Berry Consultants, LLC

Should a pharmaceutical company invest in a dedicated statistical methodology group? On this week's "In the Interim…", Scott Berry talks with Jenny Devenport and Kaspar Rufibach of Merck Group, co-authors of the paper "Statistical Methodology Groups in the Pharmaceutical Industry," and May Mo of Amgen, a co-author of a US commentary on it. Highlights from the conversation: ✅️ Why a new method adds little until it is published, built into software, and scaled ✅️ How new mechanisms of action are pushing trial design beyond the t-test and log-rank test ✅️ Keeping methodologists and project statisticians working as one team ✅️ How one better phase 3 decision can pay for a methods group ✅️ Giving leadership design options, risks, and trade-offs instead of a single answer The full episode is live now: https://lnkd.in/gdUD5rJT

"Digital twin" has become a catch-all term in clinical trials. On this week's "In the Interim…", Dr. Scott Berry makes the case for reserving it for system-level models of human biology and explains what the distinction means for trial design. Building on his presentation at a recent CSRC regulatory science hackathon on digital twins and synthetic controls in cardiovascular trials, Scott covers: ✅️ Population-level, patient-level, and system-level models, and what each can replace in a trial ✅️ ARPA-H's new CIRCLE program for digital twins in the ICU ✅️ Why virtual controls should produce a predictive distribution for each patient, not a single forecast ✅️ How prospectively enrolled controls show whether virtual controls are predicting well ✅️ How much smaller trials could become as these models improve The full episode is live now: https://lnkd.in/g5PGehWB

Every conventional clinical trial is a stadium built from scratch for a single contest, then demolished when the contest ends. Two years from synopsis to first patient. One game. Then it all comes down, and somewhere else the build starts again. Our latest blog works through what changes when the arena stays standing. ✅️ Platform and basket trials, and the modular master protocol that holds them together ✅️ Why an arm can fund 133 patients and read out against more than 200 ✅️ Four months to first patient instead of two years ✅️ 667 patients across five arms instead of 1,000 and 333 fewer people on placebo ✅️ Why the first arena is the hardest one to build, and why patients are so often the ones who build it The science has not been the bottleneck for some time. Laying the first brick has. Read the full blog for more: https://lnkd.in/g4tcu7V9

Imagine if the NFL built a brand new stadium for every game, then tore it down when the clock hit zero. Games would only be every three years and cost billions per contest. It sounds absurd and yet it’s a fair description of how we run clinical trials. This week on "In the Interim…", Dr. Scott Berry shares the talk he gave to the Speak Foundation, a limb-girdle muscular dystrophy patient organization, on building clinical trial arenas instead of one-off stadiums. ✅️ Why a conventional trial is a two-year build for a single contest ✅️ Platform and basket trials as standing arenas ✅️ What a master protocol makes modular ✅️ 667 patients in one platform versus 1,000 across five trials ✅️ Why the first contest is the hardest one to build Explaining innovative design to statisticians is one thing but explaining it to the patients who live inside these trials is the real test. Tune in today to discover more: https://lnkd.in/g2TKnjT9

This week on “In the Interim…”, Dr. Scott Berry talks with Dr. Alun Bedding, a statistician turned professional certified coach, and former global head of statistical methods at Roche, about what coaching is, and why it looks a lot like good trial design consulting. A few highlights from the conversation: ✅️ Why coaching starts by determining the goal, not accepting the first request ✅️ Why real progress requires discomfort, not reassurance ✅️ Why blind spots make self-coaching unreliable ✅️ Where AI coaching helps, and where it can't replace human judgment Tune into the full episode today which will not require discomfort: https://lnkd.in/gTYaZpeh

On the latest episode of “In the Interim…”, Dr. Scott Berry sits down with Dr. Roger Lewis, Dr. Anna McGlothlin, and Dr. Nick Berry to unpack the CHIPS trial — a Bayesian adaptive study testing how long platelets can be safely stored cold, recently published in JAMA. A few highlights from the conversation: ✅️ Why room-temperature platelets leave rural, low-volume, and military hospitals without reliable access ✅️ How a monotonic Bayesian model safely escalated the maximum cold-storage duration, interim by interim ✅️ A mid-trial curveball of an independent FDA decision that changed the rules while enrollment was underway ✅️ The result: non-inferiority out to 21 days, with a Bayesian probability greater than 99.9% Listen to their full conversation today: https://lnkd.in/gWF6kvKt