The Orphanet Nomenclature and Classification of rare diseases: a standard terminology for improved patient recognition and data interoperability

This paper presents the Orphanet Nomenclature and Classification system as a comprehensive, multilingual, and interoperable standard for rare diseases, detailing its July 2025 update which includes nearly 10,000 clinical entities with extensive mappings to major medical terminologies to facilitate global data sharing, accurate patient identification, and improved healthcare outcomes.

The Gist

Imagine the world of medicine as a massive, bustling library. For common diseases like the flu or diabetes, the library has clear, bright signs, organized shelves, and a librarian who knows exactly where everything is. But for Rare Diseases (conditions that affect very few people), the library is a dark, chaotic maze. There are no signs, the books are scattered on the floor, and the librarians often don't even know these books exist.

This paper introduces Orphanet, a project that is building a master map and a universal language to fix this chaos. Here is the story of what they did, explained simply.

1. The Problem: The "Invisible" Patients

There are over 6,500 different rare diseases. Collectively, they affect hundreds of millions of people worldwide. However, because each disease is so rare, doctors in one country might not know about a disease common in another.

In the current medical "library," these diseases are often invisible. If a doctor tries to enter a patient's diagnosis into a computer system, they might not find a specific code for it. It's like trying to check out a book that doesn't have a barcode. Without a code, the patient becomes invisible to statistics, research, and healthcare planning. We can't count them, we can't track them, and we can't find them for clinical trials.

2. The Solution: The "Orphanet Dictionary"

The authors describe Orphanet, a specialized dictionary and filing system created specifically for rare diseases. Think of it as the GPS for Rare Diseases.

Every single rare disease in this system gets a unique ID card called an ORPHAcode.

• The ID Card: Just like a passport number, this code never changes, even if the name of the disease changes.
• The Definition: Each card has a clear, simple description of what the disease looks like (symptoms) so doctors know exactly what they are dealing with.
• The Synonyms: It lists all the different names people might use for the same disease (like "Celiac Disease" vs. "Gluten Intolerance"), ensuring no one gets lost in translation.

3. The Structure: A Multi-Layered Filing System

The paper explains that Orphanet isn't just a flat list; it's a smart, 3D filing system with three levels of detail:

1. The Group (The Category): Like "Fiction" or "Science" in a library. This groups similar diseases together (e.g., "Rare Skin Disorders").
2. The Disorder (The Book): This is the specific disease itself (e.g., "Epidermolysis Bullosa"). This is the level that counts as a "Rare Disease."
3. The Subtype (The Chapter): This is the fine print. It breaks the disease down by cause (genetic) or specific symptoms. This helps researchers who need to know the tiny details.

The "Multi-Parent" Analogy:
Rare diseases are tricky because they often affect the whole body, not just one organ. A disease might affect the heart, the skin, and the brain.

• In a normal library, a book can only sit on one shelf.
• In Orphanet, a "book" (disease) can sit on multiple shelves at once. It is filed under "Heart," under "Skin," and under "Brain." This ensures that a cardiologist, a dermatologist, and a neurologist can all find the same patient's information.

4. The Translation: Speaking All Languages

Medicine is global, but languages are not. Orphanet translates this dictionary into 9 languages (English, French, German, Spanish, Italian, Dutch, Polish, Portuguese, and Czech).

• Imagine a doctor in Poland and a researcher in Brazil talking about the same patient. Without Orphanet, they might use different words and think they are talking about two different things.
• With Orphanet, they both use the same ORPHAcode. It's like a universal translator that ensures everyone is speaking the same medical language, regardless of their native tongue.

5. The Bridge: Connecting to the Big Systems

The paper highlights a major achievement: Orphanet has built bridges to the world's biggest medical filing systems (like ICD-10, ICD-11, and SNOMED CT).

• The Analogy: Think of the big medical systems as the "Old City" with narrow, confusing streets. Orphanet is the "New City" with wide, modern roads.
• Orphanet has built bridges (mappings) connecting the two. Now, when a doctor in the "Old City" enters a code, the system knows exactly which "New City" address it corresponds to. This allows data to flow freely between countries and research groups.

6. Why This Matters: From Invisible to Visible

Why do we need all this?

• Counting the Unseen: Before, we didn't know how many people had rare diseases. Now, we can count them accurately.
• Finding the Needle in the Haystack: If a new drug is being tested, researchers can use these codes to find the specific patients who need it, even if they are scattered across different countries.
• Policy and Money: Governments need to know how many people are affected to decide how much money to spend on research and care. You can't budget for something you can't count.

The Bottom Line

This paper is essentially a report card on a massive project to organize the world's most chaotic medical library. By giving every rare disease a unique ID, a clear definition, and a place on the map, Orphanet is turning invisible patients into visible people. It ensures that no matter where you live or what language you speak, if you have a rare disease, the world's medical system can finally see you, understand you, and help you.

Technical Summary

1. Problem Statement

Rare diseases (RDs) collectively affect an estimated 329–624 million people globally, yet individually, 85% of the over 6,500 known RDs affect fewer than 1 in 1 million people. This scarcity creates significant challenges:

• Data Fragmentation: Critical data for improving care and treatment can only be achieved through standardized, cross-border collection.
• Under-representation in Terminologies: RDs are poorly represented in standard medical terminologies (e.g., ICD-10, SNOMED CT). Many lack specific, unambiguous codes, making it difficult to identify patients in health information systems.
• Lack of Interoperability: Different regions use diverse semantic resources (e.g., ICD-10 vs. ICD-11, local extensions), and existing systems often fail to explicitly indicate disease rarity, hindering data sharing, public health monitoring, and research.
• Inconsistent Definitions: There is no global consensus on the definition of an RD, with prevalence thresholds varying by region (e.g., <1/2,000 in Europe vs. <200,000 total in the US).

2. Methodology

The paper details the structure, production, and analysis of the Orphanet Nomenclature and Classification of RD, a standardized system maintained by Inserm (France) since 1997.

• Data Source: Analysis was performed using the Orphanet July 2025 release (Version 1.3.42 / 4.1.8).
• Data Extraction: Metrics were computed using Python 3.8 scripts (Jupyter Notebooks) processing XML and JSON datasets from Orphadata Science. Manual curation was used for specific counts (e.g., historical entities, translation coverage).
• System Structure:
  • Clinical Entities: Each entry includes a unique, time-stable ORPHAcode, a preferred term, synonyms, a classification level, and a textual definition.
  • Classification Levels: Three levels of granularity:
    1. Group of disorders (e.g., Clinical groups, Categories).
    2. Disorder (The core level corresponding to the WHO-RDI consensus definition of an RD).
    3. Subtype of disorder (Clinical, Etiological, or Histopathological subtypes).
  • Multi-hierarchical Structure: Entities are classified into 34 hierarchies (29 distinct medical domains + 5 organizational/ad-hoc groups). A "linearization" process assigns a single preferential parent for statistical analysis, while allowing multi-parental classification to reflect the multisystemic nature of RDs.
• Curation Process:
  • Inclusion: Requires a well-delineated clinical presentation, prevalence ≤1/2,000, and description in at least two unrelated individuals (or two family members with a known genetic cause).
  • Updates: Monthly literature review, manual curation, and expert validation.
  • Lifecycle Management: ORPHAcodes are never deleted or reused. Inactivated codes (due to obsolescence or merging) are replaced by active target codes to ensure traceability.
• Mapping Strategy: ORPHAcodes are mapped to international standards (ICD-10, ICD-11, SNOMED CT, OMIM, UMLS, MeSH, MedDRA, GARD, Mondo) using qualified proximity relationships (Exact, Broader-to-Narrower, Narrower-to-Broader).

3. Key Contributions

• Standardized Counting: Provides a clear, operational count of RDs based on the consensus definition (6,527 disorders), distinguishing them from subtypes and groups to prevent overcounting in epidemiological studies.
• Multilingual Interoperability: The nomenclature is available in 9 languages (English + 8 translations), with rigorous validation by local medical experts.
• Semantic Alignment: Establishes a comprehensive mapping framework that bridges the gap between clinical definitions (Orphanet) and administrative/research coding systems (ICD, SNOMED).
• Undiagnosed Patient Coding: Introduces a specific code (ORPHA:6616874) for patients who remain undiagnosed after full investigation, allowing them to be tracked in health systems and research.

4. Key Results (As of July 2025)

• Total Entities: 9,784 active clinical entities.
  • Disorders: 6,527 (66.7% of total).
  • Subtypes: 1,084 (11.1%).
  • Groups: 2,173 (22.2%).
  • Historic Entities: 326 (3.3% of disorders) with no publications in 25 years.
• Genetic Component: 72.2% of disorders are genetic; 63.4% are mapped to at least one phenotypic OMIM number.
• Mapping Coverage:
  • ICD-10: 97.4% of disorders mapped (only 6.4% with exact relationship; others are index/inclusion/attribution terms).
  • ICD-11: 71.8% mapped (14.7% exact).
  • SNOMED CT: 94.8% mapped (all with exact relationship).
  • OMIM: 63.4% mapped (50.9% exact).
  • Other Standards: High coverage in UMLS (99.6%) and Mondo (94.9%).
• Distribution by Domain:
  • Multiclassification: 67.7% of disorders are classified in more than one medical domain, reflecting their complex nature.
  • Top Domains (Linearized): Rare developmental defects during embryogenesis (32.3%) and Rare neurologic disorders (16.8%) account for nearly 50% of all RDs.
• Growth Rate: Between 2020–2025, an average of 123 new disorders were added annually, while 44 were inactivated, resulting in a net annual increase of <80 disorders.

5. Significance

• Public Health & Policy: The Orphanet Nomenclature is the only RD-specific terminology meeting high-quality standards for medical terminologies. It enables accurate prevalence estimation, resource allocation, and policy-making by providing a stable, consensus-based definition of RDs.
• Clinical Care: Facilitates the identification of undiagnosed patients and supports the referral of patients to appropriate centers of expertise by capturing the multisystemic nature of diseases.
• Research & Innovation: By linking clinical phenotypes to genetic data (OMIM) and regulatory vocabularies (MedDRA), it accelerates patient recruitment for clinical trials and supports the development of orphan drugs.
• Global Adoption: Recognized by the WHO, the EU, and the World Health Assembly (2025 resolution) as a critical tool for global health equity. Its open licensing (CC-BY 4.0) and integration into the eHealth Digital Service Infrastructure (eHDSI) promote cross-border data interoperability.

In conclusion, the Orphanet Nomenclature serves as the foundational infrastructure for making rare diseases "visible" in health systems, transforming fragmented data into actionable knowledge for better patient outcomes.