Inferensys

Glossary

mCODE (Minimal Common Oncology Data Elements)

A consensus-driven, FHIR-based data standard created to facilitate the exchange of essential oncology data for research, treatment, and patient care.
Stylish WeWork-like workspace with hot desks and document wall, professional searching through enterprise knowledge base on a mounted ultrawide display, warm industrial pendants overhead.
ONCOLOGY INTEROPERABILITY STANDARD

What is mCODE (Minimal Common Oncology Data Elements)?

A consensus-driven, FHIR-based data standard created to facilitate the exchange of essential oncology data for research, treatment, and patient care.

mCODE (Minimal Common Oncology Data Elements) is a consensus-driven, open-source data standard built on HL7 FHIR that defines a core set of structured oncology data elements for seamless electronic exchange across health systems. It standardizes the representation of a patient's cancer diagnosis, genomic markers, treatment history, and outcomes into a common computable format, enabling interoperability between disparate electronic health records (EHRs) and research databases.

Developed by the American Society of Clinical Oncology (ASCO) in collaboration with MITRE, mCODE leverages FHIR profiles to model key clinical concepts like CancerPatient, PrimaryCancerCondition, and CancerRelatedSurgicalProcedure. By establishing a minimal yet comprehensive data layer, it directly supports use cases in federated learning, real-world evidence generation, and clinical decision support without requiring institutions to abandon their local data schemas.

MINIMAL COMMON ONCOLOGY DATA ELEMENTS

Core Characteristics of mCODE

A consensus-driven, FHIR-based data standard created to facilitate the exchange of essential oncology data for research, treatment, and patient care.

01

FHIR-Based Interoperability

mCODE is built entirely on the HL7 FHIR (Fast Healthcare Interoperability Resources) standard, using modern RESTful APIs and web technologies. This ensures seamless integration with existing Electronic Health Records (EHRs) and other FHIR-compliant systems.

  • Uses standard FHIR Resources like Patient, Condition, Observation, and Procedure
  • Defines specific FHIR Profiles that constrain these resources for oncology use cases
  • Enables plug-and-play interoperability across different vendor systems
02

Six Core Data Domains

mCODE organizes essential oncology data into six foundational domains, each representing a critical aspect of a patient's cancer journey. These domains were selected by a multidisciplinary expert panel as the minimal set necessary for care and research.

  • Patient: Demographics and baseline characteristics
  • Disease: Primary cancer diagnosis, staging, and histology
  • Genomics: Somatic mutations, biomarkers, and genomic reports
  • Treatment: Medications, surgeries, and radiation therapy
  • Outcome: Survival status, disease progression, and performance status
  • Assessment: Cancer staging evaluations and tumor marker tests
03

Consensus-Driven Governance

mCODE is developed and maintained through an open, multi-stakeholder governance process led by the American Society of Clinical Oncology (ASCO) and the MITRE Corporation. This ensures the standard reflects real-world clinical needs rather than any single vendor's priorities.

  • Governed by an Executive Committee with representatives from oncology, informatics, and patient advocacy
  • All development occurs in public GitHub repositories with community input
  • Regular balloting and comment periods ensure transparency and clinical validity
04

Real-World Data Enablement

A primary goal of mCODE is to transform unstructured clinical data trapped in EHR narrative text into structured, computable data suitable for research. By standardizing data elements, mCODE enables:

  • Pragmatic clinical trials using real-world evidence
  • Cohort discovery for identifying eligible patients across institutions
  • Quality reporting for value-based care programs
  • Federated learning by providing a common data schema across sites without requiring centralized data aggregation
05

Extensible Profiling Architecture

While mCODE defines a minimal set of data elements, it is designed to be extended. Implementers can layer additional FHIR profiles on top of the mCODE base to capture specialty-specific or trial-specific data without breaking conformance.

  • Uses FHIR's Extension mechanism for custom data elements
  • Supports US Core profiles as a foundational layer
  • Enables Implementation Guides (IGs) for specific cancer types or research protocols
  • Maintains backward compatibility as extensions evolve
06

Genomics Integration

mCODE includes a dedicated Genomics Domain that standardizes how somatic mutations and biomarkers are represented, bridging the gap between molecular pathology and clinical oncology.

  • Aligns with the Global Alliance for Genomics and Health (GA4GH) standards
  • Uses FHIR's Observation resource with specific genomic profiles
  • Captures gene studied, variant description, and clinical significance
  • Enables structured reporting of next-generation sequencing (NGS) results for treatment decision support
mCODE EXPLAINED

Frequently Asked Questions

Clear, technical answers to the most common questions about the Minimal Common Oncology Data Elements (mCODE) standard and its role in interoperable cancer care and federated research.

mCODE, or Minimal Common Oncology Data Elements, is a consensus-driven, FHIR-based data standard created to facilitate the exchange of essential oncology data for research, treatment, and patient care. It works by defining a core set of structured data elements—such as cancer diagnosis, genomic variants, and treatment history—that are mapped to specific FHIR resources like Condition, Observation, and MedicationAdministration. This creates a universal 'lingua franca' for cancer data, allowing disparate electronic health record (EHR) systems to transmit a standardized, computable summary of a patient's cancer journey without requiring a complete data dump of the entire medical record.

INTEROPERABILITY COMPARISON

mCODE vs. Other Oncology Data Standards

A feature-level comparison of mCODE against other prominent data standards used in oncology research and clinical care.

FeaturemCODENAACCROMOP CDM

Base Framework

FHIR R4

Fixed-width text files

Relational database schema

Primary Use Case

Real-time clinical care and research

Central cancer registry reporting

Large-scale observational analytics

Data Exchange Mechanism

RESTful API

Batch file transfer

ETL and bulk export

Granularity

Individual data elements

Summary case records

Standardized clinical concepts

Extensibility

Terminology Binding

Mandatory (SNOMED CT, LOINC)

NAACCR codes

Standardized vocabularies (SNOMED, RxNorm)

Patient-Facing Access

Real-Time Updates

Prasad Kumkar

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.