What “reached Phase 3” actually filters

This distinction trips people up: Drug & Asset separates highest phase from current status, and they answer different questions.

Filtering on highest_phase=["phase_3"] alone — as this tutorial does — returns every drug that has touched Phase 3 or beyond at some point, not only the ones still moving. For a genuinely “active pipeline” view, add a dev_status filter alongside it.

Why this needs Drug & Asset

Claude doesn’t maintain a live drug database — it can’t retrieve current pipeline status, development milestones, or sponsor assignments for commercial assets on its own. Drug & Asset provides structured pipeline records filterable by target, disease, organization, phase, and milestone date across three tools: drug_search for filtering and listing, drug_milestone_fetch for timeline detail, and drug_fetch for batch profile retrieval — each returning current records from Patsnap’s curated life sciences intelligence.

Prepare your query

Decide which attributes define the set you want to track — common filters are target name, disease indication, sponsoring organization, highest development phase, and country. This tutorial’s question is “Which PD-1-targeted drugs in China have reached Phase 3 or beyond?”, using target “PD-1”, minimum phase “phase_3”, and country code “CN”. You don’t need to look up numeric identifiers or construct query syntax — Claude translates your natural-language request into the tool call once Drug & Asset is connected.

1Connect Drug & Asset

1. Get a Patsnap Open Platform API key.
2. Copy the connection URL or JSON config from the Connect panel on the Drug & Asset page.
3. Add the server to your AI assistant or coding agent or another MCP-compatible client, then restart it.
4. Confirm the drug search, milestone, and fetch tools are available before you ask.

2Run the example query

Show me PD-1 drugs in China that have reached Phase 3 or beyond.

Claude recognizes the request and calls drug_search(target=["PD-1"], highest_phase=["phase_3"], country=["CN"]), retrieving drugs filtered by target, highest phase reached, and country of activity.

3What the result tells you

The query returned 22 total hits. The top four:

Each row is a distinct asset meeting all three filters. IBI-363, for instance, is a fusion protein targeting both PD-1 and IL-2, developed by Innovent. The remaining 17 assets each carry their own mechanism and pipeline detail.

4Go further with follow-up questions

From the same session:

• “Retrieve the development milestones for IBI-363.” — traces the timeline via drug_milestone_fetch.
• “What other PD-1 drugs from Innovent are in the system?” — compares a sponsor’s portfolio.
• “Show me PD-1 drugs in China at Phase 2 instead of Phase 3.” — widens or shifts the phase filter.
• “Which of these 22 are still actively in development?” — layers a dev_status filter onto the original result.