FDA Catalyst Research, Newsletters, and Event Data

A disciplined source stack for FDA events, with calendars, source documents, event timestamps, and post-event research kept separate.

Biotech calendars look precise because they contain dates. The precision can be misleading. An advisory-committee meeting, a PDUFA goal date, a Complete Response Letter, a clinical hold, and a later resolution are different events with different public records and different market clocks.

This guide shows how I would combine official FDA sources, specialist commentary, issuer disclosures, and structured event data. The central rule is simple: use the official source to establish the event, use research to understand it, and use point-in-time data to test comparable cases.

Start with the official FDA record

The FDA Advisory Committee Calendar publishes meeting notices and links to the individual meeting pages. Those pages can contain the agenda, Federal Register notice, briefing materials, webcast details, and later meeting material. The calendar is the primary source for whether a public advisory meeting is scheduled.

FDA's product-review explanation says advisory committees provide independent advice and that their recommendations are nonbinding. That distinction belongs in every event record. A favorable vote is evidence about the panel's view on a defined question. It is not the agency's final approval decision.

• Save the meeting page URL and the time you first observed it.
• Preserve every voting question instead of collapsing the meeting into one label.
• Record yes, no, and abstain counts for each question when available.
• Link the briefing documents and note when each version appeared.
• Keep later FDA action as a separate event.

There is no single complete public catalyst page

The official advisory calendar covers advisory meetings, not every PDUFA goal date or company catalyst. Goal dates often enter the public record through sponsor disclosures, filing exhibits, earnings materials, or specialist calendars. Dates can move, so each observed date needs a source and retrieval timestamp.

Alphanume does not provide a complete PDUFA calendar. Its biotech datasets cover two narrower jobs: structured FDA response events and advisory-committee vote records. Use a separate, verified calendar workflow when your strategy requires upcoming goal dates, and confirm each date against a primary issuer or agency document.

What Alphanume's event data covers

The FDA Response Events dataset organizes disclosed Complete Response Letters, clinical holds, refuse-to-file decisions, and resolution events into dated records with source links and issuer identifiers when available. The Advisory Committee Votes dataset preserves committee questions and vote counts for analysis.

Coverage should be read from the data contract, not inferred from the product name. FDA Response Events starts on January 2, 2024 and structures model-extracted events disclosed in SEC 8-K filings, so multiple filing rows can refer to one underlying event. Advisory Committee Votes starts on January 1, 2024 and covers drug and biologic meetings, but it does not itself supply stock prices, ticker mapping, or the later final FDA decision. The FDA response-event documentation describes the served fields and caveats before you build a cohort.

Alphanume publishes this guide and sells access to these datasets, which is our disclosed affiliation. The fit is event-study construction and historical comparison. The product is not a recommendation on a drug, a forecast of FDA action, or a substitute for the official record.

Newsletters should explain the mechanism

A useful biotech newsletter earns its place after the event has been verified. BioPharma Dive offered a weekday Daily Dive plus specialist weekly newsletters when checked on September 4, 2026. BIO.News maintained a daily public archive focused on biotech policy and industry developments. BioPharmCatalyst's FDA calendar serves the separate monitoring job. These sources are listed by workflow fit, not claimed returns, and Alphanume has no stated affiliation with them.

Read the public methodology and sample work before paying. Look for links to trial registries, FDA documents, filings, and company materials. Treat unnamed sources, price targets, and confident approval odds as hypotheses. A writer can have deep domain expertise and still supply no reproducible probability estimate.

When a publication quotes an approval probability, write down how you could score it later. A calibrated forecaster who assigns 70% should be right roughly seven times in ten across comparable calls, while a string of isolated winners proves very little. Save the forecast before the decision and keep withdrawals, delays, and changed endpoints in the audit.

• Clinical fit: Does the writer explain the endpoint and patient population in plain language?
• Regulatory fit: Are advisory votes, goal dates, holds, and final actions kept separate?
• Market fit: Does the analysis discuss expectations, financing, and competing readouts?
• Evidence fit: Can you trace important claims to primary documents?
• Research fit: Can you define a comparable historical cohort after reading?

Build an event record before measuring returns

The event timestamp is usually the hardest field. A company can receive an agency communication privately, announce it before the open, file an 8-K later, and appear in a vendor feed after both. Your strategy could only act after the first source it actually monitored became observable.

FDA event record

issuer:
ticker_at_event:
product_or_program:
event_type:
event_status:
decision_or_meeting_date:
first_public_timestamp:
source_url:
source_type:
vote_question:
yes_no_abstain:
prior_related_event_id:
next_related_event_id:
market_open_status:
confounding_disclosure:
retrieved_at:

Keep the agency date, disclosure timestamp, and first tradeable session as separate columns. A pre-market release and an after-hours release share a calendar date while creating different entry rules. If only the date is known, downgrade the timing precision and avoid pretending you could trade at the prior close.

Test reactions without rewriting history

Once the record is clean, define the event study before looking at returns. Choose an entry convention, benchmark, holding window, duplicate policy, and treatment for halts and delistings. Match by point-in-time market cap and, when possible, development stage or financing profile.

Confounders deserve their own fields. A biotech company may announce financing, trial data, guidance, or a strategic review near the same regulatory event. Keep those rows and tag them, then report a clean subset beside the full cohort. Silently dropping complicated events leaves you with a sample of companies that disclosed less, which is a different selection rule.

1. Freeze the source event table and its retrieval timestamp.
2. Deduplicate repeated disclosures while preserving true status changes.
3. Enter at the next executable price after the public timestamp.
4. Report mean, median, hit rate, dispersion, and the worst individual paths.
5. Split results by event type instead of pooling every negative or positive FDA development.
6. Run the rule on a later period that did not influence its design.

our strategy track-record review provides a useful warning: a historical FDA-setback sleeve can disappoint even when the event sounds directionally obvious. Markets may incorporate a public setback quickly, and a next-session entry can arrive after the repricing.

Use a source stack, not one magic calendar

A robust workflow uses the FDA calendar and Drugs@FDA for agency records, issuer filings and releases for sponsor-disclosed dates, a specialist publication for scientific and regulatory context, and structured data for historical comparison. Each layer has a distinct job and its own failure mode.

Check active catalysts again on the morning of the event because meeting details, documents, and company guidance can change. Save the version you used. Then keep any newsletter interpretation in a separate field from the factual event label, so a later analyst can rebuild the cohort without inheriting your view.

Start with one narrow question, such as the next-session reaction after a first disclosed Complete Response Letter. Pull the sourced records, inspect the worst cases, and document where the timestamp is only a date. That modest study will teach you more about the data than a giant catalyst spreadsheet with uncertain provenance.