One in a Million: A Nationwide Diagnostics Strategy for Ultrarare Diseases
Leveraging integrated healthcare frameworks can significantly reduce diagnostic delays
Zahraa Chorghay, PhD, specialized in neuroscience during her undergraduate (University of Toronto) and doctoral studies (McGill University). She continues to explore her passion for neuroscience and for making science accessible and inclusive.
ViewFull ProfileLearn about ourEditorial Policies.
How do you diagnose a disease that is as rare as one in a million? It often takes years. In a study published in the journal Nature Genetics earlier this year, Schmidt et al. report how they successfully implemented a novel diagnostic concept for ultrarare diseases in a nationwide study in Germany. The new concept uses a standardized approach that combines clinical assessments and advanced sequencing methods like exome sequencing to reduce diagnostic delays. Their efforts decreased time-to-diagnosis to less than a year and established molecular genetic diagnoses for an impressive one-third of the study’s cohort. The research demonstrates the power of harnessing integrated frameworks to improve the diagnosis and management of rare diseases.
What is an ultrarare disease?
An “ultrarare disease” is one that has an extremely low prevalence in the general population, e.g., less than 1 in 50,000, but for an estimated 80 percent of the 5,000+ known rare genetic diseases, this number is closer to a prevalence of one in a million. As a group, ultrarare diseases occur in 3–6 percent of the global population, representing a significant global health burden. Many rare diseases are related to or caused by a single gene (i.e., monogenic), but single gene analyses or small gene panels to detect gene variants often fail to establish a diagnosis, especially for phenotypes with high genetic heterogeneity.
A nationwide strategy for reducing time-to-diagnosis
With the aim of decreasing time-to-diagnosis for rare diseases, researchers in Germany launched a three-year prospective study called TRANSLATE NAMSE. In the study, multidisciplinary teams (MDTs) across 10 university hospital-based centers for rare diseases used standardized structures and procedures to diagnose as many patients enrolled in the study as possible.
To do so, the MDTs reviewed patient records to conduct clinical assessments and select appropriate diagnostic procedures for each study participant, including exome sequencing and additional methods as needed (multi-omics sequencing and facial analysis), and evaluated all findings. Between 2018 and 2020, the researchers analyzed exosome sequencing data for 1,577 individuals with a suspected rare disorder that were enrolled in TRANSLATE NAMSE. Notably, Germany has a health care system where most people have statutory health insurance, creating the perfect conditions to carry out this type of multisite study.
Exome sequencing led to molecular diagnoses
From the analyzed data, a molecular diagnosis was established for 32 percent of patients using exome sequencing. The study identified 549 disease-causing variants in 362 different genes and structural variants in 14 genomic regions, including 34 novel and 23 candidate genotype–phenotype associations.
"We are particularly proud of the discovery of 34 new molecular diseases, which is a great example of knowledge-generating patient care at university hospitals," said Theresa Brunet, MD, one of the study authors from the Institute of Human Genetics at the Klinikum rechts der Isar of the Technical University of Munich, in a recent press release about the study.
Exome sequencing revealed de novo mutations underlying 45 percent of the diagnosed cases. Parental mosaicism was seen in just over 1 percent of diagnosed cases, a frequency that falls within the expected range. Almost 15 percent of the diagnosed cases involved autosomal recessive (AR) inheritance, which has important considerations for family planning. Schmidt et al. also identified cases with more than one phenotype, as well as medically actionable variants that were unrelated to the present phenotype.
| Exome sequencing: sequencing the protein-coding regions of the genome. This method can identify variants that fully or partially explain the rare disease phenotype. |
| De novo mutations: mutations that are not inherited as they arose spontaneously. |
| Mosaicism: presence of two or more populations of genetically distinct cells within an individual. |
For undiagnosed cases, the researchers assessed the non-clinical exome for potentially deleterious variants and performed multi-omics assay(s), including methylome, proteome, and transcriptome analyses. While 13 of the cases remained unsolved, the successful diagnosis of nearly 500 rare disease patients in a single study is unprecedented.
Machine learning for phenotyping ultrarare diseases
To assess the links between phenotype and the diagnostic yield of exome sequencing, Schmidt et al. applied the YieldPred machine learning model. The model found clinical annotations of certain rare disease groups like ataxia, hematological abnormalities, and cognitive dysfunction were more predictive of exome sequencing establishing a molecular diagnosis, whereas groups like autism, anterior eye chamber anomalies, and skin abnormalities were least predictive. The model’s performance improved significantly when facial image analysis was included. Therefore, rare disease diagnosis can be enhanced by supplementing case annotations and molecular diagnosis with facial analysis.
Accurate diagnoses enable targeted and effective therapy
Schmidt et al.’s integrated, national framework for diagnosis significantly enhanced the detection of ultrarare genetic diseases, reducing time-to-diagnosis from years to less than a year. Quick and accurate diagnosis is crucial for individualized treatment plans directed against the mechanism of the disease, such as prescribing creatine supplementation for an individual with the SLC6A8 variant. Therefore, such an improvement in rare disease diagnosis can greatly impact clinical outcomes for patients.
Importantly, the success of this novel diagnostic approach for quickly and accurately establishing a diagnosis for otherwise highly challenging cases depended upon the active participation of MDTs across Germany, a platform for systematic data collection, and the ability to perform data analysis on independent and combinatorial sources of data. Together, this study demonstrates the remarkable power of implementing integrated frameworks within healthcare systems to treat individuals with complex diseases.
