bio-clinical-biostatistics-bayesian-trials
Designs Bayesian clinical trials including Phase I dose-finding (BOIN, CRM, EWOC, mTPI-2), meta-analytic-predictive (MAP) priors with robust mixtures for external data borrowing, EXNEX for basket trials, hierarchical models for safety AE (Berry-Berry), Bayesian platform trials (I-SPY 2, GBM AGILE, REMAP-CAP), and posterior probability stopping rules. Covers FDA Bayesian Devices Guidance (2010), FDA Bayesian Methodology in Drugs Draft (January 2026), BOIN Fit-for-Purpose qualification (December 2021), and Project Optimus dose-optimisation. Use when designing dose-finding studies, platform trial
npx skills add BioTender-max/awesome-bio-agent-skills --skill bayesian-trials --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Designs Bayesian clinical trials including Phase I dose-finding (BOIN, CRM, EWOC, mTPI-2), meta-analytic-predictive (MAP) priors with robust mixtures for external data borrowing, EXNEX for basket trials, hierarchical models for safety AE (Berry-Berry), Bayesian platform trials, and posterior probability stopping rules. Covers FDA Bayesian Devices Guidance (2010), FDA Bayesian Methodology in Drugs Draft (January 2026), BOIN Fit-for-Purpose qualification (December 2021), and Project Optimus dose-optimisation. Use when designing dose-finding studies, platform trials, or sensitivity analyses with informative priors.
How it works
- Recommends Bayesian design approaches by scenario: Phase I MTD with BOIN; multi-arm or dose-optimisation via BOIN-12/gBOIN-ET; borrowing via MAP priors using RBesT (gMAP, robustify); EXNEX for basket trials; and safety multiplicity via Berry-Berry hierarchical models.
- Provides concrete code sketches and parameters for BOIN, CRM, EWOC, and MAP/RBesT workflows, including how to calibrate priors, simulate operating characteristics, and interpret posterior or predictive probabilities for decision rules.
- Outlines workflows for Bayesian platform trials (I-SPY 2, REMAP-CAP) with adaptive graduation/dropping rules based on posterior predictive or posterior probability thresholds.
- Describes regulatory context, recommended software, and compatibility notes for alternative packages (BOIN, dfcrm, OncoBayes2, RBesT, psborrow2, brms, Stan).
When to use it
- Use when designing dose-finding studies, platform trials, or sensitivity analyses with informative priors.
What it can touch
- Software: R; primary tool: RBesT. See examples referencing R packages: BOIN, dfcrm, OncoBayes2, bhmbasket, c212, bayesDP, psborrow2, brms, Stan. It includes code snippets and package names exactly as shown in the skill.
Caveats
- Includes notes on regulatory guidance, calibration via simulation, and prior-data conflict considerations (robust MAP, EXNEX weight choices, gamma for power priors).
- Mentions required verification of package versions and compatibility before using code patterns; emphasizes simulation-based operating characteristics for regulatory pivots.
## Version Compatibility Reference examples tested with: R `RBesT` 1.7+ (Roche), `OncoBayes2` 0.8+ (Novartis), `BOIN` 2.7+, `dfcrm` 0.2-2+, `escalation` 0.1+, `trialr` 0.1.6+, `bayesDP`, `psborrow2` (FDA-supported), `rstan` / `cmdstanr`, `brms`. Legacy: `JAGS`, `WinBUGS`. Before using code patterns, verify installed versions match. If versions differ: - R: `packageVersion('<pkg>')` then `?function_name` - Confirmatory regulatory work: validate against pinned package versions in submission If code throws an error, introspect the installed package and adapt the example to match the actual API rather than retrying. # Bayesian Clinical Trials **"Design a Bayesian clinical trial"** -> Specify a prior, likelihood, and decision rule with frequentist operating characteristics demonstrated via simulation; for dose-finding use FDA-endorsed BOIN; for borrowing use robust MAP priors; for adaptive platforms use posterior probability of efficacy stopping with simulation-calibrated thresholds. ## Regulatory Status -- The 2024-2026 Bayesian Pivot **FDA 2010 CDRH Bayesian Devices Guidance** (Feb 5 2010): the only Bayesian-specific FDA guidance until January 2026. Why devices were ahead: CDRH's PMA
- Version Compatibility
- Regulatory Status -- The 2024-2026 Bayesian Pivot
- Algorithmic Taxonomy
- Decision Tree by Scenario
- Phase I Dose-Finding -- BOIN, CRM, mTPI-2
- BOIN (FDA-preferred operational)
- CRM with calibrated skeleton
- EWOC (overdose control)
- MAP Priors and RBesT
- EXNEX for Basket Trials
- Bayesian Platform Trials
- I-SPY 2 (Park-Liu 2016 NEJM 375:11)
- REMAP-CAP (Angus 2020 JAMA)
- Drop-the-loser vs promising-the-winner
What does the bio-clinical-biostatistics-bayesian-trials skill do?
Designs Bayesian clinical trials including Phase I dose-finding (BOIN, CRM, EWOC, mTPI-2), meta-analytic-predictive (MAP) priors with robust mixtures for external data borrowing, EXNEX for basket trials, hierarchical models for safety AE (Berry-Berry), Bayesian platform trials (I-SPY 2, GBM AGILE, REMAP-CAP), and posterior probability stopping rules. Covers FDA Bayesian Devices Guidance (2010), FDA Bayesian Methodology in Drugs Draft (January 2026), BOIN Fit-for-Purpose qualification (December 2021), and Project Optimus dose-optimisation. Use when designing dose-finding studies, platform trial
How do I install it?
Run `npx skills add BioTender-max/awesome-bio-agent-skills --skill bayesian-trials --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From BioTender-max/awesome-bio-agent-skills, a repository with 135 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.