TL;DR
The New York Blood Center Enterprises and the Chan Zuckerberg Biohub are collaborating to derive induced pluripotent stem cell lines from cord blood, using NYBCe’s inventory of over 30,000 units, including rare HLA-homozygous donors, and Biohub’s expertise in reprogramming immune cells. The resulting cell lines will be shared research resources for regenerative medicine, disease modelling, and cell therapy, addressing the field’s core bottleneck: the gap between laboratory biology and scalable, immune-compatible treatments
Most of the umbilical cord blood collected at birth is discarded. The tissue is clamped, cut, and disposed of as medical waste, along with the stem cells it contains, cells that are immunologically naive, genetically diverse, and capable of being reprogrammed into virtually any cell type in the human body. The New York Blood Center Enterprises, which operates the world’s oldest public cord blood bank, and the Chan Zuckerberg Biohub, the biomedical research network funded by Mark Zuckerberg and Priscilla Chan, announced on 30 April that they are collaborating to turn those discarded cells into something more durable: a library of induced pluripotent stem cell lines derived from cord blood, designed to be broadly compatible with human immune systems and available as shared research resources for regenerative medicine, disease modelling, and cell therapy. The collaboration is not large by the standards of pharmaceutical deal-making. There is no billion-dollar licensing agreement, no IPO filing, no press release with a valuation attached. What it represents is an attempt to solve a structural problem that has constrained the field of cell therapy since its inception: the gap between the biology that researchers can create in a laboratory and the biology that can be manufactured, stored, and delivered to patients at scale.
The biology
Induced pluripotent stem cells are adult cells that have been reprogrammed to behave like embryonic stem cells, capable of differentiating into any cell type. The technology was pioneered by Shinya Yamanaka, who won the Nobel Prize in 2012 for demonstrating that mature cells could be returned to a pluripotent state by introducing four specific genes. Since then, iPSCs have become the foundation of a growing pipeline of experimental therapies: iPSC-derived neurons for Parkinson’s disease, iPSC-derived cardiomyocytes for heart failure, iPSC-derived immune cells for cancer. More than 115 clinical trials involving pluripotent stem cell therapies are now under way globally, with over 1,200 patients treated and no significant safety concerns reported. The FDA has granted its Regenerative Medicine Advanced Therapy designation to more than 60 products, including the first iPSC-derived therapy to receive both Fast Track and RMAT status, a treatment for Parkinson’s disease developed by iRegene. The global iPSC market is projected to grow from $2.6 billion in 2026 to $4.1 billion by 2031.
The problem is not whether iPSCs work. It is where they come from. Most iPSC lines used in research are derived from skin cells or blood cells taken from individual donors, a process that produces lines with highly variable genetic backgrounds and immune profiles. When these cells are transplanted into a patient, the recipient’s immune system treats them as foreign tissue and attacks them, requiring lifelong immunosuppressive drugs that carry their own risks. The alternative, creating patient-specific iPSC lines for each individual, is prohibitively expensive and slow. The field needs something in between: a set of standardised, well-characterised iPSC lines that are compatible with large segments of the population without requiring personalised manufacturing for each patient. This is where cord blood enters the picture.
The resource
The National Cord Blood Program, housed within NYBCe’s Lindsley F. Kimball Research Institute, was established in 1992 by Dr Pablo Rubinstein with funding from the National Heart, Lung and Blood Institute. It was the world’s first public cord blood bank and for decades was the largest, maintaining an inventory of over 30,000 cord blood units consented for clinical transplantation and research. The programme produced HEMACORD, the first FDA-licensed cord blood product, and its units have been used in thousands of transplants for patients with blood cancers and other disorders who lack a matched bone marrow donor. Cord blood is immunologically privileged: its stem cells are less likely to trigger graft-versus-host disease than adult bone marrow, which is why cord blood transplants can tolerate a greater degree of HLA mismatch between donor and recipient. A phase 2 trial published in the Journal of Clinical Oncology in April 2026 by Fred Hutch Cancer Center showed that a pooled cord blood product achieved 96 per cent one-year survival in leukaemia patients with zero cases of severe graft-versus-host disease.
Among NYBCe’s inventory are cord blood units from donors who are homozygous at key HLA loci, meaning they carry identical copies of the immune-compatibility genes on both chromosomes. These units are rare and valuable because iPSC lines derived from them would be compatible with a disproportionately large share of the population. Research in Japan, where the iPSC haplobanking concept originated, demonstrated that lines derived from approximately 50 homozygous donors could match over 90 per cent of the Japanese population. The NYBCe-Biohub collaboration will use these homozygous units, along with LFKRI’s expertise in high-resolution cell sorting and haematology, to create iPSC lines with the broadest possible translational potential.
The approach
The collaboration is structured as a multi-phase scientific effort. CZ Biohub New York, which focuses on engineering immune cells for early disease detection and treatment, contributes expertise in reprogramming defined immune cell subsets into iPSCs. The key innovation is specificity: rather than reprogramming a random population of cells from cord blood, the team will isolate defined immune cell populations, T cells, B cells, natural killer cells, and others, sort them using high-resolution techniques, and reprogram each subset individually. This matters because iPSCs retain epigenetic memory of their cell of origin, meaning an iPSC derived from a T cell will behave differently from one derived from a B cell, even after reprogramming. By preserving these genetic signatures, the collaboration aims to create iPSC lines that encode specific immune functions, effectively immortalising donor immune cells for long-term research use.
The resulting cell lines will be made available as shared research resources, a detail that distinguishes this effort from the proprietary iPSC platforms being built by pharmaceutical companies and venture-backed startups. The open-resource model reflects CZ Biohub’s broader mission: by the end of 2026, researchers across the Biohub network plan to genetically modify approximately 100 million individual cells and systematically record the resulting molecular and functional changes. The iPSC lines from this collaboration will feed into that larger programme, enabling systematic study of how a cell’s origin affects its behaviour after reprogramming, a question that has significant implications for the growing use of AI in drug development, where the quality and standardisation of biological inputs directly determines the reliability of computational predictions.
The context
The collaboration arrives at a moment when cell therapy is transitioning from a research curiosity to an industrial proposition, and encountering the manufacturing constraints that transition implies. The field’s commercial leaders are pursuing two parallel strategies to achieve immune compatibility at scale. The first is hypoimmunogenic engineering: using CRISPR to knock out HLA genes from iPSCs, rendering the cells invisible to the recipient’s immune system. The second is haplobanking: building curated libraries of iPSC lines from HLA-homozygous donors that collectively cover large populations. Japan’s CiRA has established a GMP-grade haplobank of 27 lines from homozygous donors, already used in 12 clinical trials. Spain has built a similar bank covering its population. The NYBCe-Biohub effort adds a distinctive element: by deriving iPSCs specifically from immune cell subsets in cord blood, it creates lines that are not only broadly compatible but also carry the functional signatures of the immune system, a feature that could prove critical for developing cell therapies that need to interact with, rather than evade, the patient’s immune response.
The broader landscape is shaped by the convergence of AI and biology that is redefining both fields. DeepMind’s Isomorphic Labs, which is preparing to dose its first patients in oncology trials this year, represents one pole of this convergence: using AI to design drugs computationally before they enter the laboratory. The iPSC platform represents the other: creating standardised biological materials that can serve as inputs for AI-driven research and development. Eleven billion dollars flowed into AI-enabled drug discovery in the first quarter of 2026 alone. But AI in healthcare is only as reliable as the data it trains on, and in biology, the data is cells. Standardised, well-characterised iPSC lines with known immune profiles and defined cell-of-origin histories are the biological equivalent of clean training data: they reduce noise, improve reproducibility, and make computational predictions more trustworthy. The NYBCe-Biohub collaboration is building infrastructure for a field that is increasingly dependent on both computational and biological platforms working in concert.
The gap
The distance between a cord blood unit in a freezer in New York and a cell therapy delivered to a patient remains vast. iPSC-derived therapies face regulatory hurdles, manufacturing complexity, and the fundamental biological uncertainty of whether cells that behave predictably in a laboratory will behave predictably in a human body over decades. No iPSC-derived drug has yet received FDA approval. The clinical trials are promising, but they are early-stage, small, and concentrated in a handful of indications. The iPSC market’s projected growth to $4.1 billion by 2031 is significant for a nascent industry but modest compared to the tens of billions flowing into AI infrastructure every quarter. The Chan Zuckerberg Initiative has committed to spending at least $10 billion on basic scientific research over the next decade, roughly $1 billion per year, which is substantial for philanthropy but marginal by the standards of the pharmaceutical industry, where a single drug can cost $2 billion to bring to market.
What the NYBCe-Biohub collaboration offers is not a product but a platform: a set of well-characterised, broadly compatible, immune-cell-derived iPSC lines that any researcher can use to study how cells develop, differentiate, and respond to disease. The value of a platform is not in what it produces immediately but in what it enables others to produce over time. The world’s oldest public cord blood bank stopped collecting new donations in 2020, a decision driven by funding constraints that affect health infrastructure far more than they affect technology companies. But the units it collected over three decades, carefully typed, consented, and frozen, turn out to be precisely the raw material that the next generation of cell therapy requires. The biology was already in the freezer. The question was always whether anyone would build the platform to use it.
