Merck & Co. will partner with Protillion Biosciences to discover multiple new therapy candidates through a collaboration that could generate up to $510 million in milestone payments for the artificial intelligence-based drug design company whose “lab-in-the-loop” approach combines AI with a continuous feedback loop of experimental wet-lab data.

The collaboration, launched through a multi-target discovery collaboration and license agreement, is designed to combine Merck’s global expertise in discovering novel therapeutics with Protillion’s Prot-MaP on-chip antibody discovery platform.

Prot-MaP, short for Protein Display on a Massively Parallel Array, is designed to facilitate AI-based optimization of therapeutic antibodies through the quantitative analysis of protein libraries with unprecedented speed, scale, and precision, characterizing millions of variants per run and avoiding the common pitfalls of model overfitting.

The result, according to Protillion, is the identification of optimized biologics with sophisticated therapeutic profiles such as pH-dependent sweeping and multi-target specificity—profiles that are difficult to achieve with traditional methods.

Protillion says Prot-MaP is intended to enable the engineering of novel biologics by generating megascale, just-in-time quantitative antibody binding datasets for protein design AI. The platform enables the generation of tens of millions of clusters of immobilized proteins directly on an Illumina DNA sequencing flow cell through efficient tethered in situ transcription and translation.

Prot-MaP was invented by the company’s CEO and co-founder, Curtis Layton, and co-founder Will Greenleaf, PhD, a professor of genetics at Stanford University School of Medicine and a member of Protillion’s Scientific Advisory Board. After receiving his PhD in computational biology from Duke University, Layton studied with Greenleaf as a postdoctoral fellow in the genetics department at Stanford Medicine.

Layton developed Prot-MaP while working in Greenleaf’s lab, then organized Protillion in 2019 to commercialize the technology. Layton’s work pioneered a new approach to high-throughput interrogation of biochemical systems, tackling ultra-high-impact technology approaches for drug discovery by uniting fields that included protein engineering, next-generation sequencing technology, molecular biology, in vitro transcription and translation, computational biology, software development, and various engineering disciplines.

Days rather than months

“Prot-MaP is a technology platform that allows us to test millions of protein interactions simultaneously, generating an unprecedented amount of data in a matter of days rather than months. For example, we can rapidly evaluate large libraries of therapeutic protein candidates to see how they bind to different targets and how they behave under different biological conditions,” Robert Hollingsworth, PhD, Protillion’s CSO, told GEN.

“We then combine that data with proprietary AI and machine learning tools to understand what drives the best-performing proteins and quickly design improved candidates. This gives us the ability to engineer antibodies with highly specific characteristics, such as stronger and more precise target binding, the ability to engage multiple targets, or the ability to activate only under certain physiological conditions,” Hollingsworth explained. “In practical terms, Prot-MaP helps us discover and optimize better drug candidates faster, with a level of insight and precision that has not previously been possible at this scale.”

Prot-MaP allows Protillion to test up to one million protein variants simultaneously in a single experiment and generate results in as little as 48 hours.

“Because we operate multiple proprietary platforms in parallel, we can rapidly scale that capability and generate enormous amounts of experimental data on demand,” Hollingsworth explained.

What makes Prot-MaP unique, he continued, is not just its scale, but its combination of scale, speed, and AI.

“The platform is designed to seamlessly connect high-throughput protein testing with proprietary machine learning models, allowing us to quickly identify promising drug candidates, understand what makes them work, and design improved versions,” Hollingsworth said. “This enables us to tackle everything from discovering entirely new therapeutic molecules to optimizing existing candidates for multiple desired characteristics. In practical terms, Prot-MaP helps us find and develop better biologic medicines faster and more efficiently than traditional approaches.

Opposite approach

How does Prot-MaP overcome the complexity of protein molecules, which has long been a hurdle in protein and biologics design?

“Many companies start with AI and then look for data. We took the opposite approach,” Hollingsworth said.

Rather than relying primarily on computer predictions of protein structure, he elaborated, Protillion can apply Prot-MaP to directly generate large-scale functional data and identify the best therapeutic candidates based on real-world experimental results.

“What makes this especially powerful is the sheer scale of the data we can generate. While much of the industry has focused on applying AI to relatively limited biological datasets, we believe that the biggest challenge in drug discovery is obtaining enough high-quality data to truly understand the complexity of protein function. Prot-MaP was built to solve that problem,” Hollingsworth said.

By generating millions of protein measurements in parallel, Protillion says, it can create the kind of rich, large-scale datasets needed to train more powerful and predictive machine learning models. Those models, in turn, help design and optimize better therapeutic candidates faster and with greater precision.

“Prot-MaP combines high-throughput experimentation with AI. The platform allows us to rapidly generate the data, and the AI helps us learn from it—creating a cycle that accelerates the discovery of next-generation biologic medicines,” Hollingsworth said.

Speaking with GEN, Layton said Protillion and Merck have charted a course for the start of their drug discovery collaboration.

“Our first two programs focus on inflammatory diseases, where we see significant unmet medical need and strong opportunities for differentiation,” he said.

Immune-mediated inflammatory disorders are Merck’s specialty within its therapeutic area of focus in immunology. Merck focuses on several other therapeutic areas, which include oncology, vaccines, infectious diseases, cardiometabolic and respiratory diseases, neuroscience, and ophthalmology.

“However, the Prot-MaP platform’s capabilities extend far beyond inflammation, enabling the discovery and development of novel biologics across a broad range of therapeutic areas,” Layton added. “As we continue to advance the platform, we expect to expand into additional disease areas where its unique capabilities can have the greatest impact.”

Tech-focused pipeline collaborations

Merck has launched several tech-focused pipeline collaborations in recent months aimed at replenishing its cancer and immunology pipelines, with the goal of developing new therapies that can recoup the billions of dollars in sales the pharma giant will lose as patent exclusivity expires in the United States and elsewhere for its aging blockbusters, including cancer immunotherapy Keytruda® (pembrolizumab) and Gardasil® 9 (Human Papillomavirus 9-valent Vaccine, Recombinant).

In March, Merck inked an up-to-$2.2 billion collaboration with Quotient Therapeutics to apply Quotient’s somatic genomics platform to discover novel drug targets in inflammatory bowel disease (IBD). Also that month, Merck launched a partnership with Infinimmune to apply its Anthrobody® discovery platform and GLIMPSE antibody language model to identify and develop antibody candidates against multiple undisclosed Merck-designated targets. Merck agreed to pay Infinimmune an undisclosed upfront payment and up to $838 million in milestone payments.

Merck has also launched several tech-focused partnerships, with partners that include:

• Google Cloud—An up-to-$1 billion collaboration announced in April to deploy an agentic platform across Merck’s R&D, manufacturing, commercial, and corporate functions. Google Cloud engineers are working alongside Merck teams to deploy Google Cloud’s most sophisticated AI, including Gemini Enterprise.
• Tempus AI—An expanded, multi-year collaboration of undisclosed value announced in March, aimed at accelerating discovery and development of precision medicine biomarkers, and supporting Merck’s oncology and potentially broader therapeutic portfolios.
• Mayo Clinic—An R&D agreement of undisclosed value to apply AI, advanced analytics, and multimodal clinical data to support drug discovery and development. The agreement integrates Mayo Clinic’s Platform architecture, as well as clinical and genomic datasets, with Merck’s ambition of harnessing AI-enabled virtual cell technologies to enhance disease understanding, improve target identification, and drive early development decisions.

In its latest collaboration, Merck has agreed to pay Protillion an undisclosed upfront payment, plus up to $510 million in payments tied to achieving research, development, and commercial milestones toward the successful development of an unspecified number of therapies.

“Compelling opportunity”

“Powerful emerging technologies offer the potential to transform the speed and precision with which we characterize protein landscapes and identify novel therapeutic candidates,” Juan Alvarez, PhD, vice president, discovery biologics at Merck Research Laboratories, said in a statement. “Protillion’s platform offers a compelling opportunity, and we look forward to working with the team to advance these programs.”

Illumina’s venture capital arm, Illumina Ventures, is among investors in Protillion, having joined ARCH Venture Partners in 2022 to co-lead an $18 million financing in 2022.

Based in Carlsbad, CA, Protillion has grown rapidly to a workforce of 30 people. In March, Protillion hired Robert Hollingsworth, PhD, a drug development executive with more than 30 years’ experience in biopharma, as CSO through a placement by executive search firm CollectiveMinds. Before joining Protillion, Hollingsworth was CSO at Shoreline Therapeutics, and earlier held positions in companies that included Pfizer (as vp and CSO of cancer vaccines and immunotherapeutics), Pharmacia & Upjohn (since absorbed into Pfizer), GlaxoSmithKline (GSK), and MedImmune (acquired by AstraZeneca).

Protillion says it is continuing to expand its team and facilities, with the aim of supporting its internal pipeline and high-value strategic partnerships.

“We plan to hire six more FTEs [full-time equivalents] by the end of the year,” Layton said.

The post Merck, Protillion Launch AI Drug Discovery Collaboration with Up-to-$510M in Milestone Payments appeared first on GEN – Genetic Engineering and Biotechnology News.

Pancreatic cancer remains one of the most lethal malignancies, notorious for its late detection, rapid progression, and stubborn resistance to many therapeutic strategies clinicians have tried. Despite decades of effort, standard treatments have delivered only incremental gains, and the disease is projected to become the second leading cause of cancer‑related death within this decade. Now, researchers at the University of Cologne’s Center for Molecular Medicine Cologne (CMMC) have uncovered a surprising vulnerability in KRAS‑mutant pancreatic tumors—one that primes them for a potent form of programmed cell death.

In a study published in Nature Communications titled “Oncogenic KRAS-driven type I interferon signaling primes pancreatic cancer for necroptosis,” the team reported that oncogenic KRAS, the defining driver mutation in roughly 90% of pancreatic ductal adenocarcinomas (PDAC), activates a type I interferon signaling program that inadvertently primes tumor cells to necroptosis, an inflammatory form of regulated cell death. However, “KRAS‑mutated tumors have a previously unknown Achilles heel,” said senior author Silvia von Karstedt, PhD. “By switching off the tumor cells’ defense mechanisms, we can significantly kill these tumors.”

The defense mechanism in question is caspase‑8, a protein long known for its role in apoptosis but increasingly recognized as a gatekeeper that prevents necroptosis. The Cologne team found that KRAS‑driven interferon signaling induces high expression of necroptosis‑related interferon‑stimulated genes—including MLKL—creating a state in which tumor cells become heavily dependent on caspase‑8 for survival.

Using genetically engineered mouse models, the researchers showed that deleting caspase‑8 specifically in KRAS‑driven pancreatic lesions triggered widespread necroptotic cell death and eliminated most precursor lesions. “Cancer cell-specific deletion of caspase‑8 is sufficient to trigger necroptotic cell death, eliminating most pancreatic precursor lesions,” the authors reported in their paper.

Furthermore, in aggressive PDAC mouse models and human patient‑derived tumor organoids, pharmacologic caspase inhibition significantly reduced tumor burden.

First author Sofya Tishina, PhD, emphasizes the translational potential: “The findings provide strong evidence that certain forms of pancreatic cancer could be specifically targeted for treatment based on their dependence on caspase‑8. In the long term, this could help develop new therapies for patients who currently have very limited treatment options.”

Beyond pancreatic cancer, the study’s pan‑cancer transcriptomic analysis revealed that tumors with high Ras pathway activity and strong interferon signatures also exhibit elevated necroptosis gene expression, hinting at broader applicability. As the authors concluded in their paper, their work “reveals a KRAS-induced IFN program that sensitizes tumor cells to necroptosis, highlighting a therapeutic vulnerability in PDAC with broader relevance across IFN-activated cancers.”

The post Pancreatic Cancer Cell Death Triggered by Caspase‑8 Blockade in Preclinical Models appeared first on GEN – Genetic Engineering and Biotechnology News.

The Danaher Bioprocessing Summit, “The Next Era of Bioprocessing: From Promise to Patient Impact,” took place in London earlier this month. The event brought together officials from Danaher companies (Cytiva, Pall, Beckman Coulter Life Sciences, IDBS, and Leica Microsystems), along with biopharma manufacturers and researchers to discuss how the industry can accelerate the transition from scientific breakthroughs to commercial therapies.

Key themes included:

• AI-driven bioprocess development and manufacturing
• Intensified and continuous bioprocessing
• Digitalization and connected data ecosystems
• Improving productivity and reducing cost of goods
• Cell and gene therapy manufacturing challenges
• Scaling production of high-demand biologics, including GLP-1 therapies
• Advanced analytics and process control
• Sustainability in biomanufacturing

The conference emphasized that future competitive advantage in biomanufacturing will come less from building additional capacity and more from increasing productivity, speed, and process intelligence across the development-to-manufacturing workflow.

Based on the formal presentations, roundtable discussion groups, and conversations among speakers, panelists, and attendees, six key takeaway ideas emerged.

The old manufacturing playbook no longer fits

Biomanufacturing was built for large batches of standardized therapies. The next generation of medicines—cell and gene therapies, targeted and complex biologics, N-of-1 treatments—doesn’t fit that mold. As molecular diversity increases, the field is shifting toward smaller, parallel and distributed systems that can flex to meet the complexity of individualized medicine. This means faster decision-making, new investment models, and process designs that are data-driven and purpose-built from the start rather than adapted from previous playbooks.

Manufacturing can no longer be treated as a downstream problem. It has to be part of the scientific conversation from day one.

Automation and AI are making personalized scale possible

For years, the promise of personalized medicine ran into a hard wall: you can’t manufacture one patient’s therapy the same way you manufacture a million doses of a traditional drug. Integrating automation, AI and high-throughput experimentation is changing that equation.

These tools are enabling a shift from large-batch production to small-batch and even patient-specific manufacturing while improving efficiency, regulatory consistency, and access to advanced therapies. AI and digital tools are also compressing process development timelines, making it possible to design more tailored, adaptive manufacturing approaches without sacrificing rigor.

Prediction is becoming a competitive advantage

The organizations gaining ground are not merely reacting to problems faster but actually anticipating them. Digital twins powered by integrated data are helping teams model outcomes before committing resources, accelerating timelines and reducing risk. And real-time, molecular and submolecular-level data—shared across interoperable systems—are enabling more precise, proactive decision-making at every stage of development and manufacturing. Investing in this area with critical infrastructure is essential.

Breakthroughs require collaboration across the whole ecosystem

It turned out that the most consistent theme across both days of the Summit was this: no single organization can accomplish what’s needed alone. Partnerships across academia, industry, and regulators are accelerating how all kinds of therapies, especially gene therapies, move from discovery to approved treatment. Earlier alignment between developers, manufacturers, and regulators is reducing friction and compressing timelines.

What makes these partnerships work is not goodwill alone but transparency, shared incentives, and data-driven collaboration that keeps everyone oriented around the same outcomes. The organizations making the most progress are those treating collaboration as a core capability.

Regulatory models are evolving with the science, and sustainability is now a procurement requirement, not a values statement 

Most of the frameworks that currently govern drug development were built for an earlier era of medicine. As therapies grow more complex and more personalized, those frameworks must change. Early engagement and risk-based approaches are helping bring complex therapies to patients faster without compromising scientific rigor. For rare diseases, which collectively affect an estimated 300 million people worldwide, tailored regulatory pathways are becoming essential.

Summit speakers agreed that regulators are not the obstacle they’re sometimes assumed to be. They want to move faster too, and building the right collaborative infrastructure makes that possible.

Environmental and social criteria are moving from corporate commitments into day-to-day supplier decisions. Organizations are integrating sustainability standards into procurement frameworks, requiring verified supplier commitments and shared performance targets as part of doing business.

Far from being separate from operational strategy, this is part of building supply chains resilient enough to support long-term innovation at scale. Progress toward net-zero goals, Summit participants agreed, accelerates when sustainability is wired into commercial relationships rather than managed alongside them.

The post Six Takeaways from the Danaher Bioprocessing Summit appeared first on GEN – Genetic Engineering and Biotechnology News.