Cellular Therapies in General Surgery: An Evidence-Based Review

Executive Summary

Cellular therapies — the therapeutic use of living cells, their derivatives or cell-instructive biomaterials — have moved from the edge of experimental biology into selected areas of surgical practice. In general surgery the clearest translational successes sit in three places: skin and burn reconstruction (cultured epidermal autografts, autologous skin-cell suspension), surgical oncology (the first tumour-infiltrating lymphocyte therapy approved for a solid tumour) and the β-cell replacement that concerns pancreatic and endocrine surgery (the first approved allogeneic islet-cell therapy, and stem-cell-derived islets that achieved sustained insulin independence in early trials).

Alongside these sits a large preclinical and early-phase literature covering mesenchymal stromal cells (MSCs), adipose- and bone-marrow-derived populations, induced pluripotent stem cells (iPSCs), organoids, extracellular vesicles (exosomes), platelet-rich plasma (PRP) and three-dimensional bioprinted constructs: wound healing, anastomotic reinforcement, fibrosis modulation, hepatic and pancreatic regeneration, whole-organ bioengineering.

The central message is measured optimism: a handful of cell products have earned a permanent place in surgical care, a few more are advancing quickly, but most surgical applications remain investigational. The withdrawal from the European Union in 2024 of the first approved allogeneic MSC product (darvadstrocel) for perianal fistula in Crohn's disease, after its confirmatory trial failed, is a striking reminder that early promise is no guarantee of durable clinical benefit.

Introduction

Surgery removes, repairs and reconstructs tissue; yet to finish the job it has historically depended on the body's own healing capacity. Cellular therapy proposes to augment or replace that capacity: it offers living cells, cell-derived signals or cell-laden scaffolds in order to accelerate healing, restore lost function, remodel scar or attack residual malignancy.

The premise is compelling for a surgeon: many of the most feared outcomes in general surgery — anastomotic leak, non-healing wounds, incisional hernia, fistula, hepatic failure after major resection and locoregional cancer recurrence — are failures or limitations of tissue biology that a well-designed cell product could in principle address.

The field sits at the intersection of three converging currents: stem cell biology, which explains how self-renewing and multipotent cells build and repair tissue; tissue engineering, which rebuilds three-dimensional structures by pairing cells with scaffolds and signals; and immuno-oncology with adoptive cell therapy, which uses engineered or expanded immune cells as weapons against tumours. General surgery draws on all three.

Even so, the translational record is uneven. For every approved product there are dozens of promising preclinical reports that failed rigorous human testing, and an unproven stem-cell and exosome grey market that regulators actively police. This review therefore takes a deliberately critical stance: it credits the genuine successes, situates them within the mechanisms that explain them, and draws a clear line between what is approved, what is promising in trials and what remains hypothesis.

Questions addressed

The text is organised to answer the following: which cellular therapies are used in general surgery and which remain experimental; how stem cell populations compare biologically and clinically; which molecular mechanisms drive regeneration; whether cell therapies improve wound, anastomotic and postoperative outcomes and reduce fibrosis; whether immune-cell therapy improves cancer surgery outcomes; how exosomes compare with cells; what the tumourigenesis risks are; how biomaterials and gene editing strengthen these therapies; and what can reasonably be expected over the next 5 to 20 years.

Historical Development of Cellular Therapy

The lineage of cellular therapy runs from blood transfusion, systematised in the early twentieth century, to the archetypal curative cell therapy: haematopoietic stem cell transplantation (bone marrow transplant; pioneered by E. Donnall Thomas in the 1950s–1970s, Nobel Prize 1990). Skin biology advanced in parallel: Rheinwald and Green established serial culture of human keratinocytes in 1975, which made possible the cultured epidermal autografts used to close massive burns in the 1980s and later commercialised as Epicel.

Two conceptual milestones defined the modern era. Friedenstein described bone-marrow-derived fibroblastic colony-forming cells in the 1970s; in 1991 Arnold Caplan reframed these marrow stromal cells as multipotent progenitors for connective tissue and coined the term "mesenchymal stem cells". In 1993 Langer and Vacanti codified tissue engineering in Science, proposing cells seeded onto biodegradable scaffolds as the route to functional tissue. In 1998 Thomson and colleagues derived human embryonic stem cells, which both opened and politicised the pluripotent-cell field. The transformative breakthrough came in 2006–2007, when Takahashi and Yamanaka reprogrammed somatic cells into induced pluripotent stem cells with four transcription factors (Oct4, Sox2, Klf4, c-Myc) — sidestepping much of the ethical debate around embryonic cells and winning the 2012 Nobel Prize.

The 2010s brought clinical inflection points relevant to surgery: the first organoids (Sato and Clevers, intestinal crypt organoids, 2009) established patient-derived three-dimensional "mini-organs"; CAR-T cells received their first approvals in haematological malignancies in 2017; and the 2006 ISCT minimal criteria standardised MSC identity. Most recently, in 2023–2024, regulators approved the first allogeneic islet-cell therapy (Lantidra/donislecel) and the first cell therapy for a solid tumour (Amtagvi/lifileucel) — both of which bear directly on surgical oncology and endocrine/pancreatic surgery — and in 2025 stem-cell-derived islets crossed the threshold of sustained insulin independence in humans.

Fundamentals of Cell Biology

A few principles of cell biology clarify why cell therapies behave as they do.

The potency hierarchy

Cells differ in the breadth of differentiation available to them: totipotent (the zygote, which can form both embryo and placenta), pluripotent (embryonic stem cells and iPSCs; all three germ layers), multipotent (tissue-specific stem cells such as MSCs and haematopoietic stem cells; a restricted set of lineages) and unipotent (committed progenitors). Broader potency means greater regenerative flexibility but — critically — higher tumourigenic risk: pluripotent cells can form teratomas if undifferentiated cells remain in a graft.

Self-renewal versus differentiation

Stem cells balance symmetric division (expanding the stem pool) against asymmetric division (producing differentiated daughters), governed by niche signals. Therapeutically we exploit either expansion, to generate enough cells, or directed differentiation, to make the desired cell type.

Niche and microenvironment

Stem cell behaviour is dictated by a local niche of matrix stiffness, oxygen tension, growth factors and neighbouring cells. That is why where and how cells are delivered often matters more than which cells they are — a graft dropped into a hostile, ischaemic, inflamed surgical bed frequently dies before it can act.

Paracrine signalling

A striking insight of the past two decades is that many therapeutic cells — MSCs above all — act less by engrafting and differentiating than by secreting a bioactive milieu of growth factors, cytokines and extracellular vesicles that instructs host tissue. This "hit-and-run" paracrine mechanism reframes MSCs as drug-delivering cells and motivates directly cell-free approaches (exosome, secretome).

Immunobiology

Autologous cells avoid rejection but require patient-specific, slow and expensive manufacture. Allogeneic cells allow "off-the-shelf" products but carry a risk of immune rejection; MSCs are relatively — not absolutely — immune-evasive and immunomodulatory, which is the basis of their allogeneic use. Gene editing to delete HLA or add immune-cloaking genes is an active strategy for creating truly universal donor cells.

Types of Cellular Therapy

The cellular therapies relevant to general surgery fall into several families:

• Adult (somatic) stem and stromal cells — MSCs from bone marrow (BM-MSC), adipose tissue (ASC/AD-MSC) and umbilical cord (UC-MSC); haematopoietic stem cells; tissue-resident progenitors. The workhorses of surgical regenerative research.
• Pluripotent stem cells — embryonic stem cells (ESCs) and iPSCs; used mostly as a manufacturing source of differentiated cells (islets, hepatocytes) rather than administered raw.
• Differentiated cell products — cultured keratinocytes, chondrocytes, hepatocytes and pancreatic islets.
• Organoids — self-organising three-dimensional tissue units grown from stem or progenitor cells, for disease modelling, drug testing and, experimentally, transplantable tissue.
• Immune-cell therapies — tumour-infiltrating lymphocytes (TILs), chimeric antigen receptor (CAR) T and NK cells, and engineered macrophages (CAR-M) in oncology.
• Cell derivatives (cell-free) — extracellular vesicles and exosomes, conditioned-medium secretome, and platelet-rich plasma (PRP).
• Engineered tissues and bioprinted constructs — cells combined with scaffolds, hydrogels or bioinks.

Stem Cell Biology

Mesenchymal stromal cells dominate surgical regenerative research and deserve emphasis. By ISCT criteria they are plastic-adherent, express CD73/CD90/CD105, lack haematopoietic markers (CD34/CD45/CD14) and differentiate in vitro into bone, cartilage and fat. Their therapeutic effect, however, is now largely attributed not to engraftment but to trophic and immunomodulatory secretion: MSCs sense inflammatory cues (IFN-γ, TNF-α) and respond by secreting anti-inflammatory, pro-regenerative mediators (IL-10, TGF-β, PGE2, IDO, HGF, VEGF and extracellular vesicles). They shift macrophages from the pro-inflammatory (M1) to the reparative (M2) phenotype, suppress effector T cells and induce regulatory T cells — an ideal profile for an inflamed, fibrotic or ischaemic surgical bed.

Source matters. Adipose-derived cells are abundant, easily harvested by liposuction and strongly pro-angiogenic; bone marrow cells are the best characterised but the harvest is invasive and potency declines with age; umbilical cord cells are young, potent and suited to allogeneic banking. Manufacturing conditions — passage number, oxygen tension, priming — substantially alter potency and are a principal source of the field's reproducibility problems.

Pluripotent cells are rarely given directly; instead they are differentiated into the required cell type. The salient surgical example is stem-cell-derived pancreatic islets, where directed differentiation now yields glucose-responsive β-cells at scale. The central safety task is to ensure complete differentiation and remove residual pluripotent cells to avoid teratoma.

Organoids — intestinal, hepatic, pancreatic and others — reproduce tissue architecture and function from the patient's own cells. Their near-term surgical value lies in precision oncology (testing chemotherapy sensitivity on a patient's tumour organoid before or after resection) and in disease modelling; transplantable organoid therapy is limited principally by the absence of an intrinsic blood supply and remains experimental.

Mechanisms of Tissue Regeneration

Regeneration after surgical injury proceeds through overlapping phases — haemostasis, inflammation, proliferation and remodelling — each governed by conserved signalling pathways that cell therapies aim to modulate.

Angiogenesis

Angiogenesis is often rate-limiting: a graft or wound bed must be revascularised to survive. VEGF (via VEGFR2/PI3K–AKT and MAPK) drives endothelial sprouting; HIF-1α couples the hypoxic stimulus to a pro-angiogenic response; PDGF recruits pericytes to stabilise vessels; FGF-2 promotes endothelial proliferation and matrix remodelling. Pro-angiogenic secretion is the principal mechanism by which ASCs and MSCs help ischaemic wounds and flaps.

Proliferation, growth control and survival

Wnt/β-catenin signalling sustains stem and progenitor proliferation in skin, gut and liver and is central to intestinal crypt and hepatic regeneration. Notch governs cell-fate decisions and lateral inhibition and helps maintain the stem cell niche. Hedgehog patterns tissue and influences fibroblast and epithelial behaviour. PI3K–AKT–mTOR integrates growth factor input with cell survival, size and metabolism, and is a key pro-survival axis for cells delivered into hostile beds. Hippo–YAP/TAZ translates mechanical and architectural cues (matrix stiffness, cell density) into proliferative or quiescent states — relevant both to how scaffold mechanics determine graft behaviour and to liver-size regulation after resection.

Inflammation and fibrosis — a double-edged sword

NF-κB is the master pro-inflammatory switch; the MSC secretome tends to suppress it. TGF-β is pivotal and bidirectional: it is required for wound closure and matrix deposition, yet excess TGF-β1 drives fibroblast-to-myofibroblast conversion, pathological collagen deposition and the formation of scar, stricture and adhesions. BMPs direct osteogenesis and epithelial differentiation. A recurring therapeutic goal is to rebalance healing away from fibrosis and towards regeneration — shifting the TGF-β1/TGF-β3 ratio, promoting M2 macrophages, and remodelling rather than merely accumulating extracellular matrix through matrix metalloproteinases.

Extracellular matrix remodelling links all the phases: provisional fibrin gives way to collagen III, which cross-links and reorganises over months into collagen I. Cellular homing supports endogenous repair: the CXCR4/SDF-1α axis draws endogenous repair cells to the site of injury. Scarless fetal healing — low inflammation, high TGF-β3, rapid and orderly matrix turnover — is the biological north star for anti-fibrotic cell strategies.

Applications in General Surgery

The clinical maturity of cell therapy varies widely across surgical domains. The sections below place each domain along a spectrum from approved and established (burns, TILs in surgical oncology, islet replacement) through early-phase clinical (chronic wounds, fistula) to preclinical (anastomosis, hepatic and pancreatic regeneration, bioprinted organs).

Wound healing

Wound healing is the most translationally advanced of the non-oncological applications. Bioengineered skin substitutes — living cell-containing constructs such as cultured epidermal autografts, bilayered keratinocyte-fibroblast products (e.g. Apligraf) and fibroblast-seeded matrices (e.g. Dermagraft) — have long been approved for chronic wounds, particularly venous leg ulcers and diabetic foot ulcers. These are genuine, if incremental, cell-based standard-of-care options.

Platelet-rich plasma is an autologous concentrate delivering a bolus of platelet growth factors (PDGF, TGF-β, VEGF, EGF). Several meta-analyses of randomised trials show that PRP increases both the likelihood and the speed of diabetic foot ulcer healing compared with standard care — one pooled analysis (10 randomised trials, 550 patients) reported a healing-rate risk ratio of about 1.38 (95% CI 1.05–1.82) — but the evidence base is heterogeneous and methodologically limited, so effect sizes should be read cautiously. PRP is best regarded as a low-risk, modestly beneficial adjunct rather than a definitive treatment.

MSC and ASC therapy in chronic and ischaemic wounds acts predominantly through pro-angiogenic and immunomodulatory paracrine effects; early-phase trials in diabetic foot ulcer and critical limb ischaemia are encouraging but not definitive. Autologous fat grafting — the delivery of ASCs within adipose tissue — is used clinically to improve scar quality and soft-tissue contour and to treat radiation-damaged tissue.

Surgical oncology

Surgical oncology hosts the most striking recent success of cell therapy. Tumour-infiltrating lymphocyte therapy takes T cells from the patient's resected tumour, expands them ex vivo with IL-2 and reinfuses them after lymphodepletion. In February 2024 the FDA granted accelerated approval to lifileucel (Amtagvi) for advanced melanoma refractory to anti-PD-1 (and, if BRAF-mutant, to BRAF/MEK inhibitors) — the first cell therapy approved for a solid tumour, based on the phase II C-144-01 trial with an objective response rate of about 31%. Health Canada followed in 2025.

Because the therapeutic starting material is a surgical specimen, TIL therapy functionally couples oncological surgery to cell manufacturing: the quality of the surgeon's excision directly affects the viability of the product. TIL therapy carries a boxed warning (treatment-related mortality, severe cytopenia, infection, cardiopulmonary and renal impairment) and requires specialised infrastructure.

CAR-T and CAR-NK cells, transformative in haematological malignancy, have so far translated poorly to the solid tumours general surgeons resect (colorectal, gastric, pancreatic, hepatobiliary). Barriers include the scarcity of tumour-specific surface antigens, an immunosuppressive tumour microenvironment, poor cell trafficking and persistence, and antigen heterogeneity. Numerous trials targeting CEA, mesothelin, GPC3 and Claudin18.2 are ongoing; results are early. CAR-macrophages and armoured or logic-gated constructs are being developed specifically to penetrate solid-tumour biology. One clear and powerful surgical use of cell technology today is patient-derived tumour organoids in ex vivo chemosensitivity testing, to personalise adjuvant therapy after resection.

Hernia repair and abdominal surgery

In hernia and abdominal wall reconstruction cell therapy is largely preclinical or early-stage. The rationale is to improve mesh and biological scaffold integration and repair durability by seeding them with MSCs or ASCs — encouraging vascularised, well-organised tissue ingrowth instead of chronic inflammation and fibrotic encapsulation, and thereby reducing recurrence and mesh-related complications. Decellularised extracellular matrix scaffolds (biological meshes) are in clinical use; cell-seeded and stem-cell-enhanced meshes remain an experimental field supported mostly by animal data showing increased neovascularisation and collagen maturation. Cellular therapies are also being investigated for the prevention of intra-abdominal adhesions by modulating mesothelial cell transitions and reducing fibrin deposition.

Colorectal surgery: anastomotic healing and fistula

Two axes dominate. First, perianal fistula in Crohn's disease has produced the field's cautionary tale. Darvadstrocel (Alofisel) — locally injected expanded allogeneic adipose-derived MSCs — was approved in the EU and Japan in 2018 after the ADMIRE-CD trial showed superior combined remission. However, the confirmatory phase III ADMIRE-CD II trial failed its primary endpoint, and in December 2024 Takeda voluntarily withdrew the EU marketing authorisation. Observational data remain broadly favourable and the product's biology is sound, yet this reversal underlines that a positive pivotal trial and real-world use do not guarantee durable, reproducible efficacy.

Second, anastomotic leak — the most feared complication of gastrointestinal surgery, roughly 8% in colorectal and higher in oesophago- and pancreatico-jejunal anastomoses — has attracted extensive preclinical cell-therapy work. In rodent models, peri-anastomotic MSCs (and MSC plus PRP, and MSC-derived extracellular vesicles) increase collagen deposition and hydroxyproline, reduce inflammation, and improve anastomotic bursting pressure and histological healing scores. Critically, these models mostly measure histological surrogate endpoints in low-leak settings rather than clinically meaningful leak prevention, and no robust human randomised trial has yet demonstrated efficacy. Delivery is a central problem: the cells must survive and act in a region that is often ischaemic, contaminated and mechanically stressed.

Hepatobiliary surgery

The liver's remarkable natural regenerative capacity makes it both an obvious target and a high bar for cell therapy. Hepatocyte transplantation — infusion of isolated hepatocytes, usually via the portal vein — has been used as a bridge in acute liver failure and experimentally in metabolic liver disease, but is limited by cell source, engraftment and persistence. MSC therapy is being studied to modulate liver regeneration after hepatectomy, reduce ischaemia–reperfusion injury and treat hepatic fibrosis and cirrhosis, with the aim of reducing the risk of post-hepatectomy liver failure after major resection; human evidence is early-phase and mixed. iPSC- and organoid-derived hepatocytes and bioengineered liver tissue are advancing preclinically as future sources for both transplantation and drug testing. A pragmatic near-term contribution is improving the viability of marginal donor livers through cell- or secretome-based conditioning during machine perfusion.

Pancreatic surgery

Beyond oncological resection, the pancreas is the stage for one of cell therapy's clearest gains: β-cell replacement. Of specific relevance to pancreatic surgery is islet auto-transplantation after total pancreatectomy for chronic pancreatitis, which reinfuses the patient's own islets to preserve endogenous insulin production — a long-standing cell therapy embedded in surgical practice. Research directions include MSC co-transplantation to improve islet engraftment and survival, and stem-cell-derived islets to overcome the shortage of donor tissue. Postoperative pancreatic fistula is the target of the same anti-inflammatory and pro-healing strategies studied for anastomoses, and remains preclinical.

Breast surgery

In breast reconstruction and oncoplastic surgery, autologous fat grafting — largely characterised as ASC delivery — is used to correct contour deformities and improve the quality of irradiated tissue; cell-assisted lipotransfer, which enriches grafts with the stromal vascular fraction, aims to increase graft retention. Evidence for improved volume retention is mixed, and an ongoing, unresolved concern is the theoretical oncological safety question: could ASC paracrine factors influence dormant tumour cells or local recurrence? Current clinical data show no clear increase in recurrence; nevertheless caution and careful patient selection are required, and this remains an area of active surveillance.

Endocrine surgery

The principal development is β-cell replacement for type 1 diabetes. In June 2023 the FDA approved donislecel (Lantidra) — the first allogeneic pancreatic islet-cell therapy, prepared from cadaveric donor pancreata and infused into the hepatic portal vein — for adults with type 1 diabetes and severe, intractable hypoglycaemia. The approval rests on two small single-arm studies totalling 30 patients; some achieved sustained insulin independence, but the treatment requires lifelong immunosuppression and is constrained by donor cell supply and quality.

The transformative frontier is stem-cell-derived islets, which decouple the therapy from cadaveric donors. Vertex's zimislecel (VX-880) — fully differentiated stem-cell-derived islet cells — was reported in the New England Journal of Medicine in June 2025: in the phase 1/2 portion of the FORWARD trial, 10 of 12 patients were insulin-independent at one year, all reached target HbA1c and target time-in-range, and there were no treatment-related serious adverse events; a pivotal phase 3 is enrolling, with global regulatory submissions expected in 2026. Encapsulation approaches aim to eliminate immunosuppression altogether. Cell therapy in thyroid and parathyroid surgery is minimal; however, parathyroid auto-transplantation during thyroidectomy is a classic, established autologous tissue-transfer technique.

Burn surgery

Burns, along with oncology and islets, are where cell therapy has genuinely arrived. Cultured epidermal autografts (Epicel) — sheets of the patient's expanded keratinocytes — hold an FDA Humanitarian Device Exemption for providing permanent coverage of deep dermal and full-thickness burns involving at least 30% of total body surface area when donor skin is scarce; a recognised caveat is fragility and reports of squamous cell carcinoma in burn scars. The RECELL system delivers a non-cultured autologous skin-cell suspension (keratinocytes, fibroblasts and melanocytes) prepared at the bedside from a small skin sample, dramatically reducing the donor-site requirement. Initially approved for thermal burns in 2018, RECELL's indications expanded to full-thickness skin defects in 2023 and to stable vitiligo repigmentation, followed by approval of a simplified version. StratGraft, composed of allogeneic keratinocytes, received FDA approval for deep partial-thickness burns in 2021. These are true point-of-care cellular therapies and an integral part of modern burn surgery. Bioprinted skin and MSC-enhanced dermal regeneration are advancing but remain investigational.

Trauma surgery

In trauma and emergency surgery, cell therapy is mostly aspirational, constrained by the need for immediate, off-the-shelf intervention: autologous manufacture is far too slow for acute care. Research interest centres on MSC infusion to modulate the systemic inflammatory response after severe trauma, haemorrhagic shock and traumatic brain injury — exploiting immunomodulation that blunts cytokine storm and organ injury. Early-phase trials exist, particularly in traumatic brain injury and ARDS, but efficacy in trauma is unproven. Allogeneic, cryopreserved, banked MSC or extracellular vesicle products are the plausible route to acute-care feasibility.

Transplant surgery

Transplant surgery intersects with cell therapy along three axes:

1. Immune tolerance induction — infusing donor haematopoietic cells or regulatory cells (Tregs, regulatory MSCs, tolerogenic dendritic cells) to reduce or eliminate the need for lifelong immunosuppression. An active and promising area of clinical trials.
2. Organ reconditioning — using MSCs, their secretome or extracellular vesicles during ex vivo machine perfusion to repair marginal donor organs and reduce ischaemia–reperfusion injury before implantation.
3. Manufacturing organs — addressing donor scarcity through decellularisation–recellularisation, three-dimensional bioprinting and xenotransplantation with gene-edited pig organs (such as the gene-edited pig kidney and heart transplants performed between 2022 and 2025). The last of these is a cell-engineering approach rather than a cell therapy, but it belongs to the same regenerative ambition.

Biomaterials and Tissue Engineering

Cells rarely act alone; where they are placed and what holds them there determines the outcome. Biomaterials function as scaffold (three-dimensional architecture and mechanical support), delivery vehicle (retaining and protecting cells at the target) and instructive matrix (presenting signals that direct cell behaviour).

Natural polymers (collagen, fibrin, hyaluronic acid, alginate, decellularised extracellular matrix) offer excellent biocompatibility and native cell-binding cues but variable mechanics and batch consistency. Synthetic polymers (PLGA, PCL, PEG) provide tunable, reproducible mechanics and degradation but lack intrinsic bioactivity and can trigger inflammation as they degrade. Decellularised extracellular matrix — tissue stripped of cells but retaining native architecture and biochemistry — is a powerful scaffold in clinical use. Hydrogels are particularly important for cell delivery: injectable, in-situ gelling systems can carry cells or extracellular vesicles to a surgical site (a fistula tract, an anastomosis, a wound bed) and shield them from mechanical and immune stress.

Three-dimensional bioprinting deposits bioinks (cells plus hydrogel) layer by layer to rebuild patient-specific geometry from imaging data, using extrusion, jet or light-based methods. The field has produced impressive constructs — including, in 2024, perfusable vascular networks with distinct endothelial and smooth muscle layers embedded in cardiac tissue — but vascularisation at scale remains the central unsolved obstacle: without a perfusable microvascular network, thick constructs cannot survive. Reviews estimate that fewer than 5% of laboratory bioprinting innovations reach human studies. Four-dimensional bioprinting (constructs that change shape or function over time in response to stimuli) and AI-assisted bioink and print-path optimisation are emerging developments.

Clinical Trials

The trial landscape is dense but uneven: hundreds of registered studies cluster in wound healing, fistula, oncology and metabolic/endocrine applications, while far fewer rigorous studies exist in anastomotic healing, hepatobiliary surgery or hernia surgery. A recurring methodological weakness across the field is reliance on small, single-arm or surrogate-endpoint studies; adequately powered, blinded, placebo- or sham-controlled randomised trials with clinically meaningful endpoints remain the exception.

Currently Approved Therapies

The set of cell and cell-based products with regulatory approval relevant to general surgery is small but real, and clusters in three areas: skin and burns, surgical oncology, and β-cell replacement, with cartilage repair adjacent. Notably, the field's first approved allogeneic MSC product has now been withdrawn in the EU, leaving that jurisdiction with no approved MSC product for a core general-surgical indication — which underscores the fragility of early approvals.

Experimental Therapies

The experimental frontier is broad. It includes iPSC-derived cell products approaching or entering trials (islets, hepatocytes, cardiomyocytes); organoid transplantation; extracellular vesicle and exosome therapeutics — no product is FDA-approved, and the agency actively polices an unapproved grey market; gene-edited cells (HLA-modified universal donor MSCs and islets, CRISPR-enhanced immune cells, hypoimmune cell lines); CAR-T/NK/M for solid tumours; bioprinted tissues; and whole-organ engineering with xenotransplantation. Most are supported by preclinical data or early-phase, single-arm studies.

Complications, Safety and Limitations

Safety is the field's central discipline. The principal hazards are as follows.

Tumourigenicity

This is the primary concern for pluripotent-cell-derived products: residual undifferentiated ESCs or iPSCs can form teratomas, and reprogramming and expansion can introduce genomic instability or oncogenic mutations. Mitigation rests on rigorous purification, suicide-gene safety switches and extended follow-up. For MSCs the direct tumourigenic risk is low, but a theoretical concern persists: their pro-angiogenic, immunomodulatory secretome could support existing tumours — this is the crux of the fat-grafting oncological safety debate in breast surgery. For that reason, administering MSCs to patients with active malignancy remains a hotly contested contraindication.

Immunogenicity and rejection

Allogeneic products carry a risk of rejection and sensitisation; islet and organ-cell therapies typically require lifelong immunosuppression, itself a source of infection and malignancy risk. This drives the move towards encapsulation and hypoimmune engineering.

Immune effector toxicities

Adoptive immune therapies (TIL, CAR-T) carry cytokine release syndrome, neurotoxicity, prolonged cytopenias and infection; TIL regimens add the toxicity of high-dose IL-2 and lymphodepletion.

Delivery-related and procedural risks

Intraportal islet infusion carries risks of bleeding and portal thrombosis; intravenous MSC infusion carries pulmonary first-pass entrapment and a rare thromboembolic risk (the instant blood-mediated inflammatory reaction). Local delivery into a hostile surgical bed often results in poor engraftment and survival — frequently the true limiting factor rather than the intrinsic potency of the cells.

Contamination and manufacturing failure

Cell products are biologically complex and hard to standardise; unapproved products have caused serious infections. Assurance of potency, identity and sterility is far from trivial.

Overarching limitations: heterogeneous and often low-quality evidence; lack of standardisation (cell source, dose, timing, potency assays); manufacturing cost and logistics; delivery and engraftment failure; uncertain long-term safety; and translational attrition from animal models that frequently overstate benefit.

Ethics

Ethical considerations span the entire pipeline. Embryonic stem cells raise moral-status concerns that iPSCs partly circumvent — although iPSCs create their own questions about consent for cell-line derivation and possible germline applications. Unproven commercialisation — clinics marketing unapproved stem-cell and exosome "treatments" — exploits desperate patients and is a significant ethical and regulatory problem. Equity and access are a major issue: bespoke autologous and bioprinted products are expensive and infrastructure-intensive, risking a future in which advanced regenerative care is available only to the wealthy or to patients near specialised centres. Xenotransplantation raises animal-welfare, zoonosis and informed-consent questions. Informed consent in early-phase cell therapy trials must clearly convey uncertainty, irreversibility and the difference between research and treatment. Finally, donor procurement (cadaveric islets, cord blood, gametes) requires robust consent and governance.

Regulatory Frameworks and Health Economics

Cellular therapies are regulated as biological or advanced therapies through pathways designed to balance access against the risks of complex, hard-to-characterise living products.

• United States (FDA/CBER): cell and gene therapies are regulated under Section 351 of the PHS Act and the FD&C Act; expedited pathways include RMAT, Fast Track and Breakthrough designations. The FDA distinguishes minimally manipulated, homologous-use tissue products from those requiring full Biologics License Application approval. It actively enforces against unapproved exosome and stem-cell products — warning letters, injunctions and in some cases criminal referrals — and emphasises that no exosome product is FDA-approved and that therapeutic use requires an authorised investigational new drug application.
• European Union (EMA): Advanced Therapy Medicinal Products are assessed centrally by the Committee for Advanced Therapies, with PRIME for priority medicines. The withdrawal of darvadstrocel in 2024 shows that an authorisation can be reversed.
• United Kingdom (MHRA): post-Brexit routes include the Innovative Licensing and Access Pathway and the Innovation Passport.
• Japan (PMDA): a notably flexible conditional and time-limited approval framework for regenerative medicine, permitting earlier market access subject to subsequent efficacy confirmation — a model that accelerates access but is debated on evidentiary grounds.

Health economics

Cellular therapies are expensive to manufacture and deliver. Bespoke autologous products (TIL, CAR-T, cultured autografts) carry a high per-patient cost and require specialised infrastructure; islet and organ therapies add the cost of lifelong immunosuppression. Cost-effectiveness depends on durability: a one-time treatment producing lasting cure or complication avoidance — functional cure of type 1 diabetes, or a healed limb-threatening ulcer that averts amputation — can be economically justified, whereas modest and transient benefits are far harder to defend. Scalable allogeneic, off-the-shelf and cell-free products are the most plausible route to affordability.

Future Technologies and Emerging Therapies

The pipeline converges on solutions to the field's core problems — potency, safety, immunogenicity, delivery, vascularisation and cost:

• AI-assisted cell therapy design — machine learning for differentiation protocol optimisation, bioink formulation, print-path planning, potency prediction and patient selection.
• CRISPR-edited and hypoimmune cells — deleting HLA and adding immune-cloaking genes to create universal donor islets, MSCs and immune cells that evade rejection without immunosuppression.
• Synthetic biology and programmable cells — logic-gated CAR constructs, sense-and-respond "smart" cells, and synthetic gene circuits for controllable therapeutic output.
• Smart biomaterials and four-dimensional bioprinting — stimulus-responsive scaffolds and constructs that mature or remodel after implantation.
• Digital twins — patient-specific computational models to simulate and personalise regenerative interventions.
• Organoid transplantation and whole-organ bioengineering — from decellularisation–recellularisation to fully bioprinted, vascularised organs.
• Xenotransplantation with multi-gene-edited pigs — an increasingly serious near-term bridge for organ scarcity.
• Nanotechnology-assisted delivery and in vivo reprogramming — targeted nanoparticles and in-situ conversion of the patient's own cells.
• Bioartificial organs — hybrid device-plus-cell systems such as encapsulated islets and bioartificial liver support.

Outlook

• 5 years: consolidation and expansion of what already works — likely approval and refinement of stem-cell-derived islets; more TIL and adoptive-cell indications in oncology; improved skin and wound products; the first well-controlled human studies in anastomotic and hepatobiliary applications; and a growing but still contested vesicle evidence base under tighter regulation. Hypoimmune off-the-shelf cell lines enter the clinic.
• 10 years: encapsulated or hypoimmune islet and cell grafts without immunosuppression; routine tumour-organoid-guided oncology; bioprinted tissue patches in selected clinical use (skin, abdominal wall, vascular grafts); xenografts as an accepted bridge for kidney and heart; maturing vesicle therapeutics.
• 20 years: the plausible — not guaranteed — arrival of bioengineered, vascularised solid organs for transplantation; and broad personalised regenerative surgery integrating AI, gene editing and biofabrication. All of it contingent on solving vascularisation, maturation, cost and durable safety.

Research Gaps

1. The scarcity of adequately powered, blinded, controlled randomised trials with clinically meaningful endpoints — as against small, single-arm, surrogate-endpoint studies.
2. Lack of standardisation in cell source, dose, potency assays, timing and delivery.
3. Poor understanding of engraftment and in vivo fate.
4. The translational gap between animal models — particularly low-leak anastomotic models — and human disease.
5. Unresolved long-term safety data (tumourigenicity, immunogenicity).
6. Manufacturing scalability and cost.
7. The vascularisation problem in engineered tissues.
8. Weak health-economic evidence.

Anastomotic healing, hepatobiliary regeneration and hernia repair are particularly underserved in terms of rigorous human evidence.

Expert Appraisal

Read as a whole, the evidence supports a layered view. Cells work, and have earned approval, where they clearly replace a missing function or provide durable coverage — cultured and suspension skin cells for burns, islets for type 1 diabetes and tumour-derived T cells for melanoma all share the logic of replacing or repositioning a defined cell population. Where cells are asked to modulate a complex, multifactorial process — anastomotic healing, fibrosis, systemic inflammation — results are biologically plausible and often favourable in animals but have repeatedly failed to translate into robust human benefit, as the darvadstrocel reversal shows.

Three intersecting lessons stand out:

1. Paracrine biology reframes the field. If cells act mainly by secreting signals, then cell-free products (vesicles, secretome) and in-situ strategies may deliver most of the benefit at lower risk and cost — provided characterisation and potency can be standardised and regulated.
2. The bottleneck is often not cell identity but delivery and engraftment; biomaterials and smart delivery matter as much as the cells.
3. Evidentiary discipline is decisive. The gap between a promising preclinical report and a durable clinical therapy is wide, littered with translational losses, and — in a field that also contains unproven commercial clinics — demands exceptional rigour and honesty.

Conflicting evidence is common: PRP trials with opposing results, or darvadstrocel's favourable observational data against its failed confirmatory randomised trial. These discordances usually rest on heterogeneity in product, dose, patient selection, endpoints and bias — reinforcing the need for standardised, controlled trials.

Clinical Recommendations

A pragmatic, evidence-graded stance for the practising general surgeon:

1. Adopt what is approved and durable. Use cultured and suspension autologous skin cells (Epicel, RECELL) in appropriate burn and full-thickness defect care; refer suitable patients to centres offering TIL therapy (melanoma) and islet or β-cell replacement (selected type 1 diabetes).
2. Use low-risk adjuncts judiciously. PRP and autologous fat grafting are reasonable, low-harm adjuncts for chronic wounds and reconstruction, with explicit counselling that the benefit is modest and the evidence heterogeneous.
3. Keep investigational applications within trials. MSC and vesicle therapy for anastomotic healing, hepatobiliary regeneration, hernia and trauma should be offered within registered clinical trials, not as off-label practice.
4. Counsel against unproven commercial stem-cell and exosome clinics. These are unapproved, potentially harmful and unsupported by evidence; direct patients to regulated trials instead.
5. Individualise and follow up. Weigh the unresolved oncological safety question for breast fat grafting; ensure informed consent for any cell therapy conveys the uncertainty; and follow patients for long-term safety.
6. Integrate cell technology now as a diagnostic tool. Patient-derived tumour organoids for chemosensitivity testing are a near-term, low-risk route by which cell biology can improve surgical oncology decisions.

Future Perspectives

The trajectory points towards personalised, off-the-shelf and increasingly cell-free regenerative surgery. The most transformative near-term possibility is functional cure of type 1 diabetes through stem-cell-derived, immune-protected islets — a genuine paradigm shift already visible in trial data. On a longer horizon, solving vascularisation opens the way to bioprinted tissues and ultimately organs, easing the transplant scarcity that constrains much of surgery. Gene editing (hypoimmune cells) and AI-driven design are the enabling technologies most likely to accelerate progress. Realistically, the coming decade will bring incremental, domain-specific gains rather than a wholesale reinvention of surgery — but those gains (burns, oncology, β-cells, tolerance induction, tissue patches) are already changing patients' lives.

Conclusion

Cellular therapy has moved from promise to practice in a narrow but real set of general-surgical applications: burn and skin reconstruction, one solid-tumour immunotherapy, and β-cell replacement — the last of these poised for dramatic expansion through stem-cell-derived islets. Across the much wider terrain of anastomotic healing, fibrosis modulation, hepatobiliary and pancreatic regeneration, hernia, trauma and whole-organ engineering, the science is biologically compelling and the preclinical literature vast, yet rigorous human evidence remains thin; the recent withdrawal of the first approved allogeneic MSC product is a reminder that promise must be demonstrated, not assumed.

The field's future will be won not by discovering ever more remarkable cell behaviours in the laboratory, but by standardising them, delivering them, proving them and paying for them — and by preserving the evidentiary and ethical discipline that separates durable therapy from hopeful hype.

Evidence Quality Summary

References

1. European Medicines Agency. (2024, 13 December). Alofisel (darvadstrocel): Withdrawal of EU marketing authorisation. https://www.ema.europa.eu/en/medicines/human/EPAR/alofisel
2. Takeda. (2024). Update on Alofisel (darvadstrocel) — voluntary EU withdrawal. https://www.takeda.com/newsroom/statements/2024/takeda-alofisel-update-2024/
3. Taxonera, C., et al. (2025). Darvadstrocel for complex perianal fistulas in Crohn's disease: A systematic review and meta-analysis. United European Gastroenterology Journal. https://doi.org/10.1002/ueg2.12673
4. Burke, J. P., et al. (2021). Mesenchymal stromal cells in animal models for gastrointestinal anastomotic leak: A systematic review. Colorectal Disease, 23, 3123–3140. https://doi.org/10.1111/codi.15864
5. U.S. Food and Drug Administration. (2023, 28 June). FDA approves donislecel (Lantidra), the first allogeneic islet cell therapy for type 1 diabetes. Prescribing information: https://www.fda.gov/media/169920/download
6. Vertex Pharmaceuticals. (2025, June). Positive zimislecel (VX-880) data — ADA 2025; concurrent New England Journal of Medicine publication. https://news.vrtx.com/news-releases/news-release-details/vertex-presents-positive-data-zimislecel-type-1-diabetes
7. AVITA Medical / U.S. FDA. (2023). FDA approval of RECELL for full-thickness skin defects. FDA summary: https://www.fda.gov/media/175574/download
8. U.S. Food and Drug Administration. Epicel (cultured epidermal autografts), HDE H990002 — Summary of safety and probable benefit. https://www.accessdata.fda.gov/cdrh_docs/pdf/H990002b.pdf
9. American Association for Cancer Research / U.S. FDA. (2024, 16 February). Lifileucel (Amtagvi) — first tumour-infiltrating lymphocyte therapy approved for a solid tumour (melanoma). https://www.aacr.org/patients-caregivers/progress-against-cancer/fda-approves-first-tumor-infiltrating-lymphocyte-therapy/
10. Deng, J., et al. (2023). Efficacy and safety of autologous platelet-rich plasma for diabetic foot ulcer healing: A systematic review and meta-analysis of randomised controlled trials. Journal of Orthopaedic Surgery and Research, 18, 370. https://doi.org/10.1186/s13018-023-03854-x
11. Del Pino-Sedeño, T., et al. (2019). Platelet-rich plasma for the treatment of diabetic foot ulcers: A meta-analysis. Wound Repair and Regeneration. https://doi.org/10.1111/wrr.12690
12. Regulatory status of exosome products (2023–2026): FDA warning letters and enforcement summaries — no exosome product is FDA-approved; ISEV MISEV 2023 guidelines.
13. Three-dimensional bioprinting for transplantation — status reviews (2025); United Therapeutics UKidney first-in-human (NYU Langone, 2025).
14. Langer, R., & Vacanti, J. P. (1993). Tissue engineering. Science, 260(5110), 920–926.
15. Takahashi, K., & Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 126(4), 663–676. (Human cells: Cell, 2007, 131(5), 861–872.)
16. Caplan, A. I. (1991). Mesenchymal stem cells. Journal of Orthopaedic Research, 9(5), 641–650.
17. Friedenstein, A. J., et al. (1966). Osteogenesis in transplants of bone marrow cells. Journal of Embryology and Experimental Morphology, 16(3), 381–390.
18. Dominici, M., et al. (2006). Minimal criteria for defining multipotent mesenchymal stromal cells: The International Society for Cellular Therapy position statement. Cytotherapy, 8(4), 315–317.
19. Thomson, J. A., et al. (1998). Embryonic stem cell lines derived from human blastocysts. Science, 282(5391), 1145–1147.
20. Sato, T., et al. (2009). Single Lgr5 stem cells build crypt–villus structures in vitro without a mesenchymal niche. Nature, 459(7244), 262–265.
21. Rheinwald, J. G., & Green, H. (1975). Serial cultivation of strains of human epidermal keratinocytes. Cell, 6(3), 331–343.