ClinicalTrials.gov Material Change Monitor

Monitor recruiting clinical studies. Pay only when a study is first seen (NEW_TRIAL) or its record materially changes later (SOURCE_CORRECTION).

• buyer: clinical operations, site-feasibility, competitive-intelligence, and research teams
• result: one source-linked trial record — either a first-seen recruiting study or a material correction to an already-tracked record
• official source: ClinicalTrials.gov API v2
• first useful run: use the bounded public example with baselineMode=emit_matches
• maximum charge: $0.25 with maxEvents=1
• unchanged replay: $0.00; the identical stateful replay emits no dataset row

buyer value

Enroll conditions and recruitment statuses once, then receive evidence-backed record deltas instead of repeatedly comparing bulk exports. Two distinct event types keep the signal honest:

• NEW_TRIAL — a study matching your watchlist that has never been seen by this state store before. First-seen recruiting trials. Priced at $0.25.
• SOURCE_CORRECTION — a study already tracked by this state store whose sponsor-submitted record changed materially (fingerprint mismatch). Later material record changes. Priced at $0.05.

Stable SHA-256 fingerprints, idempotent settlement, and hard event/spend caps prevent replay billing. Cap-deferred records are preserved in state for later replay when budget allows.

source and scope

The Actor reads sponsor-submitted records from the official ClinicalTrials.gov API v2. Every delivered row links to the exact record at https://clinicaltrials.gov/study/{nctId} and includes a source fingerprint and screening disclaimer. A listing or status change is not proof of safety, efficacy, approval, enrollment availability, or medical advice.

60-second quick start

The public example is an intentionally bounded preview. It can emit at most one current melanoma recruiting record for a maximum event charge of $0.25.

{
  "watchlistId": "melanoma-recruiting-preview",
  "conditions": ["melanoma"],
  "statuses": ["RECRUITING"],
  "lookbackDays": 30,
  "baselineMode": "emit_matches",
  "maxEvents": 1,
  "maxTotalChargeUsd": 0.25,
  "maxSourceRecords": 100,
  "stateStoreName": "melanoma-recruiting-preview",
  "resetState": false,
  "qualityTestMode": false
}

For recurring production monitoring, use a stable watchlist/store and baselineMode=silent for the initial baseline. Subsequent material source changes then emit as SOURCE_CORRECTION without rebilling unchanged records.

Illustrative output shape

Illustrative — not a run receipt. The row shape below is illustrative; the named study evidence is independently source-verified.

verified source-backed example

Verified against ClinicalTrials.gov API v2 on November 3, 2026: NCT07890415, “Using an Artificial Intelligence (AI) Tool for Melanoma Diagnosis,” was RECRUITING, first posted October 22, 2026, and sponsored by Memorial Sloan Kettering Cancer Center. The bounded 30-day preview emits it as NEW_TRIAL on fresh state. This is source evidence, not proof of safety, efficacy, approval, or enrollment availability.

A delivered row contains eventType (NEW_TRIAL or SOURCE_CORRECTION), NCT identifier, title, recruitment status, sponsor, updated date, exact study link as sourceRecordUrl, sourceFingerprint, and disclaimer.

Pricing

• NEW_TRIAL (first-seen study) — $0.25
• SOURCE_CORRECTION (material record change) — $0.05

The bounded preview allows one primary event. Unmatched records, unchanged replays, silent baselines, and failed source fetches create no paid event; Apify platform usage may still apply under the user's plan.

Schedule, webhook, and API recipe

After the preview, create a production watchlist with a stable stateStoreName, run a silent baseline once, and schedule the same input daily or weekly. Route dataset-item webhooks to a clinical-operations queue. API clients should preserve state identifiers and set an explicit maximum run charge.

Empty-run troubleshooting

An empty dataset usually means no current record matched the condition/status/lookback, the silent baseline intentionally stored state, or the same record was unchanged. Inspect REPORT, broaden one condition deliberately, and widen the lookback within its schema limit. Do not reset state merely to manufacture a paid row.

Buyer bundle

Pair with Clinical Trial Fit Monitor: the fit monitor finds recently posted recruiting opportunities; this Actor follows enrolled trial records for material changes. Add the drug-safety monitor for a broader life-sciences intelligence workflow.