Translation of cell therapies to treat autoimmune disorders.

A variety of cell-based therapies have been developed to target dysregulated immunity in AD and restore tolerance toward self-antigens. Hematopoietic stem cell transplantation (HSCT) was first explored by Medawar in the 1950s with his seminal experiments exploring the induction of tolerance, for which he ultimately received the Nobel Prize in Medicine [ 213 ]. In 1955, the first experiments exploring the transplantation of allogeneic bone marrow into lethally irradiated mice followed by skin grafts demonstrated the first induction of immunological chimerism. From here, the concept of allogeneic and autologous HSCT arose, and the first major clinical application was to treat hematological malignancies [ 214 ]. Following the successes of HSCT and the discovery of distinct cell subsets involved with regulating immunity, such as regulatory T cells (Treg), DC, and myeloid cell subsets, further cell therapies have evolved to restore tolerance in AD. Here we review the state of the art of these approaches and recent preclinical and clinical results ( Table 2 ). HSCT is the standard of care for hematological malignancies and disorders of the hematopoietic system, with the goal of eradicating malignant hematopoietic cells using chemotherapeutic agents, followed by transplantation of bone marrow-derived HSCT [ 214 , 292 ]. The source of HSC can be autologous or allogeneic, depending on the application; however, allogeneic grafts present a risk of graft-versus-host disease and require systemic immunosuppression [ 292 , 293 ]. It is hypothesized that HSCT alters the immune system composition by eliminating memory cells that are pathogenic in AD [ 215 ], essentially a “reset” of the immune system that alleviates symptoms in the short and, occasionally, long term. It is important to note, however, that HSCT is associated with a high rate of adverse events and the development of secondary AD in nearly 10% of patients [ 294 ]. Autologous HSCT is currently considered the most effective treatment for MS [ 217 ], where high-dose chemotherapy is used to destroy existing immune cells and the immune system is reestablished via infusion of autologous hematopoietic stem and progenitor cells [ 295 ]. HSCT was initially investigated in numerous phase I/II trials worldwide, resulting in over 600 treated patients [ 296 ]. These initial successes led to a randomized, controlled trial of 110 patients that showed substantial disease improvement and no disease progression over a period of five years, compared to worsening of disease in the standard therapy arm and a time to disease progression of 24 months [ 218 ]. HSCT leads to a prolonged disease-free state in at least 80% of patients; studies in patients have shown that an early recovery of memory T cell populations with delayed naive T cell maturation contributes to the efficacy of HSCT [ 297 ]. Additionally, studies in human patients show that depleted B cells recover 3 months after autologous HSCT therapy, and B cells shift to a predominantly naive phenotype, with memory B cells recovering slowly and to below normal levels at 1 year post-transplant. This shift in B cell populations leaves patients susceptible to infection but is likely an important mechanism of action in an MS disease-free state [ 298 ]. The application of HSCT in T1D is challenging, as newly diagnosed stage 3 patients typically possess less than 15% reserve β cells once symptomatic [ 39 ]. Nevertheless, several small clinical trials have evaluated the efficacy of HSCT in reversing insulin dependence in newly diagnosed T1D patients [ 299 - 302 ]. Among four trials with a cumulative total of 80 patients, insulin independence was achieved in the range of 23–54% of patients for a maximum of 54 months following treatment. Patients who remained insulin dependent experienced temporary increases in C-peptide concentrations and a reduction in exogenous insulin requirements. The limited durability of HSCT and the risks associated with the treatment, including susceptibility to infection, limit the utility of this method in the treatment of T1D. HSCT has been explored in other AD, such as SLE, RA, and CeD. In one phase I trial of 7 patients with SLE, all patients receiving autologous HSCT were free of active signs of lupus at the 25 month follow-up, with minimal adverse events [ 303 ]. In another phase I trial of 15 patients with SLE followed for 5 years post-autologous HSCT treatment, 10 patients had discontinued immunosuppressive medicines by 1 year post-treatment, and only one patient experienced a recurrence of active lupus requiring immunosuppression [ 304 ]. A review of multiple trials of HSCT in SLE found that treatment resulted in the elimination of autoantibodies in some patients and a significant increase in patient survival, though 31% experienced serious infection as an adverse event [ 305 ]. In cases of severe RA, about two-thirds of patients respond to autologous HSCT treatment; however, like HSCT in T1D, responses are not durable, and remission lasts less than 12 months [ 306 ]. In refractory CeD, which is unresponsive to available therapies and carries a high risk of transition into enteropathy-associated T-cell lymphoma, a phase I clinical study evaluated autologous HSCT transplantation in 7 patients and observed significant clinical improvements in hematologic and biochemical markers and a reduction in aberrant T cells [ 307 ]. Another study of 13 refractory CeD patients observed that most patients showed clinical improvements, with complete histological remission in 5 patients. A four-year follow-up revealed that one patient had developed enteropathy-associated T-cell lymphoma despite HSCT [ 204 ]. HSCT broadly works to reset the immune system to a state that eliminates peripheral autoreactive immune cells, restoring tolerance toward self-antigens long-term in MS and temporarily in RA, T1D, and SLE. The evidence suggests it is possible that HSCT would be effective for an even broader range of AD, given its broad mechanism of action. However, the significant risks involved with a transplant procedure, particularly using allogeneic cell sources, and the high rate of adverse events make this an underutilized approach even in conditions with a significant track record of efficacy, such as MS. T cells play a critical role in AD and are the key mechanism of disease pathogenesis for the majority of AD. One avenue to combat immune dysregulation in AD is to use T cells themselves as a cell therapy, either by engineering CD8 T cells to target and destroy pathogenic self-reactive immune cells or by increasing the prevalence of Tregs to restore immune tolerance against self-antigens [ 308 ]. Tregs are a unique regulatory T cell lineage that uses a diverse TCR repertoire that is primarily aimed at autoantigen recognition [ 309 ]. The Treg T cell subset (CD4+CD25+FoxP3+) develops in the thymus and populates the periphery as long-lived cells that regulate immune responses and control autoimmunity by altering the activation and differentiation of pathogenic T cells through an effect on antigen-presenting cells [ 310 , 311 ]. Due to this unique role, there has been an increasing interest in using Tregs as a cell therapy to combat autoimmunity, either through the adoptive transfer of ex vivo expanded autologous polyclonal or antigen-specific Tregs or through the engineering of chimeric antigen receptor (CAR) Treg cells [ 219 ]. Tregs can be isolated from patients and expanded in vitro or ex vivo to be readministered to patients as an adoptive immunotherapy. Tregs can be isolated by the surface markers CD4 and CD25, but FoxP3 is an internal transcription factor that requires cell permeabilization and cannot be used on cells intended to be kept alive for adoptive transfer [ 312 , 313 ]. Some groups use CD127-low or negative cells as an additional marker of specificity, as CD127 is inversely correlated with FoxP3 expression [ 314 ]. Adoptive transfer of both polyclonal Tregs, which are not specific to a single antigen, and antigen-specific Treg clones has demonstrated efficacy in preclinical models of SLE [ 220 ], RA [ 221 , 222 ], MS [ 223 ], and T1D [ 224 - 226 ]. While antigen-specific Tregs are more effective than polyclonal T regs in diseases with specific, discrete self-antigens such as T1D or MS, polyclonal Tregs may be useful in systemic diseases with a myriad of diffuse antigens such as RA or SLE [ 219 , 315 ]. While antigen-specific Tregs necessitate fewer cells to induce a response, it can be challenging to generate sufficient cell numbers either by isolation (using tools such as tetramers) and expansion or de novo generation in vitro; to date, there are no clinical trials exploring antigen-specific Tregs to treat AD. Some studies have demonstrated that polyclonal Tregs are required in very large numbers to generate efficacy in vivo, likely given that a small proportion of this population is specific to the pathogenic self-antigen, and it is unclear whether this Treg dosage would result in adverse events such as infection and malignancies due to systemic immunosuppression [ 219 ]. To date, there have been several clinical trials evaluating polyclonal Tregs in kidney transplantation [ 227 , 228 ], to delay the progression of T1D [ 231 , 232 , 234 ], and to reduce disease activity in Crohn’s disease [ 229 ] and SLE ( NCT02428309 ). A small study using polyclonal ex vivo expanded Tregs in newly diagnosed T1D patients demonstrated potential efficacy, with a possible reduction in exogenous insulin and maintenance of c-peptide levels and no serious adverse events [ 232 , 233 ]. Another study administering polyclonal Tregs in T1D patients showed a comparable safety profile but did not evaluate improvements in metabolic function, and infused Tregs were undetectable 3 months post transfer [ 231 ]. A follow-up trial evaluated low-dose IL-2 to enhance Treg survival and proliferation after infusion in T1D patients and found an increase in Tregs over 3 months post infusion; unfortunately, IL-2 also increased NK and CD8 cell expansion, raising concerns that the immune balance in patients could be inadvertently shifted toward activation rather than tolerance, potentially exacerbating rather than ameliorating disease [ 234 ]. A phase I study evaluating polyclonal Tregs in SLE ( NCT02428309 ) enrolled only one patient and was discontinued after recording serious adverse events and no improvement in disease. Finally, several clinical trials using polyclonal Tregs to treat Crohn’s disease have demonstrated tolerability and initial evidence of a reduction in disease activity [ 229 ]; however, maximum improvement in one study was observed at week 5, suggesting successful treatment may require multiple regular injections [ 230 ]. Recently, a phase I study evaluated antigen-specific Tregs in 3 patients receiving kidney transplants using a modified immunosuppression protocol without induction therapy. Treatment was well tolerated and no rejection episodes were observed [ 316 ]. Chimeric antigen receptors (CAR) are modified T cell receptors designed to target specific antigens and were originally developed to target cancer cells [ 235 ]; this approach has proved wildly successful in targeting hematological cancers, with remission rates as high as 80% and resulting in six currently approved CAR-T cell therapies [ 317 ]. With the success of targeted CAR-T cell therapies in cancer, several preclinical studies have explored the utility of CAR-T cells in suppressing or reversing AD by targeting pathogenic auto-reactive immune cell populations or enhancing antigen-specific Treg populations [ 236 , 318 , 319 ]. In one study in a mouse model of T1D, a CAR was designed to target APCs bearing the pathogenic antigen peptide B:9–23 [ 237 ]. B:9-23 APCs were effectively targeted by the CAR-T cells; however, only a slight delay was observed in disease onset. A similar approach was developed to target MBP-carrying antigen-presenting cells in a mouse model of MS; CAR-T cells successfully prevented MS progression, and this was found to be due to the CD8 CAR-T subset [ 238 ]. Preclinical studies evaluating MBP-specific [ 239 ] or MOG-specific [ 240 ] CAR-Tregs, or a combination of the two antigens [ 241 ], demonstrated a moderate reduction or reversal of disease in a murine MS model. Cumulatively, the preclinical and clinical efforts to use Tregs and CAR-T or CAR-Treg-based cell therapies to ameliorate AD have demonstrated the challenges associated with these approaches. Generating sufficient doses of cells to produce an effect is challenging, and programming antigen specificity is often insufficient to produce dramatic improvements. While clinical data using antigen-specific Tregs is sparse and no data exists yet for AD, the cumulative evidence suggests that polyclonal Tregs are of limited use in reversing AD, but antigen-specific Treg treatment may result in more durable and targeted responses and improved outcomes. Future studies will likely focus on improving the efficacy and specificity of Treg approaches using CAR-T and antigen-specific Treg strategies. Mesenchymal stromal cells (MSC), initially termed mesenchymal stem cells, generated substantial excitement in the scientific community due to their capacity to differentiate into multiple lineages in vitro and their ease of isolation from adult tissues such as adipose tissue, the umbilical cord, bone marrow, and blood. This excitement has led to more than 1050 registered clinical trials [ 320 ]. Early preclinical and clinical studies demonstrated the potential of these cells as an allogeneic “off-the-shelf” cell therapy in a variety of regenerative medicine and immune-related applications due to their ease of expansion and immune evasive properties [ 321 ]. Nonetheless, studies have noted that in vitro culture and expansion may induce HLA class II expression and enhance MSC immunogenicity [ 322 ]. With the failure of many phase III clinical trials using a variety of cell sources and disease applications [ 323 ], more recent investigations have concluded that the therapeutic mechanism of MSCs is due primarily to the secretion of trophic factors rather than direct differentiation in vivo, and substantial interest remains in using MSCs as an immunomodulatory cell therapy [ 242 ]. MSCs have been explored preclinically in numerous AD applications, with promising results that formed the basis for clinical trials. A meta-analysis of preclinical studies evaluating MSCs for the treatment of RA found that 42 of 48 studies reported improvement in recipients, assessed via clinical score or histological score, regardless of the MSC source (e.g., human, murine, rat; adipose, umbilical cord, bone marrow) [ 243 ]. Several preclinical studies have explored the use of MSCs to reverse or delay autoimmune diabetes in mouse models, with allogeneic MSCs demonstrating an average delay in T1D onset of 20 days [ 244 ], out to 36 weeks in 40% of recipients [ 245 ], and reversing recent diabetes onset in 80% of recipients [ 246 ]. In a meta-analysis of 28 studies where MSCs were applied in murine SLE models, MSC therapy reduced double-stranded DNA, anti-nuclear autoantibodies, serum creatinine, and proteinuria levels [ 247 ]. In the clinic, MSCs have demonstrated moderate improvements in RA [ 324 - 326 ], T1D [ 327 , 328 ] and SLE [ 329 ], though most studies lack randomization and placebo controls [ 330 , 331 ]. MSCs for RA treatment have been shown to be safe and well tolerated, and in uncontrolled phase I studies, MSCs reduced disease activity scores at 4 weeks post-treatment [ 325 ] and up to 3 years post-treatment when delivered alongside disease-modifying drugs [ 324 ]. A placebo-controlled trial with dose escalation of adipose-derived MSCs showed no statistically significant improvements in RA but was well tolerated [ 326 ]. Autologous MSCs have been explored in the clinic in a placebo-controlled phase I trial with newly diagnosed T1D patients, showing a small statistical benefit in glycosylated hemoglobin at 12 months post-treatment, but no other measurable benefits [ 271 ]. Another placebo-controlled phase I/II trial using allogeneic umbilical cord MSCs demonstrated significantly less decline in c-peptide at 12 months relative to placebo [ 272 ]. Two uncontrolled phase I trials demonstrated a positive safety profile for allogeneic MSC treatment in SLE, with evidence of improvement in disease scores and biomarkers at 12 [ 268 ] and 17 months [ 269 ] of follow-up. However, a placebo-controlled study found allogeneic MSCs provided no evidence of SLE disease improvement over the placebo group [ 270 ]. Extensive preclinical studies have evaluated the efficacy of MSCs in MS models in mouse, rat, and dog models, with impressive results and typically ameliorating disease [ 248 ]. Despite the promise of preclinical studies, none of the numerous clinical studies investigating MSCs for the treatment of MS showed clinical benefit [ 249 ], including 7 placebo-controlled studies [ 250 - 255 ] and 12 uncontrolled studies [ 256 - 267 ] using either autologous bone marrow MSCs or allogeneic umbilical cord MSCs. To date, there are no FDA-approved MSC-based therapies in the US, though other countries such as Japan, South Korea, Canada, Europe, India, and New Zealand [ 332 ] have granted approval, including for applications relevant or adjacent to damage in AD such as fistulas in Crohn’s disease, articular cartilage defects, spinal cord injury, and sclerosis. MSC-based therapies have sought approval in the US, such as Remestemcel-L in 2020 (targeted to pediatric graft-versus-host disease) after demonstrating clinical benefit in trials, but the FDA requested further data on efficacy, validated potency assays, and product mechanism of action [ 333 , 334 ]. Cumulatively, the data surrounding the utility of MSCs in AD is not particularly strong, with multiple confounding factors including MSC tissue source, allogeneic vs. autologous, and route of administration. The repeated lack of clinical efficacy of MSCs in MS contributes to the growing body of work questioning the efficacy of MSCs in vivo and their therapeutic mechanism [ 335 ]. DCs are antigen-presenting innate immune cells of a heterogeneous phenotype with a critical role in stimulating T cells and directing their activity against self-antigens or depleting autoreactive T cells during selection in central and peripheral tolerance. DCs with a stable, semi-mature, and tolerogenic phenotype are termed tolerogenic DCs and can be identified by the expression of CD163, CD141, CD16, and CD14 [ 273 ]. Preclinical studies explored the therapeutic efficacy of monocyte-derived tolerogenic DCs in the context of T1D [ 274 ], MS [ 275 , 276 ], and RA [ 277 ]. Successes in these preclinical studies have led to ongoing clinical trials in a range of AD, despite a lack of clarity about the mechanism by which tolerogenic DCs induce a therapeutic effect. The first clinical trial using tolerogenic DCs was evaluated in T1D and demonstrated a good safety and tolerability profile, with a possible increase in the number of regulatory B cells post-administration [ 278 ]. There are currently 3 more phase I or II trials underway in T1D (NTR5542, NCT02354911 , and NCT03895996 ). A phase Ib trial [ 279 ] in 8 patients with MS found tolerogenic DCs were safe and well tolerated, and after 12 weeks of treatment there were reduced memory CD8 T cell populations and significantly increased type 1 Treg populations [ 279 ]. Two-phase I/IIa trials using tolerogenic DCs in MS are underway ( NCT02903537 , NCT02618902 ) [ 280 ]. Finally, four phase I trials have evaluated tolerogenic DCs in RA, demonstrating safety and tolerability, improvement [ 281 ] or stabilization [ 282 ] of disease symptoms, and a decrease in T effector cells [ 283 ]. Overall, current clinical data points to the potential of tolerogenic dendritic cells as a cell therapy to arrest AD, but placebo-controlled trials and investigation into the mechanism of restored tolerance are critical. Myeloid-derived suppressor cells (MDSC) are a relatively newly identified type of myeloid cell distinct from classical neutrophils and monocytes that arise under conditions of unresolved inflammation in vivo, such as infection, cancer, and chronic conditions. Neutrophils and monocyte-derived subsets are classified as polymorphonuclear-MDSC and monocytic-MDSC, respectively [ 336 , 337 ]; in chronic inflammatory conditions, they are pathologically activated, leading to the inhibition of adaptive immunity rather than classical activation [ 338 ]. MDSCs are of interest in the pathogenesis of AD, with conflicting evidence as to their role in disease progression and recovery. Studies have found that peripheral MDSC numbers are increased in T1D and reduced within the pancreas [ 284 ]. In RA, peripheral MDSC are also identified in higher numbers than in control patients [ 285 , 286 ], but their role may contribute to the pathogenesis of RA rather than the suppression of the disease [ 286 ]. Similarly, peripheral MDSCs were increased in patients with SLE, and MDSC increases correlated with disease severity [ 339 ]; however, a mouse model of SLE demonstrated that MDSCs expand regulatory B cells and ameliorate autoimmunity [ 340 ]. In a mouse model of MS, MDSCs were recruited to the site of injury and interacted with B cells; upon MDSC ablation, B cell accumulation increased in the central nervous system and prevented remission from disease [ 341 ]. Several conflicting preclinical studies have investigated the therapeutic potential of MDSCs in AD. In some experiments, the adoptive transfer of MDSC in RA reduced disease severity by suppressing T cell proliferation and autoantibody production [ 287 ] and by inhibiting TH17 cells and inflammation [ 288 ], but in other studies, it exacerbated disease [ 289 ]. Similarly, the adoptive transfer of MDSC significantly reduced the onset of T1D in an autoimmune T1D mouse model in one study [ 290 ], whereas other studies found no effect [ 291 ]. Overall, the data supporting MSDCs as a distinct cell therapy in AD to date is sparse.

As diverse new cell therapies are developed, it is critical to consider from the early stages of development how to translate these innovations to the clinic as products. Early in product development, researchers and manufacturers should consider barriers in manufacturing, regulation, and reimbursement to ensure product success in the healthcare marketplace. Additionally, product development should comply with guidelines for good pharmaceutical practices at every stage, such as preclinical (good laboratory practice (GLP) and good manufacturing practice (GMP)), clinical (good clinical practice (GCP)), and post-market approval (good distribution practice (GDP)) [ 342 ]. Approaches to translation will vary with the type of product, such as autologous or allogeneic products, or how the product is delivered (e.g., the use of biomaterials as carriers). For example, cell products that require significant manipulation (e.g., ex vivo culture, transformation, or combination with biomaterials) require different regulatory and manufacturing considerations than those requiring minimal manipulation (e.g., autologous HSCT). Advancements in cell product manufacturing technology and streamlining of regulatory and reimbursement pathways will result in improved patient accessibility to and adoption of cell therapy products in the coming decades. Cell therapy product manufacturing is a critical consideration due to the novelty of cell products in the healthcare marketplace and the minimal existing industry standards for diverse cell product types and has traditionally been an area of underinvestment [ 343 ]. Recent initiatives in the US toward expanding biomanufacturing capabilities will hopefully result in greater investment in this critical issue and spur the development of novel technologies and industry standards, which could ultimately lower cell therapy product costs [ 344 ]. One consideration in cell manufacturing is the process of product manufacturing. In many cases, cells are handled, cultured, and manipulated in a similar manner as in academic labs, with workers manually manipulating products [ 345 ]. This is labor-intensive and allows for the potential for human error at every stage of the production process; additionally, this results in a considerable cost for the final product [ 346 ]. As such, there is considerable interest in the development of automated and enclosed manufacturing processes that enable the fabrication of products with minimal manual labor [ 347 ]. These automated approaches can be engineered with product quality control and monitoring at intermediate steps. One challenge is that these manufacturing systems must be designed for individual products. For example, the process of stem cell differentiation into β cells over 30 days in culture requires significantly different procedures from an autologous DC product. Autologous and allogeneic cell therapies are currently best served by decentralized and centralized manufacturing approaches, respectively, which results in great disparities in product costs. As an example, the five FDA-approved autologous CAR-T cell products range from $373–475,000 for a single treatment [ 348 ]; models suggest this cost could be reduced to approximately $5000 per dose using allogeneic products and a centralized manufacturing model [ 346 ]. Autologous approaches require quick manipulation or transformation of a patient’s own cells due to limited stability and shelf life, whereas allogeneic products can be manufactured in a mass-production model and disseminated from a central location. Improvements in product stability and cryopreservation techniques may enable the centralization of autologous product manufacturing and improve allogeneic product capabilities, all of which could result in a substantial reduction in product cost and greater patient accessibility [ 346 ]. Global cell therapy regulations are primarily concerned with ensuring the safety and efficacy of any new cell therapy product. Regulatory considerations for cell-based therapies requiring significant manipulation were pioneered in the cancer immunotherapy space [ 349 ]. Regarding safety, regulatory agencies pay careful consideration to cell sources (e.g., screening for communicable diseases in allogeneic products) and preventing the introduction of microorganisms during the manufacturing process. Cells must be tested for bacterial, mycoplasma, and endotoxin contamination [ 350 ]. Cells must be counted, and measured for viability [ 351 ], where a common release criterion is a minimum of 70% viability [ 352 ]. All manufacturing materials must be qualified by the vendor or tested by the end user to ensure quality and safety. Additionally, stem-cell-derived products require methods to verify cell identity and product attributes that reliably predict not just safety but efficacy, both in vitro and in appropriate preclinical animal models [ 353 ]. Prior to administration, cell therapies must be tested for quality with clearly defined quality control criteria and release assays that comprise characterization and functional studies. Product potency assays can be particularly challenging, where manufacturers must develop a set of predictive tests defining product characteristics such as morphology, secretome, and gene expression profile [ 353 , 354 ]. Flow cytometry is a common method used to characterize the phenotype of cell products and the degree of contamination or purity, particularly in immune cell and stem cell-derived cell therapies. Cells can typically be further characterized for functionality using enzyme-linked immunoassays to measure cytokine (e.g., IFNγ in T cell lots) or hormone secretion levels (e.g., insulin-secreting β cells) [ 354 ]. Due to the novelty and diversity of many cell therapy products emerging in the marketplace, there is a lack of international standards for many of these quality and potency assays, which means many new products lack a defined roadmap for obtaining regulatory approval. Finally, some cell therapy products have faced challenges with regulatory approval despite clinical evidence of efficacy (e.g., MSC cell therapy Remestemcel-L), due to a lack of a defined mechanism of action [ 333 , 334 ]. As such, careful consideration for product characterization, potency, and defined mechanism of action is crucial to minimizing the cost of translation of any novel cell therapeutics. Clinical success and regulatory approval of cell therapies or regenerative medicine products are only two aspects of determining reimbursement and health system adoption for a therapeutic. The payer perspective must be considered when developing new therapies to ensure success in the healthcare marketplace. The payer can be self-pay (i.e., patient out-of-pocket), health systems (e.g., nationalized health systems such as in the EU and Canada, or the Centers for Medicare and Medicaid Services (CMS) in the US), or private insurance companies. In the US, most insurers follow the CMS approach, which assigns unique alphanumeric codes to a product or service to be reimbursed. For most novel therapies, including cell therapies, existing reimbursement codes do not exist, and code refinement can lag for years behind the emergence of new therapeutics. Inevitably, this impacts the sales of products and creates the largest burden for the first product on the market in a new category of therapeutics [ 355 ]. The high cost of new cell-based therapies must be balanced with therapeutic benefits to enable payers to justify reimbursement [ 356 ]. For example, in the United Kingdom, reimbursement rejection increases above GB £20,000 per quality-adjusted life year (a measure of quality and quantity of life lived) [ 357 ]. A recent example is CAR-T cell therapy, which demonstrates exceptional success rates in patients with hematological cancers, but the high price tag of more than $300,000 per dose has resulted in exceptional challenges in reimbursement. Initially, CMS assigned CAR-T cell therapy a reimbursement code that resulted in reimbursement of only $43,094, not nearly enough for the product itself, let alone associated treatment costs and inpatient stays, resulting in hospitals absorbing significant proportions of the treatment cost. In 2021, this was revised to a new code allowing for a base payment rate of $239,929, a significant improvement but still insufficient to cover all inpatient costs [ 358 ]. While CAR-T cell therapies are pricey, they promise a significant return on the cost of treatment when used to treat life-threatening diseases, such as cancer, so there is a strong demand for this product. Comparatively, many cell-based therapies in the regenerative medicine space need to demonstrate significant improvements over less costly treatments to justify adoption and reimbursement. One example where this bar was not met is ChondroCelect ® , which was withdrawn from the EU market due to reimbursement issues [ 359 ]. Similarly, the first-generation product CARTICEL ™ (cells only) was phased out when MACI entered the market, which delivers cells in a material and is easier to implant [ 360 ]. MACI ® was approved in the EU in 2014 but ceased manufacturing due to a lack of sales, though it is still on the US market [ 361 ]. As such, cell therapy value assessment is a critical consideration for access to major healthcare markets, and there are analytical methods to assess this value early in the process, such as headroom analysis and cost-effectiveness modeling [ 343 , 362 ]. Such practices can maximize the success of new products with payers in the reimbursement ecosystem as well as in the healthcare marketplace in general. Finally, as the costs of cell therapy production decrease with manufacturing technology maturation, these calculations will change and likely ease the path of new products into the marketplace.

Cell therapies are an expanding group of therapeutics with great potential to improve AD patients’ quality of life, whether used in the reversal of AD pathogenesis or to repair tissues damaged during the course of disease. Many stem cell-based regenerative medicine approaches are in their infancy, but select applications in advanced stages, such as T1D, or related applications, such as musculoskeletal disorders, may pave the way to advance the repair and replacement of diverse tissues damaged in AD. Similarly, great successes have been achieved with select cell therapies to halt AD progression, such as HSCT in MS. However, the majority of immune-modifying cell therapies cannot advance until more targeted approaches are developed, which may enhance the safety and efficacy of these products. Additionally, for many cell therapies, the underlying mechanism is not well understood or defined, which will limit translation to the clinic and regulatory approval. Finally, as the field advances, it is critical to consider how cell products are manufactured and to innovate in cell product manufacturing to reduce product costs and ensure patient access once products enter the healthcare marketplace.

Autoimmune diseases (AD) are heterogeneous, complex, and chronic disorders that can onset in childhood or adulthood, with more than 80 distinct diseases currently identified affecting a range of organs and organ systems [ 1 ]. Most AD have clear autoantigens that elicit an adaptive immune response, but not all have a clear etiology [ 2 , 3 ], complicating efforts to develop effective and targeted therapeutics. The AD healthcare burden in the US is estimated at $100 billion [ 4 ], and while AD impacts both men and women, many AD disproportionately impact women [ 5 , 6 ]. Additionally, select AD impact specific genetic ancestries, though epidemiological data for many AD are sparse [ 1 ]. The National Institutes of Health (NIH) invests several billion research dollars per year to study and treat AD, with a broad range of per-patient research dollar spending ranging from approximately $2000 per patient for multiple sclerosis (MS) and systemic lupus erythematosus (SLE) to less than $6 per patient for Hashimoto’s disease, Graves’ disease, and endometriosis ( Table 1 ). Current first-line approaches to treating AD include immunomodulatory approaches that are broadly acting, non-disease-specific, and of limited efficacy; thus, there is a pressing need to develop therapeutic approaches that target the underlying autoimmune mechanism and restore tolerance [ 7 ]. Additionally, there are limited means to regenerate or replace the tissues and organs targeted by AD, resulting in a permanent, debilitating loss of function if the disease cannot be arrested prior to tissue destruction. Cell therapies are a promising new approach to the treatment of a broad range of diseases, including cancer, degenerative diseases, immune diseases, and AD. Cell therapies can be implemented in AD treatment both for the suppression or correction of errant immune processes in the course of the disease as well as for replacing the function of cells and tissues damaged by disease progression. In this review, we provide an overview of cell therapies that are approaching or in the clinic to (1) regenerate or replace damaged cells and tissues in AD and (2) suppress, reset, and correct errant adaptive immune responses in autoimmunity. As cell therapies are a new approach with unique challenges to translation, we also provide a perspective on translating cell therapy products and approaches to the clinic.

AD can be tissue- or organ-specific, like type 1 diabetes (T1D) that affects the islet cells of the pancreas and Grave’s disease that affects the thyroid gland, or systemic, like SLE or rheumatoid arthritis (RA), where autoantibodies are generated against multiple cell types ( Figure 1 ) [ 22 ]. AD occurs through the loss of peripheral T and B cell tolerance toward “self” antigens within specific tissues. Tolerance of T cells toward self-antigens is maintained through central and peripheral tolerance, which occurs in the thymus and peripheral tissues, respectively. Central tolerance arises when immature T lymphocytes travel from their origin in the bone marrow to the thymus and self-reactive T cells are eliminated; however, not all tissue antigens are expressed in the thymus, requiring peripheral tolerance mechanisms to eliminate mature circulating self-reactive T lymphocytes. Self-reactive T lymphocytes that escape these central and peripheral mechanisms give rise to self-reactive B cells, resulting in autoantibodies against self-antigens; it is believed that loss of peripheral tolerance leads to autoimmunity [ 23 ]. The innate immune system, comprising antigen-presenting cells (APC) such as dendritic cells (DC) and macrophages, plays a critical role in breaking this tolerance by priming autoreactive T cells [ 24 ]. As self-reactive T cells are challenging to measure clinically, autoantibodies are typically used to detect and diagnose AD alongside the clinical manifestation of characteristic symptoms. In some AD, organ damage is mediated primarily by T cells, such as in T1D and multiple sclerosis (MS), whereas damage in other AD are mediated primarily by autoantibodies, such as in SLE. Regardless, autoimmunity typically results in irreversible organ damage, necessitating cell or tissue replacement to restore lost function. Demand for cadaveric donor organs and tissues outpaces availability, and as the global population ages and the incidence of diseases, including AD, increases, the organ donor waiting list will continue to increase, resulting in increasingly high mortality. Cell-based regenerative medicine approaches have arisen in the past few decades to replace a wide array of tissues and organ systems and reduce demand for organ donation. These approaches typically harness the plasticity and proliferative capacity of stem cells to generate large numbers of cells that can replace or encourage the regeneration of damaged tissues, and the final products may be terminally or partially differentiated stem cells [ 25 ]. Terminally differentiated products may avert the risks associated with stem cells, such as teratoma formation; however, terminally differentiated cells also have limited proliferative and self-renewal capacity in vivo [ 26 ]. Many regenerative cell therapy approaches are in their infancy, but some, including T1D and cartilage applications, are reaching a state of maturity and approaching the clinic. In this section, we discuss the pathogenesis and resulting tissue damage of common AD and discuss cell therapy-based strategies to replace AD-damaged cells, tissues, and organs. The global prevalence of T1D is 15 per 100,000 people and increasing [ 8 ], with the highest diagnoses in populations of European origin before or at puberty and a comparable incidence between males and females [ 8 , 27 ]. In T1D, autoimmunity results in the catastrophic loss of β cells in the pancreatic islet and subsequent hyperglycemia, necessitating exogenous insulin administration to maintain blood glucose within the normal range [ 28 , 29 ]. This autoimmunity is primarily believed to be mediated by cytotoxic CD8 + T cell destruction of β cells ( Figure 1A ) and is marked by dramatic insulitis and infiltration of islets by adaptive and innate immune cells. B cell involvement results in one or more measurable autoantibodies against insulin, zinc transporter 8, glutamic acid decarboxylase, and islet antigen 2 that are detected prior to and at the time of diagnosis in 70–80% of patients [ 30 - 35 ]. T1D progression is a continuum, with stages that progress from presymptomatic normoglycemia with evidence of autoimmunity via the presence of 2 or more autoantibodies (stage 1), presymptomatic with evidence of dysglycemia (stage 2), and the onset of symptomatic disease (stage 3) [ 36 , 37 ]. Patients with genetic risk factors (e.g., a close relative with T1D) are able to screen for autoantibodies via research programs such as TrialNet or commercially available home test kits, and patients that reach stage 1 have a 5-year and 10-year risk of symptomatic disease of 44% and 70%, respectively. Identification of patients in stage 1 or 2 T1D has enabled enrollment in prevention trials and led to the approval in 2022 of the first treatment to delay T1D onset in stage 2 patients [ 38 ]. However, 85% of new diagnoses have no family history; therefore, the majority of at-risk patients do not undergo early autoantibody screening (ideally at age 2.5 years), and diagnosis typically occurs in symptomatic stage 3 after loss of >85% of β cell mass ( Figure 1B ), when patients present with symptoms of hyperglycemia, diabetic ketoacidosis, weight loss, fatigue, and polyuria [ 39 ]. The pathogenesis of T1D is influenced by both genetic susceptibility (human leukocyte antigen (HLA) haplotype confers 30 –50% of the genetic risk of T1D [ 40 ]) and suspected environmental factors [ 41 , 42 ], such as the microbiome, diet, and viral infections; enteroviruses are a particular virus of interest [ 43 , 44 ]. Interestingly, the recent coronavirus disease 2019 (COVID-19) pandemic resulted in a significant increase in pediatric T1D diagnoses, suggesting severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) as a possible viral trigger for autoimmunity in T1D [ 45 - 48 ]. If patients are diagnosed in stages 1 or 2 (i.e., prior to symptomatic disease), there are several immunomodulatory treatments currently in clinical trials to delay or stop the progression of autoimmunity and β cell loss. Several trials have investigated methods to induce tolerance toward islet antigens using methods that have been successful in allergy immunotherapy [ 49 ]. Oral insulin therapy has been explored in placebo-controlled trials, with positive results only in patients diagnosed after 20 years of age and acceleration of disease in younger patients [ 50 ]. Vaccines have also been explored, such as an intralymphatic vaccine against glutamic acid decarboxylase; however, a randomized controlled phase IIb trial demonstrated limited efficacy in preventing the progression of T1D [ 51 ]. Overall, vaccine approaches to limit the progression of T1D in humans are limited in that there are multiple potential autoantigens driving disease progression that are not well defined [ 52 ]. Teplizumab, approved by the FDA in 2022 under the commercial name TZIELD ® , is a humanized anti-CD3 monoclonal antibody where the Fc region has been engineered to have non-binding properties, resulting in activated T effector cell anergy or apoptosis. TZIELD ® has been shown to delay the onset of stage 3 T1D in adults and pediatric patients over age 8 by about 2 years [ 53 ]. Current treatments for the majority of patients who have progressed to symptomatic, insulin-dependent stage 3 T1D are limited to insulin replacement therapy and blood glucose monitoring. However, a subset of patients are unable to achieve glycemic control with intensive monitoring, resulting in a high risk of secondary complications and death, necessitating more invasive interventions such as cell replacement therapy [ 54 ]. Cell replacement therapy is the most advanced in T1D of all the AD, with allogeneic islet transplantation boasting a 50-year track record of success. With advancements in transplant techniques and immunosuppression protocols, many patients remain insulin-independent for ten years or more [ 55 ]. In this procedure, allogeneic donor islets are isolated from the cadaveric pancreata and infused into the patient’s hepatic portal vein. Islets become trapped in the branching hepatic vasculature, where they can respond to blood glucose levels. After decades of experimental use, in 2023, cadaveric donor islet replacement therapy via the portal vein was Food and Drug Administration (FDA)-approved under the name Lantrida ™ for patients unable to approach average blood glucose levels due to repeat episodes of severe hypoglycemia [ 56 , 57 ]. While this technique has demonstrated the feasibility of reversing T1D in patients, it has a number of drawbacks that limit its widespread application among the T1D patient population: (1) it requires the use of chronic systemic immunosuppression that carries substantial risks of infection and malignancy, resulting in short term risks that outweigh the long-term benefit of the procedure for all but a subset of patients with highly labile disease; (2) infusion of islets within the hepatic vasculature results in the loss of ~50% of islet mass due to stresses on the cells and an instant blood-mediated inflammatory reaction against the infused cells, requiring the administration of multiple pancreata to restore normoglycemia in most patients; and (3) the limited availability of donor pancreata, estimated at ~1500 per year, restricts the number of patients that can receive this treatment [ 54 ]. In recent years, stem cell-derived insulin-secreting cell sources have been developed to address the limited availability of donor pancreata in cell therapy to treat T1D [ 58 ]. Induced pluripotent stem cells (iPSC) or embryonic stem cells (ESC) have been successfully induced to differentiate into insulin-secreting cells using complex induction protocols that take as many as 35 days to complete ( Figure 1C ) [ 59 ]. Typical protocols generate somewhat immature cells; however, further differentiation in vivo has demonstrated that these cells are capable of reversing T1D. Recent clinical trials by ViaCyte [ 55 , 60 - 62 ] and Vertex [ 63 ] pharmaceuticals have demonstrated promise in restoring normoglycemia, though under the cover of chronic systemic immunosuppression to combat the preexisting autoimmunity and graft rejection. New approaches are targeted at eliminating the requirement of immunosuppression, using biomaterial approaches [ 64 - 69 ] or engineering hypoimmune stem cells [ 70 ]; other strategies are targeting the development of alternative transplant sites [ 71 , 72 ]. These engineering advancements can likely be expanded to enhance regenerative cell therapy approaches in other AD. Multiple sclerosis (MS) is the most prevalent degenerative disease of the central nervous system, affecting 35.9 per 100,000 people worldwide [ 10 ], and is characterized by physical, cognitive, and neurological problems [ 74 , 75 ] caused by the development of lesions within the brain and spinal cord. MS is categorized into relapsing-remitting (RR), secondary progressive (SP), primary progressive (PP), and progressive relapsing (PR) subtypes [ 76 ], where RR is the most common clinical subtype in 90% of patients [ 77 ]. The prevalence of MS is increasing, affecting three times as many women as men [ 78 ], with disease onset in the age range of 20–40 years. As with most AD, the etiology of MS is complex, and a combination of genetic predisposition and environmental factors play a role in acquiring the disease. A recent study using data from more than 10 million US Army personnel over 20 years provided substantial evidence for the role of the Epstein-Barr virus in the development of MS [ 79 , 80 ]. Activated immune cells of both the innate and adaptive immune systems play a role in disease progression [ 81 ], and studies have shown that ablation of immune cells early in the course of disease can reduce long-term disability [ 82 , 83 ]. In MS, antigen-presenting cells, autoreactive CD4 T helper cells, cytotoxic CD8 T lymphocytes, and B cells all play a role in disease progression ( Figure 2A ). CD4 T helper cells play a central role in MS pathogenesis, where TH17 cells weaken the blood-brain barrier [ 84 ], contributing to neuroinflammation, and the antigen specificity of CD4 T cells toward myelin basic protein (MBP), myelin oligodendrocyte glycoprotein (MOG), and proteolipid protein result in the downstream generation of autoantibodies against myelin proteins by B cells in the cerebrospinal fluid [ 85 , 86 ]. Neurodegeneration occurs with the proliferation of astroglial cells, and inflammation and oligodendritic cell apoptosis lead to myelin degeneration [ 87 - 89 ]. Immune suppression and reduction of inflammation are two major current approaches to the treatment of MS, and disease-modifying drugs are selected based on disease category [ 90 ]. Several disease-modifying therapies aimed at regulating the immune response have been highly successful, including anti-CD20, anti-CD52, anti-VLA4, and cladribine, which can reduce the relapse rate by up to 60% [ 91 ]. Additional FDA-approved therapeutics for the treatment of MS are interferon (IFN)-β and glatiramer acetate [ 92 ]. IFN-β possesses neuroprotective properties, induces anti-inflammatory effects by suppressing the production of IFN-γ and TGF-β [ 93 ], and prevents the crossover of immune cells across the blood-brain barrier (BBB). GA mimics the myelin autoantigen, competes with MHC molecules, and has immunomodulatory effects [ 92 ]. Various monoclonal antibodies that suppress B cell antibodies are also approved, such as ocrelizumab, which prevents B cells from producing antibodies [ 94 ], and natalizumab, which prevents antigen leakage to the CNS by maintaining the integrity of the BBB [ 95 ]. These therapeutic agents may improve patient symptoms but are unable to completely stop or reverse neurodegeneration. Cell therapies to regenerate and replace demyelinated nervous tissue have been explored preclinically since the 1970s [ 96 ]. Remyelination persists throughout adulthood, perpetuated by a type of abundant adult neural stem cell called oligodendrocyte precursor cells (OPC) [ 97 ]. Studies suggest that chronic MS lesions persist due to inhibition of local OPC [ 98 ]. Myelin-forming cells, such as OPC, Schwann’s cells, or olfactory ensheathing cells, possess limited growth and expansion capabilities in vitro and contribute to remyelination only when delivered in close proximity to the CNS. Preclinical studies evaluating the delivery of OPC have demonstrated the potential for remyelination in mice ( Figure 2B ) [ 99 , 100 ]. Alternatively, neural stem/precursor cells (NPC), hematopoietic stem cells (HSC), ESC, and mesenchymal stem cells have greater potential for expansion and self-renewal, but delivery to CNS lesions is challenging, particularly when lesions are multifocal, such as in MS. Additionally, sourcing NPC are challenging, and studies using ESC and NPC are challenging and ethically fraught due to embryonic and fetal cell sources, respectively. NPC have been transplanted into preclinical models [ 101 - 106 ] of neurological disorders and demonstrated efficacy through the promotion of neuroprotection and remyelination [ 105 , 106 ]. NPC delivered systemically, either intravenous or intrathecal, can home to the lesion site via the bloodstream or cerebrospinal fluid circulation and chemoattractant gradients stemming from the lesion [ 104 , 107 - 110 ]. However, functional recovery rarely correlates with engrafted and differentiated neuronal cells [ 104 ]. Low terminal differentiation of NPC in preclinical models of MS (e.g., experimental autoimmune encephalitis) points to an indirect therapeutic benefit from this cell therapy, potentially due to the local delivery of neurotrophins and immunomodulatory molecules [ 111 ]. Interestingly, NPC have been shown to reduce the proliferation of antigen-specific T cells in preclinical models of MS [ 109 ]. Recently, a non-randomized open-label phase I clinical trial was conducted delivering human fetal NPC intrathecally in 12 patients with progressive MS; they found that NPC transplantation was safe and tolerable, and secondary analyses showed a reduced rate of brain atrophy in the group receiving the highest dosage, with measurement of increased levels of anti-inflammatory and neuroprotective molecules in cerebrospinal fluid ( Figure 2C ) [ 112 ]. HSC have the capacity to differentiate into a variety of cell types, and preclinical models of demyelination of the spinal cord have demonstrated remyelination after localized delivery of HSC to the spinal cord or brain. It is unclear whether HSC contribute to remyelination via differentiation into neural cells or contribute to recovery via immunomodulation [ 113 ]. Preclinical studies have evaluated ESC-derived OPC transplantation in mouse models of demyelination and shown further differentiation into oligodendrocytes and remyelination of the spinal cord [ 114 ]. Similarly, MSC have been shown to differentiate into a variety of cells, including neural lineage cells [ 115 - 117 ]. However, it is unclear whether MSC delivered in vivo are capable or inclined to differentiate into new neural tissue. Like NPC, current evidence indicates the regenerative effects of MSC derive from trophic effects (e.g., secretion of cytokines and chemokines), which may suppress inflammation or enhance the proliferation and differentiation of endogenous progenitor cells. Preclinical data indicate that MSC may exert beneficial effects in vivo [ 118 - 120 ] primarily through trophic factors and immunomodulatory action rather than differentiation into myelinating cells [ 121 - 123 ]. However, clinical trials implementing MSC to date have been disappointing, as summarized in greater detail in Section 3.3 . RA is a chronic inflammatory illness characterized by synovial inflammation and swelling of joint tissues, autoantibody production, and systemic features, including skeletal, pulmonary, and cardiovascular disorders [ 125 , 126 ]. As with most AD, RA affects more female than male patients at a ratio of 2-3:1, affecting 1% of the global population [ 12 ], and both genetic (HLA-DR) and environmental factors (e.g., smoking, infectious agents) influence disease risk. In the pathogenesis of RA, synovitis occurs with leukocyte infiltration of the joint space, facilitated by endothelial activation in synovial microvessels [ 127 ]. Infiltrating macrophages release the cytokines TNF-α, IL-1, and IL-6 that stimulate fibroblast-like synoviocyte (FLS) activation, proliferation, and migration [ 128 ], which leads to bone erosion [ 129 ] and matrix metalloprotease (MMP) secretion, which results in cartilage destruction [ 130 ] ( Figure 3A ). Activated T lymphocytes migrate into the synovium and secrete IL-17, which further activates macrophages and FLS [ 131 ], and synovial fluid NK cells play a role in bone erosion in RA patients [ 132 ]. B cells also contribute to inflammation in RA [ 133 ] through the release of cytokines and the generation of autoantibodies, including rheumatoid factor (RF) [ 134 , 135 ] and anti-citrullinated protein antibody (ACPA) ( Figure 3B ) [ 136 , 137 ]. There exists a wide range of drug treatment options for rheumatoid arthritis, collectively called disease-modifying anti-rheumatic drugs (DMARD) [ 138 ]. Generally, disease remission is the goal and can be achieved in 17-33% of patients if therapy begins within 3 months to 3 years of disease onset [ 139 ]. Methotrexate, an immunosuppressant anti-metabolite, is typically the first-line treatment [ 138 ], and recent studies demonstrated that combination therapy of methotrexate, sulfasalazine, and hydroxychloroquine increased remission rates from 33% to 66% [ 140 ]. Additionally, biological DMARDs have been developed to target IL-6 inhibition (Tocilizumab [ 141 ]), B cell depletion (Rituximab [ 142 ]), and Janus kinase inhibition (Tofacitinib, Baricitinib [ 143 ]); these alternative approaches are more costly but exhibit reduced side effects relative to methotrexate while demonstrating comparable remission rates when effective [ 138 ], though it can be hard to predict which patients will respond to individual biologics [ 144 ]. With poor prediction of patient response rates to traditional or biological DMARD and the substantial risks associated with immune suppressive drugs [ 145 ], there is a critical need to develop treatment approaches that restore tolerance toward self-antigens in RA. RA results in damage and degradation of the cartilage, subchondral bone, and soft tissues of the joint [ 146 ]. There are currently no approved cell therapy products to regenerate bone or cartilage in RA patients, likely due to the progressive nature of the disease. If disease can be arrested, many of the existing bone [ 147 ] and cartilage [ 148 ] regenerative cell therapy-based approaches may be applicable in RA. Regenerative approaches in cartilage have been investigated for several decades, resulting in a few approved cartilage cell therapy products currently on the market in the United States (US), Europe, and Korea, including MACI ® (US, EU) [ 149 ], Spherox (EU), and CARTISTEM ® (Korea) [ 150 ]. Spherox and MACI products are composed of autologous chondrocytes obtained by biopsy of the patient’s tissue, readministered after ex vivo expansion and either reaggregation into spheroids or delivery using a porcine-derived collagen membrane, respectively ( Figure 3C - D ) [ 151 ]. As of 2018, there were 79 registered clinical trials using stem cells to regenerate cartilage, with the majority of cell sources coming from autologous bone marrow, cartilage, adipose tissue, and allogeneic umbilical cord [ 151 ]. While no stem-cell-based cell therapies are currently approved in the US, MSC-based therapies dominate the clinical trial field, and products such as CARTISTEM ® (allogeneic cord blood MSC), which received approval in South Korea in 2012 and plans a phase III clinical trial in the US to pursue FDA approval ( Figure 3E - F ). Delivery of cartilage cells within a biodegradable matrix is considered a third-generation method for implanting chondrocytes, with numerous products under investigation outside of the US using a range of matrices including collagen, fibrin, and hyaluronic acid [ 152 ]. While there are no approved bone regeneration cell therapy products, as of 2020, there were 25 completed or ongoing clinical trials investigating cell therapy for the regeneration of bone using autologous or allogeneic MSC, bone marrow, or osteoblast progenitors [ 153 ]. In celiac disease (CeD), chronic inflammation of the small intestine is caused by the ingestion of gluten [ 154 ]. CeD is prevalent worldwide (0.5–1% [ 14 ]), and genetic predisposition (HLA-DQ2 and 8) and environmental factors are major determinants [ 155 , 156 ]. This HLA phenotype is high risk for other AD, resulting in a high comorbidity of CeD with T1D and other autoimmune disorders. While most food-related hypersensitivities are immunoglobulin (Ig)E-mediated, in CeD, gluten and similar high proline and glutamine-containing proteins found in barley (hordeins), wheat (gliadins and glutenins), and rye (secalins) [ 154 , 157 ] form immunogenic polypeptides after activation by the enzyme tissue transglutaminase (tTG) [ 158 ]. The intestinal wall is covered with mucosa containing IgA, which protects the mucosa from infection [ 159 ]. Beneath the mucosa lie the enterocytes, with a crypt and villi structure for increased nutrient absorption, followed by the lamina propria, which contains antigen-presenting cells ( Figure 4A ) [ 160 ]. In CeD, gluten polypeptides traverse the lamina propria via paracellular and transcellular routes in the jejunum, causing cytotoxicity to the epithelia and aiding in the entry of the gluten proteins [ 161 ]. After activation by tTG, gluten antigens are presented by APCs to CD4 T cells, which activate B lymphocytes and stimulate autoantibody production in mesenteric lymph nodes and Peyer’s patches, specialized intestinal immune structures [ 162 , 163 ]. This immune reaction results in mucosal and intestinal villi destruction, reducing patients’ ability to absorb nutrients. The current gold standard therapy for CeD is a gluten-free diet (a diet with no wheat, rye, or barley proteins), and 66% of patients who successfully adhere to a gluten-free diet achieve complete histological recovery within 1 year [ 164 ]. However, gluten-free diet adherence can be challenging due to gluten contamination and its presence in many non-food commercial products, resulting in accidental exposure. As such, several nondietary therapies for CeD are being explored [ 165 ], such as methods to prevent gluten peptides from reaching the lamina propria [ 166 , 167 ], blocking the immune response by preventing tTG deamidation of gluten peptides and subsequent presentation to T cells [ 168 ], and approaches to induce tolerance toward gluten [ 169 ]. While there are no approved cell therapy products specific to CeD, there is an approved allogeneic MSC-based product in the EU targeted at patients with Crohn’s disease with perianal fistulas [ 170 ]. Additionally, there have been several preclinical studies exploring the regeneration of intestinal tissues using PSC that have implications for restoring normal function in AD affecting the intestines. Adult colonic stem cells [ 171 , 172 ], fetal intestinal progenitors [ 173 ], and ESC [ 174 , 175 ] are cultured as single cells or in matrices such as collagen [ 171 ], Matrigel [ 173 ], alginate [ 174 ], and synthetic hydrogels [ 175 ] to generate intestinal organoids prior to transplantation, where they demonstrate engraftment in murine colon injury models and recapitulation of the normal intestinal epithelium ( Figure 4B ). The development of these stem cell-based approaches to regenerate damaged intestinal tissues could be of great benefit to patients with advanced or refractory CeD; however, these regenerative cell therapy approaches may be of limited utility until the underlying AD can be arrested. Autoimmune thyroid diseases (AITD) are T-cell-mediated, thyroid-specific AD [ 176 ], with a cumulative incidence of 5% [ 16 , 177 ]. The two main clinical presentations are Graves’ disease and Hashimoto thyroiditis, and both are characterized by lymphocytic infiltration of the thyroid parenchyma [ 178 , 179 ], resulting in hypothyroidism in Hashimoto thyroiditis [ 180 ] and hyperthyroidism in Graves’ disease [ 181 ]. Anti-thyroid antibodies increase with age, peaking at 45–55 years, and AITD are much more prevalent in women (4–10 times the incidence in men), as thyroid self-antigens are located on the X-chromosome [ 182 , 183 ]. The major thyroid antigens are thyroglobulin (Tg), thyroid peroxidase (TPO), and thyroid-stimulating receptor; in Graves’ disease, autoantibodies are primarily generated against Tg, and in Hashimoto thyroiditis, autoantibodies are generated against TPO, leading to thyroid lesions, fibrotic changes, and parenchymal atrophy ( Figure 5A ) [ 184 ]. Hormone replacement therapy with levothyroxine is currently the gold standard for Hashimoto thyroiditis and other hypothyroidism conditions; however, it has a narrow therapeutic index, resulting in dosing challenges [ 185 ]. In Graves’ disease, inhibition of thyroid hormone production via antithyroid drugs is the first-line treatment, in conjunction with β-blockers, which aid in ameliorating adrenergic symptoms. For both Hashimoto thyroiditis and Graves’ disease, total thyroidectomy is another common strategy, and in Graves’ disease, it has comparable quality of life and cost to antithyroid drugs [ 186 - 189 ]. AITD are ideal for the development of antigen-specific immunotherapies given that there is a dominant self-antigen [ 189 ]. Patients undergoing thyroidectomy to treat AITD can manage hormonal deficiencies with hormonal replacement therapy; however, this approach has dosing challenges, as outlined above. As such, these patients may be good candidates for cell therapy approaches to replace damaged or excised thyroid cells. The past decade has seen the development of several approaches to generating thyroid cells from embryonic [ 190 - 195 ] and endodermal [ 196 ] stem cells. A few preclinical models of thyroid replacement with cell therapies have shown promise. Antonica et al. generated thyroid follicular cells from embryonic stem cells that assembled into functional thyroid follicles, and in vivo delivery into athymic mice demonstrated recovery of T4 hormone levels to pre-injury levels and relieved symptoms of thymus loss [ 195 ]. Similarly, Kurmann et al. delivered ESC-derived thyroid follicular organoids to athymic mice, which restored pre-injury T3, T4, and thyroid stimulating hormone levels. They additionally demonstrate the potential to use iPSC to generate thyroid follicular organoids, which has implications for using autologous cell sources or banked allogeneic cell sources ( Figure 5B ) [ 197 ]. While there are no approved cell therapy-based products for the treatment of AITD, there are allogeneic thymic tissue products approved for congenital athymia, where children are born without a functional thymus. Rethymic, approved by the FDA in 2021, consists of allogeneic thymus slices transplanted with the goal of restoring immune function [ 198 ]. Regulatory approval of allogeneic thyroid products is likely to ease the regulatory pathway for future cell therapy products to replace the dysfunctional thymus in AITD [ 190 ]. SLE and endometriosis are AD with systemic features and organ damage that are more difficult to target with a single regenerative cell therapy approach, but we include them here to describe their pathogenesis and impact. Both endometriosis and SLE patients experience significant delays in diagnosis due to heterogeneity in clinical manifestation and have a significant impact on patient quality of life [ 199 , 200 ]. SLE presents with multi-organ system involvement and a relapse-remitting course, with a broad spectrum of clinical features ranging from mild cutaneous lesions to severe organ damage to the kidney, heart, and lungs. It presents between the ages of 15 and 45, is 10 times more frequent in women, and its prevalence in individuals of African ancestry in the US is twice that of the Caucasian population [ 20 ]. Biomarkers used for diagnosis include anti-nuclear and anti-double-stranded DNA autoantibodies. The pathogenesis of SLE involves a complex interaction between genetic factors and environmental factors such as UVB radiation, infections (particularly Epstein-Barr virus), and environmental toxins [ 201 ]. Patients with SLE are at an increased risk of cardiovascular disease, likely due to circulating autoantibodies and immune complexes that damage the vascular endothelium [ 202 ]. SLE management is complex, requiring a multidisciplinary approach, and often involves corticosteroids and immune suppressants, though few patients experience long-term remission and many suffer from chronic remission-relapse [ 203 ]. With insufficient response to first-line treatment, monoclonal antibody treatments targeting B cells, such as belimumab and rituximab, are effective alternatives [ 204 ]. Endometriosis is a highly prevalent gynecological disease that impacts 1 in 10 women, resulting in chronic pain and infertility that results in health care costs estimated at $70 billion, comparable to T1D and RA [ 205 ]. The disease is characterized by the presence of endometrial tissue (glandular epithelium and stroma) outside the uterine cavity and uterine musculature [ 18 ]. These endometrial lesions can grow and invade abdominal tissues, potentially interfering with the function of the kidneys, bladder, colon, and intestines, and, in rare cases, extrapelvic organs such as the lungs, heart, and even brain. Endometriosis has only recently been recognized as an AD given the presence of autoantibodies, inflammatory cytokines, the therapeutic response of the disease to immunomodulators, and comorbidity with other ADs [ 206 ]. First-line treatment for endometriosis is primarily symptomatic management using nonsteroidal anti-inflammatory drugs for pain or hormonal treatment with contraceptives to prevent menstruation [ 18 ]. While hormonal management can help manage pain, hormonal treatment does not treat the underlying lesions, and the majority of patients do not respond to treatment; as such, surgical intervention is currently considered the most effective means of eliminating disease [ 207 , 208 ]. Novel therapies to treat endometriosis aim to target endometriotic lesion vasculogenic signaling and inflammatory pathways such as TNFα [ 209 ] and nuclear factor kappa B (NF-κB) signaling [ 210 ]. A recent preclinical study investigating rituximab showed a dramatic reduction in endometriosis lesion volume in rats [ 211 ], adding support to the premise that endometriosis is autoimmune in nature and that immunomodulatory strategies in treating other AD may be successfully applied in endometriosis. AD with multi-system organ damage such as endometriosis and SLE present a unique challenge and opportunity in the development of regenerative cell therapies, particularly given the prevalence of these AD. As the field of regenerative medicine develops approaches to reverse diseases of the kidneys, heart, lungs, blood vessels, and intestines [ 212 ], these approaches may be readily adapted to restore function in patients with severe SLE or endometriosis. As with other AD discussed above, it will be critical to address the underlying immune dysfunction in SLE and endometriosis to enable the application of regenerative cell therapies in these patient populations.