🧬 NCI Thesaurus Concept Scraper

🚀 Export biomedical and cancer concepts in seconds. Pull NCI Thesaurus concepts straight from the official NCI EVS REST API by keyword or concept code.

🕒 Last updated: 2026-06-08 · 📊 Up to 16 fields per record · live terminology · 1,600+ concepts for a single search like "melanoma"

Turn the NCI Thesaurus into clean, structured records you can drop into an ontology, a clinical data dictionary, a research database, or a terminology mapping pipeline. Search a keyword or list exact NCIt codes and get one tidy record per concept with names, synonyms, definitions, semantic types, and the hierarchy around each term.

Coverage is the NCI Thesaurus (NCIt) as published through the NCI Enterprise Vocabulary Services (EVS) REST API: preferred names, synonyms from contributing sources like CDISC and GDC, authoritative definitions, semantic types, concept status, and the parent and child concepts that define each term's place in the cancer and biomedical hierarchy.

🎯 Target Audience	💡 Primary Use Cases
Clinical data and biomedical informatics teams	Map local terms onto standardized NCIt codes
Cancer and life-science researchers	Build concept lists for a study or registry
Ontology and vocabulary engineers	Seed or audit a terminology with parents and children
Health-tech and EHR developers	Power code lookups and synonym resolution

📋 What the NCI Thesaurus Concept Scraper does

This Actor calls the public NCI EVS REST API and returns one clean record per concept, with two ways to select what you want:

• Search by term — find every NCI Thesaurus concept matching a keyword such as melanoma or BRAF gene.
• By concept codes — fetch specific concepts by their NCIt code, for example C3224 or C16612.

Choose a detail level per run. Summary returns names, synonyms, definitions, semantic type, and status. Full adds the parent and child concepts plus extra properties. Every record carries a scrapedAt timestamp.

🎬 Full Demo (🚧 Coming soon)

⚙️ Input

Field	Type	Description
mode	select	How to select concepts. One of search or codes. Defaults to search.
term	string	Keyword to search the NCI Thesaurus. Used when mode is search.
codes	array	List of NCIt concept codes (for example C3224). Used when mode is codes.
include	select	Detail level per concept. One of summary or full. Defaults to full.
maxItems	integer	How many records to return. Free plan is capped at 10.

Example 1 — search by keyword

{
    "mode": "search",
    "term": "melanoma",
    "include": "full",
    "maxItems": 25
}

Example 2 — fetch specific codes

{
    "mode": "codes",
    "codes": ["C3224", "C16612", "C12434"],
    "include": "full",
    "maxItems": 3
}

⚠️ Good to Know: parent and child concepts are populated only when include is set to full. With summary they come back as empty arrays, which keeps the records lighter when you only need names, synonyms, and definitions.

📊 Output

Each concept record looks like this:

Field	Description
📌 name	Preferred concept name
🆔 code	NCIt concept code
🔗 url	NCI Thesaurus browser page for the concept
🏷 displayName	Display name for the concept
🧬 semanticTypes	Semantic type labels (for example Neoplastic Process)
📝 definition	Authoritative concept definition
📊 conceptStatus	Concept status flag
✅ active	Whether the concept is active
🍃 leaf	Whether the concept has no children
🏷 terminologyVersion	NCIt version the data came from
🧫 neoplasticStatus	Neoplastic status when present
🔗 umlsCui	Linked UMLS concept identifier
🗂 contributingSources	Source vocabularies that contribute terms
🏷 synonyms	Distinct synonym names
⬆ parents	Parent concepts (code, name)
⬇ children	Child concepts (code, name)
🕒 scrapedAt	Collection timestamp
❌ error	Null on success

Real sample — concept (full detail)

{
    "name": "Melanoma",
    "code": "C3224",
    "url": "https://ncithesaurus.nci.nih.gov/ncitbrowser/ConceptReport.jsp?dictionary=NCI_Thesaurus&code=C3224",
    "displayName": "Melanoma",
    "semanticTypes": ["Neoplastic Process"],
    "definition": "A malignant, usually aggressive tumor composed of atypical, neoplastic melanocytes. Most often, melanomas arise in the skin (cutaneous melanomas) and include the following histologic subtypes: superficial spreading melanoma, nodular melanoma, acral lentiginous melanoma, and lentigo maligna melanoma...",
    "conceptStatus": "DEFAULT",
    "active": true,
    "leaf": false,
    "terminologyVersion": "26.05d",
    "neoplasticStatus": "Malignant",
    "umlsCui": "C0025202",
    "contributingSources": ["CDISC", "Cellosaurus", "CPTAC", "CTEP", "CTRP", "GDC", "MedDRA", "NICHD"],
    "synonyms": ["Melanoma", "Malignant Melanoma", "MELANOMA, MALIGNANT", "Melanoma, NOS", "melanoma"],
    "parents": [
        { "code": "C9305", "name": "Malignant Neoplasm" },
        { "code": "C7058", "name": "Melanocytic Neoplasm" }
    ],
    "children": [
        { "code": "C3802", "name": "Amelanotic Melanoma" },
        { "code": "C3510", "name": "Cutaneous Melanoma" },
        { "code": "C8925", "name": "Metastatic Melanoma" }
    ],
    "scrapedAt": "2026-06-08T18:00:39.770Z",
    "error": null
}

Real sample — pathway concept

{
    "name": "Melanoma Pathway",
    "code": "C91477",
    "url": "https://ncithesaurus.nci.nih.gov/ncitbrowser/ConceptReport.jsp?dictionary=NCI_Thesaurus&code=C91477",
    "displayName": "Melanoma Pathway",
    "semanticTypes": ["Functional Concept"],
    "definition": "Five distinct stages have been proposed in the evolution of melanoma on the basis of histological criteria...",
    "conceptStatus": "DEFAULT",
    "active": true,
    "leaf": true,
    "terminologyVersion": "26.05d",
    "neoplasticStatus": null,
    "umlsCui": "C2984289",
    "contributingSources": [],
    "synonyms": ["Melanoma", "Melanoma Pathway"],
    "parents": [{ "code": "C91436", "name": "Cancer Pathway" }],
    "children": [],
    "scrapedAt": "2026-06-08T18:02:18.402Z",
    "error": null
}

Real sample — code lookup

{
    "name": "Blood",
    "code": "C12434",
    "url": "https://ncithesaurus.nci.nih.gov/ncitbrowser/ConceptReport.jsp?dictionary=NCI_Thesaurus&code=C12434",
    "displayName": "Blood",
    "semanticTypes": ["Body Substance"],
    "definition": "A liquid tissue; its major function is to transport oxygen throughout the body...",
    "conceptStatus": "DEFAULT",
    "active": true,
    "leaf": false,
    "terminologyVersion": "26.05d",
    "neoplasticStatus": null,
    "umlsCui": "C0005767",
    "contributingSources": ["CDISC", "GDC", "NCI-GLOSS"],
    "synonyms": ["Blood", "BLOOD", "blood"],
    "parents": [{ "code": "C13236", "name": "Body Fluid or Substance" }],
    "children": [],
    "scrapedAt": "2026-06-08T18:02:18.402Z",
    "error": null
}

✨ Why choose this Actor

• One clean record per concept, with consistent field names that map onto a database schema.
• Two selection modes in one Actor, so keyword discovery and exact code lookup share the same output shape.
• Synonyms are de-duplicated and definitions favor the authoritative NCI source.
• Parent and child concepts come fully named, not as bare codes.
• No login, no API key, and direct from the official NCI EVS REST API.

📈 How it compares to alternatives

Approach	Effort	Structured fields	Hierarchy included	Maintenance
This Actor	One run	Yes	Yes	None on your side
Hand-copying from the NCIt browser	Hours	Inconsistent	Manual	Constant
Writing your own EVS API client	Days	Depends	Manual	You own the upkeep

🚀 How to use

1. Create a free Apify account using this sign-up link.
2. Open the NCI Thesaurus Concept Scraper.
3. Choose a mode (search a term or list concept codes).
4. Set include to summary or full, and maxItems to the number of records you want.
5. Click Start and grab your results when the run finishes.

💼 Business use cases

Clinical data and informatics

Goal	How this helps
Standardize local terms	Map free text onto NCIt codes and synonyms
Build a study code list	Search a topic and capture every matching concept

Ontology and vocabulary work

Goal	How this helps
Audit a terminology branch	Pull a concept with its parents and children
Resolve synonyms	Use the de-duplicated synonym arrays

Research and analytics

Goal	How this helps
Assemble a cancer concept set	Search by disease and export the records
Link to other vocabularies	Use the UMLS CUI and contributing sources

Product and engineering

Goal	How this helps
Power a code lookup feature	Fetch concepts by NCIt code on demand
Seed a reference table	Snapshot a topic into your database

🔌 Automating NCI Thesaurus Concept Scraper

Connect runs to the tools you already use:

• Make and Zapier to trigger runs and route records into sheets or databases.
• Slack to post a summary when a run finishes.
• Airbyte to load results into a warehouse.
• GitHub Actions to schedule periodic snapshots of a concept set.
• Google Drive to archive each run's output.

🌟 Beyond business use cases

• Research: assemble concept lists for a systematic review or registry.
• Personal: explore the cancer terminology hierarchy around a diagnosis.
• Non-profit: power a patient-education glossary with authoritative definitions.
• Experimentation: prototype a terminology app without writing an API client.

🤖 Ask an AI assistant

Paste your results into ChatGPT, Claude, Perplexity, or Microsoft Copilot and ask it to group concepts by semantic type, summarize a definition, or build a concept hierarchy from the parent and child fields.

❓ Frequently Asked Questions

Do I need an NCI or NIH account? No. The Actor reads the public NCI EVS REST API, which needs no login.

Do I need an API key? No key is required.

What is the NCI Thesaurus? It is the National Cancer Institute's reference terminology of cancer and biomedical concepts, published as NCIt through NCI Enterprise Vocabulary Services.

How do I find a concept code? Run a search by term first, then read the code field, or enter codes you already know in codes mode.

What is the difference between summary and full detail? Summary returns names, synonyms, definitions, semantic type, and status. Full adds parent and child concepts plus extra properties.

Why are parents or children empty on some records? They are populated only when include is full, and some concepts genuinely have no children (a leaf) or sit near the top of the hierarchy.

How many concepts can a search return? A broad search like "melanoma" matches over 1,600 concepts; you control how many come back with maxItems.

Are synonyms de-duplicated? Yes. Synonym names are collapsed to a distinct list per concept.

Which definition is shown? The Actor favors the authoritative NCI definition and falls back to another source when NCI does not provide one.

How fresh is the data? Each run pulls live from the NCI EVS API and records the NCIt version in terminologyVersion.

🔌 Integrate with any app

Results are available through the Apify API, so you can pull them into any app, database, or workflow you already run.

🔗 Recommended Actors

• Have I Been Pwned Breaches Catalog Scraper
• More reference and data Actors in the ParseForge collection

💡 Pro Tip: browse the complete ParseForge collection.

🆘 Need Help? Open our contact form

⚠️ Disclaimer: independent tool, not affiliated with the National Cancer Institute or the National Institutes of Health. Only publicly available data is collected.