One of my favorite Sagan moments

A chance to listen to J. Jack Lee comment on FDA's guidance on Bayesian statistics in trials www.linkedin.com/feed/update/... Here's his letter, alongside co-authors Harrell & Spiegelhalter et al in JAMA: jamanetwork.com/journals/jam...

Regulatory round-up: A month of key FDA approvals for cancer drugs www.pharmaceutical-technology.com/news/regulat...

Peter Thall: Practical Bayesian Guidelines for Small Randomized Oncology Trials (2025) www.mdpi.com/2072-6694/17...

You can see me at the upper left corner playing chess on my phone haha. It was a really great event. I love the LSHTM community!

Letter to POTUS supporting Dr. Richard Pazdur as the next FDA Commissioner www.nopatientleftbehind.org/fda-recommen...

Just learned Prof. @f2harrell.bsky.social is visiting Imperial soon. 🤩 So excited!!!

Response to the FDA Dosage Optimization Draft Guidance for Radiopharmaceutical Therapies jnm.snmjournals.org/content/earl...

CDC director called Test-Negative study design ‘crap.’ www.forbes.com/sites/johndr...

Academic clinical cancer trials to improve patient outcomes - The Lancet Oncology www.thelancet.com/journals/lan...

FDA ODAC's comments (rejection, 6-3 voting) on AstraZeneca's application for Camizestrant (in combination with a CDK4/6 inhibitor) for advanced HR+ breast cancer when they show an ESR1-mutation on 1L prior to PD (SERENA-6; phase 3 trial) www.fda.gov/media/192156... (PDF)

We are recruiting! Research Associate/Senior Research Associate in Medical Statistics. Exciting role working with Pavel Mozgunov and colleagues on novel dose optimisation designs for combination trials with application in oncology and infectious diseases. 👇 www.cam.ac.uk/jobs/researc...

Pharma-Stat meeting in Basket Trial Designs in early-phase oncology. The presenters will discuss a computationally-cheap approach (power prior) for borrowing. Date & Time: Saturday, May 16, 2026. 10:00am-12:00pm, EST. www.linkedin.com/posts/yuanye...

What a story (2007): a trial participant was suicidal due to a breakup with his gf, and took 29 pills of IMP (antidepressant). Ended up pale and hypotensive in ER seeking help and received medical attention, only to find out that it was the placebo www.sciencedirect.com/science/arti...

UK’s NIHR Clinical Trials Toolkit has been updated: www.ct-toolkit.ac.uk/routemap

ggplot2 is one of those tools that completely changes how you think about data visualization. A great introduction if you want cleaner and more expressive charts: ggplot2-book.org #RStats #DataVisualization #DataScience #ggplot2 #DataViz

Webinar: Biostatistical Considerations When Using RWD and RWE in Clinical Studies for Regulatory Purposes Thursday, 21 May 2026 16:00 – 17:00 GMT+1 events.phuse.global/hub/events/b...

Retraction note in NEJM: "To the Editor: We were unaware of Journal policies on image manipulation and had altered our submission by using an AI tool to move the ruler to the top of the image. We therefore wish to retract our image and case report." www.nejm.org/doi/full/10....

I felt a strange burning sensation over my entire body when I entered this building

Trump plans to fire FDA chief Marty Makary: report www.fiercepharma.com/pharma/trump...

Webinar by International Chinese Statistics Association on "Oncology Drug Development: A Practical Overview for Statisticians" by Qi Xia, Head of Oncology Statistics at AbbVie. 22 May, 8 PM London us06web.zoom.us/webinar/regi...

Malignant Adult Ovarian Germ Cell Tumors: An International Multicenter Study to Identify Relevant Prognostic Risk Factors for Stage IC and Beyond | Journal of Clinical Oncology ascopubs.org/doi/10.1200/...

A review from Stuart Pocock on covariate adjustment in randomized trials journals.sagepub.com/doi/10.1177/...

Mixed feelings about this, but good FDA is trying to speed things up I'd love to hear how it works for the piloting in AZ and Amgen's early-phase cancer trials www.fda.gov/news-events/...

PHAROS is a single arm trial that probably did not address OS intercurrent events handling strategies a priori ascopubs.org/doi/10.1200/...

‘Do not be surprised if statisticians obsess about sample size determination for as regards its importance “they know it and they know that very powerfully”’ Sayings of Confuseus

RIP

Just learned that Oscar Sheynin died a couple of years ago. I always enjoyed his pieces on the history of statistics. As an associate editor for Biometrika in the min 1990s I had no hesitation in recommending this academic.oup.com/biomet/artic... coauthored with Pfanzagl for publication.

Apes, monkeys and their relatives that live in despotic societies are less likely to clown around as adults go.nature.com/48vHVTL

The 8th birthday for @f2harrell.bsky.social's roadmap on choosing between Statistical Modeling vs Machine Learning for prediction studies. Very helpful: www.fharrell.com/post/stat-ml/

New review in Cancer Medicine: Modern Clinical Trials: Seamless Designs and Master Protocols onlinelibrary.wiley.com/doi/10.1002/...

A Love Letter to Clinical Trials: Introducing JAMA+ Trials jamanetwork.com/journals/jam...

"...industry funded trials written by professional writers are at a higher risk of spin and bias. Conversely, I would put higher trust on academia-sponsored trials written by the authors." 😁 ascopubs.org/doi/10.1200/...

Vacancy: Professor in Medical Statistics and Academic Lead for Clinical Trials in Oncology, University of Sheffield jobsite.sheffield.ac.uk/job/Professo...

Boston University Seminar: Improving Decision-Making in Oncology Basket Clinical Trials: Challenges of Small Sample Sizes and Heterogeneous Treatment Effects Presented by Haolin Li 12 May 2026 Zoom Meeting ID: 998 0913 5732 Passcode: 201194 www.linkedin.com/posts/leohao...

R 4.6.0 "Because it was There" is set for release on 24 April. @russhyde.bsky.social writes up the more interesting changes — %notin% in base R, neater character column summaries, and more. Full breakdown with code: www.jumpingrivers.com/blog/whats-n... #rstats #datascience #rprogramming

New presentation: www.fharrell.com/talk/qrct/ #Statistics #StatsSky #clinicaltrials

A good case study on how: (1) OS is the most important endpoint in cancer trials (2) Prespecifying strategy for handling intercurrent events + sensitivity/supplementary analyses are crucial (3) Above strategies (e.g. ?hypothetical estimand with RPSFT) are not without limitations among others

Bayesian analysis in confirmatory clinical trials: A narrative review and discussion of current practice journals.sagepub.com/doi/full/10....

FDA novel oncology drug approvals in 2025 Table includes requirement for randomised dose finding and post-marketing dose optimization www.parexel.com/insights/blo...

If you would like to get started on oncology trial endpoints and stats, Linh Dinh has collated this nice Wiki here stat-wiki.linhdinh-dl1989.workers.dev

Trials of TIGIT inhibitors as a case study of herding in oncology drug development bmjoncology.bmj.com/content/5/1/...

Just noted this peculiar approach from survival::quantile.survfit() in estimating median time: "Since S(t) is a step function, it is possible for the curve to have a horizontal segment at exactly 1-k, in which case the midpoint of the horizontal segment is returned." ?

I can't stop laughing at this post. It's perfect.

Causal by Design: new blog article on Statistical Thinking. fharrell.com/post/causal/ #Statistics #StatsSky #EpiSky

Not really, but thanks 😂

✅ #Artemis II update: the spacecraft entered the lunar sphere of influence this morning at 05:37 BST/06:37 CEST, 6 April, at about 51 000 km from the Moon (the region of space where the Moon’s gravitational pull is stronger than that of Earth’s). www.nasa.gov/image-detail... @exploration.esa.int

Two recent FDA rejections due to concerns with the unconvincing use of external control arms onbiostatistics.blogspot.com/2025/11/when...

Charité's @bihatcharite.bsky.social peer review summer course "Peerspectives" is open for applications until April 10: iph.charite.de/en/academic_...