👋 Hey, let’s get into it. The U.S. government would like scientists to know that artificial intelligence may soon be less of a suggestion and more of a requirement.

HHS is joining the White House’s Genesis Mission, a more than $5 billion effort connecting federal datasets, supercomputers, autonomous laboratories, and AI models to tackle chronic disease, pediatric cancer, and drug discovery.

For researchers preparing their next grant application, the message seems fairly clear: having an AI strategy probably won’t hurt. We go deeper on the Genesis Mission in today’s Deep Dive, halfway down the newsletter.

🧪 FDA Advisers Back Six Influencer-Famous Peptides for Compounding
Biohackers may want to chill the celebratory peptide injections…for now. An FDA advisory committee narrowly recommended adding six unapproved peptides to the 503A Bulk Drug Substances List, potentially allowing specialized pharmacies to compound them for individual patients. The endorsed ingredients included BPC-157, TB-500, KPV, MOTs-C, Epitalon, and Semax, which are promoted for everything from wound healing and weight loss to insomnia and cognitive performance. FDA scientists repeatedly cited weak evidence and safety concerns, while patients and compounders pushed for access. The agency makes the final call and is not required to follow the panel.

💉 Lilly’s Retatrutide Delivers Up to 22.6% Weight Loss but Misses Clear Cardiovascular Win
Eli Lilly’s triple-agonist retatrutide hit the primary endpoints in two Phase 3 obesity trials, producing 20.8% weight loss in patients with type 2 diabetes and 22.6% in patients with severe obesity and cardiovascular disease after 80 weeks. Those results trailed the roughly 28% reductions seen in earlier studies involving different populations, while discontinuations reached 13.5% at the highest dose in TRIUMPH-3. The cardiovascular analysis also proved less decisive, with 44 major events across retatrutide groups versus 52 on placebo. Lilly still plans an FDA submission in early 2027.

💰 Argenx Buys Forte Biosciences for $2.2B After Vitiligo Data
Argenx is turning an existing investment into a full acquisition, paying $2.2 billion for Forte Biosciences and its Phase 2-stage anti-CD122 antibody FB102. Earlier this month, Forte linked the therapy to a 29.6% improvement in facial vitiligo during a Phase 1b study, while a Phase 2 celiac disease readout is expected later this year. Argenx believes FB102 could also address alopecia areata and other autoimmune diseases, giving it the increasingly fashionable “pipeline-in-a-product” profile. At $77 per share, the deal represents a 40.5% premium to Forte’s previous closing price.

🎯 J&J Bispecific Combo Cuts Myeloma Progression Risk by 89%
Johnson & Johnson’s Tecvayli and Talvey combination delivered the kind of Phase 3 result that makes oncology investors reread the press release. In the MonumenTAL-6 trial, the dual bispecific regimen reduced the risk of disease progression or death by 89% and lowered the risk of death by 62% compared with standard combinations in earlier-line relapsed or refractory multiple myeloma. Roughly 80% of patients were refractory to Darzalex, making the result particularly notable. The efficacy is enormous, although combining two T-cell engagers may reserve the regimen for high-risk patients treated at larger academic centers.

🧬 Yescarta Study Finds Genetic Clues Behind CAR-T Toxicity
Why do some CAR-T patients experience severe side effects while others do not? Part of the answer may be written into their DNA. Researchers analyzing Kite Pharma’s Zuma-1 and Zuma-7 trials identified inherited genetic variants associated with Yescarta toxicity and cell expansion. Mutations that disabled STXBP2 were linked to toxicity in the more severely ill Zuma-1 population, while ADAMTSL3 variants appeared protective. Versions of PTPN22 were associated with CAR-T proliferation across both trials, raising the possibility that future therapies could deliberately modify the gene to improve expansion, lower dosing requirements, and potentially reduce exhaustion.

BioNucleus Briefing lands in the inboxes of 1,300+ scientists, founders, investors, operators, and industry decision-makers twice a week.

If you'd like to sponsor a future issue, just hit reply and let's talk.

Alright, back to the newsletter.

👔 Legend Biotech CEO Abruptly Resigns
Ying Huang resigned without a permanent successor, leaving Alan Bash as interim CEO while Legend conducts a comprehensive executive search.

🫁 GSK Wins Its First Lung Cancer Approval
The FDA approved Jideytro for previously treated ROS1-positive lung cancer, giving GSK a rapid payoff from its $10.6 billion Nuvalent acquisition.

😬 FDA Rejects Capricor’s Phase 3 Victory Lap
Regulators said deramiocel missed its prespecified Duchenne endpoints despite Capricor claiming success, sending the biotech’s shares down more than 65%.

💪 Outlook Wins Approval After Three FDA Rejections
Lytenava becomes the first FDA-approved bevacizumab formulation for wet AMD, giving Outlook access to a U.S. market it estimates at $8.5 billion.

📉 MapLight Falls 60% on Mixed Schizophrenia Data
ML-007C-MA succeeded at one twice-daily dose but missed at the once-daily dose, complicating the company’s effort to challenge established antipsychotics.

🛡️ Roche Defends Vabysmo’s $7B Ambition
Quarterly sales missed expectations by 7.1%, but Roche still believes its eye drug can reach roughly 6 billion Swiss francs in peak revenue.

💊 FDA Approves Otsuka’s Once-Daily ADHD Drug
Simtriyo won approval for patients aged six and older, positioning Otsuka’s first-in-class NDSRI for peak sales exceeding $610 million.

🔄 Gossamer Reclaims Seralutinib From Chiesi
The companies ended their $486 million partnership after a Phase 3 miss, leaving Gossamer to pursue FDA approval with worldwide rights restored.

The White House is attempting something considerably larger than handing federal scientists a chatbot subscription.

The Genesis Mission is expanding into a whole-of-government effort backed by more than $5 billion in commitments and contributions from over 15 agencies. Its goal is to connect federal datasets, Department of Energy supercomputers, scientific facilities, and AI tools through a shared national research infrastructure.

For biotech, the marquee projects are ambitious. HHS plans to combine longitudinal health cohorts with environmental data to investigate chronic disease. Pediatric cancer researchers will gain access to integrated clinical datasets and massive computing power. Drug discovery programs will bring molecular, genomic, phenotypic, clinical, and real-world data together to identify new uses for existing medicines and accelerate translation.

Then things get more futuristic. The initiative includes autonomous laboratories capable of running AI-guided experiments, models designed to make living systems more predictable, and shared biological infrastructure meant to turn discoveries into commercial-scale products.

NIH says more than $1.2 billion in fiscal 2026 obligations and planned 2027 funding already aligns with these areas, with additional Bio Genesis opportunities expected.

The opportunity is obvious: federal science holds enormous datasets that private companies would love to access. The challenge is making those datasets clean, interoperable, secure, and useful. These four words have defeated many ambitious platforms before.

Still, Genesis signals where U.S. research policy is heading. AI is no longer being treated as a useful side project. Washington wants it embedded into the machinery of scientific discovery.

For grant applicants, consider this your gentle reminder to update the proposal template.

That’s how much BeOne Medicines will invest to add small-molecule manufacturing at its New Jersey campus. The expansion brings BeOne’s total U.S. production commitment above $1 billion, adds a 145,000-square-foot building, and is expected to create 120 manufacturing jobs as pharma’s onshoring spree continues.

Here’s what biotech Redditors are talking about:

🤔 Would Biotech Workers Choose Biotech Again?
A final-year zoology student asked whether industry veterans would still pursue biotech if they could start over. Many responses landed somewhere between “the science was worth it” and “please study accounting.” Commenters appreciated the friendships, technical skills, and meaningful work, but repeatedly cited low pay, unstable careers, and the long Ph.D. path. Several recommended pairing a STEM background with an MBA instead. Then again, Reddit rarely attracts people logging in to announce that everything is going wonderfully.

🤚 Raise Your Hand if You’re Burned Out
A principal scientist with a decade in biotech asked whether anyone else felt exhausted by reorganizations, politics, shaky morale, and compensation that rarely reflects the required education. The responses were sympathetic but not exactly inspiring. Former pivoters warned that tech, medicine, law, sales, and education each come with their own specialized form of misery. The practical advice: move into a different function, change environments periodically, and build a life outside work. Apparently the grass isn’t greener. It just comes with a different flavor of stress.

📺 Novo Tries to Stop Lilly’s GLP-1 Ads
Novo plans to seek an injunction blocking Lilly advertisements it says falsely portray its obesity and diabetes medicines as more effective.

🥬 Food Consolidation Complicates Outbreaks
Experts warn that fewer companies handling more produce can spread Cyclospora contamination nationwide, although larger suppliers may offer stronger traceability systems.

💼 Sarepta Hires Former AbbVie Executive as CEO
Michael Severino will lead Sarepta through Elevidys safety concerns and commercial challenges after previously running Tessera Therapeutics and serving as AbbVie president.

😴 Apnimed Targets $160M IPO for Sleep Apnea Pill
The biotech hopes public investors will fund Oxnimbi’s potential launch as the first prescription medicine designed specifically for obstructive sleep apnea. I guess sleep really is valuable.

💻 Dimension Raises $800M to Keep Digitizing Drug Discovery
Venture firm Dimension has closed an $800 million third fund, bringing total assets under management to $1.65 billion. Founded in 2023, the firm backs startups where biotechnology and computing increasingly blur together, including Chai, NewLimit, and Earendil Labs. Its bets range from single-digit millions to more than $50 million, and one portfolio company, Coefficient Labs, was acquired by Anthropic for $400 million. Dimension’s thesis is straightforward: tomorrow’s biotech teams may employ as many computer scientists as biologists.

💉 July 28, 1925: The Scientist Behind the Hepatitis B Vaccine Is Born
Dr. Baruch Blumberg discovered the hepatitis B virus, developed the first diagnostic blood test for it, and co-invented the first vaccine targeting a major cause of cancer. His work transformed blood transfusion safety and prevented millions of cases of chronic liver disease. The World Health Organization later designated his birthday as World Hepatitis Day. Happy World Hepatitis Day to those who celebrate. Just remember that opening champagne at a liver-health event may require a brief explanation.

Thinking back to the FDA peptide meeting story, wellness influencers everywhere are probably preparing the world’s first compounding pharmacy afterparty.

BPC-157 for the dance-floor injuries. MOTs-C for anyone worried about the snack table. Semax for remembering what happened the next morning.

The FDA still has to make the final call, so maybe don't book the venue for Peptide Palooza just yet.

That’s all for today. See you Thursday for the next issue. 👋

✍️ Today’s email was brought to you by Josh Martin.

If you’re enjoying BioNucleus, share it with a friend or coworker who’d get value from it. We’re a small operation, and every share makes a bigger difference than you think.

Was this email forwarded to you? Sign up here 👇

Subscribe for free

Follow us on social and stay one step ahead