HBB: c.399A>T)]: a mosaic to be put together.

OBJECTIVES: Artifactually altered glycated hemoglobin (HbA1c) concentrations are frequently linked to hemoglobin (Hb) variants. Their expression and detection require in-depth analysis. METHODS: Cation exchange high performance liquid chromatography (HPLC) (Bio-Rad Variant™ II; Trinity Biotech Premier Hb9210 Resolution), capillary electrophoresis (CE) (Sebia Capillarys 2 Flex Piercing) and mass spectrometry (MS) (Waters) were used for variant detection; Sanger sequencing, multiplex ligation-dependent probe amplification (MLPA) and next generation sequencing (NGS) were used for DNA analysis; HbA1c was measured with cation exchange HPLC (Bio-Rad Variant™ II; Arkray Adams HA-8180V; Tosoh HLC-723 G7), CE (Sebia Capillarys 2 Flex Piercing), boronate affinity HPLC (Trinity Biotech Hb9210 Premier), immunoassay (Cobas c501 Tina-quant HbA1c Gen. 3; Nihon Kohden CHM-4100 Celltac chemi HbA1c HA-411V) and enzymatic assay (Abbott Architect c 8000 HbA1c). RESULTS: Hb Yamagata [ß132(H10)Lys→Asn; (HBB: c.399A>T)] was identified in the proband by MS after the observation of an abnormal peak in HPLC and CE. A mosaic expression of this variant was detected by NGS (mutant: 8%; wild type: 92%), after negative results in Sanger sequencing. Hb Yamagata interfered with HbA1c measurements by cation exchange HPLC and CE whereas immuno and enzymatic assay values showed good agreement with boronate affinity HPLC measurement. CONCLUSIONS: A mosaicism of Hb Yamagata was found in a patient with altered HbA1c values. This rare gene variant was detected only by advanced technologies as MS and NGS. The variant interfered with common HbA1c determination methods.

Usefulness of NGS for Diagnosis of Dominant Beta-Thalassemia and Unstable Hemoglobinopathies in Five Clinical Cases.

Unstable hemoglobinopathies (UHs) are rare anemia disorders (RADs) characterized by abnormal hemoglobin (Hb) variants with decreased stability. UHs are therefore easily precipitating, causing hemolysis and, in some cases, leading to dominant beta-thalassemia (dBTHAL). The clinical picture of UHs is highly heterogeneous, inheritance pattern is dominant, instead of recessive as in more prevalent major Hb syndromes, and may occur de novo. Most cases of UHs are not detected by conventional testing, therefore diagnosis requires a high index of suspicion of the treating physician. Here, we highlight the importance of next generation sequencing (NGS) methodologies for the diagnosis of patients with dBTHAL and other less severe UH variants. We present five unrelated clinical cases referred with chronic hemolytic anemia, three of them with severe blood transfusion dependent anemia. Targeted NGS analysis was performed in three cases while whole exome sequencing (WES) analysis was performed in two cases. Five different UH variants were identified correlating with patients' clinical manifestations. Four variants were related to the beta-globin gene (Hb Bristol-Alesha, Hb Debrousse, Hb Zunyi, and the novel Hb Mokum) meanwhile one case was caused by a mutation in the alpha-globin gene leading to Hb Evans. Inclusion of alpha and beta-globin genes in routine NGS approaches for RADs has to be considered to improve diagnosis' efficiency of RAD due to UHs. Reducing misdiagnoses and underdiagnoses of UH variants, especially of the severe forms leading to dBTHAL would also facilitate the early start of intensive or curative treatments for these patients.

A roadmap for the standardization of hemoglobin A2.

BACKGROUND: Standardization of laboratory tests can be a long process, and this is the case with regards to the methods used to measure hemoglobin A2 (HbA2), an important marker for beta-thalassemia and other thalassemic conditions. The IFCC standardization project started in 2004, and it took at least 15 years before developing a reference measurement procedure, defining and producing calibrators and certified reference materials. METHODS: A series of steps have to be undertaken in order to promote the standardization in the field, a process involving a number of stakeholders (manufacturers, scientific societies, national health bodies, laboratory professionals, clinicians). In this work we describe some possible process indicators, in order to assure that the standardization will have internal and external validity and be effective for a long time. These indicators concern the inter-method studies, elaboration of External Quality Assessment Schemes, and the evaluation of the yearly distributions of HbA2 measurements collected in selected laboratories. RESULTS: Preliminary results are reported concerning the yearly distributions of HbA2, collected in two different locations, and using different analytical methods. Median yearly values were found very constant over the years, but different between methods. On the other side, results obtained on the same specimens using two different techniques, proved that results by capillary electrophoresis in 2 out of the 3 years of observation, were significantly lower than those by HPLC. CONCLUSION: In this document we report what has been done so far, and what has to be done to achieve the standardization of the measurement of HbA2 worldwide.

Further evaluation of the world health organization international reference reagent for Haemoglobin A2 measurement.

INTRODUCTION: The accurate measurement of HbA2 is essential for the detection of ß-thalassaemia carriers and as no single calibrant is used by the various manufacturers of analysers, differences are seen in results obtained. The World Health Organization International Reference Reagent for HbA2 (WHO IRR 89/666) was made available to diagnostic laboratories in the 1980s and remains the only international reference material available. A previous study (2015) demonstrated that the WHO IRR remained suitable for use as an HbA2 standard as tested by 52 participants in the UK NEQAS Haematology Abnormal Haemoglobins Programme. This study was undertaken to include simultaneous analysis of three whole blood specimens over a range of HbA2 values with the WHO IRR and to include participants from laboratories outside of the UK. METHOD: Three whole blood specimens with HbA2 levels ranging from 2.4% to 5.7% and the WHO IRR were distributed to 56 laboratories located in 14 different countries. Participants were requested to test the specimens at defined intervals and return results accompanied by chromatograms or electropherograms produced. RESULTS: Differences found in results from different analyser groups reflect the bias found in the 2015 study in that bias is seen according to the methodology used and also varies in relation to the level of analyte being measured. CONCLUSION: Results of measurements from whole blood specimens and the lyophilized WHO IRR standard did not show any deterioration of the IRR, and it remains suitable for use. Linearity and calibration of analysers remain a problem.

A Woman with Missing Hb A2 Due to a Novel (εγ)δß0-Thalassemia and a Novel δ-Globin Variant Hb A2-Gebenstorf (HBD: c.209G>A).

A woman completely lacking Hb A2 on the high performance liquid chromatography (HPLC) analysis, presented with a novel deletional (εγ)δß0-thal and a δ-globin gene variant. This combination causes a ß-thalassemia (ß-thal) minor phenotype. The woman was referred by a hematologist due to abnormal blood counts. Multiplex ligation-dependent probe amplification (MLPA) and microarray analysis showed a heterozygous, 177 kb long deletion that removed the locus control region enhancer plus the ε, Gγ and Aγ genes. Additional sequencing revealed a novel variant HBD: c.209G>A, p.Gly70Asp in the heterozygous state, called Hb A2-Gebenstorf. The combination of the two variants explains the lack of Hb A2 in this woman.