Unlearn.AI

Advancing AI to eliminate trial and error in medicine.
Read our blog:

[Unlearn.AI @UnlearnAI]

How do you generate interpretable evidence from a small, open-label single-arm #ALS trial when there's no placebo arm to compare against?

In a new post, our CEO Steve Herne walks through how @ProJenX is approaching this challenge in PRO-101, their Phase 1 study of prosetin: pairing each participant with their own #digitaltwin.

Steve unpacks what that enables, from surfacing candidate subgroups to identifying which endpoints carry the strongest signal for Phase 2.

Read it here: https://t.co/l7b3Hmh4H4

[Unlearn.AI @UnlearnAI]

A $535K protocol amendment. 80% of trials chronically delayed. 9 in 10 drugs failing.

These are symptoms of one root cause: Decision Debt, the unresolved tradeoffs and lost rationale that compound every time a clinical trial decision leaves its context behind.

Our team's latest whitepaper, The Scientific Intelligence Layer for Connected Clinical Trial Decisions, lays out where Decision Debt accrues across the trial lifecycle, and what a connected alternative looks like in practice. Grounded in regulator-validated methods (EMA qualified, FDA supported) and published results from work with AbbVie, ProJenX, and the ADCS DHA Alzheimer's reanalysis.

Read it: https://t.co/7F8setW6lE

[Unlearn.AI @UnlearnAI]

If you’re at #ASCO26 this week, we’d love to tell you how we help oncology teams:

➡️ Pressure-test trial design before protocol finalization, in hours instead of months.
➡️ Run a single-arm study in a narrow biomarker-defined population when data matching turns up a handful of patients and acquiring more RWD isn’t viable.
➡️ Compare two established treatments when there's no head-to-head trial data.

We just wrote about that last one — FOLFIRINOX vs. gem/nab-paclitaxel in first-line pancreatic cancer: https://t.co/eApgc0fsPy

If you're there, be sure to connect with our team 🚀
———
@ASCO

[Unlearn.AI @UnlearnAI]

#UnlearnerSpotlight 🚀 “I love working with such a supportive and motivated team, tackling challenging problems together, and seeing the real-world impact our work can have on patients and clinical research.”

[Unlearn.AI @UnlearnAI]

🚀 New on the blog: we wrote about how one of our partners, VectorY, used #digitaltwins to build interpretability into a 12-patient gene therapy trial for #ALS, no concurrent placebo arm required: https://t.co/2VQX4J9OTh

[Unlearn.AI @UnlearnAI]

The Unlearn team was proud to be at SCOPE X in Boston this week, connecting with leaders across #clinicaldevelopment and sharing the work we've been doing to advance #AI in #clinicaltrials. Among the highlights was our VP of Product Kwame Marfo's presentation, "The Trust Dividend: How Regulatory-Grade AI Compounds Value Across the Trial Lifecycle" — a talk that spoke to the heart of what we build at Unlearn and why regulatory-grade AI matters for the future of #drugdevelopment. Thank you to everyone who attended his session and took the time to connect with our team. We look forward to continuing the conversation. 🚀
———
@SCOPE_365

[Unlearn.AI @UnlearnAI]

We're excited to be attending the 5th ALS Drug Development Summit in Boston on June 2. We'll be presenting a poster: Digital Twins for Synthetic Control Arms in ALS Clinical Trials — exploring how AI-generated #digitaltwins can serve as patient-level external comparators in single-arm #ALS studies, helping sponsors generate stronger, more interpretable evidence from early-phase #trials.

This builds on our recent partnerships with VectorY Therapeutics and SOLA Biosciences, where we're applying digital twins as participant-level external comparators to support evidence generation in their early-phase ALS programs.

If you're attending, we'd love to connect 👋
———
@HansonWade

[Unlearn.AI @UnlearnAI]

💡 Many decisions in #clinicaldevelopment depend on combining patient-level data with aggregate #trial summaries. The reasons vary: the published summary is the only public source, standard of care has shifted since the original trial ran, or no head-to-head comparison was ever conducted.

A new preprint from our research team (Franklin Fuller, Daniele Bertolini, Samantha Liang, Jason Christopher, Aaron Smith) introduces FRESH, a method for calibrating generative models to published trial evidence. The paper demonstrates the approach with an in-silico head-to-head in metastatic pancreatic cancer using our solid tumor foundation model.

Read it here: https://t.co/eHsRtzSYij

[Unlearn.AI @UnlearnAI]

Single-arm trials are everywhere now: early-phase programs in rare disease, Phase 1/2 gene therapy studies, and indications where a placebo arm is unethical or impossible to enroll. And the same question keeps surfacing. What do you compare the treated patients to?

The standard answer has been propensity score matching against historical patients who look enough like the trial population. It works, sometimes. But it depends on historical data that's broad enough to overlap with the trial population and narrow enough not to introduce confounders, which is a hard line to walk.

In a new paper out today, our team (Daniele Bertolini, Franklin Fuller, Aaron Smith, Jon Walsh, and Run Zhuang) argues that model-based synthetic control arms built on #digitaltwins are a more robust path. The paper presents sample-size formulas for the doubly-robust AIPW estimator, applies the FDA's recent draft guidance on #AI in #drugdevelopment, and reanalyzes #ALS and #HuntingtonsDisease trials to show the methods at work.

Use data when you can, models when you must. For most single-arm trials, that means models should be in the room.

📄 Read it here: https://t.co/eg06XSpzhO

[Unlearn.AI @UnlearnAI]

Most #AI investment in #biopharma has gone into #drugdiscovery. Once a promising treatment is discovered, there can still be many years, if not a decade or more, of in-human #clinicaltrials, where the treatment faces its ultimate test: proving its efficacy and safety profile in diverse populations. Yet AI hasn’t received as much attention in #clinicaldevelopment, even though it’s the more expensive part of the work and where many high-stakes programs fail. That's where useful AI work waits to be done, and it's what Unlearn founder and Head of AI Aaron Smith covers in our latest whitepaper: https://t.co/Z53Zxfm5fc

[Unlearn.AI @UnlearnAI]

See you at SCOPE X in a week! Don't miss our VP of Product Kwame Marfo's presentation, "The Trust Dividend: How Regulatory-Grade AI Compounds Value Across the Trial Lifecycle" on May 19 at 10am. Every #clinicaldevelopment team will adopt #AI — the harder question is which AI is trustworthy enough to drive key decisions. Kwame will be drawing on Unlearn's journey to regulatory qualification to show how trust earned once compounds across planning, monitoring, and analysis. 🚀
———
#SCOPEsummit

[Unlearn.AI @UnlearnAI]

FOLFIRINOX vs. gem + nab-paclitaxel is one of the most common first-line decisions in advanced pancreatic cancer — but no RCT has ever directly compared them. Indirect methods like NMA and target trial emulation each have limitations: transitivity assumptions break down when trial populations differ, and RWD-based approaches are resource-intensive and frequently study a different population than the trials in question.

In Part 4️⃣ of our #oncology trial design series, we show how trial-calibrated generative modeling offers a third path—producing patient-level simulations that respect published RCT results while adjusting for baseline covariate imbalances. Applied to PRODIGE4 vs. MPACT, the approach softens the apparent 1-year RMST advantage of FOLFIRINOX by about half.

Read the full post: https://t.co/iqIQzdMnjz

[Unlearn.AI @UnlearnAI]

We're heading to #SCOPEsummit in Boston later this month, and the conversations happening there are ones we think about every day. How do we design smarter #trials, increase confidence, and build #AI that's trustworthy enough to drive key decisions across #clinicaldevelopment? Our VP of Product Kwame Marfo will be addressing exactly that on May 19 at 10am: The Trust Dividend: How Regulatory-Grade AI Compounds Value Across the Trial Lifecycle. See you there! 🚀

[Unlearn.AI @UnlearnAI]

May is #ALSAwarenessMonth. #ALS is relentlessly progressive, with no cure and painfully limited treatment options — making confident, efficient #clinicaldevelopment not just valuable, but essential.

We're proud to partner with ProJenX, QurAlis, Trace Neuroscience, SOLA Biosciences, VectorY Therapeutics, and others at the forefront of ALS #drugdevelopment helping them reach go/no-go decisions with greater confidence — so promising therapies can move forward faster for people who are waiting.
———
@alsassociation

[Unlearn.AI @UnlearnAI]

#UnlearnerSpotlight 🚀 "I'm surrounded by sharp, thoughtful people who care about the work and are very supportive of each other, all working toward smarter trials and faster treatments for patients.”

[Unlearn.AI @UnlearnAI]

We predicted the BREAKWATER control arm without ever training on BREAKWATER data. That's the kind of question most #oncology teams want to answer during the design phase but can't, because the work to run each scenario could take months.

Part 3️⃣ of our oncology series is about what changes when that's no longer the constraint. When you can ask "what does the control arm look like under current standard of care?" and get a calibrated answer before the protocol is finalized.

❓ What would your team do differently with that kind of flexibility during the design phase?

https://t.co/HGPU4n25sL

[Unlearn.AI @UnlearnAI]

🎉 We're proud to announce that Unlearn has been named the winner of the 2026 Fierce Biotech Outsourcing Award for AI & Advanced Analytics. This recognition is a testament to our team's dedication to transforming how #clinicaltrials are designed, analyzed, and advanced — empowering sponsors to make faster, more confident decisions that get treatments to patients sooner. We also want to recognize the finalists in our category — the depth of innovation across this field speaks to the incredible momentum #AI is bringing to #drugdevelopment. 🚀 Read more: https://t.co/Eupydx4WGE

[Unlearn.AI @UnlearnAI]

Unlearn team at #DISS2026! 🎉 Last week our team had a great time at the Duke Industry Statistics Symposium, exploring the latest in AI/ML and data innovation in pharmaceutical development — from adaptive trial designs to real-world evidence and beyond. Grateful for the insightful conversations and connections!
———
@DukeBiostats

[Unlearn.AI @UnlearnAI]

Behind every #clinicaltrial is a person waiting for a breakthrough. That's why we use AI-generated #digitaltwins to make PD trials more efficient, reducing sample sizes by up to 23% and control arms by up to 38%, helping promising therapies reach patients faster. This #ParkinsonsAwarenessMonth, see what that looks like it our latest PD case study ➡️ https://t.co/sgkBeReJgt

[Unlearn.AI @UnlearnAI]

5️⃣ patients.

That's how many in our available datasets fully matched the eligibility criteria for a recent BRAF-mutant colorectal #cancer trial. A data-matching approach stops there.

Part 2 of our #oncology series explains what a modeling approach makes possible when the #data runs out, including how we predicted that trial's control arm overall survival without ever training on its data.

https://t.co/AEC0jxdImB