A machine-learning method for biobank-scale genetic prediction of blood group antigens.

A key element for successful blood transfusion is compatibility of the patient and donor red blood cell (RBC) antigens. Precise antigen matching reduces the risk for immunization and other adverse transfusion outcomes. RBC antigens are encoded by specific genes, which allows developing computational methods for determining antigens from genomic data. We describe here a classification method for determining RBC antigens from genotyping array data. Random forest models for 39 RBC antigens in 14 blood group systems and for human platelet antigen (HPA)-1 were trained and tested using genotype and RBC antigen and HPA-1 typing data available for 1,192 blood donors in the Finnish Blood Service Biobank. The algorithm and models were further evaluated using a validation cohort of 111,667 Danish blood donors. In the Finnish test data set, the median (interquartile range [IQR]) balanced accuracy for 39 models was 99.9 (98.9-100)%. We were able to replicate 34 out of 39 Finnish models in the Danish cohort and the median (IQR) balanced accuracy for classifications was 97.1 (90.1-99.4)%. When applying models trained with the Danish cohort, the median (IQR) balanced accuracy for the 40 Danish models in the Danish test data set was 99.3 (95.1-99.8)%. The RBC antigen and HPA-1 prediction models demonstrated high overall accuracies suitable for probabilistic determination of blood groups and HPA-1 at biobank-scale. Furthermore, population-specific training cohort increased the accuracies of the models. This stand-alone and freely available method is applicable for research and screening for antigen-negative blood donors.

Hippocampal sclerosis is associated with celiac disease type immunity in patients with drug-resistant temporal lobe epilepsy.

BACKGROUND: A prior small-scale single center study suggested an association between celiac disease (CD)-type immunity and refractory temporal lobe epilepsy (TLE) with hippocampal sclerosis (HS). The present study addresses this putative association in a large, well-characterized group of drug-resistant epilepsy (DRE) patients. These patients were grouped based on the spectrum of CD and gluten sensitivity-associated antibodies. METHODS: In this cross-sectional study, 253 consecutive adult epilepsy patients (135 females, 118 males; age 16-76 years) were categorized into three groups: (i) CD-positive group with either prior diagnosis of CD or CD-specific TG2/EmA antibodies, (ii) AGA-positive group with antigliadin antibodies (AGA) but without CD, and (iii) CD/AGA-negative group without any gluten sensitivity-associated antibodies or CD. Clinical and immunological findings were then compared among the groups. RESULTS: TLE with HS was more common in the CD-positive group compared to CD/AGA-negative group (31.8% versus 11.9%, P = 0.019). Autoimmune disorders were more common in the AGA-positive group than in the CD/AGA-negative group (P = 0.025). Considering HS lateralization; left lateralization was more common in CD-positive group compared to CD/AGA-negative group (71.4% versus 25%, P = 0.030). TG6 seropositivity did not differ among the groups (P > 0.05). CONCLUSIONS: This study provides further evidence linking TLE with HS and CD-type autoimmunity suggesting that CD-type immune response to gluten can be one potential mechanism as a disease modifier leading to DRE and HS. Understanding these immunological factors is imperative for developing immunomodulatory or dietary treatments for DRE potentially preventing HS progression.

A novel nonsense variant in RHAG underlies a Nordic Rhnull phenotype.

BACKGROUND AND OBJECTIVES: The extremely rare Rhnull phenotype is characterized by the absence of all Rh antigens on erythrocytes. It is divided into the regulator and amorph types based on the underlying genetic background. The more common regulator type depends on critical variants silencing RHAG, which encodes RhAG glycoprotein, necessary for RhD/RhCE expression. Rhnull cells have altered expression of glycophorin B and LW glycoprotein. MATERIALS AND METHODS: Four unrelated Rhnull individuals were investigated. Serological testing was performed according to standard blood bank practice. RHD/RHCE and S/s allele-specific Polymerase chain reaction (PCR) genotyping was done on genomic DNA using in-house PCR assays. RHAG, and in some cases also RHD/RHCE, were sequenced. Initial s phenotyping results triggered additional serological investigation. RESULTS: Anti-Rh29 was identified in all four individuals. Extended typing with anti-S and anti-s showed that the three samples predicted to type as s+ failed to react with 2 of 5 anti-s. Sequence analysis of all 10 RHAG exons and the immediate intron/exon boundaries revealed a single nucleotide variant in the 3'-end of intron 6, c.946 -2a>g in all samples. RHD/RHCE showed no alterations. CONCLUSION: A novel Nordic Rhnull allele was identified. In addition, it was shown that s+ Rhnull red blood cells are not only U- but also have qualitative changes in their s antigen expression.

Rh Blood Group D Antigen Genotyping Using a Portable Nanopore-based Sequencing Device: Proof of Principle.

BACKGROUND: Nanopore sequencing is direct sequencing of a single-stranded DNA molecule using biological pores. A portable nanopore-based sequencing device from Oxford Nanopore Technologies (MinION) depends on driving a DNA molecule through nanopores embedded in a membrane using a voltage. Changes in current are then measured by a sensor, thousands of times per second and translated to nucleobases. METHODS: Genomic DNA (gDNA) samples (n = 13) were tested for Rh blood group D antigen (RHD) gene zygosity using droplet digital PCR. The RHD gene was amplified in 6 overlapping amplicons using long-range PCR. Amplicons were purified, and the sequencing library was prepared following the 1D Native barcoding gDNA protocol. Sequencing was carried out with 1D flow cells R9 version. Data analysis included basecalling, aligning to the RHD reference sequence, and calling variants. Variants detected were compared to the results acquired previously by the Ion Personal Genome Machine (Ion PGM). RESULTS: Up to 500× sequence coverage across the RHD gene allowed accurate variant calling. Exonic changes in the RHD gene allowed RHD allele determination for all samples sequenced except 1 RHD homozygous sample, where 2 heterozygous RHD variant alleles are suspected. There were 3 known variant RHD alleles (RHD*01W.02, RHD*11, and RHD*15) and 6 novel RHD variant alleles, as previously seen in Ion PGM sequencing data for these samples. CONCLUSIONS: MinION was effective in blood group genotyping, provided enough sequencing data to achieve high coverage of the RHD gene, and enabled confident calling of variants and RHD allele determination.

Recommendation for validation and quality assurance of non-invasive prenatal testing for foetal blood groups and implications for IVD risk classification according to EU regulations.

BACKGROUND AND OBJECTIVES: Non-invasive assays for predicting foetal blood group status in pregnancy serve as valuable clinical tools in the management of pregnancies at risk of detrimental consequences due to blood group antigen incompatibility. To secure clinical applicability, assays for non-invasive prenatal testing of foetal blood groups need to follow strict rules for validation and quality assurance. Here, we present a multi-national position paper with specific recommendations for validation and quality assurance for such assays and discuss their risk classification according to EU regulations. MATERIALS AND METHODS: We reviewed the literature covering validation for in-vitro diagnostic (IVD) assays in general and for non-invasive foetal RHD genotyping in particular. Recommendations were based on the result of discussions between co-authors. RESULTS: In relation to Annex VIII of the In-Vitro-Diagnostic Medical Device Regulation 2017/746 of the European Parliament and the Council, assays for non-invasive prenatal testing of foetal blood groups are risk class D devices. In our opinion, screening for targeted anti-D prophylaxis for non-immunized RhD negative women should be placed under risk class C. To ensure high quality of non-invasive foetal blood group assays within and beyond the European Union, we present specific recommendations for validation and quality assurance in terms of analytical detection limit, range and linearity, precision, robustness, pre-analytics and use of controls in routine testing. With respect to immunized women, different requirements for validation and IVD risk classification are discussed. CONCLUSION: These recommendations should be followed to ensure appropriate assay performance and applicability for clinical use of both commercial and in-house assays.

Time points and risk factors for RhD immunizations after the implementation of targeted routine antenatal anti-D prophylaxis: A retrospective nationwide cohort study.

INTRODUCTION: Targeted routine antenatal anti-D prophylaxis was introduced to the national prophylaxis program in Finland in late 2013. The aim of this study was to assess the incidence, time-points, and risk factors for Rhesus D immunization after the implementation of routine antenatal anti-D prophylaxis, in all women in Finland with antenatal anti-D antibodies detected in 2014-2017. MATERIAL AND METHODS: In a nationwide population-based retrospective cohort study, the incidence, time-points, and risk factors of anti-D immunizations were analyzed. Information on antenatal screening was obtained from the Finnish Red Cross Blood Service database, and obstetric data from hospital records and the Finnish Medical Birth Register. RESULTS: The study included a total of 228 women (197 with complete data for all pregnancies). After the implementation of routine antenatal anti-D prophylaxis, the prevalence of pregnancies with anti-D antibodies decreased from 1.52% in 2014 to 0.88% in 2017, and the corresponding incidence of new immunizations decreased from 0.33% to 0.10%. Time-points for detection of new anti-D antibodies before and after 2014 were the first screening sample at 8-12 weeks of gestation in 52% vs 19%, the second sample at 24-26 weeks in 20% vs 50%, and the third screening at 36 weeks in 28% vs 32%. CONCLUSIONS: The incidence of new anti-D immunizations decreased as expected after the implementation of routine antenatal anti-D prophylaxis. True failures are rare and they mainly occur when the prophylaxis is not given appropriately, suggesting a need for constant education of healthcare professionals on the subject.

Next-generation sequencing of 35 RHD variants in 16 253 serologically D- pregnant women in the Finnish population.

Fetal RHD screening for targeted routine antenatal anti-D prophylaxis has been implemented in many countries, including Finland, since the 2010s. Comprehensive knowledge of the RHD polymorphism in the population is essential for the performance and safety of the anti-D prophylaxis program. During the first 3 years of the national screening program in Finland, over 16 000 samples from RhD- women were screened for fetal RHD; among them, 79 samples (0.5%) containing a maternal variant allele were detected. Of the detected maternal variants, 35 cases remained inconclusive using the traditional genotyping methods and required further analysis by next-generation sequencing (NGS) of the whole RHD gene to uncover the variant allele. In addition to the 13 RHD variants that have been previously reported in different populations, 8 novel variants were also detected, indicating that there is more variation of RHD in the RhD- Finnish population than has been previously known. Three of the novel alleles were identified in multiple samples; thus, they are likely specific to the original Finnish population. National screening has thus provided new information about the diversity of RHD variants in the Finnish population. The results show that NGS is a powerful method for genotyping the highly polymorphic RHD gene compared with traditional methods that rely on the detection of specific nucleotides by polymerase chain reaction amplification.

Targeted antenatal anti-D prophylaxis program for RhD-negative pregnant women - outcome of the first two years of a national program in Finland.

INTRODUCTION: The aim of this study was to assess the accuracy of the non-invasive fetal RHD test at 24-26 weeks of gestation as part of the national antenatal screening program to target routine antenatal anti-D prophylaxis (RAADP) at 28-30 weeks at women carrying an RhD-positive fetus. MATERIAL AND METHODS: A prospective cohort study involving all maternity care centers and delivery hospitals in Finland between February 2014 and January 2016. Fetal RHD genotyping using cell-free fetal DNA in maternal plasma was performed with real-time polymerase chain reaction in a centralized setting. The results were systematically compared with the serological newborn RhD typing. The main outcome measure was the accuracy of the fetal RHD assay; the secondary variable was compliance with the newly introduced RAADP program. RESULTS: Fetal RHD was screened from 10 814 women. For the detection of fetal RHD, sensitivity was 99.99% [95% confidence interval (CI) 99.92-99.99] and specificity 99.81% (95% CI 99.60-99.92). One false-negative and seven false-positive results were reported by the delivery hospitals in two years. The negative predictive value of the test was 99.97% (95% CI 99.81-99.99). At the end of the study period, over 98% of the RhD-negative women participated in the new screening program. CONCLUSIONS: The targeted RAAPD program was implemented effectively in the national maternity care program in Finland. An accurate fetal RHD screening test allows discontinuation of newborn testing without risking the postnatal prophylaxis program. In the future, the main area to investigate will be the clinical effect of RAADP on subsequent pregnancies.

Maternal HLA genotyping is not useful for predicting severity of fetal and neonatal alloimmune thrombocytopenia.

Lack of reliable laboratory parameters is the main challenge in the management of fetal and neonatal alloimmune thrombocytopenia (FNAIT). Despite the long-known association between the HLA-DRB3*01:01 allele and human platelet antigen 1a (HPA-1a) alloimmunisation, maternal human leucocyte antigen (HLA) typing has been of little clinical value. Recently, other DRB3 allele variants have been suggested to predict the severity of FNAIT. In this nationwide population-based retrospective cohort study, we performed extensive HLA typing of 96 women, accounting for 87% of our cohort of 110 families with confirmed or possible HPA-1a-immunisation. The HLA type was compared with anti-HPA-1a levels, severity of neonatal disease and responsiveness to maternally administrated intravenous gammaglobulin (IVIG). HLA haplotypes were constructed to investigate further HLA associations. Despite significantly lower anti-HPA-1a levels in DRB3*01:01-negative women, the carrier status of this particular allele could not be used to confirm or rule out FNAIT in the absence of detectable antibodies. In the haplotype analysis, the DRB3*01:01 allele was the actual factor associated with FNAIT. No other HLA allele was shown to be of additional value as a predictor of severe FNAIT or non-responsiveness to IVIG treatment. Thus, HLA genotyping was not found useful in differentiating high- and low-risk pregnancies or in guiding antenatal treatment in affected families.

[Where to find a suitable red cell product, if the patient has a rare blood group?]. / Mistä sopiva punasoluvalmiste, jos potilaalla on harvinainen veriryhmä?

A rare blood group and the associated anti-red cell antibody against a very common blood group factor make it significantly more difficult to find suitable red cell products, possibly endangering the additional treatment of the patient. The need of rare blood should be anticipated early enough. The availability of red cells is secured by extensive blood group studies among blood donors, by donor registers, freezing of rare red cells and through international collaboration. The determination of rare blood groups by genotyping provides a longed-for addition to the investigational repertoire of both patients and blood donors. Systematic freezing of red cells was started in Finland in 2010.

Case report: Severe hemolytic disease of the fetus and newborn due to anti-C+G.

Anti-G is commonly present with anti-D and/or anti-C and can confuse serological investigations. in general, anti-G is not considered a likely cause of severe hemolytic disease of the fetus and newborn (HDFN), but it is important to differentiate it from anti-D in women who should be administered anti-D immunoglobulin prophylaxis. We report one woman with three pregnancies severely affected by anti-C+G requiring intrauterine treatment and a review of the literature. In our case, the identification of the correct antibody was delayed because the differentiation of anti-C+G and anti-D+C was not considered important during pregnancy since the father was D-. In addition, anti-C+G and anti-G titer levels were not found to be reliable as is generally considered in Rh immunization. Severe HDFN occurred at a maternal anti-C+G antibody titer of S and anti-G titer of 1 in comparison with the critical titer level of 16 or more in our laboratory. close collaboration between the immunohematology laboratory and the obstetric unit is essential. In previously affected families, early assessment for fetal anemia is required even when titers are low.

Degree of damage to the small bowel and serum antibody titers correlate with clinical presentation of patients with celiac disease.

BACKGROUND & AIMS: In patients with celiac disease, gluten-induced lesions of the small-bowel mucosa develop gradually. However, it is not clear whether clinical presentation correlates with the degree of mucosal damage based on histology analysis. We investigated whether the degree of mucosal damage to the small bowel correlates with clinical presentation and serum markers of celiac disease. METHODS: We collected results from serology tests and mucosal biopsy samples from 638 consecutive patients with celiac disease and compared them with reported gastrointestinal symptoms, health-related quality-of-life scores, results from laboratory tests, and bone mineral densities of patients. We assessed mucosal injury based on the ratio of villous height to crypt depth, numbers of intraepithelial CD3(+) cells, and semiquantitative Marsh classification criteria. Correlations were established based on the Pearson or Spearman coefficients. RESULTS: The ratio of the villous height to crypt depth correlated with the severity of gastrointestinal symptoms, quality-of-life scores, laboratory test results, numbers of intraepithelial CD3(+) cells, and serum levels of antibodies associated with celiac disease. There was no correlation between the ratio of villous height to crypt depth and bone mineral density. The number of intraepithelial CD3(+) cells was not associated with symptoms, whereas the Marsh classification and serum levels of antibodies associated with celiac disease correlated with gastrointestinal symptoms, laboratory test results, and numbers of intraepithelial CD3(+) cells. CONCLUSIONS: The ratio of small-bowel villous height to crypt depth and results from serology tests correlate with reported symptoms and quality of life of patients with celiac disease. Patient-reported outcomes are therefore of value, in addition to histology findings, in assessing patients with celiac disease.

Analysis of the adequate size of a cord blood bank and comparison of HLA haplotype distributions between four populations.

The number of units and especially the number of different HLA haplotypes present in a cord blood (CB) bank is a crucial determinant of its usefulness. We generated data relevant to the development of our national CB in Finland. The HLA haplotype distribution was examined between specific populations. We developed graphical ways of data presentation that enable easy visualization of differences. First, we estimated the optimal size of a CB bank for Finland and found that approximately 1700 units are needed to provide a 5/6 HLA-matched donor for 80% of Finnish patients. Secondly, we evaluated HLA haplotype distributions between four locations, Finland, Japan, Sweden and Belgium. Our results showed that the Japanese Tokyo Cord Blood Bank differs in both the frequency and distribution of haplotypes from the European banks. The European banks (Finnish Cord Blood Registry, The Swedish National Cord Blood Bank, and Marrow Donor Program-Belgium) have similar frequencies of common haplotypes, but 26% of the haplotypes in the Finnish CB bank are unique, which justifies the existence of a national bank. The tendency to a homogenous HLA haplotype distribution in banks underlines the need for targeting recruitment at the poorly represented minority populations.

Methods for diagnostic HLA typing in disease association and drug hypersensitivity.

This chapter describes the application of diagnostic HLA typing for disease association and five methods used for specific HLA genotypes. The methods utilise a combination of polymerase chain reaction (PCR) amplification to detect sequence polymorphism by the presence or absence of amplification, nucleotide sequencing of the PCR product, and hybridisation of the PCR product with labelled probes. The probes are specific for sequence polymorphism associated with the genotype and are attached to either a Micro Bead or a Solid Phase. In addition, the detection of single nucleotide polymorphism(s) which "tag" for the genotype using a real-time PCR is described.

Endomysial antibodies predict celiac disease irrespective of the titers or clinical presentation.

AIM: To investigate the association between serum antibody levels and a subsequent celiac disease diagnosis in a large series of children and adults. METHODS: Besides subjects with classical gastrointestinal presentation of celiac disease, the study cohort included a substantial number of individuals with extraintestinal symptoms and those found by screening in at-risk groups. Altogether 405 patients underwent clinical, serological and histological evaluations. After collection of data, the antibody values were further graded as low [endomysial (EmA) 1:5-200, transglutaminase 2 antibodies (TG2-ab) 5.0-30.0 U/L] and high (EmA 1: ≥ 500, TG2-ab ≥ 30.0 U/L), and the serological results were compared with the small intestinal mucosal histology and clinical presentation. RESULTS: In total, 79% of the subjects with low and 94% of those with high serum EmA titers showed small-bowel mucosal villous atrophy. Furthermore, 96% of the 47 EmA positive subjects who had normal mucosal villi and remained on follow-up either subsequently developed mucosal atrophy while on a gluten-containing diet, or responded positively to a gluten-free diet. CONCLUSION: Irrespective of the initial serum titers or clinical presentation, EmA positivity as such is a very strong predictor of a subsequent celiac disease diagnosis.

High-density SNP mapping of the HLA region identifies multiple independent susceptibility loci associated with selective IgA deficiency.

Selective IgA deficiency (IgAD; serum IgA<0.07 g/l) is the most common form of human primary immune deficiency, affecting approximately 1∶600 individuals in populations of Northern European ancestry. The polygenic nature of IgAD is underscored by the recent identification of several new risk genes in a genome-wide association study. Among the characterized susceptibility loci, the association with specific HLA haplotypes represents the major genetic risk factor for IgAD. Despite the robust association, the nature and location of the causal variants in the HLA region remains unknown. To better characterize the association signal in this region, we performed a high-density SNP mapping of the HLA locus and imputed the genotypes of common HLA-B, -DRB1, and -DQB1 alleles in a combined sample of 772 IgAD patients and 1,976 matched controls from 3 independent European populations. We confirmed the complex nature of the association with the HLA locus, which is the result of multiple effects spanning the entire HLA region. The primary association signal mapped to the HLA-DQB1*02 allele in the HLA Class II region (combined P = 7.69×10(-57); OR = 2.80) resulting from the combined independent effects of the HLA-B*0801-DRB1*0301-DQB1*02 and -DRB1*0701-DQB1*02 haplotypes, while additional secondary signals were associated with the DRB1*0102 (combined P = 5.86×10(-17); OR = 4.28) and the DRB1*1501 (combined P = 2.24×10(-35); OR = 0.13) alleles. Despite the strong population-specific frequencies of HLA alleles, we found a remarkable conservation of these effects regardless of the ethnic background, which supports the use of large multi-ethnic populations to characterize shared genetic association signals in the HLA region. We also provide evidence for the location of association signals within the specific extended haplotypes, which will guide future sequencing studies aimed at characterizing the precise functional variants contributing to disease pathogenesis.

Persistently positive gliadin antibodies without transglutaminase antibodies in the elderly: gluten intolerance beyond coeliac disease.

BACKGROUND: The specificity of the conventional gliadin antibody test is considered low. AIMS: We explored whether gliadin antibody(AGA)-positivity without tissue transglutaminase antibodies (tTGA) is persistent in the elderly population and whether such positivity indicates overt or potential coeliac disease in genetically predisposed individuals. METHODS: AGA and tissue transglutaminase antibody were measured in 2089 elderly individuals twice with a three-year interval. AGA-positive but tissue transglutaminase antibody-negative subjects with coeliac-type human leucocyte antigen (HLA) were examined and underwent gastroduodenal endoscopy (cases). Small-bowel mucosal villous morphology and densities of CD3+ and γδ+ intraepithelial lymphocytes and the occurrence of tissue transglutaminase-specific IgA deposits were analysed. Randomly selected persistently AGA-negative age- and sex-matched subjects served as controls. RESULTS: AGA-positivity was persistent in 81% of those initially positive. Amongst the 49 clinically studied and 36 endoscopied cases only one (2.8%) had coeliac disease. Many (54%) showed signs of inflammation in the biopsy, without villous atrophy. Coeliac-type HLA was not over-represented in the persistently AGA-positive compared to the general population. Persistently AGA-positive coeliac-type HLA-positive subjects had more gastrointestinal symptoms than AGA-negative controls. CONCLUSIONS: AGA-positivity is often persistent. Overt coeliac disease is seldom found behind persistent AGA-positivity, but this characteristic is associated with mucosal inflammation and gastrointestinal symptoms at least in HLA-positive individuals.

IgA-class autoantibodies against neuronal transglutaminase, TG6 in celiac disease: no evidence for gluten dependency.

BACKGROUND: It has been suggested that antibodies against transglutaminase (TG) 6 could serve as a biomarker to identify a subgroup of gluten-sensitive patients who may be at risk of developing neurological disease. We here investigated whether TG6-targeted autoantibodies are a characteristic feature of celiac patients, especially those with neurological symptoms, and further, whether such antibodies are gluten-dependent. METHODS: Serum IgA-class TG6 autoantibodies were measured in untreated and treated celiac patients with and without neurological manifestions and in non-celiac controls. The results were compared to TG2 autoantibody levels. RESULTS: During a gluten-containing diet the number of TG6 autoantibody-positive celiac patients with neurological problems (25%) did not significantly differ from that of TG6-seropositive patients without neurological impairment (16%) or from non-celiac controls (15%). This was in contrast to our finding in TG2 autoantibody-positive individuals, whose numbers differed significantly between all three study groups. On a gluten-free diet the levels of TG6 autoantibodies remained unchanged. CONCLUSIONS: We conclude that the serum IgA-class TG6 autoantibody assay is not able to distinguish gluten-sensitive patients with neurological manifestations from celiac patients without neurological problems or from control subjects, and further, that TG6 autoantibodies are not gluten-dependent.