aiwiki.page
English
Health / Medicine & Treatment / adoptive-cell-therapy

Adoptive Cell Therapy

Adoptive cell therapy transfers selected or engineered immune cells into patients to treat cancer, infections, and other disorders.

39 keywords4 linked from7 not yet writtenWritten by AI
Cancer Immunothe…CAR T-Cell Thera…Hematopoietic St…Graft-versus-Leu…Graft-versus-Hos…VaccineT CellCytokineAdoptive C…

Adoptive cell therapy (ACT) is the administration of immune cells selected, expanded, activated, or genetically modified outside the body to produce a therapeutic effect after transfer into a patient. Its principal established application is cancer immunotherapy, although related approaches are used against certain viral infections and are being investigated for autoimmune and transplant-related disorders. The transferred cells may come from the patient, termed autologous therapy, or from another person, termed allogeneic therapy.

Unlike a conventional drug with a fixed administered dose and relatively predictable elimination, an adoptive cellular product can proliferate, migrate between tissues, change its functional state, and persist for years. These properties can produce durable disease control, but they also create distinctive manufacturing requirements and potentially serious toxicities. ACT includes tumor-infiltrating lymphocyte therapy, genetically engineered T-cell receptor therapy, chimeric antigen receptor T-cell therapy, and several forms of natural killer-cell and virus-specific T-cell therapy.

The field brings together immunology, genetic engineering, transplantation, and clinical oncology. Its development has depended on understanding how immune cells recognize targets, learning to manufacture living products reproducibly, and establishing treatment systems capable of managing rapid changes in both disease and immune activity.

Historical development and boundaries of the field

The conceptual basis of ACT is that immunity can be transferred through cells rather than only through soluble molecules such as antibodies. Animal experiments in the mid-twentieth century demonstrated that transferred lymphocytes could convey antigen-specific immune responses. These findings helped establish the importance of cellular immunity and provided an experimental basis for attempting to transfer antitumor activity.

Clinical transplantation supplied an important parallel line of evidence. After allogeneic hematopoietic stem-cell transplantation, donor immune cells can attack residual malignant cells, producing a graft-versus-leukemia effect. Donor lymphocyte infusions subsequently made this principle more explicit: lymphocytes collected from a transplant donor could be administered separately to enhance antileukemic activity. However, the same cells could also attack normal recipient tissues and cause graft-versus-host disease. These observations demonstrated both the therapeutic power and the selectivity problem of transferred immunity.

In the 1980s, Steven Rosenberg and colleagues at the US National Cancer Institute developed treatments using lymphokine-activated killer cells and interleukin-2. Such cells were generated by exposing peripheral blood lymphocytes to activating conditions, but their antitumor specificity was limited. A more targeted strategy used lymphocytes recovered from tumors. Reports in 1988 described objective regressions in metastatic melanoma after infusion of expanded tumor-infiltrating lymphocytes with interleukin-2. Subsequent work showed that preparative lymphodepletion could improve the conditions for transferred cells to expand and function.

Genetic engineering created another route to specificity. Early chimeric receptor studies in the late 1980s and early 1990s joined antibody-derived recognition domains to intracellular T-cell signaling machinery. Later receptor designs incorporated costimulatory domains, supporting stronger proliferation and persistence. Clinical studies in the early 2010s demonstrated substantial activity against CD19-positive leukemias and lymphomas. In 2017, tisagenlecleucel became the first CAR T-cell product approved by the US Food and Drug Administration.

ACT is narrower than cell therapy as a whole. Transfusion of red blood cells, replacement of pancreatic islets, and regenerative administration of structural cells are not ordinarily classified as adoptive immunotherapy. Stem-cell transplantation is closely related but usually aims initially to establish or restore blood formation. ACT more specifically emphasizes the therapeutic activity of transferred immune effector or regulatory cells. It also differs from vaccination, which primarily seeks to induce an immune response within the recipient, and from checkpoint blockade, which modifies the activity of immune cells already present.

Immune recognition and the behavior of transferred cells

Most established ACT approaches use T cells, whose natural receptors recognize short protein fragments presented on cell surfaces. These fragments, or peptides, are displayed by molecules encoded within the human leukocyte antigen system. A conventional T-cell receptor therefore recognizes a combined peptide–HLA structure rather than an intact protein alone. This allows T cells to detect proteins produced inside a cell, including viral proteins and proteins altered by cancer-associated mutations.

Recognition is only the beginning of an effective response. T cells integrate signals from their antigen receptor, costimulatory receptors, inflammatory mediators, and inhibitory pathways. Activated cytotoxic T cells can release perforin and granzymes, which initiate target-cell death, and can engage death-receptor pathways. They also secrete cytokines, including interferon-gamma, that influence neighboring immune and stromal cells. Consequently, the effects of a transferred population extend beyond direct contact with malignant cells.

A chimeric antigen receptor changes the recognition step. Its extracellular domain commonly derives from an antibody and binds an accessible surface antigen without requiring peptide presentation by HLA. A membrane-spanning region connects this recognition domain to intracellular signaling components, usually including CD3-zeta and a costimulatory domain such as CD28 or 4-1BB. CARs can therefore bypass some defects in antigen presentation, but conventional CARs cannot directly recognize most intracellular proteins.

Cellular state is as important as receptor specificity. Less-differentiated T-cell populations can retain proliferative capacity and generate long-lived descendants, whereas extensively stimulated cells may have limited persistence. Persistent antigen exposure can induce T-cell exhaustion, a state involving altered transcriptional programs and reduced function. Manufacturing methods attempt to balance the immediate killing capacity of effector cells against the expansion and durability associated with memory-like populations.

Transferred cells must also reach their targets. This requires passage through blood vessels, responses to tissue-derived chemotactic signals, and survival in local metabolic conditions. A cell that efficiently kills a tumor line in laboratory culture may fail in a patient because it cannot enter the tumor or remain functional there.

Natural killer cells use a different recognition architecture. Their behavior reflects the balance between activating and inhibitory receptors, including receptors that detect normal HLA class I molecules. Reduced expression of these molecules can remove inhibitory signals, enabling recognition of “missing self.” This biology provides a basis for using natural killer cells against some tumors and infected cells, while also explaining why results obtained with T-cell platforms do not automatically apply to other immune-cell products.

Principal cellular platforms

Tumor-infiltrating lymphocyte therapy uses immune cells recovered from a surgically removed tumor deposit. The tumor is fragmented or dissociated, lymphocytes are grown under controlled conditions, and the resulting population is expanded before infusion. Some of these cells recognize tumor-associated antigens or mutation-derived neoantigens. Because a product may contain numerous T-cell clones, it can potentially recognize several targets without introducing an artificial receptor. However, not every lymphocyte found in a tumor is tumor-reactive, and successful expansion does not by itself establish useful specificity.

T-cell receptor-engineered T-cell therapy introduces genes encoding a selected antigen-specific receptor into a patient’s T cells. This approach can target intracellular proteins represented by peptide–HLA complexes. Its clinical applicability is constrained by both tumor antigen expression and the patient’s HLA type. Receptors may be selected from naturally occurring T cells or modified to alter their binding properties, but stronger binding does not necessarily mean greater safety: recognition of a similar peptide in healthy tissue can cause severe injury.

CAR T-cell therapy instead targets cell-surface molecules. CD19 has been especially useful in B-cell malignancies because it is widely expressed across many malignant B-cell populations. B-cell maturation antigen, or BCMA, is a major target in multiple myeloma. The acceptable consequences of destroying normal target-bearing cells depend on the tissue involved. Loss of normal B cells can often be supported clinically; analogous destruction of essential heart or lung tissue would be fundamentally different.

Natural killer-cell therapy includes products derived from peripheral blood, umbilical cord blood, established cell lines, or induced pluripotent stem cells. Some are genetically engineered to express CARs. These approaches seek to exploit innate killing mechanisms and facilitate donor-derived production. Their limitations include variable persistence, susceptibility to suppression, and the need to demonstrate reproducible clinical benefit. Natural killer-cell platforms remain less clinically established than approved CAR T-cell treatments.

Other approaches occupy more specialized positions. Virus-specific T-cell therapy enriches cells that recognize viral antigens, particularly after transplantation. Regulatory T-cell therapy aims to suppress rather than intensify immune responses. Donor lymphocyte infusion generally transfers a broader donor immune population without the receptor engineering characteristic of CAR or TCR products.

These categories overlap. A tumor-infiltrating lymphocyte product can undergo additional selection or engineering, and a donor-derived T cell can be modified to reduce alloreactivity while expressing a CAR. Product identity therefore depends on several independent features: cellular lineage, donor source, antigen-recognition mechanism, manufacturing process, and intended biological function.

Collection, manufacture, and treatment delivery

Autologous blood-derived ACT commonly begins with apheresis, which separates and collects circulating white blood cells while returning other blood components to the patient. Tumor-infiltrating lymphocyte therapy instead requires suitable tumor tissue obtained through surgery. Starting material varies substantially between patients because of previous treatment, disease burden, age, and the composition and functional condition of the collected cells.

Manufacturing generally involves cell selection, activation, expansion, and formulation. For engineered T-cell products, receptor genes may be introduced with integrating viral vectors or nonviral systems. The cells are then cultured under conditions designed to generate enough viable, functional cells for treatment. Culture duration, cytokine exposure, activation strength, and the relative proportions of T-cell subsets can all influence the final product. A larger cell yield is not necessarily a more effective product if expansion produces a poorly persistent or dysfunctional population.

Release testing evaluates attributes such as identity, viability, purity, sterility, and biological potency. Engineered products also require controls related to the introduced genetic material and its delivery system. Potency is particularly challenging because no single laboratory assay fully captures the ability of a living cell population to expand, migrate, recognize heterogeneous targets, and function within a patient. For autologous treatment, chain-of-identity controls ensure that each product returns to the correct recipient; chain-of-custody controls document its handling throughout collection, transport, manufacture, and delivery.

The interval between collection and infusion can be clinically significant. Some patients receive bridging treatment to control disease while their cells are manufactured. Such treatment can alter tumor burden and the patient’s condition, complicating interpretation of outcomes. Manufacturing failure, disease progression, infection, or declining organ function may prevent infusion even after collection has occurred. Analyses restricted to infused patients therefore describe a different population from analyses beginning at enrollment or cell collection.

Many protocols use lymphodepleting chemotherapy, often fludarabine and cyclophosphamide, shortly before cell infusion. This reduces competing lymphocytes and changes the availability of homeostatic cytokines that support transferred-cell expansion. It also causes toxicities of its own. Some tumor-infiltrating lymphocyte regimens additionally administer interleukin-2 after infusion to support cell activity.

Autologous production reduces certain donor-recipient compatibility problems but requires an individualized manufacturing pathway. Allogeneic production offers the possibility of preparing multiple doses from one donor source and maintaining ready-to-use inventory. That advantage comes with other challenges, including recipient rejection, donor-cell alloreactivity, and the consequences of additional genetic modifications intended to control those processes.

Established applications in blood cancers

The clearest clinical successes of ACT have occurred in selected hematologic malignancies. These diseases often provide targets that are relatively consistent within a cell lineage and are accessible to circulating immune cells. CD19-directed CAR T cells have produced substantial remissions in B-cell acute lymphoblastic leukemia, large B-cell lymphoma, and several other B-cell malignancies. BCMA-directed products have extended the approach to multiple myeloma.

The ELIANA study of tisagenlecleucel in children and young adults with relapsed or refractory B-cell acute lymphoblastic leukemia reported an overall remission rate of 81% within three months among 75 infused patients. This figure combined complete remission with complete remission accompanied by incomplete blood-count recovery. It demonstrated major activity in a heavily treated population, but it did not mean that every remission was permanent. Relapse could occur through loss of the target antigen, inadequate cellular persistence, or other disease-related mechanisms.

In the initial phase 2 ZUMA-1 analysis, axicabtagene ciloleucel produced an objective response in 82% of 101 treated patients with refractory large B-cell lymphoma, including complete responses in 54%. Subsequent follow-up established that a proportion of patients experienced prolonged disease control. Such results were important because they showed that a single cellular treatment could produce durable remissions in some patients whose disease had resisted conventional therapy.

Randomized studies subsequently tested CAR T cells earlier in treatment. In ZUMA-7, involving 359 patients with early relapsed or refractory large B-cell lymphoma, median event-free survival was 8.3 months with axicabtagene ciloleucel and 2.0 months with a standard-care strategy based on salvage therapy and transplantation in responders. Results have not been identical across all products and trial designs, underscoring that “CAR T-cell therapy” is not one interchangeable intervention.

In multiple myeloma, idecabtagene vicleucel and ciltacabtagene autoleucel established BCMA as a clinically useful target. Their activity also illustrated a recurrent issue: high initial response rates can coexist with later relapse because myeloma remains biologically heterogeneous and capable of immune escape.

Clinical interpretation depends on the endpoint and denominator. Objective response, complete response, measurable residual disease negativity, progression-free survival, and overall survival answer different questions. A randomized controlled trial can estimate comparative benefit more directly than a single-arm study, whereas long-term follow-up is needed to establish durability and delayed harms. Results from one disease, treatment line, or manufacturing platform cannot be assumed to apply to another.

Solid tumors and the expansion of clinical indications

Solid tumors present obstacles that are less prominent in many blood cancers. Malignant cells may be embedded within dense extracellular material, supplied by abnormal blood vessels, and surrounded by suppressive immune and stromal populations. Different regions of the same tumor can express different antigens. An effective cellular product must therefore combine appropriate recognition with tissue access and sustained function.

Tumor-infiltrating lymphocytes have provided some of the strongest evidence in this setting, particularly in melanoma. In a randomized phase 3 study published in 2022, 168 patients with advanced melanoma were assigned to tumor-infiltrating lymphocyte therapy or ipilimumab. Median progression-free survival was 7.2 months with the cellular therapy and 3.1 months with ipilimumab; objective responses occurred in 49% and 21%, respectively. The trial supported the therapeutic relevance of expanded tumor-derived lymphocytes while also documenting the substantial treatment burden associated with the regimen.

In February 2024, the FDA granted accelerated approval to lifileucel for certain adults with unresectable or metastatic melanoma previously treated with a PD-1-blocking antibody and, when relevant, BRAF-directed treatment. The approval marked the entry of a tumor-derived T-cell product into US regulatory practice. It also illustrated the distinction between a product’s activity and the full clinical pathway: tumor procurement, manufacturing time, lymphodepletion, cell infusion, and interleukin-2 support all contribute to feasibility and risk.

Engineered T-cell receptors offer another route into solid tumors. In August 2024, the FDA granted accelerated approval to afamitresgene autoleucel for selected adults with unresectable or metastatic synovial sarcoma after prior chemotherapy. Eligibility depends on specified HLA-A*02 alleles and tumor expression of MAGE-A4. This restriction follows directly from the receptor’s mechanism: the engineered cells recognize a MAGE-A4-derived peptide presented by particular HLA molecules, not MAGE-A4 protein independently of HLA.

CAR approaches to solid tumors face a particularly difficult target-selection problem. Many candidate surface antigens are present at some level on normal tissues. Other targets occur only on a fraction of malignant cells, allowing antigen-negative populations to survive. Investigational strategies include regional delivery into body cavities or the central nervous system, targeting more than one antigen, and engineering cells to respond to local tumor conditions.

Clinical responses in an early trial do not by themselves establish a broadly applicable treatment. Solid-tumor studies often involve small, molecularly selected populations and specialized centers. Confirmatory evidence must distinguish the contribution of the transferred cells from accompanying chemotherapy, cytokine support, and differences in patient selection.

Acute toxicities, delayed effects, and safety monitoring

ACT toxicity reflects the combined effects of the cellular product, preparative therapy, disease burden, and supportive treatments. Some complications arise from vigorous immune activation; others result from destruction of normal cells bearing the target antigen. Infection, organ injury, and prolonged blood-count abnormalities may have multiple overlapping causes. Safety profiles therefore differ substantially between products and treatment regimens.

Cytokine release syndrome is a characteristic complication of several immune effector-cell therapies. Activation of transferred cells and host immune populations can generate systemic inflammatory signaling, including substantial interleukin-6 activity. Clinical manifestations range from fever to low blood pressure, oxygen requirement, and organ dysfunction. The syndrome often develops during the period of early cellular expansion, but its timing and severity vary. Clinical management uses standardized assessment, supportive care, and, when indicated, cytokine-directed treatment such as the interleukin-6 receptor blocker tocilizumab.

Immune effector cell-associated neurotoxicity syndrome, commonly abbreviated ICANS, can involve impaired attention, language disturbance, altered consciousness, seizures, and, rarely, severe cerebral swelling. Its biology involves interactions among inflammatory mediators, vascular endothelium, and the blood–brain barrier. ICANS can accompany cytokine release syndrome or appear after systemic inflammatory symptoms have improved. The two syndromes overlap clinically but are not identical, and suppression of interleukin-6 signaling does not reliably control isolated neurotoxicity.

Hematologic and infectious complications can persist beyond the initial admission. Lymphodepleting treatment and subsequent inflammatory processes may contribute to prolonged cytopenias, including neutropenia. CD19-directed treatment commonly depletes normal B cells, sometimes producing sustained antibody deficiency. BCMA-directed treatment can also substantially impair humoral immunity. These effects help explain vulnerability to infection even when the malignancy is responding.

Other toxicities are product-specific. TCR-engineered cells can recognize unintended peptide targets, while CAR-bearing cells can injure normal tissue expressing the intended surface antigen. Allogeneic T-cell products may cause graft-versus-host disease unless their alloreactive potential is adequately controlled. Some treatment-associated immune activation can develop into an immune effector cell-associated hemophagocytic syndrome, a serious hyperinflammatory complication distinct from uncomplicated early cytokine release syndrome.

Long-term surveillance addresses delayed cytopenias, immune recovery, secondary malignancies, and risks associated with genetic modification. In 2024, US regulators required strengthened warnings concerning T-cell malignancies after approved autologous CD19- and BCMA-directed CAR T-cell therapies. Identifying a malignancy after treatment does not automatically establish that the engineered receptor or vector caused it; previous therapies, underlying disease, and molecular findings must also be considered. Such events nevertheless require systematic reporting and long-term evaluation.

Treatment failure, resistance, and engineering strategies

Failure can occur before a cellular product reaches its target, during the initial response, or after a prolonged remission. Manufacturing failure and clinical deterioration before infusion are different from biological resistance after treatment. Among infused patients, inadequate expansion, short persistence, poor trafficking, target heterogeneity, and suppression within diseased tissue can each limit activity.

Antigen escape is a well-established mechanism. After CD19-directed treatment, some B-cell leukemias relapse with little or no detectable CD19. Underlying mechanisms can include alterations in the target gene, changes in RNA processing, selection of pre-existing antigen-low populations, and, in some leukemias, lineage change. A receptor that remains fully functional cannot eliminate cells that no longer display a recognizable target. BCMA expression and the genetic pathways controlling it can likewise change under treatment pressure.

TCR-based therapies face an additional dependency on antigen processing and presentation. A tumor may retain the source protein but cease to display the relevant peptide–HLA complex. Loss of an HLA allele, defects in presentation machinery, or altered antigen expression can each interrupt recognition. Conversely, increasing receptor affinity to compensate for weak recognition can raise the risk of detecting related peptides in healthy tissues.

The tumor microenvironment can inhibit cells that successfully arrive at the tumor. Low oxygen, restricted nutrients, suppressive metabolites, inhibitory ligands, and regulatory immune populations can impair proliferation and cytotoxicity. Persistent stimulation can reinforce exhaustion-associated programs. These barriers are interconnected: a poorly perfused tumor may simultaneously restrict cell entry, limit nutrients, and favor suppressive cell populations.

Engineering strategies attempt to address particular failure modes. Dual-target and multi-target receptors seek to reduce escape through loss of one antigen. Logic-gated designs aim to make activation depend on combinations of signals, potentially distinguishing malignant from normal tissue more precisely. Other approaches alter chemokine receptors to improve trafficking, introduce cytokine-support circuits, or modify inhibitory signaling. Combining ACT with an immune checkpoint inhibitor is another investigational strategy, although a plausible mechanism does not guarantee improved clinical outcomes.

Safety engineering includes inducible suicide switches, drug-controlled activity, and receptors designed for limited persistence. Allogeneic platforms may disrupt the native T-cell receptor to reduce graft-versus-host activity and alter other genes to influence immune rejection. Each additional modification creates a need to establish genetic integrity, manufacturing consistency, and functional behavior. Improvements in laboratory assays or animal models require clinical confirmation, particularly because changes that strengthen antitumor activity may also increase inflammation or normal-tissue injury.

Applications beyond cancer

Adoptive transfer can restore immune functions that are absent or insufficient rather than attack a malignancy. Virus-specific T-cell therapy is particularly relevant after hematopoietic stem-cell or solid-organ transplantation, when immunosuppressive treatment can permit viral reactivation. Products may recognize one virus or several, including cytomegalovirus, Epstein–Barr virus, adenovirus, and BK polyomavirus.

These cells can be obtained from the original transplant donor or from a bank of partially HLA-matched third-party donors. Their function still depends on suitable antigen recognition and presentation, so donor selection is not arbitrary. The goal is to supply antiviral activity without introducing excessive alloreactivity. Clinical interpretation is complicated by concurrent antiviral drugs, changes in immunosuppression, and spontaneous or transplant-related immune recovery.

Epstein–Barr virus-associated post-transplant lymphoproliferative disease provides an example at the boundary between antiviral immunity and cancer treatment. In this disorder, EBV-infected B cells can proliferate under conditions of impaired immune surveillance. EBV-specific T cells target viral antigens expressed by the abnormal population. Tabelecleucel, an allogeneic EBV-specific T-cell product, received authorization in the European Union in 2022 for specified patients with relapsed or refractory EBV-positive post-transplant lymphoproliferative disease after prior treatment.

ACT is also being investigated in autoimmune disease. CD19-directed CAR T cells can profoundly deplete B cells, including populations that may be difficult to eliminate with antibody-based treatment. Small clinical series involving systemic lupus erythematosus and other severe autoimmune disorders have reported substantial clinical improvement after treatment. These findings establish an important research direction, but limited cohort size, patient selection, and incomplete long-term follow-up restrict conclusions about durability, comparative benefit, and uncommon harms.

Regulatory T-cell approaches pursue the opposite functional goal: increasing antigen-specific or tissue-directed immune suppression. Such cells are being studied in transplantation, type 1 diabetes, and other immune-mediated disorders. Key questions include whether the cells retain a stable regulatory state, reach the relevant tissue, and suppress the unwanted response without broadly compromising protective immunity.

These uses demonstrate why ACT cannot be equated simply with stronger immunity. Depending on the product, the intended effect may be cytotoxic elimination, restoration of pathogen control, depletion of a disease-driving immune population, or reinforcement of immune tolerance. The relevant measures of success therefore differ from oncology response rates and may include infection control, graft function, withdrawal of immunosuppressive medication, or preservation of tissue-specific function.

Evidence, regulation, and access

A cellular therapy is defined not only by its receptor or cell type but also by its manufacturing process. Changes in collection, culture conditions, genetic modification, cryopreservation, or release testing can affect its biological properties. Regulatory evaluation therefore links clinical evidence to a particular product and production system. Evidence for one CAR construct or one tumor-infiltrating lymphocyte process does not automatically validate another.

Early studies frequently enroll patients with few remaining treatment options and use single-arm designs. Large and durable responses can support regulatory decisions in such settings, including accelerated approval, but uncertainty remains about comparative benefit and long-term outcomes. Confirmatory studies may be required. The distinction between an approved indication and an investigational use is especially important because efficacy and toxicity can change with disease type, tumor burden, prior therapy, and patient fitness.

Study design must account for the interval between enrollment and infusion. Reporting only patients who receive cells can omit those who progress, die, or experience manufacturing failure beforehand. An intention-to-treat analysis preserves the consequences of assignment in randomized comparisons, including practical difficulties in delivering the treatment. Real-world registries complement trials by documenting outcomes in broader populations, but observational comparisons remain vulnerable to differences in eligibility, referral, supportive care, and disease severity.

Access depends on more than the availability of a manufactured dose. Treatment requires coordinated referral, collection facilities, specialist laboratories, transport systems, trained clinical teams, and rapid access to intensive support. Geographic distance, caregiver requirements, time away from employment, and insurance or reimbursement arrangements can all affect who receives treatment. These factors create health equity concerns even where a product has regulatory approval.

Economic evaluation must consider collection, manufacturing, bridging treatment, conditioning, hospitalization, toxicity management, and subsequent care. The relationship between high initial expenditure and long-term value depends heavily on response durability, avoided treatments, survival, and quality of life. Estimates expressed as cost per quality-adjusted life-year are sensitive to assumptions about how long benefits persist, particularly when trial follow-up is short.

Patient-derived products also raise practical issues concerning consent, tissue use, genetic information, long-term follow-up, and the handling of unused material. Donor-derived products add questions about donor screening and the use of one donation to manufacture treatments for multiple recipients. International differences in regulation and reimbursement shape how these products move from specialized trials into routine care, while long-term registries remain important for tracking uncommon toxicities and outcomes that emerge years after infusion.