Home › Research

Research & data studies

Updated September 20, 2026·next FDA decision on the calendar

Original, sourced studies on how biotech catalysts behave: linkable data for analysts and journalists.

SLS (SELLAS) deep dive: the REGAL 80th-event analysis

What the 80th event mechanically triggers, the measured pace of event accrual (60 → 72 → 78 deaths), every dated catalyst since 2024 with its actual day-of and +5d stock reaction, and the documented bull/bear facts. Primary sources cited.

What actually happens when a clinical trial reads out

1,752 readouts with real stock returns. Median move 3.8%; 57% land within ±5%, but 7.6% crash 30%+. And the third proof that there is no systematic run-up.

Short interest into FDA decisions, and why the obvious conclusion is wrong

FINRA short interest, point-in-time, matched to 1,753 FDA decisions. The most-shorted names move 3× more, and it’s entirely a market-cap artifact. We publish the debunk, not the headline.

Conference run-up study (ASCO, ASH, ESMO, AACR)

How presenting biotechs actually traded into the meeting: median move D-30→D-1 by market-cap tier, with n and quartiles. 1,425 presentations, 2017-2026.

PDUFA run-up by year (2020-present)

Median price path into FDA decisions, T-120→T+5, by year: 1,852 events, plus current in-flight names.

PDUFA stock run-up by market cap

Decision-day move & run-up by size tier (cohort study).

Questions

What datasets does pdufa.bio publish?

Four studies: PDUFA stock run-up by market cap (763 decisions), readout reactions (1,752 readouts), conference run-ups (1,425 presentations), and short interest into FDA decisions.

Can the data be reused?

Yes. The study data tables are published under CC BY 4.0; reuse with attribution to pdufa.bio.

Do the studies predict outcomes?

No. They measure what happened, with n and interquartile ranges stated; we do not publish predictions or approval probabilities.