Clinical applications of deep learning in neuroinflammatory diseases: A scoping review.

BACKGROUND: Deep learning (DL) is an artificial intelligence technology that has aroused much excitement for predictive medicine due to its ability to process raw data modalities such as images, text, and time series of signals. OBJECTIVES: Here, we intend to give the clinical reader elements to understand this technology, taking neuroinflammatory diseases as an illustrative use case of clinical translation efforts. We reviewed the scope of this rapidly evolving field to get quantitative insights about which clinical applications concentrate the efforts and which data modalities are most commonly used. METHODS: We queried the PubMed database for articles reporting DL algorithms for clinical applications in neuroinflammatory diseases and the radiology.healthairegister.com website for commercial algorithms. RESULTS: The review included 148 articles published between 2018 and 2024 and five commercial algorithms. The clinical applications could be grouped as computer-aided diagnosis, individual prognosis, functional assessment, the segmentation of radiological structures, and the optimization of data acquisition. Our review highlighted important discrepancies in efforts. The segmentation of radiological structures and computer-aided diagnosis currently concentrate most efforts with an overrepresentation of imaging. Various model architectures have addressed different applications, relatively low volume of data, and diverse data modalities. We report the high-level technical characteristics of the algorithms and synthesize narratively the clinical applications. Predictive performances and some common a priori on this topic are finally discussed. CONCLUSION: The currently reported efforts position DL as an information processing technology, enhancing existing modalities of paraclinical investigations and bringing perspectives to make innovative ones actionable for healthcare.

Real-world multicentre cohort of first-line pembrolizumab alone or in combination with platinum-based chemotherapy in non-small cell lung cancer PD-L1 ≥ 50.

INTRODUCTION: Pembrolizumab alone (IO-mono) or in combination with platinum-based chemotherapy (CT-IO) is first-line standard of care for advanced non-small cell lung cancer (NSCLC) patients with PD-L1 ≥ 50%. This retrospective multicentre study assessed real-world use and efficacy of both strategies. METHODS: Patients with advanced NSCLC PD-L1 ≥ 50% from eight hospitals who had received at least one cycle of IO-mono or CT-IO were included. Overall survival (OS) and real-word progression-free-survival were estimated using Kaplan-Meier methodology. Cox proportional hazards regression models were used to estimate hazard ratios (HRs) and 95% CIs, and a Cox model with inverse propensity treatment weighting was carried out. RESULTS: Among the 243 patients included, 141 (58%) received IO-mono and 102 (42%) CT-IO. Younger patients, those with symptomatic disease and brain metastases were more likely to be proposed CT-IO. With a median follow-up of 11.5 months (95% CI 10.4-13.3), median OS was not reached, but no difference was observed between groups (p = 0.51). Early deaths at 12 weeks were 11% (95% CI 4.6-16.9) and 15.2% (95% CI 9.0-20.9) in CT-IO and IO groups (p = 0.32). After adjustment for age, gender, performance status, histology, brain metastases, liver metastases and tobacco status, no statistically significant difference was found for OS between groups, neither in the multivariate adjusted model [HR 1.07 (95% CI 0.61-1.86), p = 0.8] nor in propensity adjusted analysis [HR 0.99 (95% CI 0.60-1.65), p = 0.99]. Male gender (HR 2.01, p = 0.01) and PS ≥ 2 (HR 3.28, p < 0.001) were found to be negative independent predictive factors for OS. CONCLUSION: Younger patients, those with symptomatic disease and brain metastases were more likely to be proposed CT-IO. However, sparing the chemotherapy in first-line does not appear to impact survival outcomes, even regarding early deaths.

Factors associated with early relapse of infantile haemangioma in children treated for at least six months with oral propranolol: A case-control study using the 2014-2021 French Ouest DataHub.

BACKGROUND: The factors associated with early relapse of infantile haemangioma (IH) after a first course of treatment with oral propranolol for at least six months (initiated after the marketing authorization had been granted) have not previously been investigated. OBJECTIVES: To identify factors associated with the risk of early relapse in children with IH treated with oral propranolol according to the current prescribing guidelines. METHODS: We performed a multicentre, retrospective, case-control study, using the Ouest Data Hub database. All children treated for at least 6 months with oral propranolol for IH between 31 June 2014 and 31 December 2021, and with a follow-up visit at least three months after treatment discontinuation were included. A case was defined as relapse of IH within three months of treatment discontinuation; each case was matched for age at treatment initiation and for centre, with four (relapse-free) controls. The association between relapse and treatment or IH characteristics was expressed as an odds ratio (OR) from univariate and multivariate conditional logistic regressions. RESULTS: A total of 225 children were included. Of these, 36 (16%) relapsed early. In a multivariate analysis, a deep IH component was a risk factor for early relapse [OR = 8.93; 95%CI: 1.0-78.9, p = 0.05]. A propranolol dosage level of less than 3 mg/kg/day protected against early relapse [OR = 0.11; 95%CI: 0.02-0.7, p = 0.02]. Tapering before propranolol discontinuation was not associated with a lower risk of early relapse. CONCLUSION: The risk factors for late and early relapse are probably different. Investigation of the risk factors for early vs. late IH relapse is now warranted.

Protocol of a multicentric prospective cohort study for the VALIDation of the IBD-disk instrument for assessing disability in inflammatory bowel diseases: the VALIDate study.

BACKGROUND: Inflammatory Bowel Diseases (IBD) affect psychological, family, social and professional dimensions of patients' life, leading to disability which is essential to quantify as part of Patient-Reported Outcomes (PROs) newly included in the targets to reach in IBD patients. Up to now, the IBD-Disability Index (IBD-DI) was the only validated tool to assess disability, but it is not appropriate for use in clinical practice. The IBD Disk was developed, a shortened and self-administered tool, adapted from the IBD-DI, in order to give immediate representation of patient-reported disability. However, the IBD Disk has not been validated yet in clinical practice. The aims of the VALIDate study are to validate this tool in a large population of IBD patients and to compare it to the already validated IBD-DI. METHODS: The VALIDate study is an ongoing multicentric prospective cohort study launched in April 2018 in 3 French University Hospitals (Nantes, Rennes, Angers), with an objective to reach a sample of 400 patients over a period inclusion of 6 months. Each patient will fill in the two questionnaires IBD Disk and IBD-DI at baseline, then between 3 and 12 months later, during a follow-up visit. Clinical and socio-demographic data will also be collected. During these two consultations, gastroenterologists and patients will evaluate disease activity thanks to a semi-quantitative 4-grade scale, named respectively PGA (Physician Global Assessment) and PtGA (Patient Global Assessment). This cohort will allow to evaluate the validity of the IBD Disk with respect to the IBD-DI in order to generalize its use for clinical practice. Other psychometric criteria of the IBD Disk will also be analysed as its reliability or its discriminant capacity. Close attention will nonetheless be needed to minimize the number of lost to follow-up patients between baseline and follow-up. DISCUSSION: The VALIDate study is the study designed to validate the IBD Disk, a visual tool easily useable in daily practice to assess disability in IBD patients. The results of this trial should enable the diffusion of this tool. TRIAL REGISTRATION: The trial is registered in ClinicalTrials.Gov with registration number NCT03590639. First posted: July 18, 2018.

Significant variation between SNP-based HLA imputations in diverse populations: the last mile is the hardest.

Four single nucleotide polymorphism (SNP)-based human leukocyte antigen (HLA) imputation methods (e-HLA, HIBAG, HLA*IMP:02 and MAGPrediction) were trained using 1000 Genomes SNP and HLA genotypes and assessed for their ability to accurately impute molecular HLA-A, -B, -C and -DRB1 genotypes in the Human Genome Diversity Project cell panel. Imputation concordance was high (>89%) across all methods for both HLA-A and HLA-C, but HLA-B and HLA-DRB1 proved generally difficult to impute. Overall, <27.8% of subjects were correctly imputed for all HLA loci by any method. Concordance across all loci was not enhanced via the application of confidence thresholds; reliance on confidence scores across methods only led to noticeable improvement (+3.2%) for HLA-DRB1. As the HLA complex is highly relevant to the study of human health and disease, a standardized assessment of SNP-based HLA imputation methods is crucial for advancing genomic research. Considerable room remains for the improvement of HLA-B and especially HLA-DRB1 imputation methods, and no imputation method is as accurate as molecular genotyping. The application of large, ancestrally diverse HLA and SNP reference data sets and multiple imputation methods has the potential to make SNP-based HLA imputation methods a tractable option for determining HLA genotypes.

Decrease of blood anti-α1,3 Galactose Abs levels in multiple sclerosis (MS) and clinically isolated syndrome (CIS) patients.

The etiology of multiple sclerosis (MS) remains elusive. Among the possible causes, the increase of anti-Neu5Gc antibodies during EBV primo-infection of Infectious mononucleosis (IMN) may damage the integrity of the blood-brain barrier facilitating the transfer of EBV-infected B cells and anti-EBV T cell clones in the brain. We investigated the change in titers of anti-Neu5Gc and anti-α1,3 Galactose antibodies in 49 IMN, in 76 MS, and 73 clinically isolated syndrome (CIS) patients, as well as age/gender-matched healthy individuals. Anti-Gal and anti-Neu5Gc are significantly increased during IMN (p=0.02 and p<1.10-4 respectively), but not in acute CMV primo-infection. We show that, whereas there was no change in anti-Neu5Gc in MS/CIS, the two populations exhibit a significant decrease in anti-Gal (combined p=2.7.10-3), in contrast with patients with non-MS/CIS central nervous system pathologies. Since anti-Gal result from an immunization against α1,3 Gal, lacking in humans but produced in the gut, our data suggest that CIS and MS patients have an altered microbiota or an altered response to this microbiotic epitope.

The killer immunoglobulin-like receptor KIR3DL1 in combination with HLA-Bw4 is protective against multiple sclerosis in African Americans.

We investigated the role of the KIR loci and their HLA class I ligands in a large cohort of African American multiple sclerosis (MS) patients (N=907) and controls (N=1456). No significant differences in carrier frequencies for any KIR locus or haplotype were observed between cases and controls. However, examination of KIR in the context of their cognate HLA ligands revealed a strong protective effect for KIR3DL1 in combination with HLA-A and -B alleles bearing the Bw4 motif (P=10(-8); odds ratio (OR)=0.60, confidence interval (CI)=0.50-0.71) and the Bw4 ligand alone (P<10(-6); OR=0.63, CI=0.53-0.75). The observed effect cannot be explained by either a specific HLA-B allele or by linkage disequilibrium with HLA-DRB1 or HLA-A. The protective effect was observed only in individuals who were not positive for the MS risk allele HLA-DRB1*15:01 (P<10(-6); OR=0.61, CI=0.51-0.74). Our study, the first investigation of KIR and MS in African Americans, confirms and refines previous findings in a European cohort.

Enrichment for Northern European-derived multiple sclerosis risk alleles in Sardinia.

BACKGROUND: The list of genomic loci associated with multiple sclerosis (MS) susceptibility outside the major histocompatibility complex (MHC) in patients of Northern European (NE) ancestry has increased to 103. Despite the extraordinarily high MS prevalence in the isolated Sardinian population, the contribution of genetic risk factors to MS in Sardinia is largely not understood. OBJECTIVE: The objective of this paper is to examine the relevance of non-MHC MS susceptibility variants in Sardinia. METHODS: We examined a log-additive MS-specific genetic burden score (MSGB) using 110 NE-derived risk alleles in a dataset of 75 Sardinian cases, 346 Sardinian controls and 177 cases and 1967 controls from the United States (US). RESULTS: Sardinian cases demonstrate a heavier non-MHC MSGB load than Sardinian controls and US cases (p = 2E-06, p = 1E-06, respectively). Furthermore, Sardinian controls carry a heavier burden than US controls (p = 2E-14). Our results confirm the limited ability of the 110-SNP MSGB to predict disease status in Sardinia (AUROC = 0.629). CONCLUSIONS: Risk alleles discovered in samples of NE ancestry are relevant to MS in Sardinia. Our results suggest a general enrichment of MS susceptibility alleles in Sardinians, encouraging the pursuit of further studies of MS in this population.

HLA phenotypes of candidates for HSCT: comparing transplanted versus non-transplanted candidates, resulting in the predictive estimation of the probability to find a 10/10 HLA matched donor.

In order to study the impact of human leucocyte antigen (HLA) polymorphism distribution in identifying a matched haematopoietic stem cells unrelated donor (UD), we performed a multi-centric retrospective analysis with the aim of comparing the HLA-A, HLA-B, HLA-C, HLA-DRB1 and HLA-DQB1 phenotypes of 2126 patients (772 patients for whom a donor search failed to identify a matched UD, and 1354 patients who received a 10/10 allele level matched UD). Our results showed that rare HLA-C is often responsible for difficulty in identifying a donor. This locus may add a degree of complexity to a supposed 'frequent' HLA-A HLA-B and HLA-DRB1 phenotype, turning this phenotype into a less frequent one. For example, 32.5% of the phenotypes in the non-transplanted patients could not be explained by any of the pairs of known HLA-A, HLA-B, HLA-C and HLA-DRB1 haplotypes while this percentage dropped to less than 2% if combinations of only HLA-A, HLA-B and HLA-DRB1 haplotypes were considered. Such situations can be anticipated by computing an index, based on HLA haplotype frequency, the average registry sample size (ARS). ARS is defined as the inverse of the phenotype frequency computed using all corresponding pairs of haplotype frequencies. ARS confirmed that the most significant difference between transplanted and non-transplanted patients was correlated with the introduction of the locus HLA-C in the analysis (median: 8.3e + 4 vs 3.1e + 6, P < 0.0001). The higher the ARS the lower the likelihood of finding a 10/10 match UD reflecting the rareness of the patient's HLA. The area under receiver operator characteristics (AUROC) values of the ARS computation for HLA-A, HLA-B and HLA-DRB1 was 0.82 (0.80; 0.84) at a low-resolution level (two digits). Overall, our study promotes the use of haplotype frequency-based computations to develop computer-assisted donor search.

Validation of statistical imputation of allele-level multilocus phased genotypes from ambiguous HLA assignments.

Genetic matching for loci in the human leukocyte antigen (HLA) region between a donor and a patient in hematopoietic stem cell transplantation (HSCT) is critical to outcome; however, methods for HLA genotyping of donors in unrelated stem cell registries often yield results with allelic and phase ambiguity and/or do not query all clinically relevant loci. We present and evaluate a statistical method for in silico imputation of HLA alleles and haplotypes in large ambiguous population data from the Be The Match(®) Registry. Our method builds on haplotype frequencies estimated from registry populations and exploits patterns of linkage disequilibrium (LD) across HLA haplotypes to infer high resolution HLA assignments. We performed validation on simulated and real population data from the Registry with non-trivial ambiguity content. While real population datasets caused some predictions to deviate from expectation, validations still showed high percent recall for imputed results with average recall >76% when imputing HLA alleles from registry data. We simulated ambiguity generated by several HLA genotyping methods to evaluate the imputation performance on several levels of typing resolution. On average, imputation percent recall of allele-level HLA haplotypes was >95% for allele-level typing, >92% for intermediate resolution typing and >58% for serology (low-resolution) typing. Thus, allele-level HLA assignments can be imputed through the application of a set of statistical and population genetics inferences and with knowledge of haplotype frequencies and self-identified race and ethnicities.

Genetic burden in multiple sclerosis families.

A previous study using cumulative genetic risk estimations in multiple sclerosis (MS) successfully tracked the aggregation of susceptibility variants in multi-case and single-case families. It used a limited description of susceptibility loci available at the time (17 loci). Even though the full roster of MS risk genes remains unavailable, we estimated the genetic burden in MS families and assess its disease predictive power using up to 64 single-nucleotide polymorphism (SNP) markers according to the most recent literature. A total of 708 controls, 3251 MS patients and their relatives, as well as 117 twin pairs were genotyped. We validated the increased aggregation of genetic burden in multi-case compared with single-case families (P=4.14e-03) and confirm that these data offer little opportunity to accurately predict MS, even within sibships (area under receiver operating characteristic (AUROC)=0.59 (0.55, 0.53)). Our results also suggest that the primary progressive and relapsing-type forms of MS share a common genetic architecture (P=0.368; difference being limited to that corresponding to ± 2 typical MS-associated SNPs). We have confirmed the properties of individual genetic risk score in MS. Comparing with previous reference point for MS genetics (17 SNPs), we underlined the corrective consequences of the integration of the new findings from GWAS and meta-analysis.

Comparative validation of computer programs for haplotype frequency estimation from donor registry data.

Estimation of human leukocyte antigen (HLA) haplotype frequencies from unrelated stem cell donor registries presents a challenge because of large sample sizes and heterogeneity of HLA typing data. For the 14th International HLA and Immunogenetics Workshop, five bioinformatics groups initiated the 'Registry Diversity Component' aiming to cross-validate and improve current haplotype estimation tools. Five datasets were derived from different donor registries and then used as input for five different computer programs for haplotype frequency estimation. Because of issues related to heterogeneity and complexity of HLA typing data identified in the initial phase, the same five implementations, and two new ones, were used on simulated datasets in a controlled experiment where the correct results were known a priori. These datasets contained various fractions of missing HLA-DR modeled after European haplotype frequencies. We measured the contribution of sampling fluctuation and estimation error to the deviation of the frequencies from their true values, finding equivalent contributions of each for the chosen samples. Because of patient-directed activities, selective prospective typing strategies and the variety and evolution of typing technology, some donors have more complete and better HLA data. In this setting, we show that restricting estimation to fully typed individuals introduces biases that could be overcome by including all donors in frequency estimation. Our study underlines the importance of critical review and validation of tools in registry-related activity and provides a sustainable framework for validating the computational tools used. Accurate frequencies are essential for match prediction to improve registry operations and to help more patients identify suitably matched donors.

Copy number variation in pediatric multiple sclerosis.

BACKGROUND: Pediatric onset multiple sclerosis (MS) accounts for 2-4% of all MS. It is unknown whether the disease shares the same underlying pathophysiology found in adult patients or an extreme early onset phenotype triggered by distinct biological mechanisms. It has been hypothesized that copy number variations (CNVs) may result in extreme early onset diseases because CNVs can have major effects on many genes in large genomic regions. OBJECTIVES AND METHODS: The objective of the current research was to identify CNVs, with a specific focus on de novo CNVs, potentially causing early onset MS by competitively hybridizing 30 white non-Hispanic pediatric MS patients with each of their parents via comparative genomic hybridization (CGH) analysis on the Agilent 1M CGH array. RESULTS AND DISCUSSION: We identified 10 CNVs not overlapping with any CNV regions currently reported in the Database of Genomic Variants (DGV). Fifty-five putatively de novo CNVs were also identified: all but one common in the DGV. We found the single rare CNV was a private variation harboring the SACS gene. SACS mutations cause autosomal-recessive spastic ataxia of Charlevoix-Saguenay (ARSACS) disease. Additional clinical review revealed that the patient with the SACS gene CNV shared some features of both MS and ARSACS. CONCLUSIONS: This is the first reported study analyzing pediatric MS CNVs. While not yielding causal variation in our initial pediatric dataset, our approach confirmed diagnosis of an ARSACS-like disease in addition to MS in the affected individual, which led to a more complete understanding of the patient's disease course and prognosis.

16(th) IHIW: global analysis of registry HLA haplotypes from 20 million individuals: report from the IHIW Registry Diversity Group.

This project has the goal to validate bioinformatics methods and tools for HLA haplotype frequency analysis specifically addressing unique issues of haematopoietic stem cell registry data sets. In addition to generating new methods and tools for the analysis of registry data sets, the intent is to produce a comprehensive analysis of HLA data from 20 million donors from the Bone Marrow Donors Worldwide (BMDW) database. This report summarizes the activity on this project as of the 16IHIW meeting in Liverpool.

SNP-based analysis of the HLA locus in Japanese multiple sclerosis patients.

Although several major histocompatibility complex (MHC)-wide single-nucleotide polymorphism (SNP) studies have been performed in populations of European descent, none have been performed in Asian populations. The objective of this study was to identify human leukocyte antigen (HLA) loci associated with multiple sclerosis (MS) in a Japanese population genotyped for 3534 MHC region SNPs. Using a logistic regression model, two SNPs (MHC Class III SNP rs422951 in the NOTCH4 gene and MHC Class II SNP rs3997849, susceptible alleles A and G, respectively) were independently associated with MS susceptibility (204 patients; 280 controls), two (MHC Class II SNP rs660895 and MHC Class I SNP rs2269704 in the NRM gene, susceptible alleles G and G, respectively) with aquaporin-4- (AQP4-) MS susceptibility (149 patients; 280 controls) and a single SNP (MHC Class II SNP rs1694112, susceptible allele G) was significant when contrasting AQP4+ against AQP4- patients. Haplotype analysis revealed a large susceptible association, likely DRB1*04 or a locus included in the DRB1*04 haplotype, with AQP4- MS, which excluded DRB1*15:01. This study is the largest study of the HLA's contribution to MS in Japanese individuals.

A community standard for immunogenomic data reporting and analysis: proposal for a STrengthening the REporting of Immunogenomic Studies statement.

Modern high-throughput HLA and KIR typing technologies are generating a wealth of immunogenomic data with the potential to revolutionize the fields of histocompatibility and immune-related disease association and population genetic research, much as SNP-based approaches have revolutionized association research. The STrengthening the REporting of Genetic Association studies (STREGA) statement provides community-based data reporting and analysis standards for genomic disease-association studies, identifying specific areas in which adoption of reporting guidelines can improve the consistent interpretation of genetic studies. While aspects of STREGA can be applied to immunogenomic studies, HLA and KIR research requires additional consideration, as the high levels of polymorphism associated with immunogenomic data pose unique methodological and computational challenges to the synthesis of information across datasets. Here, we outline the principle challenges to consistency in immunogenomic studies, and propose that an immunogenomic-specific analog to the STREGA statement, a STrengthening the REporting of Immunogenomic Studies (STREIS) statement, be developed as part of the 16th International HLA and Immunogenetics Workshop. We propose that STREIS extends at least four of the 22 elements of the STREGA statement to specifically address issues pertinent to immunogenomic data: HLA and KIR nomenclature, data-validation, ambiguity resolution, and the analysis of highly polymorphic genetic systems. As with the STREGA guidelines, the intent behind STREIS is not to dictate the design of immunogenomic studies, but to ensure consistent and transparent reporting of research, facilitating the synthesis of HLA and KIR data across studies.

Are we giving patients enough information on how to use topical treatments? Analysis of 767 prescriptions in psoriasis.

BACKGROUND: Unclear instructions probably contribute to the suboptimal efficacy and adherence to topical agents in psoriasis. OBJECTIVES: To analyse the quality of prescriptions for topical therapy in psoriasis and to determine factors associated with high-quality prescription writing. METHODS: We made a systematic analysis of 767 topical prescriptions written by dermatologists and general practitioners (GPs). The following parameters were recorded: writing mode (electronic vs. hand written), indication of formulation, frequency of administration, duration of treatment, indication of areas to be treated, and indication of amount of product to be used. We considered prescriptions of high quality to be those including at least four of the five prospectively defined quality parameters. RESULTS: Only 35·7% of prescriptions fulfilled the definition of a high-quality prescription. Quality of prescription writing was significantly influenced by two factors: electronic writing [odds ratio (OR) 3·04, 95% confidence interval (CI) 2·2-4·21; P < 10(-4) ] and specialty of the prescriber, dermatologists writing higher quality prescriptions compared with GPs (OR 1·61, 95% CI 1·54-2·14; P < 10(-4) ). CONCLUSIONS: Almost two-thirds of topical prescriptions are not adequately written and do not include the required information to help patients manage their topical treatment in psoriasis correctly. The quality of topical prescriptions could be improved by making the use of electronic prescriptions widespread and by the development of aids for easy evaluation of the right amount of topical treatment to be applied according to body surface area involved.

Genetic susceptibility to multiple sclerosis: interactions between conserved extended haplotypes of the MHC and other susceptibility regions.

BACKGROUND: To study the accumulation of MS-risk resulting from different combinations of MS-associated conserved-extended-haplotypes (CEHs) of the MHC and three non-MHC "risk-haplotypes" nearby genes EOMES, ZFP36L1, and CLEC16A. Many haplotypes are MS-associated despite having population-frequencies exceeding the percentage of genetically-susceptible individuals. The basis of this frequency-disparity requires explanation. METHODS: The SNP-data from the WTCCC was phased at the MHC and three non-MHC susceptibility-regions. CEHs at the MHC were classified into five haplotype-groups: (HLA-DRB1*15:01 ~ DQB1*06:02 ~ a1)-containing (H +); extended-risk (ER); all-protective (AP); neutral (0); and the single-CEH (c1). MS-associations for different "risk-combinations" at the MHC and other non-MHC "risk-loci" and the appropriateness of additive and multiplicative risk-accumulation models were assessed. RESULTS: Different combinations of "risk-haplotypes" produce a final MS-risk closer to additive rather than multiplicative risk-models but neither model was consistent. Thus, (H +)-haplotypes had greater impact when combined with (0)-haplotypes than with (H +)-haplotypes, whereas, (H +)-haplotypes had greater impact when combined with a (c1)-haplotypes than with (0)-haplotypes. Similarly, risk-genotypes (0,H +), (c1,H +), (H + ,H +) and (0,c1) were additive with risks from non-MHC risk-loci, whereas risk-genotypes (ER,H +) and (AP,c1) were unaffected. CONCLUSIONS: Genetic-susceptibility to MS is essential for MS to develop but actually developing MS depends heavily upon both an individual's particular combination of "risk-haplotypes" and how these loci interact.

Multiple sclerosis susceptibility alleles in African Americans.

Multiple sclerosis (MS) is an autoimmune demyelinating disease characterized by complex genetics and multifaceted gene-environment interactions. Compared to whites, African Americans have a lower risk for developing MS, but African Americans with MS have a greater risk of disability. These differences between African Americans and whites may represent differences in genetic susceptibility and/or environmental factors. SNPs from 12 candidate genes have recently been identified and validated with MS risk in white populations. We performed a replication study using 918 cases and 656 unrelated controls to test whether these candidate genes are also associated with MS risk in African Americans. CD6, CLEC16a, EVI5, GPC5, and TYK2 contained SNPs that are associated with MS risk in the African American data set. EVI5 showed the strongest association outside the major histocompatibility complex (rs10735781, OR=1.233, 95% CI=1.06-1.43, P-value=0.006). In addition, RGS1 seems to affect age of onset whereas TNFRSF1A seems to be associated with disease progression. None of the tested variants showed results that were statistically inconsistent with the effects established in whites. The results are consistent with shared disease genetic mechanisms among individuals of European and African ancestry.