Stem cell therapy for endocrine and metabolic disorders

Abstract

Stem cell therapy is a highly promising avenue in regenerative medicine. In the domain of endocrine and metabolic diseases, it has shown therapeutic potential for conditions like hypothyroidism, hypopituitarism, primary adrenal insufficiency, osteoporosis, and diabetes and its complications. This treatment modality leverages the unique self-renewal and differentiation capabilities of stem cells into specific cell types, aiming to restore damaged tissue function and alleviate symptoms of hormonal imbalances, thus offering an alternative to long-term pharmacological treatments that often yield suboptimal results. Currently, research on the aforementioned diseases primarily concentrates on pluripotent and mesenchymal stem cells. Pluripotent stem cells primarily exert their therapeutic effects through cell replacement, making them the preferred choice for functional substitution of damaged cells in conditions like type 1 diabetes, hypothyroidism, hypopituitarism, and adrenal insufficiency. Conversely, mesenchymal stem cells offer distinct advantages via multifaceted mechanisms, such as anti-inflammatory, immune-regulating, and anti-apoptotic mechanisms, making them applicable to diseases rooted in inflammatory responses and immunoregulatory disturbances, encompassing autoimmune thyroiditis, primary adrenal insufficiency, osteoporosis, and diabetes and its complications. Nevertheless, stem cell therapy overall remains in early developmental stages, and its long-term safety and efficacy require further verification. For pluripotent stem cells, unresolved issues persist, including post-transplantation immune rejection, oncogenic potential, and ethical concerns. For mesenchymal stem cells, clinical benefit has been predominantly observed in early-stage patients with diabetes, and the optimal sources for distinct disease entities warrant further investigation. This review provides a comprehensive overview of the current therapeutic potential of stem cells in endocrine and metabolic diseases, the indications for various stem cell types, the challenges faced in their clinical translation, and the corresponding countermeasures.

Key Words: Stem cell therapy; Diabetes; Hypothyroidism; Hypopituitarism; Primary adrenal insufficiency; Osteoporosis

Core Tip: This review provides the first systematic comparison of pluripotent and mesenchymal stem cell strategies across major endocrine and metabolic disorders, diabetes, hypothyroidism, hypopituitarism, adrenal insufficiency, and osteoporosis. We establish a clear functional framework: Pluripotent stem cells-mediated cell replacement is applicable to diseases characterized by irreversible cell loss, whereas mesenchymal stem cell-mediated anti-inflammatory and immunomodulatory actions are applicable to inflammation- or immune-driven conditions. We also identify translational barriers and propose actionable solutions, offering a roadmap from proof-of-concept toward clinical practice.

INTRODUCTION

Endocrine and metabolic disorders, including thyroid diseases, pituitary and adrenal gland diseases, osteoporosis, and diabetes and its complications, affect hundreds of millions worldwide and impose a substantial global health burden. Despite decades of pharmacological advances, current therapies remain largely palliative, they control symptoms and correct metabolic derangements but fail to reverse the underlying cellular deficits or restore tissue function. Despite requiring lifelong insulin, patients with type 1 diabetes (T1D) face glycemic instability and complications; those with hypothyroidism, hypopituitarism, or adrenal insufficiency depend on chronic hormone replacement that cannot replicate physiological feedback; and osteoporosis treatments slow bone loss without fully restoring skeletal integrity.

Stem cell therapy offers a transformative alternative. By leveraging self-renewal, differentiation, immunomodulation, anti-inflammatory, and paracrine activities, stem cells hold the potential for true tissue regeneration rather than metabolic compensation[1]. Mounting clinical findings have also shown their therapeutic potential in endocrine and metabolic disorders. Despite recent progress, several critical gaps, such as immune rejection, tumorigenic risk, and a range of disease-specific challenges, hinder clinical translation. In this review, we provide a comprehensive, comparative overview of stem cell therapy for major endocrine and metabolic disorders, critically evaluate current evidence across diseases, assess mechanisms and applicability of different stem cells to various diseases, summarize remaining obstacles, and propose practical strategies, offering a roadmap from study findings to clinical translation.

METHODOLOGY

This structured review was conducted in accordance with established guidelines for reviews. A systematic literature search was performed in PubMed, Web of Science, and Scopus for articles published between 1998 and June 2026. The search strategy combined keywords related to stem cell types (“stem cell”, “mesenchymal stem cell”, “pluripotent stem cell”, “induced pluripotent stem cell”, “embryonic stem cell”) and endocrine/metabolic diseases (“diabetes”, “diabetic complications”, “diabetic retinopathy”, “diabetic kidney disease”, “diabetic nephropathy”, “diabetic neuropathy”, “diabetic foot ulcer”, “diabetic cardiovascular complications”, “diabetic cardiomyopathy”, “hypothyroidism”, “autoimmune thyroiditis”, “Hashimoto’s thyroiditis”, “hypopituitarism”, “primary adrenal insufficiency”, and “osteoporosis”). Additional relevant studies were identified by manually screening the reference lists of included articles, as well as by searching ClinicalTrials.gov for ongoing or completed clinical trials.

Articles were included if they met the following criteria: (1) Original research (including clinical trials, cohort studies, case series, and high-quality preclinical studies) or authoritative reviews with substantial mechanistic or contextual insights; (2) Studies focused on stem cell therapy for endocrine or metabolic disorders, specifically diabetes mellitus and its complications, hypothyroidism, autoimmune thyroiditis (AIT), hypopituitarism, primary adrenal insufficiency (PAI), or osteoporosis; (3) Articles published in English; and (4) Studies with sufficient methodological detail to allow critical appraisal.

Articles were excluded if they: (1) Were not directly relevant to the therapeutic application of stem cells in endocrine or metabolic diseases; (2) Were conference abstracts, editorials, commentaries, or letters without original data; (3) Were duplicate publications of the same study; or (4) Lacked sufficient methodological detail to permit meaningful interpretation. Reviews that did not present original data or novel synthesis were also excluded unless they provided critical mechanistic or contextual insights that complemented the primary literature.

CLASSIFICATION OF STEM CELLS

Based on their differentiation potential, stem cells can be classified into five main types: Totipotent stem cells, pluripotent stem cells (PSCs), multipotent stem cells, oligopotent stem cells, and unipotent stem cells[2] (Figure 1). Currently, the hotspots in stem cell therapy include human PSCs (hPSCs), comprising human embryonic stem cells (hESCs) and human induced PSCs (hiPSCs), and multipotent mesenchymal stem cells (MSCs).

Figure 1 Classification, differentiation hierarchy, and potency of stem cells. The figure was created using the Figdraw platform.

Totipotent stem cells

Totipotent stem cells have the differentiation potential to form a complete individual. The fertilized egg and its cells from the first two divisions are totipotent stem cells; these can differentiate into the inner cell mass and the trophectoderm of the blastocyst, forming the embryo and placenta[3].

PSCs

PSCs can proliferate indefinitely and differentiate into cells of the three germ layers (endoderm, mesoderm, and ectoderm), further developing into various organs and tissues throughout the body[4]. Embryonic stem cells (ESCs) develop from the inner cell mass of a blastocyst and belong to PSCs. In 1998, Thomson et al[5] successfully isolated hESCs for the first time. However, their use in researching disease treatments was fraught with ethical concerns and immune rejection issues. In 2007, Takahashi and Yamanaka[6] were the first to innovatively synthesize hiPSCs through somatic cell reprogramming. These cells shared similar characteristics with ESCs, barring the ethical issues. Consequently, they were quickly used in research related to regenerative medicine.

Multipotent stem cells

Multipotent stem cells originate from a single germ layer and are present in several human tissues. Mesoderm-derived MSCs, a specific type of multipotent stem cells, can be extracted from various tissues, such as bone marrow, bone, adipose tissue, Wharton’s jelly, umbilical cord blood, and peripheral blood[7]. Numerous studies have confirmed that MSCs can differentiate into mesoderm-derived tissues, such as adipose tissue, muscle, bone and cartilage[7]. Besides, it has been shown that MSCs can transdifferentiate into neural cells derived from the ectoderm and pancreatic cells derived from endoderm[8].

Oligopotent stem cells

Oligopotent stem cells can self-renew and differentiate into cells of two or more lineages in a given tissue. For example, hematopoietic stem cells, which belong to this category, can self-renew for a long time and differentiate into numerous types of mature blood cells[9].

Unipotent stem cells

Unipotent stem cells have strong self-renewal ability; however, the differentiation potential is minimal and unidirectional. For instance, muscle stem cells can only differentiate into mature muscle cells[10].

STEM CELL THERAPY FOR ENDOCRINE AND METABOLIC DISORDERS

Overall, the pathogenesis of endocrine and metabolic disorders primarily involves immune dysregulation, chronic inflammation, oxidative stress, and functional cell damage resulting from various etiologies. Currently, stem cell therapy predominantly employs PSCs and MSCs; the former exert cell and tissue replacement effects, whereas the latter act through immunomodulation and anti-inflammatory mechanisms, thus corresponding to distinct pathophysiological mechanisms of endocrine and metabolic diseases. This section elaborates on the advantages and challenges of these two stem cell types in treating different diseases, with an aim to provide a reference for the clinical translation of stem cell-based therapies for endocrine and metabolic disorders (Figure 2).

Figure 2 The role of stem cells in the treatment of endocrine and metabolic disorders. The figure was created using the Figdraw platform.

Diabetes and complications

T1D: T1D is defined by absolute insulin deficiency and lifelong insulin dependence; however, despite therapeutic progress, glycemic control remains inadequate. As insulin cannot regenerate beta cells, alternative or adjunctive curative therapies are imperative. Pancreas transplantation for T1D was hampered by surgical risks, rejection, and donor shortage[11]. Islet transplantation was a more targeted approach but had poor outcomes initially. The Edmonton regimen improved success, enabling 7 patients to discontinue insulin[12]. However, reliance on multiple donors, islet apoptosis, and rejection limited its widespread use[13]. The past decade has witnessed the advent of stem cell-based approaches for T1D, using hPSCs (hESCs and hiPSCs) and MSCs. These therapies have now entered clinical investigation, with 61 trials registered on ClinicalTrials.gov.

(1) hPSCs. Poor efficacy was noted in the first clinical trial of hESC-derived pancreatic endodermal cells (PEC-01) encapsulated in a macroencapsulation device made of semi-permeable polytetrafluoroethylene membrane and external polyester mesh. After 2 years, grafts contained few endocrine cells, and no insulin secretion was detected; moreover, the device was fibrosed and nonfunctional[14]. A modified device with vascularized channels allowed nutrient supply but lacked immune isolation; early data showed PEC-01 survival, differentiation into mature beta-cell phenotypic cells, and meal-stimulated C-peptide secretion (Table 1), but levels remained subtherapeutic[15]. Nevertheless, this study provided the first proof-of-concept that stem cell-derived islet cells can safely mount glucose-responsive insulin secretion in humans.

Table 1 The clinical trials using stem cells in the treatment of type 1 diabetes.

Ref.	Year	Type of SCs	No.	HbA1c	TIR	Fasting C-peptide	Postprandial C-peptide	Insulin dose
[15]	2021	hESCs	15	< 7% over all follow-up	Increasedb	Increaseda	Both glucose and meal-responsive C-peptide increaseda	Decreasedb (1/15 patients decreased of 50%)
[16]	2025	hiPSCs	14	< 7% after 120 days, decreased by 1.81 percentage points at 1 year (7.8% at baseline)	> 70% at day 150, increased by 44.4 percentage points at day 365 (49.5% to 93.9%)	Detected at day 90 (undetectable at baseline)	With an average value of 424 pmol/L at day 90 and 1274 pmol/L at day 365 (undetectable at baseline)	The average insulin dose decreased by 92%; 10/12 patients achieved insulin independence and did not resume through day 365; insulin dose was decreased for 2/12 patients
[17]	2024	hiPSCs	1	Decreased by 1.07 percentage points at day 75 (7.57% to 6.50%) and then remained within the non-diabetic range (< 5.7%)	Increased by 55.41 percentage points (43.18% to 98.59%)	Increased to 102.6 pmol/L after 1 week; increased to a plateau after achieving insulin independence, with a fluctuation range of 407.1 to 721.6 pmol/L (undetectable at baseline)	Increased by 20% over fasting on day 56; further improved to 834.1 pmol/L at day 120 (undetectable at baseline)	Began to decline after 2 weeks; achieved complete insulin independence on day 75 and remained so during the 1-year follow-up
[25]	2015	Autologous BM-MSCs	9	No significant difference between the MSC group and control group		No significant difference between the MSC group and control group	Maintained in 6/9 patients of the MSC group, but decreased in 8/9 patients in the control group	No significant difference between MCS and control groups
[26]	2022	Autologous BM-MSCs	11	Decreased by 1.46 percentage points compared with the control group at 1 yeara; the early treatment group had a significant reduction of 2.95b, 1.69a, and 1.40a percentage points relative to the late treatment group at 3, 6, and 9 months, respectively		Higher than that in the control group, but with no significant difference; the early treatment group had a significant increase of 0.58 ng/mLa, 0.46 ng/mLa, and 0.56 ng/mLa than the late treatment group at 3, 6, and 9 months, respectively	Higher than that in the control group, but with no significant difference	Increased in both groups, with no significant difference between two groups
[19]	2013	Allogeneic WJ-MSCs	15	Decreased by 1.3 percentage points (the lowest level) at month 6 (6.8% ± 0.57% to 5.5% ± 0.67%) and maintained a small fluctuation over the following 18 months		Higher than the control group, also higher than the baseline at the end of follow-up, whereas the control group showed a gradual downward trend		3/15 patients discontinued insulin; daily insulin dose was reduced by more than 50% in 8 patients and by 15%-50% in 1 patient
[27]	2016	Allogeneic BM-MSCs	5 (1 patient was lost to follow-up)	3/4 patients maintained a similar or lower level for at least 3 years		Declined slowly in all patients	1 patient rose above the baseline for as long as 1 year	Insulin dose was reduced by 50% in 2/4 patients; 1 patient achieved insulin independence for 3 months
[28]	2022	Allogeneic UC-MSCs and BM-MSCs	14 (4 of BM-MSCs, 10 of UC-MSCs)	Decreased both in BM-MSC and UC-MSC groups, while slightly increased in the control groupa; no significant difference in the reduction between BM-MSCs and UC-MSCs group		Clinical remission (10% increase from baseline in the fasting C-peptide and/or postprandial C-peptide levels) was achieved in 4/14 patientsc, including 1 patient (1/4, 25.0%) in the BM-MSC group and 3 patients (3/10, 33.3%) in the UC-MSC groupc		Reduced relative to the control groupa; no significant difference in the reduction between BM-MSC and UC-MSC groups
[29]	2020	Allogeneic AD-MSCs	8	Lower than control group (6.47% ± 0.86% vs 7.48% ± 0.52%)a	No significant difference between the MSC and control groups		No significant difference between the MSC and control groups	Lower than control group (0.25 ± 0.19 IU/kg vs 0.56 ± 0.23 IU/kg)a; 2/8 patients achieved insulin independence for 4 and 8 weeks, respectively
[30]	2015	Autologous/allogeneic AD-MSCs with hematopoietic stem cells	10 of autologous, 10 of allogeneic	Decreased in both groups with no significant difference		Increased in both groups; higher in the autologous group than in the allogeneic group (0.93 ± 0.24 ng/mL vs 0.46 ± 0.29 ng/mL)a		The mean insulin dose decreased by 37.93% in the autologous group and by 33.11% in the allogeneic group

^aP < 0.05.

^bP < 0.001.

^cP > 0.05.

No.: Number of stem cell-treated patients; SCs: Stem cells; HbA1c: Glycated hemoglobin A1c; TIR: Time in range; hESCs: Human embryonic stem cells; hiPSCs: Human induced pluripotent stem cells; BM-MSCs: Bone marrow-derived mesenchymal stem cells; MSC: Mesenchymal stem cell; WJ-MSCs: Wharton’s jelly-derived mesenchymal stem cells; UC-MSCs: Umbilical cord-derived mesenchymal stem cells; AD-MSCs: Adipose-derived mesenchymal stem cells.

Regardless of the type of the encapsulated device, the implanted PEC-01 needed further differentiation, resulting in low efficacy. VX-880, a cluster of fully differentiated islet cells derived from hiPSCs (termed SC-islets)[13], entered phase I/II trials in 2021 via intraportal infusion without immune protection, thus requiring immunosuppression. In all 12 patients who received a single full dose, engraftment and glucose-responsive insulin production were observed by day 90; all achieved ADA-recommended glycemic targets [glycated hemoglobin A1c (HbA1c) < 7.0% and time in range (70-180 mg/dL) > 70%], with 10 patients achieving insulin independence. No hypoglycemic episodes or treatment-related serious adverse events (SAEs) occurred; two deaths were attributed to unrelated causes (severe cryptococcal meningitis secondary to skull base injury and progressive neurocognitive impairment)[16]. Compared with PEC-01, VX-880 more robustly restores endogenous islet function and glycemic stability. As an hiPSC-derived product, it offers a scalable cell source, facilitating larger and longer clinical investigations.

Recently, a Chinese team achieved a breakthrough in T1D therapy using chemically induced autologous iPSCs derived from adipose-derived MSCs (AD-MSCs), which were differentiated into islet cells in vitro and transplanted under the abdominal anterior rectus sheath. At 1-year follow-up, the patient achieved glycemic targets and discontinued insulin (Table 1)[17]. However, the small sample size, short follow-up, and concomitant immunosuppression limit interpretation of immune rejection, thus necessitating larger, longer studies.

(2) MSCs. The pathogenesis of T1D involves the breakdown of immune tolerance to beta-cell antigens, allowing autoreactive T cells to evade endogenous control mechanisms. HLA class II alleles (DR3-DQ2 and DR4-DQ8) confer the strongest risk, whereas non-HLA loci impair T cell and regulatory T (Treg) cell function, compromising central/peripheral tolerance. Low insulin gene expression in the thymus compromises negative selection of autoreactive T cells, diminishing the central tolerance. In the periphery, Treg cells exhibit functional defects that compromise their ability to suppress conventional T cells. Neoantigens, such as hybrid insulin peptides generated by cathepsin D, introduce novel epitopes that elicit pathogenic CD4+ T cell responses. Enteroviruses may trigger autoimmunity via molecular mimicry and type I interferon-mediated HLA class I upregulation on beta cells, enhancing CD8+ T cell recognition. Autoreactive CD8+ T cells show a stem-like phenotype, enabling sustained beta-cell destruction[18]. These interconnected mechanisms, including genetic susceptibility, defective tolerance, neoantigens, viral mimicry, and T cell persistence, underpin progressive beta-cell loss and justify immunomodulatory therapies.

Unlike hPSC-derived islet cells, which directly secrete insulin, MSCs treat T1D via three primary mechanisms: Immune regulation, anti-apoptosis, and cell replacement. Immune regulation: MSCs inhibit autoreactive T-cell proliferation/activation to prevent beta-cell destruction[19,20], promote Tregs for immune tolerance[20,21], and secrete extracellular vesicles or anti-inflammatory factors for paracrine effects[22]. Anti-apoptosis: MSCs inhibit beta-cell apoptosis, thus increasing endogenous insulin secretion[22]. Finally, cell replacement: MSCs can differentiate into insulin-producing cells both in vitro and in vivo, based on their developmental plasticity[23,24].

When compared with controls, intravenous infusion of autologous bone marrow-derived MSCs (BM-MSCs) in patients with newly diagnosed T1D led to higher C-peptide levels following mixed-meal stimulation at 1-year follow-up, with no adverse events, suggesting the preservation of beta-cell function[25]. In addition, early transplantation (within 1 year of diagnosis) conferred greater HbA1c reduction than late transplantation, supporting the benefit of earlier intervention (Table 1)[26].

Allogeneic MSCs have also shown promise in T1D. In newly diagnosed patients, Wharton’s jelly-derived MSCs (WJ-MSCs) improved glycemic control, increased fasting C-peptide, and enabled 3 of 15 patients to discontinue insulin over 2 years[19]. In T1D patients with diabetic ketoacidosis, allogeneic BM-MSCs led to insulin independence in 1 of 4 patients (for 3 months) and dose reduction in 2 others, with sustained HbA1c improvement noted for ≥ 3 years[27]. A subsequent comparison of allogeneic umbilical cord-derived MSCs (UC-MSCs) and BM-MSCs revealed reductions in both glucose and insulin requirements, with no significant difference in efficacy between the two sources[28]. Autologous and allogeneic AD-MSCs were also effective, with the autologous AD-MSC/bone marrow-derived hematopoietic stem cell co-transplantation regimen showing superior results[29,30]. Collectively, early transplantation (within 1 year of diagnosis) and autologous MSCs appear more beneficial. However, to our knowledge, no study has systematically compared MSCs from different sources, and the overall evidence remains preliminary, necessitating further investigation.

hPSCs (hESCs and hiPSCs) and MSCs have opened new therapeutic avenues for T1D. hESCs face immune rejection and vector challenges; hiPSCs are ethically favorable but require immunosuppression, with safety in pediatric patients unknown. MSCs offer broad availability, favorable safety, and potent immunomodulation without immunosuppressants, showing promise in both pediatric and diabetic ketoacidosis patients. However, evidence remains largely confined to newly diagnosed patients with preserved islet function, whereas efficacy in patients with long-standing disease having low C-peptide counts remains to be established.

Type 2 Diabetes: Type 2 diabetes (T2D) pathogenesis involves insulin resistance and progressive beta-cell dysfunction. Unfortunately, none of the new drugs provide a cure, and patients in advanced stages continue to rely on exogenous insulin but face glycemic fluctuations and severe hypoglycemia risk. Chronic inflammation and immune dysregulation contribute to insulin resistance and beta-cell damage, making MSCs a promising therapy beyond cell replacement. MSCs exert anti-inflammatory effects by activating M2 macrophages (key immune regulators) to alleviate peripheral insulin resistance[31], and they directly reduce resistance in skeletal muscle, liver, and adipose tissue through the upregulation of the IRS-1/PI3K/AKT/GLUT4 pathway[32]. Immunomodulation, as detailed for T1D, is also a crucial mechanism.

Clinical trials of MSCs from various sources for T2D are ongoing. Autologous BM-MSCs have shown safety, stable glycemic control, and improved insulin sensitivity without significantly enhancing endogenous islet function[33]; the efficacy was higher in patients with a disease duration of < 10 years; this may be attributable to MSC dysfunction arising from chronic hyperglycemia, inflammation, and oxidative stress in patients with a prolonged disease course[34]. Placenta-derived MSCs reduced HbA1c and insulin needs while increasing C-peptide[35]. WJ-MSCs improved glycemic control and islet function for a longer time, enabling reduction or discontinuation of medications in some patients[36]. UC-MSCs alleviated insulin resistance and reduced HbA1c and insulin requirements, though beta-cell function recovery was not observed (Table 2)[37].

Table 2 The clinical trials using stem cells in the treatment of type 2 diabetes.

Ref.	Year	Type of SCs	No.	HbA1c	TIR	Fasting C-peptide	Postprandial C-peptide	Diabetic medication dose	Other indicators
[33]	2017	Autologous BM-MSCs	10	6/10 patients reduced below 7%, but no significant improvement relative to baseline		No significant improvement relative to baseline	No significant improvement relative to baseline	6/10 patients achieved a > 50% reduction in insulin dose relative to baseline	ISI: Had a significant improvement (0.1 to 0.4 μmol/kg/minute/pmol/L)a
[34]	2021	Autologous BM-MSCs	30	Patients with duration of < 10 years and a BMI of < 23 kg/m2 showed a reduction of 0.7 percentage points (from 8.9% ± 1.6% to 8.2% ± 0.5%)a; no significant change (even increase) in patients with duration of > 10 years		Patients with duration of < 10 years and a BMI of < 23 kg/m2 showed a slight decreasec, whereas those with duration of > 10 years showed a significant declinea		2/4 patients using only insulin had an average reduction of 6 IU/day; 5/10 patients using multiple drugs had an average reduction of 868 mg/day; 1/3 patients using insulin and a single drug had a decrease in drug dose but an increase in insulin dose; 8/11 patients using insulin and multiple drugs: 5 of these 8 patients showed reduced use of oral drugs, and 2 patients had a decreased insulin dose
[35]	2011	Allogeneic PD-MSCs	10	Decreased by 3.1 percentage points (9.8% to 6.7%)a		Increased by 34.62% (2.6 ± 2.1 ng/mL to 3.5 ± 2.3 ng/mL)a	Increased by 31.71% (5.6 ± 4.0 ng/mL to 8.2 ± 3.9 ng/mL)a	Decreased by 45.53% (63.7 ± 18.7 to 34.7 ± 13.4 IU)a; insulin dose was reduced for 4/10 patients by > 50%
[36]	2014	Allogeneic WJ-MSCs	22	Decreased by 1.2 percentage points (8.20% ± 1.69% to 7.0% ± 0.60%)a		Increased to the peak value at month 6 (1.95 ± 1.3 ng/mL, increased by 51.16%)a, followed by slight decrease at month 12 (1.86 ± 1.0 ng/mL, which still represents 44.19%a increase relative to baseline (1.29 ± 0.83 ng/mL)	No significant difference compared to baseline	(1) 16/17 patients using insulin had a dose reduction of 53.06% (0.49 ± 0.22 to 0.23 ± 0.19 IU/kg/day)a, 7 patients achieved insulin independence, which was maintained until the last follow-up; 5 patients had a dose reduction of ≥ 50%; and (2) 1/5 patients using drugs became completely drug-free, which was maintained until the last follow-up, 4/5 had a dose reduction of > 50	HOMA-2B: Increased by 49.81%a at month 6 (98.86% ± 43.91%) and 30.32%a at month 12 (86.0% ± 37.1%) (65.99% ± 23.49% at baseline)
[37]	2022	Allogeneic UC-MSCs	45	Decreased 1.5 percentage points (9.02% ± 1.27% to 7.52% ± 1.07%)a in the UC-MSC group, and 0.7 percentage points in the control group (8.89% ± 1.11% to 8.19% ± 1.02%)c; with a significant difference noted between the two groupsa		No significant increase compared with the baseline (2.01 ± 0.70 to 2.07 ± 0.70 ng/mL)c; no significant difference between the two groups either	No significant change compared with the baseline (1.94 ± 0.61 to 1.92 ± 0.53 ng/mL)c; no significant difference between the two groups either	The percentage of insulin reduction was significantly higher in the UC-MSC group than in the control group (27.78% vs 15.62%)a	GIR: Increased by 52.56% in the UC-MSC group (3.12 to 4.76 mg/minute/kg)a, but no significant change was noted in the control group (3.26 to 3.60 mg/minute/kg)c; a significant difference was noted between the two groupsa
[38]	2024	hiPSCs	1	Decreased by 2.0 percentage points (6.6% to 4.6%)	(Target range: 3.9-10 mmol/L) increased by 12.3 percentage points (87.7% to 100%)	Increased to a peak of 1.09 nmol/L at week 24 (increased by 374%, 0.23 nmol/L at baseline) and then decreased but remained within the normal range	AUC·180 minutes of C-peptide: Increased by 49.64% (246.8 to 369.3 nmol/L)	Insulin and oral medications were discontinued at week 11 and 56, respectively, and were maintained until the last follow-up	TITR (target range: 3.9-7.8 mmol/L): Increased by 43.3 percentage points (56.7% to 100%)

^aP < 0.05.

^cP > 0.05.

No.: Number of stem cell-treated patients; SCs: Stem cells; TIR: Time in range; BM-MSCs: Bone marrow-derived mesenchymal stem cells; ISI: Insulin sensitivity index; BMI: Body mass index; PD-MSCs: Placenta-derived mesenchymal stem cells; WJ-MSCs: Wharton’s jelly-derived mesenchymal stem cells; UC-MSCs: Umbilical cord-derived mesenchymal stem cells; GIR: Glucose infusion rate; hiPSCs: Human induced pluripotent stem cells; AUC: Area under the curve; TITR: Time in tight range.

Although MSC-based therapy for T2D shows potential, it is largely confined to patients with preserved islet function and offers limited disease-modifying capacity. Recently, a Chinese team reported that autologous endoderm-derived islet tissue (E-islets) enabled achieved insulin independence for 33 months in a T2D patient with 25-year disease duration and beta-cell exhaustion. Patient-derived peripheral blood monocytes were reprogrammed into iPSCs, differentiated into non-tumorigenic endodermal stem cells, and further matured into E-islets in vitro, resembling cadaveric islets in morphology, composition, and function. Following intrahepatic portal infusion, glycemic control was restored without anti-diabetic medications, and no severe glycemic events or tumor formation occurred during follow-up (Table 2)[38].

Taken together, MSCs improve glycemic control and insulin sensitivity in early/intermediate T2D but cannot restore beta-cell function in advanced disease. Autologous E-islet transplantation has achieved insulin independence and functional recovery in end-stage T2D, representing a paradigm-shifting approach. However, this strategy remains investigational, and large-scale studies are required to confirm its efficacy and safety.

Diabetic retinopathy: Diabetic retinopathy (DR) is among the most common microvascular complications of diabetes and a leading cause of blindness, significantly impacting global health. DR includes three forms, namely non-proliferative DR (NPDR), proliferative DR (PDR), and diabetic macular edema. Currently, laser photocoagulation, local injection of anti-vascular endothelial growth factor (VEGF) antibodies, and vitreoretinal surgery are the main clinical management approaches. However, these methods still have some limitations. For example, laser photocoagulation does not improve vision and may even lead to long-term complications like visual field defects. Treatment with anti-VEGF antibodies is expensive and carries risks of infection and tractional retinal detachment. Surgical treatment is typically only for patients with PDR and is ineffective for diabetic macular edema. Therefore, a safe and effective method is urgently needed to prevent and treat DR.

In recent years, the efficacy of cell-based therapy for DR has been proven in several animal experiments, including the use of MSCs, endothelial progenitor cells (EPCs), ESCs, and adipose-derived stromal cells, which play a role in preventing neurovascular damage and promoting retinal regeneration[39,40]. Among the few clinical studies carried out on stem cell therapy for DR, some have shown treatment potential. Park et al[41] first administered autologous bone marrow-derived CD34+ stem cells via intravitreal injection for the treatment of ischemic and degenerative retinal disorders; after a 6-month follow-up, they observed improvement in visual function and imaging in a subset of patients, with no serious safety concerns. Over a 6-month follow-up period, Gu et al[42] found that intravenous infusion of autologous BM-MSCs for DR treatment was safe and feasible, significantly improving best-corrected visual acuity and resolving macular edema in patients with NPDR; however, it showed limited efficacy in patients with PDR, suggesting that the severe NPDR stage represents a more optimal therapeutic window. Currently, research on the efficacy and safety of hiPSCs and allogeneic UC-MSCs for the treatment of DR is ongoing, which may yield more valuable results (https://www.clinicaltrials.gov/).

Moreover, it is notable that in recent years, with the deepening research on exosomes, researchers have identified that exosomes derived from stem cells also hold promising therapeutic significance for DR, involving the role of anti-oxidative stress, anti-inflammatory effects, neuroprotection, reduction of cell apoptosis, and inhibition of angiogenesis. Compared with stem cell transplantation, exosomes show reduced immunogenicity and superior tissue targeting, conferring prominent advantages, making them a promising approach for the treatment of DR.

Although current DR treatments remain suboptimal, stem cell-based therapies (particularly MSCs) and stem cell-derived exosomes have transformative potential. Exosomes hold distinct advantages for retinal repair; however, rigorous clinical validation is essential to establish their role in preventing vision loss and reversing DR pathology.

Diabetic kidney disease: Diabetic kidney disease (DKD) is also a common microvascular complication of diabetes. Even when blood glucose levels in patients with diabetes are controlled close to the optimal range, 25%-40% of them still develop DKD, which is a major contributor of end-stage renal disease globally[43]. Currently, the treatment for DKD is still limited to delaying progression by controlling risk factors (e.g., blood glucose, blood pressure, and blood lipids) and symptomatic treatment using medications. Therefore, researchers turned their attention to stem cells.

Research on stem cell therapy for DKD has primarily focused on MSCs because of their wide range of applications, including immunomodulatory, anti-inflammatory, antioxidant, antifibrotic, and regenerative abilities. In particular, MSCs can secrete exosomes through paracrine action. These characteristics confer upon MSCs the ability to repair the damaged nephrons[44]. Preclinical studies on MSCs derived from various sources, including bone marrow, placenta, umbilical cord, Wharton’s jelly, and adipose tissue, are being widely conducted, and their therapeutic effects in patients with DKD have been confirmed. In terms of clinical studies, results of some studies have been reported. A clinical study evaluating allogeneic mesenchymal precursor cells for the treatment of moderate-to-severe DKD demonstrated that a single intravenous infusion yielded efficacy signals indicating stabilized or improved renal function at 12 weeks, with favorable tolerability and no evidence of immune sensitization, thus supporting larger, longer-term clinical trials to further validate its therapeutic efficacy[45]. Besides, infusion of autologous bone marrow-derived mononuclear stem cells (BM-MNCs) significantly reduced urinary microalbumin levels in patients with DKD (an 83.2% decrease, P = 0.021) and led to a short-term decrease in urinary neutrophil gelatinase-associated lipocalin levels at 3 days post-infusion, suggesting a potential direct renoprotective effect[46]. The NEPHSTROM trial evaluated the efficacy and safety of BM-MSCs (ORBCEL-M) for the treatment of DKD. The results suggested that ORBCEL-M treatment is safe with good immune tolerance. Although 4/12 patients receiving ORBCEL-M treatment experienced SAEs, including two deaths, the investigators believed these SAEs were not related to the use of stem cells in the trial. One patient died of congestive heart failure and the other of newly diagnosed multiple myeloma. The evaluation of efficacy awaits feedback from the results of the second and third cohorts receiving higher doses in the study[47]. In a recent randomized, double-blind, placebo-controlled trial involving patients with T2D-related CKD stages II-IV, a single intravenous infusion of BM-MSCs significantly improved glomerular filtration rate at 6 months. The therapeutic effect was independent of disease duration and background medications (angiotensin-converting enzyme inhibitors, angiotensin receptor inhibitor, or SGLT2 inhibitors), and the treatment was well tolerated with a favorable safety profile[48]. Besides, exosomes exert effects similar to those of MSCs, such as anti-oxidation, anti-inflammation, anti-fibrosis, promotion of proliferation and inhibition of apoptosis, which have undergone extensive research for DKD treatment[49]. Basic research on DKD has found that exosomes derived from various MSC sources can upregulate or downregulate target genes through various pathways to alleviate podocyte damage, protect renal tubular epithelial cells, inhibit renal fibrosis, regulate autophagy and cell cycle, and ultimately exert a renoprotective effect[50,51]. Although the research on MSC-Exo has not yet entered the clinical research stage, its role in the human body may be uncovered in the future.

MSCs exert renoprotective effects by repairing damaged nephrons. Extensive preclinical studies and two randomized controlled clinical studies have provided important evidence supporting its efficacy and safety. However, currently, there are relatively few clinical cases, and they have a relatively short follow-up duration. Whether MSCs can play a long-term and stable role in persistent renal damage in patients with diabetes still requires larger-scale studies with a longer follow-up period. In addition, in recent years, exosomes have emerged as a high-potential derivative therapy, making clinical validation even more crucial.

Diabetic neuropathy: Neuropathy is the most common and earliest complication of diabetes, affecting approximately half of all patients with diabetes. Diabetic peripheral neuropathy (DPN) is the most prevalent type of diabetic neuropathy (DN), accounting for 75% of cases, and it is the main focus of this section. Currently, there is a lack of effective treatments for diabetic nerve damage, and stem cell therapy may emerge as a new and effective strategy. Since DN is a neurodegenerative disease, MSCs emerge as an ideal choice.

Preclinical studies have explored the efficacy of MSCs derived from dental pulp, bone marrow, umbilical cord, adipose tissue, and placental tissue in DPN treatment. The researchers analyzed motor nerve conduction velocity (MNCV), sensory nerve conduction velocity (SNCV), compound muscle action potential, blood flow in the sciatic nerve, intra-epidermal nerve fiber density, capillary-to-muscle fiber ratio of skeletal muscles, and biochemical indicators. The results have consistently shown improvements in nerve function and increased blood flow after treatment with MSCs[52,53].

Alizadeh et al[54] recently conducted a systematic review and meta-analysis of human studies on the efficacy and safety of stem cell therapy for DPN, and the results showed promising efficacy. The analysis revealed: (1) Improvement in nerve function: Improved MNCV and SNCV (increased by 2.2 m/second and 1.9 m/second, respectively); and (2) Symptomatic relief: Significantly reduced vibration perception threshold and the Toronto Clinical Scoring System scores (by 2.9 and 3.6, respectively). Toronto Clinical Scoring System, a tool used to assess DN, scores patients based on their symptoms, reflexes, and sensory test results. A score of < 5 indicates no neuropathy, whereas a score of > 5 indicates neuropathy. Scores of 6-8 indicate mild neuropathy, 9-11 indicate moderate neuropathy, and 12-19 indicate severe neuropathy. In terms of complications, only mild pain and swelling at the injection site were noted, which resolved spontaneously within a few days, demonstrating good safety. The study also found the two types of stem cells to offer distinct therapeutic advantages. BM-MNCs were more effective in improving sensory nerve function (SNCV/vibration perception threshold), whereas UC-MSCs more effectively improved motor nerve function (MNCV), providing novel avenues for individualized treatment of different patients.

Stem cell therapy is positioned to redefine DPN treatment, offering an evidence-based approach to reverse nerve damage. Several relevant clinical studies are currently underway. Strategic deployment of cell-specific advantages will accelerate its transition from an investigational breakthrough to clinical standards of care, particularly in achieving precise, individualized treatment. Ultimately, realizing this therapeutic potential depends on the accumulation of more robust clinical data.

Diabetic foot ulcer: Diabetic foot ulcer (DFU) is a severe complication of diabetes, leading to high amputation rates and mortality, posing a serious threat to public health, and imposing a heavy economic burden. Treating DFU requires comprehensive management measures, such as blood glucose management, nutritional support, infection control, improvement of blood circulation, wound care, and wound repair, with wound repair being a critical step in wound healing. Current wound repair approaches rely on wound dressings, negative pressure drainage, skin substitutes, skin grafts, and flap transplantation. However, these traditional treatments do not achieve satisfactory results and fail to achieve revascularization and nerve reconstruction.

Stem cell therapy has attracted the attention of researchers as a promising treatment modality for DFU. The therapeutic mechanisms of MSCs in the treatment of DFU include the following: (1) Angiogenesis: MSCs can release various cytokines, such as epidermal growth factor, VEGF, insulin-like growth factor-1, keratinocyte growth factor-2, basic fibroblast growth factor, stromal cell-derived factor-1, and placental growth factor-2, while promoting the activation of vascular endothelial cells, thus facilitating angiogenesis, increasing local blood flow, and promoting ulcer healing[55]; (2) Regulating immunity and inflammatory response: MSCs can modulate T cells and macrophages that mediate inflammation and immunity, thus promoting the clearance of pathogens and necrotic tissue at the ulcer site in the early stage and facilitating tissue repair in the later stage[56]; (3) Inhibition of oxidative stress: MSCs play a role by reducing reactive oxygen species-producing enzymes, such as myeloperoxidase, as well as downregulating inflammatory factors like interleukin (IL), interferon-γ, and tumor necrosis factor-α[57]; and (4) Transdifferentiation: MSCs can transdifferentiate into epithelial cells and endothelial cells, thus stimulating angiogenesis and promoting wound healing[58].

Recent years have witnessed the emergence of clinical researches on the treatment of DFU, primarily focusing on AD-MSCs, BM-MSCs, and UC-MSCs. A clinical research on the use of an allogeneic AD-MSC-hydrogel composite for the treatment of DFU found that the proportion of wounds that completely healed was significantly higher in the treatment group than in the control group (treated with polyurethane film) (73% vs 47% at 8 weeks and 82% vs 53% at the end of the 12-week observation period), with no treatment-related SAEs noted[59]. Besides, bone marrow- and peripheral blood-derived monocytes have shown stem cell properties and may support DFU healing through anti-inflammatory effects, promoting epithelial regeneration, increasing collagen deposition, and facilitating blood vessel formation. Monocytes derived from both sources have shown efficacy in DFU treatment. During the follow-up period, the ulcer healing rate was evidently higher in the cell therapy group than in the control group (56% vs 16.7%, P = 0.0093), and the amputation rate in the former was apparently lower than that in the conservative treatment group (11.1% vs 50%, P = 0.009). When comparing monocytes from different sources, no significant differences were found in terms of safety and efficacy[60]. Another study compared the efficacy of BM-MSCs and BM-MNCs in DFU. The results showed a 100% ulcer healing rate with no SAEs in both groups, likely attributable to the short duration of ulcers in the patients enrolled in this study. However, the BM-MSC group showed signs of healing before than the BM-MNC group and also achieved 100% healing rate 4 weeks earlier than the latter (8 weeks vs 12 weeks). This is attributable to BM-MSCs releasing more VEGF (2.3-fold, P < 0.001) and basic fibroblast growth factor (2.0-fold, P < 0.001) than BM-MNCs after transplantation[61].

Cell therapies can help address the core pathophysiology of DFU, enabling vascularization, inflammation control, and tissue regeneration, which are otherwise not possible with conventional approaches. Available clinical results for cell therapies confirm higher healing rates, reduced amputations, and good safety. Despite the constantly deepening attempts at stem cell therapy for DFU, it remains unknown whether there are differences among different types of stem cells in the treatment of DFU and whether early treatment indeed improves the treatment effect.

Diabetic cardiovascular complications: Long-term hyperglycemia in patients with diabetes can lead to sustained oxidative stress and inflammatory responses in the body, causing damage and dysfunction to cardiomyocytes and vascular endothelium, impairing neovascularization, and causing abnormal vascular remodeling, myocardial fibrosis, and apoptosis, ultimately leading to cardiovascular complications[62,63]. Cardiovascular complications in patients with diabetes are the primary driver of mortality, accounting for > 80% of diabetes-related deaths and severely impacting the quality of life and lifespan of diabetic individuals[64]. Although existing treatments like medication therapy and vascular reconstruction surgery can partially ameliorate symptoms, they cannot effectively reverse the tissue damage that has already occurred and have certain side effects and limitations. Stem cell-based therapy, through repair or replacement of damaged myocardial and vascular tissues, represents a promising approach for revascularization and restoration of cardiac function.

Over the past decade, research has been conducted on the use of MSCs for the treatment of diabetic cardiovascular complications. Scholars initially believed that the therapeutic mechanism involved MSCs transdifferentiating into cardiomyocytes or vascular endothelial cells after transplantation to functionally replace the damaged cells. However, subsequent research revealed that this direct effect was minimal, demonstrating that the paracrine effects of MSCs play a central role in the treatment of cardiovascular diseases. Paracrine factors were crucial in neovascularization, cell protection, and endogenous cardiac regeneration[62]. Liu et al[65] investigated the efficacy and underlying mechanisms of human UC-MSCs regarding endothelial injury in a diabetic rat model. The results showed that UC-MSCs not only reduced blood glucose levels but also exerted a protective effect against diabetes-induced endothelial injury through a MAPK/ERK signaling pathway-mediated paracrine mechanism. AD-MSCs also exhibited cardioprotective effects. Yu et al[32] transplanted AD-MSCs into diabetic rats, and after 24 weeks of treatment (once weekly), echocardiography results revealed improved ejection fraction in rats of the cell-treated group. Histological examination indicated that MSC treatment alleviated myocardial hypertrophy and significantly resolved diabetes-induced myocardial fibrosis. Similarly, BM-MSCs reportedly exert cardioprotective effects in diabetic rat models, demonstrating an ability to restore heart rate, left ventricular developed pressure, and left ventricular contractility index. In addition, they prevent excessive accumulation of Ca²⁺ in vascular endothelial cells, thus alleviating endothelial cell damage[66].

Diabetic cardiomyopathy (DCM), another cardiovascular complication associated with diabetes, is characterized by myocardial fibrosis and cardiac dysfunction that occur independently of hypertension, coronary heart disease, and heart valve disease. Its pathophysiological processes involve a series of complex courses, containing cell death, alterations in energy metabolism and extracellular matrix, inflammation, oxidative stress, and neurohormonal activation[63]. MSCs also play a part in DCM therapy through paracrine actions. BM-MSCs can target and migrate to myocardial tissue. Through gene modification, they reportedly express potent anti-inflammatory factors (adiponectin), decreasing myocardial cell apoptosis through its anti-inflammatory effects[67]. It reportedly alleviated metabolic abnormalities, myocardial fibrosis, and cardiac dysfunction in diabetic rats treated with AD-MSCs. Combined with in vitro experiments, the results revealed that MSCs alleviate fibroblast proliferation and collagen synthesis via the COX-2-PGE2 pathway. The secretion of prostaglandin E2 by MSCs is a potential mechanism underlying their alleviation of myocardial fibrosis in DCM[68]. Moreover, in patients with diabetes, the decreased number of EPCs and impaired angiogenic function are important causes leading to DCM development. Therefore, EPC transplantation may offer cardiac benefits to patients with diabetes[64]. In DCM rats with significantly impaired cardiac systolic and diastolic function, after treatment with bone marrow-derived EPCs, the above indicators were significantly improved after 4 weeks, and fibrosis was reduced with a decrease in type I collagen expression[69].

These preclinical studies have shown the potential of stem cells in treating cardiovascular complications of diabetes. Furthermore, previous clinical studies have also upheld the efficacy and safety of MSCs in treating cardiovascular diseases[70,71]; however, to our knowledge, clinical research on cardiovascular complications in the context of diabetes has not yet been conducted (https://www.clinicaltrials.gov/). Thus, there is still a long way to go before stem cell therapy can be applied to cardiovascular complications of diabetes.

Hypothyroidism

Hypothyroidism is characterized by reduced systemic metabolism resulting from decreased production or impaired action of thyroid hormones. In severe cases, coma due to myxedema may occur and may be life-threatening. Hypothyroidism is a globally widespread disease, with 99% of cases being primary hypothyroidism. The existing treatment for primary hypothyroidism is thyroid hormone replacement. However, many issues persist in the clinical management of this condition. First, hormone replacement therapy requires lifelong medication and regular follow-ups. Second, when used concomitantly with L-T4, several medications can interact with it and adversely affect therapeutic efficacy (e.g., estrogen and glucocorticoids). Third, as some foods may also interact with L-T4, there are strict requirements for the timing of L-T4 administration. Finally, according to the specific situation, the dosage needs to be adjusted for special population, such as the elderly, children, and pregnant women.

Therefore, as early as the late 19^th century, it was proposed that using thyroid transplantation to improve the symptoms of myxedema caused by hypothyroidism[72]. However, subsequent detailed studies identified many complications and ethical issues associated with organ transplantation, causing a gradual decline of research on this front. Since the 21^st century, with the rise of regenerative medicine, stem cell-based therapy has provided new possibilities for hypothyroidism treatment. Many researchers have successfully generated thyroid tissue capable of secreting thyroid hormones from stem cells in vitro, primarily by inducing overexpression of Pax8 and NKX2-1 in hESCs, two transcription factors that play crucial roles in the maturation and differentiation of thyroid follicular cells, and generating functionally mature thyroid follicles under thyroid stimulating hormone stimulation[73]. Romitti et al[74] were the first to report the transplantation of hESC-derived thyroid follicles into thyroid-ablated mice, demonstrating thyroid hormone secretion and providing in vivo evidence supporting cell replacement therapy. In another study, human amnion-derived MSCs (AMSCs) were injected into elderly female mice with subclinical hypothyroidism, and follow-up results showed improvements in thyroid hormones, blood lipids, and cardiac function in the transplanted mice. Simultaneously, serum immune indicators and thyroid cell apoptosis in mice were observed, and significantly elevated levels of T helper 1 and Treg cells were noted in the peripheral blood of mice, alongside a decline in T helper 2 and T helper 17 cell counts. Furthermore, thyroid cell apoptosis was also inhibited, collectively suggesting that AMSCs can treat subclinical hypothyroidism in elderly patients via immunomodulatory and anti-apoptotic effects[75].

AIT is among the main causes of primary hypothyroidism. Thyroid cells are prone to immune attacks, which lead to dysfunction, characterized by an inability to sufficiently synthesize and secrete thyroid hormones, ultimately resulting in hypothyroidism. The most common form of AIT is Hashimoto’s thyroiditis (HT). For patients with HT with no abnormal thyroid function, thyroid hormone replacement therapy has no preventive effect and cannot halt the autoimmune damage and inflammatory process. Some studies have suggested that selenium supplementation is effective in this process; however, there is no consensus on the extent of efficacy and the underlying mechanisms[76]. Owing to their immunomodulatory and anti-inflammatory abilities, MSCs are expected to be invaluable in the treatment of AIT. Thyroid tissue from patients with HT shows increased oxidative stress and endoplasmic reticulum stress levels, an increased proportion of M1 macrophages, and activation of the STING pathway. Upon transplanting BM-MSCs into rats with experimental AIT (EAT), a decrease in ROS level in the thyroid tissue was observed. In addition, by inhibiting the STING pathway and its downstream signaling processes, the level of endoplasmic reticulum stress in the thyroid tissue was reduced, and M1 macrophage polarization was decreased, thus protecting the thyroid tissue from autoimmune damage[77]. UC-MSCs have also shown therapeutic potential by reducing the proportion of IL-17α+/CD25+ FOXP3+ cells in the thyroid tissue and the serum IFN-γ/IL-4 ratio in EAT rats, as well as through the PTPN2/STAT3 pathway, thus alleviating thyroid inflammation and CD4+ T cell imbalance in EAT and ultimately exerting therapeutic effects on AIT[78]. Furthermore, the combined application of AD-MSCs and Na₂SeO₃ was effective in repairing thyroid tissue and restoring thyroid function in rats with HT. After thyroid function tests, FT3, FT4, TPO-Ab, and TG-Ab levels were significantly reduced in both the AD-MSC group and AD-MSCs + Na₂SeO₃ group, with the AD-MSC + Na₂SeO₃ group showing a more significant decrease (except for TG-Ab). Compared with the Hashimoto control group, FT3, FT4, and TPO-Ab decreased by 38.74%, 42.72%, and 35.61%, respectively, in the AD-MSC group and by 49.34%, 56.33%, and 51.00%, respectively, in the AD-MSC + Na₂SeO₃ group. Structurally, in the AD-MSC group, the thyroid structure was almost completely restored with little lymphocyte infiltration, whereas in the AMSC + Na₂SeO₃ group, the thyroid structure was completely restored to normal, and CM-Dil-labeled AD-MSCs had migrated into the thyroid, as confirmed by fluorescence microscopy. In vitro experiments showed that Na₂SeO₃ enhanced the proliferation and cytokine release of AD-MSCs[79].

The cell replacement function of ESCs and the anti-inflammatory and immunomodulatory properties of MSCs provide feasibility for stem cell therapy in treating hypothyroidism and AIT, which has been confirmed in animal experiments. Nevertheless, some challenges remain. Clinical studies have not yet been conducted, resulting in a lack of critical data. Whether stem cell-induced improvement of hypothyroidism can facilitate complete discontinuation of drug replacement therapy remains unknown, and its efficacy in patients having HT with a longer disease history also needs further exploration through clinical research.

Hypopituitarism

The pituitary gland is an important neuroendocrine organ in the human body; it regulates processes like growth, development, metabolism, reproduction, stress response, and homeostasis by secreting various hormones and is structurally divided into the anterior and posterior lobes. The posterior lobe stores and releases oxytocin and vasopressin, whereas the anterior lobe contains various endocrine cells that synthesize and secrete several endocrine hormones, including adrenocorticotropic hormone (ACTH), thyroid stimulating hormone, gonadotropins [including luteinizing hormone and follicle-stimulating hormone], growth hormone and prolactin. These hormones then exert their effects on their respective target organs. Given its far-reaching significance throughout the human body, damage to the pituitary gland will result in systemic clinical symptoms, such as hypotension, hypoglycemia, electrolyte imbalances, growth and developmental delay, infertility, and even consciousness disorders. There is currently no cure for pituitary hypofunction, and patients must rely on lifelong hormone replacement therapy. However, due to variations in patient compliance and the need for dynamic adjustment of hormone doses based on individual conditions, cases of over- or under-replacement are common in clinical practice. Therefore, if cells or tissues with pituitary hormone secretion function can be resynthesized and implanted into the body, the recovery of pituitary function can be achieved.

Pituitary stem cells reside within pituitary tissue. After pituitary resection, an increase in chromophobe cells, named for their resistance to histological staining, has been observed in the remaining tissuey[80]. After transplanting them into rats with pituitary gland removed, they could differentiate into eosinophils that produce growth hormone and prolactin, as well as into basophils that produce ACTH and gonadotropins. These results suggest pluripotency of chromophobe cells. Although initially considered as pituitary stem cells, they have become candidates for achieving pituitary replacement therapy[80]. However, in-depth research and application have been hindered by the lack of typical molecular markers for these cells[81].

Subsequent studies found that SOX2, a pluripotency-inducing factor, not only served as a marker for pituitary stem cells but also played a crucial role in maintaining their pluripotency[82]. SOX2⁺ stem cells could self-renew and differentiate into cells that produced various pituitary hormones, with more pronounced differentiation observed in mice nearing adulthood[83]. Despite the proven existence of pituitary stem cells, there has been little research on their use in disease treatments. Suga et al[84], for the first time, established a three-dimensional (3D) culture technique to differentiate pituitary hormone-producing cells from mouse ESCs in vitro. These cultures were transplanted into pituitary-depleted mice to observe hormone production, suggesting the potential of stem cells to treat hypopituitarism. Later, they improved on the previous differentiation protocol and used organoid culture to generate pituitary cells from hESCs. The resulting cells responded to both positive and negative signals for hormone secretion, more closely mimicking the physiological function of pituitary hormone-secreting cells[85]. Another research team differentiated functional pituitary cells from iPSCs in vitro. They identified that changes in supplementation helped them distinguish cells that secreted different hormones in response, supporting targeted replenishment of pituitary hormone-producing cells[86]. Recently, Taga et al[87] made a breakthrough. They established a method for purifying hPSC-derived pituitary cells, thus substantially increasing differentiation efficiency. The purified pituitary cells could further form 3D pituitary spheres (3D-pituitaries) and showed high ACTH secretion capability. After transplantation into pituitary-removed mice, these 3D-pituitaries survived for 6 months, resulting in increased ACTH secretion. They responded to CRH stimulation and were subject to negative feedback inhibition by glucocorticoids, avoiding excessive ACTH production. The team plans to further refine their research on pituitary hormone-producing cells and extend their studies to monkeys. This endeavor aims to establish a robust foundation for the eventual clinical implementation of stem cell therapy for hypopituitarism in humans.

Stem cell-based regenerative strategies offer a promising therapy for hypopituitarism, with an aim to overcome the limitations of lifelong hormone replacement. Although the research is still in its preliminary stage, the breakthrough of differentiation and purification of pituitary cells from stem cells opens up the possibility of targeted therapy for hypopituitarism. Clinical application requires validation in primate models to establish safety and scalability.

PAI

The adrenal glands are located above both kidneys and comprise two parts: The outer cortex and the inner medulla. The cortex originates from the mesoderm, whereas the medulla originates from the neural ectoderm. The adrenal glands regulate homeostasis in the body by secreting various hormones. The cortex secretes mineralocorticoids, glucocorticoids, and sex hormones, and the medulla secretes epinephrine and norepinephrine. Adrenal gland damage from various causes can lead to PAI, leading to impaired synthesis and secretion of adrenal hormones, principally glucocorticoids and mineralocorticoids. The most common cause is autoimmune adrenalitis, with other causes being congenital adrenal hyperplasia, infections (particularly tuberculosis), medications (e.g., immune checkpoint inhibitors), and surgery. Currently, the treatment of PAI is still limited to lifelong hormone replacement therapy, which can neither mimic the physiological circadian rhythm of human glucocorticoids nor be feedback-regulated by the normal hypothalamic-pituitary-adrenal axis. Insufficient replacement is prone to adrenal crisis, and excessive replacement may cause drug-related adverse effects. Cell-based therapy offers a viable alternative for treating patients with PAI by functioning as a substitute for the impaired adrenal glands.

Researchers in Japan successfully transformed human BM-MSCs into steroidogenic cells, primarily glucocorticoid-producing cells, both in vitro and in vivo through treatment with steroidogenic factor-1 (SF-1) and cAMP[88]. Subsequently, they used the same method to differentiate steroidogenic cells from human UC-MSCs. After treatment with SF-1, the expression of steroidogenic genes was observed, accompanied by an increase in progesterone levels. Further treatment with cAMP upregulated the expression of these genes and further elevated progesterone levels; meanwhile, glucocorticoid expression remained low, indicating that these cells exhibited properties similar to those of granulosa-luteal-like cells[89]. WJ-MSCs could also be induced to differentiate into steroidogenic cells by SF-1, and they may be a more suitable source for steroidogenic cells than BM-MSCs because of their higher proliferative capacity, higher levels of steroidogenic mRNA expression, and greater secretion of steroid hormones, including glucocorticoids and testosterone[90].

Driven by encouraging results, researchers’ enthusiasm continued to grow, and the derivation of steroidogenic cells possessing hormone-secreting functions was also achieved from ESCs[91,92]. Although these findings have not been validated in vivo experiments, they offered a wider range of options for adrenal replacement therapy. Sakata et al[93] conducted even more groundbreaking research. They used an organoid culture system to differentiate hiPSCs into the mesoderm, the origin of the adrenal cortex. Then, they used multiple inducers to directly induce the generation of adrenal cortical progenitor cells, ultimately cultivating fetal zone adrenal cortex-like cells (FZLCs). Functional testing of FZLCs revealed that they could produce pregnenolone, cortisol, and aldosterone, consistent with the steroid synthesis profile of the human fetal adrenal cortex. Furthermore, upon exposure to drugs that inhibit the production of adrenal steroid hormones, FZLCs responded with decreased hormone production. This was the first time that adrenal-like tissue was cultivated using stem cells.

Currently, both MSCs and hPSCs are believed to have the potential for regenerating adrenal cells and tissues. Among MSCs originating from various tissues, WJ-MSCs may be a more appropriate source for steroidogenic cells; however, this warrants in vivo confirmation. Research involving PSCs, encompassing ESCs and iPSCs, remains in the in vitro phase. Nevertheless, the successful differentiation of iPSCs into FZLCs has established a robust basis for advancing animal and clinical studies, signifying a substantial leap towards the physiological replacement of cells and tissues in PAI.

Osteoporosis

Osteoporosis is a common endocrine disease, primarily characterized by the destruction of bone microstructure and a decrease in bone mass, which leads to reduced bone strength and increased risk of fractures. A severe consequence of osteoporosis is fragility fractures, which are fractures caused by minimal trauma. Fractures can lead to disability, incapacity, and even death, particularly in older patients. Current management strategies for osteoporosis include lifestyle improvements, basic bone health supplements, and anti-osteoporosis medications. However, these treatments have limited efficacy and are associated with adverse effects. Bisphosphonates are the most widely used bone resorption inhibitors; however, they often cause “flu-like” reactions, and their long-term use increases the risk of jaw bone necrosis, femoral fractures, and other complications. Denosumab, another commonly used bone resorption inhibitor, is a human monoclonal antibody that can block the bone resorption process mediated by receptor activator of nuclear factor-κB ligand. It also carries a risk of jaw bone necrosis and can lead to severe hypocalcemia. The bone formation promoter teriparatide can also be a treatment option for osteoporosis patients at high risk of fracture. However, it is expensive and reportedly increases the incidence of osteosarcoma, thus limiting the treatment course to 24 months.

MSCs, derived from the mesoderm, can differentiate into bone, cartilage, and fat tissues, making them an ideal choice for osteoporosis treatment. BM-MSCs are particularly noted for their high osteogenic differentiation capacity, enabling bone repair and regeneration[94]. Researchers locally transplanted autologous BM-MSCs into the distal femur of an osteoporosis rabbit model induced by ovariectomy (OVX). The results showed an amelioration of bone microstructure, the formation of new bone-like structures, and increased trabecular thickness, demonstrating the therapeutic effects of local autologous BM-MSC transplantation in osteoporosis[95]. Subsequently, the efficacy of systemic BM-MSC transplantation was also investigated. Allogeneic BM-MSCs were administered via tail vein injection to mouse models of age-related osteoporosis. Consequently, MSC deposition was observed in the femurs and tibias of the mice after transplantation. The bone turnover increased, new trabecular bone formed, and the bone microstructure was restored. These effects lasted for at least 6 months[96].

AD-MSCs, due to their low immunogenicity, ease of acquisition and production in large quantities, and their osteogenic differentiation process being less affected by age of donor and multiple passages, are also used in osteoporosis research[97]. Evidence has shown that one of the mechanisms underlying osteoporosis development was the increased differentiation of MSCs into adipocytes and decreased differentiation into osteoblasts, leading to the fat accumulation and bone mass reduction in the bone marrow[98]. Ye et al[99] demonstrated in an OVX rabbit model that AD-MSCs could promote the expression of bone morphogenetic protein-2 and bone morphogenetic protein receptor type IB, thus activating the bone morphogenetic protein-2/bone morphogenetic protein receptor type IB signaling pathway and reversing the imbalance between osteogenesis and adipogenesis that existed during osteoporosis onset, subsequently improving bone density in OVX rabbits.

UC-MSCs perform better than AD-MSCs and BM-MSCs in terms of immunogenicity, yield, source, immunoregulatory capacity, and anti-inflammatory ability, making them a superior source of MSCs[100]. UC-MSCs primarily rely on their robust paracrine effects, such as cytokine secretion, to treat diseases. In osteoporosis treatment, UC-MSCs could reportedly produce a series of secretomes, including transforming growth factor-beta, epidermal growth factor, and insulin-like growth factor-1, to enhance bone formation and delay bone loss in aging rats[101]. Based on the achievements of preclinical research, clinical studies are also gradually being conducted. Thus far, there are five clinical studies registered on ClinicalTrials.gov regarding the use of MSCs for osteoporosis treatment, including AD-MSCs (1), BM-MSCs (1), and UC-MSCs (3). One of these studies has been completed, but study results have not been published yet.

Stem cell therapy, particularly with MSCs, shows promise for osteoporosis treatment by restoring bone structure and balancing differentiation. UC-MSCs are a preferred source, leading to many clinical studies. However, clinical data on their efficacy and safety are lacking, and the optimal transplantation method remains undetermined. More clinical research is needed in this regard. Despite uncertainties, advances in stem cell research offer hope for osteoporosis patients, as these therapies could transform the management of this widespread condition.

DISCUSSION

Stem cell therapy for endocrine and metabolic disorders primarily involves PSCs and MSCs. PSCs, including ESCs and iPSCs, offer the theoretical advantage of unlimited self-renewal and differentiation ability, making them particularly suited for conditions characterized by irreversible cell injury, such as T1D, hypothyroidism, hypopituitarism, and adrenal insufficiency. Indeed, clinical studies using hPSC-derived islet cells have shown glucose-responsive insulin production and even insulin independence in some patients[16,17], whereas preclinical studies have successfully generated thyroid follicles, pituitary hormone-secreting cells, and adrenocortical-like cells from PSCs[73,74,84,87,93], providing possibility for cell replacement. Conversely, MSCs do not efficiently differentiate into endocrine cell types; their therapeutic benefits derive primarily from their anti-inflammatory, immunoregulatory, anti-apoptotic, and pro-angiogenic effects exerted via paracrine signaling and extracellular vesicle release[22,49]. This makes them applicable to diseases driven by immune dysregulation and chronic inflammation, encompassing diabetic complications, AIT, and osteoporosis. Moreover, although MSCs cannot restore beta-cell mass in advanced disease, they have shown efficacy in improving insulin sensitivity in T2D[34].

Despite these conceptual advantages, both PSC- and MSC-based therapies face the barriers of clinical translation that are partly shared across disease contexts and partly disease-specific. The common hurdles include: (1) The tumorigenic potential of undifferentiated PSCs, warranting rigorous quality control and long-term monitoring[4]; (2) The immunogenicity of allogeneic products, which necessitates either immunosuppression or immune-evasive engineering[14,15]; (3) The lack of standardized manufacturing protocols and potency assays, leading to product heterogeneity[102]; (4) The challenge of maintaining graft cell number stability and microenvironmental compatibility in elderly patients post-transplantation[103]; (5) The high cost and logistical complexity of personalized cell products; and finally; and (6) The absence of non-invasive imaging tools for tracking graft survival, biodistribution, and function in vivo. These interdisciplinary challenges must be systemically addressed to enable reproducible and affordable clinical application.

Disease-specific challenges must also be resolved before the widespread clinical application of stem cell therapy. For T1D, the major obstacles are immune rejection of hPSC-derived islet grafts and tumorigenicity. Addressing these challenges necessitates hypoimmunogenic cell engineering, immunomodulatory co-transplantation (e.g., with MSCs or Tregs), improved encapsulation devices with anti-fibrotic coatings and pro-angiogenic factors to ensure long-term graft survival without chronic immunosuppression, and rigorous and standardized regulatory requirements[14,15,104,105]. For T2D, MSCs show insufficient efficacy in cases of long-standing disease with beta-cell exhaustion[34], which may be attributed to the impairment of MSC function by prolonged exposure to hyperglycemia, oxidative stress, and inflammatory microenvironment[106]. Although autologous E-islet transplantation has shown success in individual cases, its broader application remains constrained by limited scalability and reproducibility[38]. Integrating MSCs with diabetic medication and developing ready-made iPSC-derived islet products with rigorous quality control may be key strategies. Diabetic complications suffer from protocol heterogeneity (cell source, dose, route, and patient selection), requiring comparative trials to draw conclusions[41,42,54,61], with stem cell-derived exosomes offering a promising cell-free alternative requiring efficacy and safety validation[45,46,48,49]. In particular, cardiovascular complications lack dedicated trials in diabetic populations despite abundant preclinical data, necessitating phase I/II studies with hard cardiovascular outcomes[70,71]. To our knowledge, for hypothyroidism and AIT, no clinical studies have yet been conducted, and key challenges, achieving vascularized, feedback-responsive thyroid tissue and ensuring durable protection in long-standing Hashimoto’s patients, remain unresolved. More animal validation and phase I/II trials are needed for clinical translation[75,78,79]. Hypopituitarism faces low differentiation efficiency and poor engraftment of pituitary hormone-producing cells, calling for optimized, clinically compliant protocols, non-invasive imaging biomarkers to track graft fate, and minimally invasive delivery routes that spare the blood-brain barrier[84,87]. Most of achievements about PAI remain at the in vitro stage, and the superiority of WJ-MSCs over other MSC sources remains to be verified in vivo. Critical next steps include functional validation in adrenal-ablated animal models and the development of scaffold-based transplantation strategies to achieve long-term, physiologically regulated steroidogenesis[90,93]. Osteoporosis is hindered by the preferential adipogenic differentiation of MSCs within the osteoporotic bone marrow microenvironment, the unknown optimal administration route (systemic vs local), and the lack of any published clinical trial results. Preconditioning strategies (hypoxia, growth factors, and small molecules) to enhance the osteogenic pathway, combined with well-powered fracture endpoint randomized controlled trials, are essential[95,96,99,107].

CONCLUSION

Two distinct stem cell strategies have emerged for endocrine and metabolic disorders, differing in their intrinsic properties and the pathophysiological layers they address. PSC-based cell replacement represents an effective pathway for restoring function in diseases defined by irreversible cell loss, T1D, hypothyroidism, hypopituitarism, and adrenal insufficiency. By contrast, MSCs have established their value through anti-inflammatory, immunomodulatory, and paracrine mechanisms, offering disease-modifying benefits in T2D, diabetic complications, AIT, and osteoporosis. Yet both trajectories converge on the same translational bottleneck, immune rejection, manufacturing heterogeneity, and cost, alongside their distinct hurdles, from the tumorigenic risk of PSCs to the limited efficacy of MSCs in advanced disease. Solutions are emerging: Hypoimmunogenic gene editing, next-generation encapsulation technologies, preconditioning strategies, and standardized products all represent concrete paths forward. The task ahead is to validate these approaches through rigorous, large-scale trials and translate the promise of regenerative medicine into reality, not as a distant vision, but as a transformative benefit for patients worldwide.