#79- Platforms, patients & pipelines at JPM

The coffee break biotech roundup, by SomX.

Greetings from deal season,

Another January, another JPM, and this year platforms and patients took centre stage. Illumina turned up with a disease biology atlas built for scale; AstraZeneca brought oncology AI fully in-house; Genentech explored new ways to keep drug costs from running away; J&J sketched an oncology strategy designed to hold patients across treatment lines; and Basecamp closed things out with gene insertion models that step beyond CRISPR.

Thanks for reading,

Dodo

If you’d like to sponsor a future edition of Biotech Dodo, send us a message.

Discover 🔍

🧭 Illumina launches ‘unprecedented’ disease biology atlas (pharmaphorum): The gene-sequencing specialist has set out ambitious plans for its ‘Billion Cell Atlas’, a genome-wide CRISPR perturbation dataset spanning 150M cells across 250 disease-relevant human cell types. Already the largest of its kind globally, the resource is set to scale to 5 billion cells within 3 years and is being rolled out via Illumina’s BioInsight unit, which focuses on data services and discovery partnerships. AstraZeneca, MSD, and Eli Lilly are already using the Atlas for target discovery and AI model training.

Our take: The Atlas trims early discovery risk by letting pharma interrogate perturbation biology without building the labs, cell libraries, and teams themselves. As more programmes draw from the same cellular groundwork, the dataset compounds in value, pushing Illumina beyond instruments and into the role of a standing discovery partner. That shift brings stickier relationships and revenue tied to ongoing scientific throughput, not one-off kit sales.

💰AstraZeneca acquires Modella AI to advance global AI-driven oncology R&D (OncoDaily): AstraZeneca has brought Modella AI fully in-house, formalising a partnership built around multimodal foundation models for biomarker discovery. Modella’s platform – which combines pathology imaging with multi-omic and clinical data – will support biomarker discovery, patient selection, and therapy response prediction across clinical development. AstraZeneca says the technology is aimed squarely at improving decision-making in trials, rather than early target discovery, and will now be deployed across its global oncology pipeline.

Our take: This is a wager on patient stratification as the pressure point in oncology development. Trial costs swell when response varies wildly across patients, and late surprises are brutally expensive. Models that surface likely responders earlier can keep programmes tidy before budgets and timelines run away. By pulling Modella fully in-house, AstraZeneca gets to train these systems on its own trial data, sharpening signal quality while keeping hard-won clinical know-how out of shared vendor pools.

💊 Genentech CEO says switch from one of the largest PBMs will save the company millions (Endpoints): The Roche subsidiary has moved its pharmacy benefits to Rightway, a privately held PBM startup, in a deal covering roughly 25,000 employees and dependents. CEO Ashley Magargee said the switch is projected to save about $70M through 2028 and halt the company’s 13% annual increases in pharmacy costs. Instead of a rebate-driven formulary, Genentech worked with Rightway to build an evidence-based one, and has explicitly confirmed its own drugs receive no preferential treatment.

Our take: Genentech is turning PBM reform into a live experiment rather than a lobbying slogan. Running an alternative model at employee scale generates hard cost data for legislative scrutiny. If Rightway delivers the projected savings, it strengthens the case that rebate-driven formularies inflate costs. If it underperforms, pharma weakens its own critique. Either way, the outcome feeds directly into the regulatory fight, making this an unusually high-stakes move for what should be a routine benefits switch.

🎗️J&J reveals oncology strategy (Pharmaceutical Technology): J&J expects oncology to clock $50B in sales by 2030, with multiple myeloma contributing roughly half of that haul. The combination drug Darzalex/Faspro (daratumumab) is the main revenue driver, propped up by bispecifics Tecvaley and Talvey, and cell therapy Carvykti across later lines. Beyond myeloma, J&J is advancing a trispecific TCE (ramantamig), partnering with Kelonia on in vivo CAR-T, and is advancing Rybrevant/Lacluze in EGFR+ lung cancer alongside Inlexzo in bladder cancer.

Our take: J&J is shoring up its portfolio to own the entire treatment journey, not individual regimens, keeping patients inside the franchise for years. Darzalex anchors early lines, bispecific TCEs cover relapse, and Carvykti captures late-stage disease. Bets on trispecific and in vivo CAR-T extend that grip further still, closing the gaps before competitors can elbow their way in.

And finally…

🏕️ Basecamp Research unveils programmable gene insertion AI models, as it confirms Nvidia investment (Tech.eu): The UK startup says its latest AI models can design large, programmable gene insertions using genomics data gathered directly from global biodiversity. Trained on Nvidia GPUs, the models learn the language of DNA to design pathogen-killing peptides with 97–100% lab accuracy, targeting edits beyond CRISPR’s practical limits. Basecamp also announced that Nvidia’s NVentures has invested in its pre-Series C round on undisclosed terms. The company has raised $85M to date.

Our take: Basecamp’s models tackle large gene insertions, which CRISPR typically struggles with due to delivery and payload constraints. Current gene editing tools excel at small tweaks, but thousands of monogenic diseases require whole functional genes that simply will not fit inside standard AAV vectors. The feather in its cap? Genomics data physically collected from global biodiversity, rather than lifted from databases, captures evolutionary patterns competitors can’t replicate without field expeditions. The dataset becomes the advantage, not the architecture.

Tune in 🎧

🎸 D.A. Wallach Explains Why Biotech VC Is So Different: From indie rock to drug development, this episode explores why biotech investing plays by different rules, what it takes to back real medicines, and how science, risk, and time reshape venture capital.

📉 Why the AI Bubble Hasn’t Popped: Josh Brown, CEO of Ritholtz, joins Ed Elson to unpack the 2025 market. He explains why earnings growth might hold, and suggests that young investors should welcome the next correction.

🤖 Tech in 2026: Inside the AI bubble: The FT’s tech team weighs in on what’s coming for AI, from VC funding jitters to LLM limitations to workplace shakeups. With Murad Ahmed, Melissa Heikkilä, George Hammond, and Sarah O’Connor.

Apply ✍️

🧪 Research Scientist in Antibody Discovery, RxBiologics: Phage display pro? Join the Babraham-based team to isolate antibodies, master characterisation assays, and drive drug discovery programmes in a high-growth, hands-on lab environment.

🧬 CAR-T Production Supervisor, Legend Biotech: Master of the cleanroom? Lead operational teams in Ghent to manufacture life-saving cell therapies, ensuring strict cGMP compliance and driving operational excellence in a state-of-the-art ATMP facility.

💻 Computational Protein Design, Latent Labs: Protein design visionary? Join this London-based moonshot to bridge generative AI and synthetic biology, leveraging proprietary frontier models to engineer functional proteins and drive high-impact therapeutic discoveries.

RSVP 📆

🧪 19.01 | BioSeed | London, UK: A one‑day event where cutting‑edge early‑stage life science companies pitch for seed investment and connect directly with a broad network of active sector investors.

🧬 27-29.01 | mRNA-Based Therapeutics Summit Europe | Berlin, Germany: Explore the latest in mRNA, saRNA, and circular RNA innovation – from design to delivery.

🍽️ 29.01 | BIA Gala Dinner | London, UK: This is your chance to dine in an iconic venue with 760+ of life science’s finest, whilst having the opportunity to fundraise for our new charity partner of the year 2026.

Got news, jobs or events you think are worth coo-ing over? Post an event here, or email us at biotechdodo@substack.com!

We’re SomX – a communications and creative agency trusted by biotech, pharma and healthtech pioneers. We craft strategy, content, PR and design that translate complex biology into compelling stories and investor‑ready excitement. Get in touch to amplify your science beyond the lab bench.