ChEMBL Drug Pipeline Scraper - Phases & Approvals

What does ChEMBL Drug Pipeline Scraper do?

Which molecules target which disease, at what development phase — with approval year, drug class, route of administration and safety flags.

No login. No API key. No proxies.

The join ChEMBL does not make

ChEMBL publishes 2.9 million molecules on one side and 60,055 indications on the other, and never joins them. The question a competitive intelligence team actually asks lives in the join:

What is in development for Alzheimer's, at phase III or above, and who else is there?

Give this Actor an indication and it queries the indications, collects the molecules, enriches them in batches, and returns one row per molecule-indication pair — with the phase reached for that indication specifically, which is not the same as the molecule's highest phase elsewhere.

Molecule	Indication	Phase	Class
RIVASTIGMINE	Alzheimer disease	4	cholinesterase inhibitors
DONEPEZIL	Alzheimer disease	4	acetylcholinesterase inhibitors
SEMAGACESTAT	Alzheimer disease	3	enzyme inhibitors: gamma secretase

306 molecules have been developed for Alzheimer's; 80 reached phase III or beyond.

Two fields made readable

developmentStage. ChEMBL returns max_phase: 4.0 and stops. Here that also arrives as Approved, Phase III, Phase II, Phase I or Preclinical, so nobody has to carry the scale in their head.

drugClass. Drug names encode their pharmacology in the suffix — -mab is a monoclonal antibody, -tinib a kinase inhibitor. ChEMBL records the expansion in usan_stem_definition, and almost nobody surfaces it. Here it is a plain column: monoclonal antibodies: fully human, acetylcholinesterase inhibitors, peroxisome proliferator activated receptor agonists. Measured on approved antibodies: 100 % populated.

Output

Field	Example
name, chemblId	RIVASTIGMINE, CHEMBL636
indication, indicationPhase	Alzheimer disease, 4
meshHeading, efoId	ontology identifiers
maxPhase, developmentStage	4.0, Approved
firstApproval	1997
moleculeType	Small molecule / Antibody / Protein
drugClass, usanStem	cholinesterase inhibitors, -stigmine
atcCodes, atcLevel1	N06DA03, N
routes	oral, parenteral, topical
blackBoxWarning, withdrawn, orphanDrug	safety and status flags
firstInClass, prodrug, naturalProduct
molecularWeight, alogp, polarSurfaceArea, ro5Violations, qedWeighted	drug-likeness
formula, smiles, inchiKey	structure
synonyms	development codes and trade names

Input

{
  "indication": "Alzheimer",
  "minPhase": "3",
  "maxItems": 500
}

Field	Default	Notes
indication	Alzheimer	Switches to the pipeline view
minPhase	—	3 and above is the near-term competitive set
maxItems	500	Rows, sorted most advanced first
moleculeType	all	No indication: small molecule, antibody, …
approvedSince	—	Only 391 molecules approved since 2020
withdrawnOnly	false	364 molecules withdrawn from a market
blackBoxOnly	false	936 carry the strongest FDA warning
firstInClassOnly	false	352 opened a new class

Leave indication empty and the filters below it apply to molecules directly — useful for "every approved antibody" or "every withdrawn drug".

Use cases

• Competitive intelligence — the full development set for an indication, by phase.
• Business development and licensing — molecules at phase II or III in a therapeutic area.
• Safety and pharmacovigilance — withdrawn drugs and black-box warnings, with their classes.
• Chemistry and screening — structures, drug-likeness and Lipinski violations for a filtered set.
• Market and regulatory analysis — approvals by year, class and route.

Limits, honestly

• ChEMBL is not a live clinical trial registry. Phases reflect what has been curated from the literature and regulatory sources, and lag current trial status. For active recruitment, use ClinicalTrials.gov.
• Indication search is text matching on the EFO term, so "cancer" returns many distinct conditions. Check indication on the rows.
• indicationPhase and maxPhase differ on purpose. A drug approved for one condition may be at phase II for another; the first is indication-specific, the second is the molecule's highest anywhere.
• Antibodies, proteins and cell therapies have no SMILES — the structural columns are empty for them by nature, not by omission.
• drugClass is derived from the naming stem, so molecules without a USAN stem have it blank. Measured: 65 % on a mixed Alzheimer set, 100 % on approved antibodies.
• Requests are paced out of courtesy to a free public service run by the EMBL-EBI.
• Not affiliated with ChEMBL or the EMBL-EBI.

How much does it cost?

You pay per molecule row returned: $0.006 each, that is $6.00 per 1,000. There is no start fee, and subscription plans pay less per molecule row.

The example input below asks for up to 500 molecule rows, so it costs $3.00 at most.

If a run reaches the spending limit you set, the output stops at that limit and never goes past it. You are never charged for rows that were not delivered.

Use ChEMBL Drug Pipeline Scraper as an API

Call it from your own code with the Apify client, here in Python:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("DataIO/chembl-drug-pipeline-approvals").call(run_input={'indication': 'Alzheimer', 'maxItems': 500})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

It also works from JavaScript, Make, Zapier, n8n, and from AI agents through the Apify MCP server.

Other actors you might like

• FDA 510(k) Scraper: Device Clearances Data
• FDA Recalls Scraper: openFDA Enforcement
• NPI Registry Scraper: US Healthcare Providers

FAQ

Is it legal to use this data?

The actor reads public data from its official source, without logging in and without bypassing any access control. What you do with the data, for example contacting people listed in it, is your responsibility under the laws that apply to you, such as GDPR in Europe.

Can I run it on a schedule?

Yes. Create a schedule in Apify Console, daily or weekly for example, and each run delivers a fresh dataset, which you can send by email, webhook or integration.

Can AI agents use it?

Yes. It is available through the Apify MCP server, and every input field is described in its input schema, so an agent can call it directly.