The History of Stem Cell Research: Sixty Years, One Lineage at a Time

Scope of This Review

This is a history of stem cell science: what was discovered, when, by whom, and — a distinction this page keeps throughout — in which cells.

That last point matters more than it might seem. "Stem cell research" covers several largely separate scientific programmes: haematopoietic cells that rebuild blood and immune systems, embryonic and induced pluripotent cells that can become any tissue, and mesenchymal stromal cells found in marrow, fat and umbilical cord. They were discovered by different people, they treat different diseases, and they are given by different routes.

Progress in one does not transfer to another. A history page is where that distinction is most easily lost, so it is marked at every step.

Early Discoveries (1960s–1970s)

1961–1963: The clonal proof

The modern era begins at the Ontario Cancer Institute in Toronto with James Till and Ernest McCulloch.

Their 1961 paper measured the radiation sensitivity of normal mouse bone marrow cells¹. In doing so they noticed that transplanted marrow produced visible nodules in the spleens of irradiated recipient mice, and that the number of nodules varied linearly with the number of cells injected — implying each colony arose from a single cell.

Two years later, with Andrew Becker, they proved it. Using radiation-induced chromosome markers, they showed that all the cells within a given spleen colony carried the same marker: the colonies were clonal, derived from single cells, and contained multiple blood lineages².

That is the experimental foundation of the field: the first quantitative demonstration that a single cell could both self-renew and differentiate into several cell types. Their work is among the most heavily cited in twentieth-century haematology.

(A note on a common claim: Till and McCulloch did not coin the term "stem cell", which is considerably older. What they did was harder — they made it measurable.)

1966: Friedenstein and the marrow stroma

While haematopoietic research advanced, Alexander Friedenstein in Moscow was studying a different population entirely.

His 1966 work showed that bone marrow transplants could generate bone in ectopic sites, implying that marrow contained an osteogenic precursor distinct from blood-forming cells³. Over the following two decades he and his colleagues characterised these cells: fibroblast-like, colony-forming when cultured in monolayer⁴, and capable of forming bone when transplanted in diffusion chambers¹⁰.

These are the cells that would later be called mesenchymal stem cells. The identification is 1966 — a laboratory finding about what marrow contains, not a treatment.

1968: The first clinical successes

Clinical translation came remarkably fast, and it came in two papers published consecutively in the same issue of The Lancet.

Bach and colleagues reported marrow transplantation in a patient with Wiskott-Aldrich syndrome⁵. Gatti, Meuwissen, Allen, Hong and Good reported immunological reconstitution in a boy with sex-linked lymphopenic immunological deficiency⁶ — an inherited severe combined immunodeficiency.

Both were allogeneic sibling transplants, and that is precisely the significance: they demonstrated that marrow from a non-identical donor could engraft and rebuild an immune system. Two different children, two different diseases, one shared proof of principle.

The 1970s: From proof to therapy

Turning that proof into reliable treatment took a further two decades of work on conditioning regimens, donor matching, graft-versus-host disease and infection control — led principally by E. Donnall Thomas and colleagues in Seattle, whose 1975 account in the New England Journal of Medicine set out marrow transplantation as clinical practice⁷. Thomas shared the 1990 Nobel Prize in Physiology or Medicine for this work.

This is the field's genuine triumph: a cell therapy that became curative standard of care for leukaemia and inherited immune disorders. It is also, specifically, haematopoietic stem cells given intravenously to rebuild a blood and immune system that has been deliberately ablated. That is a different intervention from anything discussed later on this page.

The Expansion Era (1980s–1990s)

1981: Mouse embryonic stem cells

Two groups reported the same breakthrough independently. Martin Evans and Matthew Kaufman at Cambridge established pluripotent cell cultures from mouse embryos⁸; Gail Martin at UCSF isolated a pluripotent line from early mouse embryos cultured in conditioned medium⁹.

For the first time, pluripotent cells could be maintained indefinitely in culture while retaining the capacity to become any cell type. The mouse embryonic stem cell system went on to underpin gene targeting and transgenic mouse models, transforming biomedical research well beyond stem cell biology.

1991: Naming the mesenchymal stem cell

Arnold Caplan at Case Western Reserve gave the field its vocabulary. His 1991 paper set out a framework for marrow-derived cells capable of forming bone, cartilage and other connective tissues, and named them mesenchymal stem cells¹¹.

The following year, Haynesworth, Goshima, Goldberg and Caplan characterised cells with osteogenic potential from human marrow¹⁶ — an important step toward clinical-scale work, building on culture methods Friedenstein had established two decades earlier⁴ ¹⁰.

Caplan later argued that "mesenchymal stem cell" was a misnomer, proposing "medicinal signalling cells" instead, on the grounds that these cells act mainly by secreting factors rather than by becoming new tissue. That revision is itself part of the history, and it anticipates what later clinical evidence would show.

1998: Human embryonic stem cells

James Thomson and colleagues at Wisconsin derived the first human embryonic stem cell lines from blastocysts, demonstrating differentiation into all three germ layers¹².

The announcement triggered simultaneous scientific investment and profound ethical controversy over the use of human embryos — a debate that shaped research funding policy in several countries for the following decade.

The Pluripotency Revolution (2006–2007)

Shinya Yamanaka and Kazutoshi Takahashi at Kyoto reported in 2006 that introducing four transcription factors could reprogram adult mouse fibroblasts into pluripotent cells resembling embryonic stem cells¹³.

The following year, two groups independently extended this to human cells: Yamanaka's group¹⁴ and Thomson's¹⁵.

The implications were immediate. Pluripotent cells could now be made from a patient's own skin, avoiding the ethical objections to embryonic sources and opening patient-specific disease modelling. Yamanaka shared the 2012 Nobel Prize in Physiology or Medicine with Sir John Gurdon for the demonstration that cellular differentiation is reversible.

This is genuinely revolutionary science. It is also, so far, principally a research tool: induced pluripotent cells are not the cells used in commercial regenerative clinics.

The Clinical Era for Mesenchymal Cells

2006: A definition, at last

For decades, "mesenchymal stem cell" meant different things in different laboratories, which made study results impossible to compare. In 2006 the International Society for Cellular Therapy published minimal defining criteria: plastic adherence, a specified surface-marker profile, and demonstrated tri-lineage differentiation²⁰.

This is less glamorous than a Nobel Prize and arguably more consequential for clinical research — it is why MSC studies from different groups can be compared at all.

The 2010s: Scale without conclusions

MSC clinical trials expanded enormously. By 2020, more than 1,050 clinical trials exploring mesenchymal stromal cells were registered with the FDA, spanning neurodegenerative and cardiac disorders, perianal fistulas, graft-versus-host disease, COVID-19 and cancer¹⁹.

The results were harder than the volume suggests. As one 2020 review put it directly: "most of the clinical-stage MSC therapies have been unable to meet primary efficacy end points", and "the innate therapeutic functions of MSCs administered to humans are not as robust as demonstrated in preclinical studies"¹⁹.

The gap between preclinical promise and clinical result is one of the defining features of this period, and it is why the approvals that did come took as long as they did.

2018: The first European approval

Darvadstrocel (Alofisel), an allogeneic adipose-derived mesenchymal stromal cell product, was approved in the European Union in 2018 for complex perianal fistulas in Crohn's disease — a specific indication with a locally administered product. Long-term follow-up has since been published²¹.

December 2024: The first FDA approval — and how it happened

Remestemcel-L (Ryoncil) became the first mesenchymal stromal cell product approved by the FDA, for steroid-refractory acute graft-versus-host disease in children.

The path there is worth telling honestly, because it is instructive.

The randomised trial did not succeed. A Phase 3 multicentre randomised study enrolled 260 patients aged six months to 70 years, assigned 2:1 to remestemcel-L or placebo added to second-line therapy for steroid-refractory acute GVHD. It did not meet its primary endpoint: durable complete response was 35% versus 30% (P = 0.42). Post-hoc analyses suggested benefit in specific groups — patients with liver involvement (29% vs 5%, P = .047), high-risk grades C and D (overall response 58% vs 37%, P = 0.03), and paediatric patients¹⁷.

The paediatric trial did. A subsequent Phase 3 single-arm study enrolled 54 children with primary steroid-refractory acute GVHD. Day-28 overall response was 70.4%, against a prespecified control value of 45% (P = .0003), sustained through day 100, with complete response rising from 29.6% at day 28 to 44.4% at day 100. Overall survival was 74.1% at day 100 and 68.5% at day 180, and day-28 responders survived substantially better than non-responders (86.8% vs 47.1% at day 100, P = .0001)¹⁸.

Approval followed. Fifty-eight years after Friedenstein's 1966 paper, one mesenchymal stromal cell product reached the market — in a specific, severe, well-defined indication, in children, after a failed randomised trial and a successful single-arm one.

Timeline

What This History Does and Does Not Establish

This section exists because the inference is easy to get wrong, and it is the most common error made with this material.

What the history establishes. Cell therapy can work. Haematopoietic stem cell transplantation is curative for leukaemia and for inherited immune deficiencies — a genuine, Nobel-recognised triumph built over decades⁵ ⁶ ⁷. Pluripotent cell biology has transformed biomedical research⁸ ¹² ¹³. And as of December 2024, one mesenchymal stromal cell product has met a regulator's evidentiary standard in a defined indication¹⁷ ¹⁸.

What it does not establish. That any of this transfers.

Marrow transplantation cures leukaemia because haematopoietic stem cells, given after the recipient's marrow has been ablated, permanently rebuild a blood and immune system. That mechanism has no counterpart in an intra-articular or intravenous mesenchymal cell infusion, where the cells are cleared by the recipient within days.

The remestemcel-L approval is for steroid-refractory acute graft-versus-host disease in children — a severe, well-defined, life-threatening condition, treated with a specified product at a specified dose. It says nothing about joints, energy, ageing or cosmetic indications, and its own randomised trial in a broader population did not meet its primary endpoint¹⁷.

And across the field as a whole, most MSC therapies tested in trials have not met their primary efficacy endpoints¹⁹.

A long and distinguished history is a reason to take a field seriously. It is not evidence for a particular treatment. For any specific intervention, the question remains what it always was: which trial, in which patients, measuring what, against what comparison.

Safety, Across Sixty Years

The accumulated safety record for mesenchymal stromal cells is one of the more reassuring things in this history, and it should be stated with both halves.

The largest safety synthesis across indications found no association with acute infusional toxicity, organ system complications, infection, death or malignancy — and a significant association with transient fever (OR 16.82, 95% CI 5.33–53.10)²².

That is a genuinely favourable profile for a cell therapy, and it is not the same as an absence of risk. Route, cell source and manufacturing standard all matter, and the safety of one product does not establish the safety of another.

Where the Field Stands

Established. Haematopoietic stem cell transplantation for haematological malignancy and inherited immune deficiency. Decades of data, curative in defined populations.

Approved, narrowly. Darvadstrocel in the EU for perianal fistulas in Crohn's disease (2018)²¹. Remestemcel-L in the US for paediatric steroid-refractory acute GVHD (December 2024)¹⁷ ¹⁸.

Investigational. Almost everything else — orthopaedic, cardiac, neurological, autoimmune and cosmetic applications of mesenchymal stromal cells, where more than a thousand trials have been registered¹⁹ and most have not met their primary endpoints.

Research tools. Embryonic and induced pluripotent stem cells, transformative for disease modelling and drug discovery, and not yet routine therapies.

The honest summary of sixty years is that the science is remarkable, the translation has been slow and partial, and the distinction between the two is where most of the confusion in this area lives.