/ writing / clinical-research

ClinicalTrials.gov has every trial on earth. The search box hides them.

2026-08-27 · cynix · 3 min read

ClinicalTrials.gov is the global registry of clinical studies, run by the US National Library of Medicine. It is also a genuinely open, free, public API — the v2 endpoint returns clean JSON with no key, no proxy, no scraping. The problem is the website: it is built for a patient reading one study at a time, not for an analyst who wants every Phase 3 melanoma trial with a PD-1 inhibitor, recruiting in Germany, with enrollment over 200.

ClinicalTrials.gov Extractor wraps that official v2 API into one actor with one flat record shape per study. You pass a query; it pages the API and hands back typed rows.

What you can query

Input	What it filters on	Example
`condition`	The medical condition (query.cond)	`type 2 diabetes`, `melanoma`
`intervention`	A drug or device (query.intr)	`pembrolizumab`
`country`	Where the trial recruits (query.locn)	`Germany`, `United States`
`overallStatus`	Recruitment stage (filter.overallStatus)	`RECRUITING`, `COMPLETED`, `TERMINATED`
`maxResults`	Cap across pages (1–500)	`25` default

Start broad with a condition, layer an intervention to isolate a mechanism, and scope geography with country. Set overallStatus: RECRUITING when active enrolment is what you care about.

Example output

One record per study — a real row from a run of the actor:

{
  "nctId": "NCT03509935",
  "briefTitle": "Use of Bedside Ultrasonography on the Incidence of Acute Renal Failure in High-risk Surgical Patients",
  "overallStatus": "COMPLETED",
  "conditions": "Acute Kidney Injury",
  "interventions": "Intervention Ultrasound Group",
  "phase": "NA",
  "studyType": "INTERVENTIONAL",
  "enrollmentCount": "111",
  "startDate": "2018-03-12",
  "completionDate": "2019-03-31",
  "leadSponsor": "Federal University of Minas Gerais",
  "url": "https://clinicaltrials.gov/study/NCT03509935",
  "fetchedAt": "2026-08-21T12:46:26.107Z"
}

Why this over the website?

Official API, not a scrape — authoritative data straight from the NLM endpoint; no HTML parsing to break when they redesign.
One record shape — NCT id, status, phase, conditions, interventions, enrollment, dates, lead sponsor, URL. Flat, typed, pipeline-ready.
Combine filters — condition + intervention + country + status in a single run; the web UI makes you click through tabs.
Auto-paging — set maxResults high and get the full cohort without manual page-walking.
No key, no proxy — it is a public US government endpoint built for programmatic access.

Use cases

Competitive intelligence — what pipelines a competitor is running, by condition and phase.
Therapeutic-area analysis — pipeline density and trial-status mix across a disease area.
Patient-recruitment research — trials actively recruiting, scoped by region.
Investor diligence — pipeline and trial-status signals ahead of a catalyst.
Academic groundwork — frame a new study against what already exists.

Try it: ClinicalTrials.gov Extractor on Apify — the official registry as typed records, no API key.