Optimistic, Not Naive: Where Artificial Blood Actually Stands
The blood supply is short again. It's short often enough that "short" undersells it, this is closer to a chronic condition than an emergency. So the question comes up, reasonably, from patients, from colleagues outside the field, sometimes from residents who assume someone must have solved this by now: why can't we just make blood?
The honest answer is that people have been trying for over a century, and it's worth being clear-eyed about why before getting excited about where things stand now.
The part that didn't work
Hemoglobin-based oxygen carriers, cell-free hemoglobin packaged to carry oxygen the way red cells do, looked like the obvious answer for decades. Several made it deep into clinical development: HemAssist, PolyHeme, Hemolink, Hemospan. All were eventually discontinued. Free hemoglobin outside a red cell membrane scavenges nitric oxide, which is the molecule that keeps blood vessels relaxed, and the resulting vasoconstriction was linked to safety concerns, including cardiac events, serious enough that none of these products gained lasting traction. There's genuine debate about how solid that signal actually was, I've written elsewhere about the meta-analysis that's often credited with sinking the field, and I won't relitigate it here, but the practical upshot for the field was the same either way: the products didn't make it to market, and momentum stalled.
That history matters as context for what follows, because the newer approaches aren't ignoring that failure. They're built directly in response to it.
ErythroMer: the encapsulation fix
ErythroMer, developed by KaloCyte, takes a different structural approach: instead of free hemoglobin in solution, the hemoglobin is encapsulated in a lipid-peptide nanoparticle designed to limit exactly the nitric oxide interaction that caused trouble before. It's also freeze-dried, meaning it can be stored stable and reconstituted when needed, which is a meaningfully different profile than a product requiring refrigeration and a short shelf life.
The Defense Advanced Research Projects Agency backed this approach with a $46 million grant aimed at a field-deployable, shelf-stable blood substitute, with ErythroMer as the core technology. This is still in development, not in routine clinical use, but the engineering logic is a direct answer to the specific mechanism that sank the earlier generation.
RESTORE: blood grown in a lab
A different approach entirely: growing actual red blood cells from donor stem cells in the lab, then transfusing them. The RESTORE trial, run by NHS Blood and Transplant and the University of Bristol, became the first in the world to transfuse laboratory-grown red cells into another person. Volunteers receive mini-transfusions of both lab-grown and standard donated cells from the same source donor, allowing a direct comparison of how long each population survives in circulation.
The important caveat, and the researchers have been explicit about this: the goal isn't replacing blood donation at scale. It's aimed at patients who are hard to transfuse safely through the conventional donor pool, people with rare blood phenotypes or significant alloimmunization, where finding compatible units the usual way is already a struggle. This is a precision tool for a specific hard problem, not a fix for Tuesday's shortage.
Platelets are getting their own version of this
Red cells aren't the only shortage. Freeze-dried platelet substitutes like Thrombosomes are in parallel development, aimed less at routine platelet support and more at hemostatic emergencies, trauma, austere environments, situations where getting a normal platelet unit to the bedside in time isn't realistic. It's a quieter research track than the red cell substitutes, but it's answering a real and distinct gap.
What's actually reached a patient
Of everything in this space, Hemopure is the one product that's actually in clinical use anywhere, and it's worth being specific about what it is. Hemopure (HBOC-201) is itself one of the original generation of hemoglobin-based oxygen carriers, the same era as HemAssist, PolyHeme, and Hemolink, but built differently. Where PolyHeme used human hemoglobin with pyridoxylation and glutaraldehyde polymerization, and Hemolink used human hemoglobin crosslinked with o-raffinose into a smaller oligomer, Hemopure uses bovine hemoglobin polymerized with glutaraldehyde into a notably larger molecule, averaging around 250 kDa. That different chemistry and different molecular weight profile is generally credited as part of why it's the one product from that generation still standing. It's approved in South Africa and available in the US under FDA expanded access for specific situations. That's a meaningfully lower bar than "solved," but it's real-world use, not just a promising abstract, and it's worth naming as the one concrete data point that artifiical blood products can reach a patient at all.
Where that leaves us
None of this fixes this month's shortage. ErythroMer is still in development. RESTORE is aimed at a narrow, difficult patient population, not the general supply problem. Hemopure's footprint is small and geographically limited. If you're standing in a blood bank today trying to find units for a patient, none of this helps you yet.
But "not yet" is a genuinely different sentence than "never," and that's the honest distinction worth holding onto. The current generation of approaches is responding directly to the specific failure modes of the last one, which is not something the field could always say. It's also not close enough to change how any of us practice this year, or probably this decade.
Optimistic, in this field, has to mean optimistic about the trajectory. It can't yet mean optimistic about the timeline.
