MOSAIC: An Artificial Intelligence-Based Framework for Multimodal Analysis, Classification, and Personalized Prognostic Assessment in Rare Cancers.

PURPOSE: Rare cancers constitute over 20% of human neoplasms, often affecting patients with unmet medical needs. The development of effective classification and prognostication systems is crucial to improve the decision-making process and drive innovative treatment strategies. We have created and implemented MOSAIC, an artificial intelligence (AI)-based framework designed for multimodal analysis, classification, and personalized prognostic assessment in rare cancers. Clinical validation was performed on myelodysplastic syndrome (MDS), a rare hematologic cancer with clinical and genomic heterogeneities. METHODS: We analyzed 4,427 patients with MDS divided into training and validation cohorts. Deep learning methods were applied to integrate and impute clinical/genomic features. Clustering was performed by combining Uniform Manifold Approximation and Projection for Dimension Reduction + Hierarchical Density-Based Spatial Clustering of Applications with Noise (UMAP + HDBSCAN) methods, compared with the conventional Hierarchical Dirichlet Process (HDP). Linear and AI-based nonlinear approaches were compared for survival prediction. Explainable AI (Shapley Additive Explanations approach [SHAP]) and federated learning were used to improve the interpretation and the performance of the clinical models, integrating them into distributed infrastructure. RESULTS: UMAP + HDBSCAN clustering obtained a more granular patient stratification, achieving a higher average silhouette coefficient (0.16) with respect to HDP (0.01) and higher balanced accuracy in cluster classification by Random Forest (92.7% ± 1.3% and 85.8% ± 0.8%). AI methods for survival prediction outperform conventional statistical techniques and the reference prognostic tool for MDS. Nonlinear Gradient Boosting Survival stands in the internal (Concordance-Index [C-Index], 0.77; SD, 0.01) and external validation (C-Index, 0.74; SD, 0.02). SHAP analysis revealed that similar features drove patients' subgroups and outcomes in both training and validation cohorts. Federated implementation improved the accuracy of developed models. CONCLUSION: MOSAIC provides an explainable and robust framework to optimize classification and prognostic assessment of rare cancers. AI-based approaches demonstrated superior accuracy in capturing genomic similarities and providing individual prognostic information compared with conventional statistical methods. Its federated implementation ensures broad clinical application, guaranteeing high performance and data protection.

Transition database for rare diseases and its use for clinical documentation.

Patients with rare diseases commonly suffer from severe symptoms as well as chronic and sometimes life-threatening effects. Not only the rarity of the diseases but also the poor documentation of rare diseases often leads to an immense delay in diagnosis. One of the main problems here is the inadequate coding with common classifications such as the International Statistical Classification of Diseases and Related Health Problems. Instead, the ORPHAcode enables precise naming of the diseases. So far, just few approaches report in detail how the technical implementation of the ORPHAcode is done in clinical practice and for research. We present a concept and implementation of storing and mapping of ORPHAcodes. The Transition Database for Rare Diseases contains all the information of the Orphanet catalog and serves as the basis for documentation in the clinical information system as well as for monitoring Key Performance Indicators for rare diseases at the hospital. The five-step process (especially using open source tools and the DataVault 2.0 logic) for set-up the Transition Database allows the approach to be adapted to local conditions as well as to be extended for additional terminologies and ontologies.

Curation and expansion of Human Phenotype Ontology for defined groups of inborn errors of immunity.

BACKGROUND: Accurate, detailed, and standardized phenotypic descriptions are essential to support diagnostic interpretation of genetic variants and to discover new diseases. The Human Phenotype Ontology (HPO), extensively used in rare disease research, provides a rich collection of vocabulary with standardized phenotypic descriptions in a hierarchical structure. However, to date, the use of HPO has not yet been widely implemented in the field of inborn errors of immunity (IEIs), mainly due to a lack of comprehensive IEI-related terms. OBJECTIVES: We sought to systematically review available terms in HPO for the depiction of IEIs, to expand HPO, yielding more comprehensive sets of terms, and to reannotate IEIs with HPO terms to provide accurate, standardized phenotypic descriptions. METHODS: We initiated a collaboration involving expert clinicians, geneticists, researchers working on IEIs, and bioinformaticians. Multiple branches of the HPO tree were restructured and extended on the basis of expert review. Our ontology-guided machine learning coupled with a 2-tier expert review was applied to reannotate defined subgroups of IEIs. RESULTS: We revised and expanded 4 main branches of the HPO tree. Here, we reannotated 73 diseases from 4 International Union of Immunological Societies-defined IEI disease subgroups with HPO terms. We achieved a 4.7-fold increase in the number of phenotypic terms per disease. Given the new HPO annotations, we demonstrated improved ability to computationally match selected IEI cases to their known diagnosis, and improved phenotype-driven disease classification. CONCLUSIONS: Our targeted expansion and reannotation presents enhanced precision of disease annotation, will enable superior HPO-based IEI characterization, and hence benefit both IEI diagnostic and research activities.

A resource to explore the discovery of rare diseases and their causative genes.

Here, we describe a dataset with information about monogenic, rare diseases with a known genetic background, supplemented with manually extracted provenance for the disease itself and the discovery of the underlying genetic cause. We assembled a collection of 4166 rare monogenic diseases and linked them to 3163 causative genes, annotated with OMIM and Ensembl identifiers and HGNC symbols. The PubMed identifiers of the scientific publications, which for the first time described the rare diseases, and the publications, which found the genes causing the diseases were added using information from OMIM, PubMed, Wikipedia, whonamedit.com, and Google Scholar. The data are available under CC0 license as spreadsheet and as RDF in a semantic model modified from DisGeNET, and was added to Wikidata. This dataset relies on publicly available data and publications with a PubMed identifier, but by our effort to make the data interoperable and linked, we can now analyse this data. Our analysis revealed the timeline of rare disease and causative gene discovery and links them to developments in methods.

New problems of a new health system: the creation of a national public policy of rare diseases care in Brazil (1990s-2010s). / Nuevos problemas de un nuevo sistema de salud: la creación de una política pública nacional de atención de enfermedades raras en Brasil (1990-2014).

This study discusses actors and institution movements leading to the disclosure in 2014 of Resolution 199 by the Brazilian Ministry of Health, which establishes the National Policy for the Comprehensive Care of Persons with Rare Diseases. Taking as sources the mainstream newspapers, drafts law, and secondary literature on the subject, we begin our analysis in the early 1990s when the first patient associations were created in Brazil - mainly for claiming more funds for research on genetic diseases - and arrive at the late 2010s when negotiations for a national policy are taking place in the National Congress. Resolution 199 is part of an ongoing process and the path towards its disclosure and the complications that followed have given us elements to discuss contemporary aspects of the Brazilian public health. Based on the references of the history of the present time and the social studies of science, we argue that two aspects have been fundamental to creating a national policy: framing different illnesses within the terminology "rare diseases" and the construction of a public perception about the right of health which is guaranteed by the 1988 Brazilian Constitution.

En este trabajo se analizan los movimientos de actores e instituciones que llevaron a la promulgación, en 2014, de la Resolución 199 del Ministerio de Salud de Brasil, que establece la Política Nacional de Atención Integral a las Personas con Enfermedades Raras. Tomando como fuentes los principales periódicos, proyectos de ley y bibliografía secundaria sobre el tema, comenzamos nuestro análisis a principios de la década de 1990 con la creación de las primeras asociaciones de pacientes en Brasil, para reclamar fundamentalmente más fondos para la investigación de enfermedades genéticas, y llegamos a fines de la década de 2010 con las negociaciones para una política nacional. La Resolución 199 es parte de un proceso en curso, en el que el camino hacia la promulgación y las complicaciones posteriores nos dan elementos para discutir aspectos actuales de la salud pública brasileña. Sobre la base de la historia del tiempo presente y los estudios sociales de la ciencia, argumentamos que hay dos aspectos que han sido fundamentales para crear una política nacional: enmarcar diferentes enfermedades en la terminología "enfermedades raras" y la construcción de una percepción pública sobre el derecho a la salud, que se garantiza en la Constitución brasileña de 1988.

Atypical and ultra-rare Usher syndrome: a review.

Usher syndrome has classically been described as a combination of hearing loss and rod-cone dystrophy; vestibular dysfunction is present in many patients. Three distinct clinical subtypes were documented in the late 1970s. Genotyping efforts have led to the identification of several genes associated with the disease. Recent literature has seen multiple publications referring to "atypical" Usher syndrome presentations. This manuscript reviews the molecular etiology of Usher syndrome, highlighting rare presentations and molecular causes. Reports of "atypical" disease are summarized noting the wide discrepancy in the spectrum of phenotypic deviations from the classical presentation. Guidelines for establishing a clear nomenclature system are suggested.

Rare cancers in Canada, 2006-2016: A population-based surveillance report and comparison of different methods for classifying rare cancers.

BACKGROUND: The cumulative burden from rare cancers has not been adequately explored in Canada. This analysis aims to characterize the occurrence of rare cancers among Canadians and estimate the probability of being diagnosed with a rare cancer among cancer patients with different demographic characteristics. METHODS: The Canadian Cancer Registry was used for this analysis. Cancer types were classified in three ways: using the SEER site recode scheme; by histology group; and by site/histology group. The age-standardized incidence rate (ASIR) and 95 % confidence intervals (CI) for each cancer type was estimated for diagnoses from 2006 to 2016. Two ASIR thresholds were used to classify cancers as rare:6/100,000/year and 15/100,000/year. Log-binomial regression was used to estimate the adjusted probability of having a rare cancer among those with cancer by age, sex and geographic region. RESULTS: Using the 6/100,000/year threshold, the incidence proportion (IP) of rare cancers ranged from 9.7 %(95 %CI:9.6,9.7 %)-17.0 %(95 %CI:16.9,17.0 %), and ranged from 19.2 %(95 %CI:19.1,19.3 %)-52.5 %(95 %CI:52.0,53.0 %) using the <15/100,000/year threshold. The adjusted probability of being diagnosed with a rare cancer was highest among those aged ≤19 years. There was higher concordance in estimates of the burden of rare cancers across methods to classify cancer types when the lower incidence rate threshold was used to define rare cancers. INTERPRETATION: This analysis yielded evidence that rare cancers comprise a substantial proportion of annual cancer diagnoses among Canadians. Findings from this analysis point to using a lower incidence rate threshold, to generate estimates of the burden of rare cancers that are robust to different cancer classification schemes.

Congenital diabetes mellitus.

Congenital diabetes mellitus is a rare disorder characterized by hyperglycemia that occurs shortly after birth. We define "Diabetes of Infancy" if hyperglycemia onset before 6 months of life. From the clinical point of view, we distinguish two main types of diabetes of infancy: transient (TNDM), which remits spontaneously, and permanent (PNDM), which requires lifelong treatment. TNDM may relapse later in life. About 50% of cases are transient (TNDM) and 50% permanent. Clinical manifestations include severe intrauterine growth retardation, hyperglycemia and dehydration. A wide range of different associated clinical signs including facial dysmorphism, deafness and neurological, cardiac, kidney or urinary tract anomalies are reported. Developmental delay and learning difficulties may also be observed. In this paper we review all the causes of congenital diabetes and all genes and syndromes involved in this pathology. The discovery of the pathogenesis of most forms of congenital diabetes has made it possible to adapt the therapy to the diagnosis and in the forms of alteration of the potassium channels of the pancreatic Beta cells the switch from insulin to glibenclamide per os has greatly improved the quality of life. Congenital diabetes, although it is a very rare form, has been at the must of research in recent years especially for pathogenesis and pharmacogenetics. The most striking difference compared to the more frequent autoimmune diabetes in children (type 1 diabetes) is the possibility of treatment with hypoglycemic agents and the apparent lower frequency of chronic complications.

Nuevos problemas de un nuevo sistema de salud: la creación de una política pública nacional de atención de enfermedades raras en Brasil (1990-2014) / New problems of a new health system: the creation of a national public policy of rare diseases care in Brazil (1990s-2010s)

RESUMEN En este trabajo se analizan los movimientos de actores e instituciones que llevaron a la promulgación, en 2014, de la Resolución 199 del Ministerio de Salud de Brasil, que establece la Política Nacional de Atención Integral a las Personas con Enfermedades Raras. Tomando como fuentes los principales periódicos, proyectos de ley y bibliografía secundaria sobre el tema, comenzamos nuestro análisis a principios de la década de 1990 con la creación de las primeras asociaciones de pacientes en Brasil, para reclamar fundamentalmente más fondos para la investigación de enfermedades genéticas, y llegamos a fines de la década de 2010 con las negociaciones para una política nacional. La Resolución 199 es parte de un proceso en curso, en el que el camino hacia la promulgación y las complicaciones posteriores nos dan elementos para discutir aspectos actuales de la salud pública brasileña. Sobre la base de la historia del tiempo presente y los estudios sociales de la ciencia, argumentamos que hay dos aspectos que han sido fundamentales para crear una política nacional: enmarcar diferentes enfermedades en la terminología "enfermedades raras" y la construcción de una percepción pública sobre el derecho a la salud, que se garantiza en la Constitución brasileña de 1988.

ABSTRACT This study discusses actors and institution movements leading to the disclosure in 2014 of Resolution 199 by the Brazilian Ministry of Health, which establishes the National Policy for the Comprehensive Care of Persons with Rare Diseases. Taking as sources the mainstream newspapers, drafts law, and secondary literature on the subject, we begin our analysis in the early 1990s when the first patient associations were created in Brazil - mainly for claiming more funds for research on genetic diseases - and arrive at the late 2010s when negotiations for a national policy are taking place in the National Congress. Resolution 199 is part of an ongoing process and the path towards its disclosure and the complications that followed have given us elements to discuss contemporary aspects of the Brazilian public health. Based on the references of the history of the present time and the social studies of science, we argue that two aspects have been fundamental to creating a national policy: framing different illnesses within the terminology "rare diseases" and the construction of a public perception about the right of health which is guaranteed by the 1988 Brazilian Constitution.

Improving rare disease classification using imperfect knowledge graph.

BACKGROUND: Accurately recognizing rare diseases based on symptom description is an important task in patient triage, early risk stratification, and target therapies. However, due to the very nature of rare diseases, the lack of historical data poses a great challenge to machine learning-based approaches. On the other hand, medical knowledge in automatically constructed knowledge graphs (KGs) has the potential to compensate the lack of labeled training examples. This work aims to develop a rare disease classification algorithm that makes effective use of a knowledge graph, even when the graph is imperfect. METHOD: We develop a text classification algorithm that represents a document as a combination of a "bag of words" and a "bag of knowledge terms," where a "knowledge term" is a term shared between the document and the subgraph of KG relevant to the disease classification task. We use two Chinese disease diagnosis corpora to evaluate the algorithm. The first one, HaoDaiFu, contains 51,374 chief complaints categorized into 805 diseases. The second data set, ChinaRe, contains 86,663 patient descriptions categorized into 44 disease categories. RESULTS: On the two evaluation data sets, the proposed algorithm delivers robust performance and outperforms a wide range of baselines, including resampling, deep learning, and feature selection approaches. Both classification-based metric (macro-averaged F1 score) and ranking-based metric (mean reciprocal rank) are used in evaluation. CONCLUSION: Medical knowledge in large-scale knowledge graphs can be effectively leveraged to improve rare diseases classification models, even when the knowledge graph is incomplete.

Morfea: revisión práctica de su diagnóstico, clasificación y tratamiento.

Morphea, or localized scleroderma, is a rare disease of the connective tissue that manifests itself with localized sclerosis of the skin and, in some cases, with extracutaneous manifestations. Its etiology is not fully understood, but it is believed that there is genetic predisposition, in addition to environmental triggering factors. Classification of the disease is not simple due to its multiple presentations; however, it is useful in order to define the treatment, which should be individualized and started early to avoid cosmetic and functional complications. In this review, we summarize the most important practical aspects of the classification, diagnostic methods and evaluation of morphea activity, as well as available therapeutic options, with an emphasis on existing clinical evidence regarding their efficacy and safety.

La morfea o esclerodermia localizada es una enfermedad poco común del tejido conectivo que se manifiesta con esclerosis localizada de la piel y, en algunos casos, con lesiones extracutáneas. Su etiología no se comprende por completo, pero se cree que hay predisposición genética, además de factores ambientales desencadenantes. La clasificación de la enfermedad no es sencilla debido a las múltiples presentaciones, sin embargo, es útil para definir el tratamiento, el cual debe individualizarse e iniciarse tempranamente para evitar complicaciones cosméticas y funcionales. En esta revisión resumimos los aspectos prácticos más importantes de la clasificación, métodos diagnósticos y de evaluación de actividad en morfea, así como las opciones terapéuticas disponibles, con énfasis en la evidencia clínica existente respecto a su eficacia y seguridad.

Refined Disease Risk Index for Hematological Malignancies, Including Rare Disorders, After Allogeneic Stem Cell Transplantation.

OBJECTIVE: The refined disease risk index (R-DRI) is a well-designed prognostic parameter that is based on only the disease type and status and is used for stratifying patients undergoing allogeneic hematopoietic stem cell transplantation (allo HSCT) into 4 risk groups. However, the application of the R-DRI for rare diseases has remained unclear. METHODS: We evaluated 135 patients who underwent allo HSCT for hematological malignancies including rare diseases, such as acute leukemia of ambiguous lineage, acute T-cell leukemia/lymphoma, extranodal natural killer T-cell lymphoma, and lymphoblastic lymphoma, at our institute. RESULTS: According to the R-DRI, overall survival (OS) and progression-free survival at 2 years for patients with the low, intermediate, high, and very high groups were 66.7% and 66.7%, 60.8% and 56.0%, 27.1% and 23.7%, and 5.9% and 5.1%, respectively (P < .0001 and P < .0001, respectively). OS showed no significant difference between B-cell non-Hodgkin lymphoma (B-NHL) and T-cell non-Hodgkin lymphoma (T-NHL) (P = .71). Moreover, OS at 1 year was 80%, 14.3%, 60%, and 0% for the intermediate risk group, the very high-risk group of B-NHL, the intermediate risk group, and the high-risk group of T-NHL, respectively (P = .035). CONCLUSION: We showed the applicability of the R-DRI for hematological malignancies, including rare disorders. However, we suggest that T-NHL patients may be better to be assigned between the nodal group and the extranodal group in the R-DRI.

Phenotypic similarity for rare disease: Ciliopathy diagnoses and subtyping.

Rare diseases are often hard and long to be diagnosed precisely, and most of them lack approved treatment. For some complex rare diseases, precision medicine approach is further required to stratify patients into homogeneous subgroups based on the clinical, biological or molecular features. In such situation, deep phenotyping of these patients and comparing their profiles based on subjacent similarities are thus essential to help fast and precise diagnoses and better understanding of pathophysiological processes in order to develop therapeutic solutions. In this article, we developed a new pipeline of using deep phenotyping to define patient similarity and applied it to ciliopathies, a group of rare and severe diseases caused by ciliary dysfunction. As a French national reference center for rare and undiagnosed diseases, the Necker-Enfants Malades Hospital (Necker Children's Hospital) hosts the Imagine Institute, a research institute focusing on genetic diseases. The clinical data warehouse contains on one hand EHR data, and on the other hand, clinical research data. The similarity metrics were computed on both data sources, and were evaluated with two tasks: diagnoses with EHRs and subtyping with ciliopathy specific research data. We obtained a precision of 0.767 in the top 30 most similar patients with diagnosed ciliopathies. Subtyping ciliopathy patients with phenotypic similarity showed concordances with expert knowledge. Similarity metrics applied to rare disease offer new perspectives in a translational context that may help to recruit patients for research, reduce the length of the diagnostic journey, and better understand the mechanisms of the disease.

Common pre-diagnostic features in individuals with different rare diseases represent a key for diagnostic support with computerized pattern recognition?

BACKGROUND: Rare diseases (RD) result in a wide variety of clinical presentations, and this creates a significant diagnostic challenge for health care professionals. We hypothesized that there exist a set of consistent and shared phenomena among all individuals affected by (different) RD during the time before diagnosis is established. OBJECTIVE: We aimed to identify commonalities between different RD and developed a machine learning diagnostic support tool for RD. METHODS: 20 interviews with affected individuals with different RD, focusing on the time period before their diagnosis, were performed and qualitatively analyzed. Out of these pre-diagnostic experiences, we distilled key phenomena and created a questionnaire which was then distributed among individuals with the established diagnosis of i.) RD, ii.) other common non-rare diseases (NRO) iii.) common chronic diseases (CD), iv.), or psychosomatic/somatoform disorders (PSY). Finally, four combined single machine learning methods and a fusion algorithm were used to distinguish the different answer patterns of the questionnaires. RESULTS: The questionnaire contained 53 questions. A total sum of 1763 questionnaires (758 RD, 149 CD, 48 PSY, 200 NRO, 34 healthy individuals and 574 not evaluable questionnaires) were collected. Based on 3 independent data sets the 10-fold stratified cross-validation method for the answer-pattern recognition resulted in sensitivity values of 88.9% to detect the answer pattern of a RD, 86.6% for NRO, 87.7% for CD and 84.2% for PSY. CONCLUSION: Despite the great diversity in presentation and pathogenesis of each RD, patients with RD share surprisingly similar pre-diagnosis experiences. Our questionnaire and data-mining based approach successfully detected unique patterns in groups of individuals affected by a broad range of different rare diseases. Therefore, these results indicate distinct patterns that may be used for diagnostic support in RD.