Public API

Interactive documentation is served by the API at /api/v1/docs. Base URL: https://www.cancerindex.io/api/v1.

Every successful response wraps its payload with the sources that contributed and the data release, so attribution travels with the data:

{
  "data": { ... } | [ ... ],
  "sources": [ { "slug": "oncotree", "name": "OncoTree", "license": "CC BY 4.0", "retrievedAt": "2026-09-08T09:07:07Z" } ],
  "dataRelease": "2026-09-08",
  "meta": { "page": 1, "pageSize": 50, "total": 865 }   // on paginated endpoints
}

Errors return { "error": { "code": "not_found", "message": "…" } } with the matching HTTP status.

• Anonymous: 60 requests per minute per IP.
• With an API key (Authorization: Bearer …): per-key limit according to tier (free 60/min, research and institutional tiers on request).
• Responses include X-RateLimit-Limit, X-RateLimit-Remaining and Retry-After on 429.

# One cancer entity with codes, aliases and counters
curl -s https://www.cancerindex.io/api/v1/cancers/diffuse-glioma | jq .

# Search across entities
curl -s "https://www.cancerindex.io/api/v1/search?q=glioma" | jq '.data[] | {type, title}'

# Current ranking snapshot for a metric and scope
curl -s "https://www.cancerindex.io/api/v1/rankings/active_trials?scope=geo=WORLD|sex=all|age=all|year=latest|level=top" | jq .

# Same ranking as CSV with attribution rows (served by the web app)
curl -sL "https://www.cancerindex.io/api/export/rankings.csv?metric=active_trials" -o active_trials.csv

Public IDs look like CI-CAN-00000123 (cancer), CI-GENE-…, CI-VAR-…, CI-DRUG-…, CI-TRIAL-…, CI-PUB-…, CI-SOURCE-…. They are minted once and never reused; merged entities redirect to the surviving record. Upstream identifiers (NCIt, OncoTree, ICD-10, DOID, HGNC, NCT, PMID) are returned alongside with their match type.