Genomic stability of human pluripotent stem cells: advances in research and screening criteria

Abstract

Genetic and acquired genomic abnormalities pose substantial risks to human pluripotent stem cells (hPSCs), presenting a critical challenge to their safe use in both research and clinical applications. Although hPSCs possess intrinsic regulatory mechanisms that support genomic stability, diverse culture conditions and extended in vitro expansion inevitably give rise to chromosomal and subchromosomal abnormalities. Genomic adaptation during long-term culture, therefore, remains an unavoidable phenomenon. This review synthesizes recent advances in understanding the determinants of hPSC genomic stability and highlights strategies to preserve stem cell quality. Furthermore, it summarizes current screening criteria for defining safe stem cell lines, providing a valuable reference for their translational use. Together, these insights lay the groundwork for improving culture systems, refining monitoring techniques, and guiding the development of safer and more reliable stem cell-based applications in regenerative medicine.

Graphical Abstract

Significance statement.

The core challenge of regenerative medicine lies in identifying clinically viable human pluripotent stem cell (hPSC) lines that exhibit both genomic stability and functional integrity. This review summarizes the factors that may contribute to genomic instability of hPSCs, and highlights current detection methods and evaluation criteria for assessing stem cell stability. Through discussing future directions in stem cell safety assessment and proposing key improvements, this review informs the optimization of culture systems and the selection of reliable stem cell lines.

Introduction

Stem cell-based therapy constitutes a cornerstone of regenerative medicine, functioning to restore tissue homeostasis and promote regeneration by stimulating or regulating endogenous stem cell populations or replenishing the stem cell pool.¹ Human pluripotent stem cells (hPSCs), including embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs), hold significant promise for disease treatment and serve as essential cell sources in regenerative medicine due to their capabilities for self-renewal and pluripotency. Notably, hPSCs can be directed to differentiate into specific functional cell types, showing remarkable therapeutic potential for neurodegenerative disorders, diabetes, spinal cord injury-induced paralysis, and other conditions.^2,3 Generating enough functional cells through in vitro expansion is a critical step in advancing cell therapy. However, the accumulation of DNA mutations and damage caused by replication stress during expansion increases the risk of genetic abnormalities. This phenomenon compromises the differentiation potential of hPSCs and increases their tumorigenic risk, posing a major barrier to clinical application.^4,5

Chromosomal abnormalities in hESCs were first described in 2004.⁶ Since then, numerous studies have identified not only karyotype abnormalities but also subchromosomal variations, including copy number variations (CNVs), insertions and deletions (INDELs), and single nucleotide variations (SNVs).⁷ Recent findings indicate that hESCs frequently harbor structural variations, approximately half of which are mosaic and confer growth advantages under in vitro conditions.⁸ The most common ones are the increase of chromosomes 1q, 12p, 17, 20 and X, the loss of chromosomes 10p, 18q and 22p, and the mutation of TP53.^9,10 Among these, genetic alterations at specific loci—such as copy number gain of 20q11.21—pose particular risks, as this aberration, together with additional epigenetic modifications, underlies the so-called “culture-adapted phenotype”.¹¹ Similarly, chromosomes 1q gain affects MDM4, a negative regulator of TP53, reducing sensitivity to DNA damage-induced apoptosis, while gains on chromosome 12 often linked to increased expression of NANOG, which promote hESCs proliferation and suppress spontaneous differentiation.^12,13 Many recurrent chromosomal abnormalities, including gains of 12p and 17q, are hallmarks of cancer and contribute to the tumorigenic potential of hPSCs.^13,14 Abnormalities in these regions may drive the clonal expansion of aberrant cells. Otherwise, loss of chromosome 18q specifically decreased the expression of SALL3 and disrupted neuroectoderm commitment.⁹

Due to numerous CNVs can be propagated, selected, and enriched until they ultimately provide growth advantages during expansion.⁸ It is now widely recognized that the genomic stability of human pluripotent stem cells (hPSCs) is strongly influenced by factors such as cell source, culture conditions, and passaging techniques.¹⁵ In this review, we systematically summarize and analyze genomic instability in hPSCs, with a focus on these factors, aiming to provide valuable references and critical insights for quality assessment and cell line selection in stem cell research and therapeutic applications.

Contribution of cell sources to the genomic instability of hPSCs

Currently, hPSCs can be derived from unused embryos through assisted reproductive technology (ART), blastocysts generated via somatic cell nuclear transfer (SCNT), or reprogrammed directly from somatic cells to generate induced pluripotent stem cells (hiPSCs).^16,17 The use of embryonic material is further constrained by strict ethical regulations, resulting in a limited supply and variable quality of embryos available for research. Low-quality embryos may reduce the efficiency of stem cell line establishment and increase genomic instability, owing to risk factors such as donor age, health status, and mosaic ratio.¹⁸ Although discarded IVF embryos are considered a valuable source for hESCs, the establishment efficiency of cell lines from morphologically abnormal embryos was only 7.7%.¹⁹ In contrast, high-quality blastocysts markedly improve efficiency. With optimized isolation and GMP-grade culture, derivation rates reach 60%-100% for grade A/B ICMs, versus ∼30% for grade C.²⁰ Chromosomal abnormalities are common in human cleavage-stage embryos before implantation and are recognized as one of the main causes a major cause of miscarriage and birth defects.^21,22 Undeniably, low-quality blastocysts constitute a major source for the derivation of hESC lines. It is reported that human embryonic stem cells (hESCs) can be successfully derived from embryos with abnormal karyotypes, the efficiency of derivation is only 3.08%, and the ratio of triploid to diploid hESC lines is approximately 1:3.^23,24 The lack of ICM genomic data, reliance on morphology, and inconsistent derivation methods hinder comparison of hESC derivation efficiency between aneuploid and euploid embryos.²⁵ Moreover, hESCs with abnormal karyotypes may still exhibit characteristics similar to those of normal hESCs, including typical morphology, long-term proliferative capacity, and pluripotent differentiation potential.⁵

As presented in the schematic of Figure 1, based on preimplantation genetic testing for aneuploidy (PGT-A) performed on clinical samples before embryo transfer, the incidence of abnormal mosaic in morphologically high-quality blastocysts is estimated to range from 14% to 59%.²⁶ Notably, PGT-A is typically performed on small-scale trophectoderm biopsies, reliably detects whole-chromosome aneuploidies but fails to capture ICM genomic integrity and offers limited sensitivity for segmental mosaicism.^27,28 This limitation extends to clinical embryo screening as well as in vitro stem cell line derivation. More recently, Zhai et al. performed single-cell sequencing of all cells from each blastocyst and found that 100% (20/20) were mosaic, suggesting that mosaicism has likely been underestimated in previous PGT-A studies.²⁹ Clinical studies further indicate that the presence of chromosomal mosaicism is significantly associated with implantation outcomes: while implantation rates are comparable, mosaic blastocysts carry a higher risk of miscarriage.^21,30 Previous studies have demonstrated that chromosomal duplications and fragmentations are strongly associated with developmental arrest, with abnormalities in chromosome 21 exhibit high arrest rates as early as day 9.³¹ It is noteworthy that mosaic embryos can also develop into live offspring. Furthermore, recurrent chromosomal anomalies in low-quality blastocysts differ fundamentally from the culture-adapted karyotypic changes that promote selective growth during stem cell expansion in vitro.^32,33 Mosaic defects present during embryogenesis often resolve or disappear following isolation and in vitro culture, particularly when they arise from the inner cell mass.^34,35 Although no significant bias has been observed in the developmental fate of abnormal cells between the trophoblast and the inner cell mass, the proportion of abnormal cells in the inner cell mass also tends to decline progressively after implantation.³⁶ These findings imply that hESC lines derived from morphologically high-quality blastocysts may still harbor intrinsic quality discrepancies. Factors such as blastocyst source and mosaic levels exert a significant contribution to the genomic instability of hESCs. Consequently, even under identical culture conditions and passaging techniques, certain hESC lines exhibit an inherently greater susceptibility to chromosomal instability.

Figure 1.

Contribution of cell sources to the genomic instability of hPSCs. This diagram illustrates the risk factors influencing hPSCs quality prior to cell line derivation. The quality of donor embryos or somatic cells, mosaic ratio and the reprogramming process may increase the heterogeneity and instability of hPSCs. Although it is unlikely to obtain completely normal hPSCs, their quality can be improved by eliminating abnormal cells. Graphical elements are annotated. ICM, inner cell mass; TE, trophectoderm; hPSCs, human pluripotent stem cells; PGT, preimplantation genetic testing.

Genomic instability is also evident during the early passaging of induced pluripotent stem cells (hiPSCs), potentially originating from rare parental cell subsets and reprogramming process.³⁷ It has been reported that approximately 5%-46% of phenotypic heterogeneity in hiPSCs reflects donor-associated genetic backgrounds. Additionally, somatic CNVs are present in approximately 41% of hiPSCs, with a mean length of ∼7.5 Mb, of which 15%-22% are donor-derived and the remainder emerge during reprogramming or in vitro propagation.³⁸ Such donor-specific variations have been documented in hiPSCs from particular patient populations; for example, hiPSCs derived from patients with chronic myeloid leukemia (CML) exhibit complex chromosomal translocations,³⁹ while those from patients with mitochondrial disorders harbor point mutations in oxidative phosphorylation complexes.^40,41 Reprogramming techniques themselves also influence the genomic integrity of hiPSCs. Integrated reprogramming strategies are prone to non-targeted genetic modifications, frequently resulting in increased mutations characterized by aneuploidy and epigenetic alterations.^42,43 Nonetheless, some studies have challenged this view.⁴⁴ Compared to parental cells, hiPSCs display increased CNVs and accumulated mutations.⁴⁵ Copy-number losses, typically affecting tumor suppressors, arise during reprogramming, while gains in oncogenic regions increase with passaging. Certain deletions are progressively eliminated, implying positive selection during reprogramming followed by negative selection in culture, whereas cancer-related mutations confer a growth advantage and perturb differentiation.^46,47 Paniza et al. quantified replication stress and demonstrated that hiPSCs exhibit significantly higher frequencies of genomic instability, manifested as abnormal DNA replication, although the mechanistic relationship between these factors remains unclear.⁴⁸ It cautions stem cell researchers to rigorously assess the genetic integrity of parental cells as a prerequisite for reliable hiPSC derivation. Additionally, compared with hESCs, hiPSCs face fewer ethical constraints but retain epigenetic memory and more genomic variability. Recently, Buckberry et al. introduced transient naïve treatment (TNT), a strategy designed to reset DNA methylation and chromatin states in hiPSCs. TNT effectively overcomes epigenetic barriers that limit differentiation and eliminates abnormalities acquired during reprogramming or long-term culture, thereby improving the suitability of hiPSCs for regenerative applications.^49,50

Effect of culture conditions on the genomic stability of hPSCs

The disparity between in vitro culture conditions and the in vivo context may substantially contribute to the genomic instability observed in hPSCs. Ensuring stem cell quality, therefore, depends on developing stable culture systems capable of efficiently producing high-quality hPSCs in vitro.

Current hPSC culture systems are broadly categorized into naïve and primed states. Naïve hESCs exhibit characteristics resembling those of the inner cell mass or pre-implantation epiblast, whereas primed hESCs are more similar to post-implantation epiblast cells, with the possibility of interconversion between these states.⁵¹ Compared with primed hESCs, naïve hESCs display global DNA hypomethylation, including a marked reduction in H3K27me3 levels at both promoter and gene regions. This state is associated with higher proliferation rates, improved survival during single-cell passaging, and enhanced potential for chimeric integration.⁵² Genomic instability has been reported in naïve hESCs maintained under low-oxygen conditions such as the 5iLAF and t2iLGöY systems.^53,54 For example, UCLA20n hESCs cultured in 5iLAF developed trisomies of chromosomes 3, 7, 12, and 20 by the 14th passage, with progressive karyotypic abnormalities emerging upon extended culture.⁵⁴ This instability may be linked to impaired imprinted gene function caused by the global hypomethylation characteristic of naïve hESCs.⁵⁵ Normal karyotype maintenance can be achieved by modifying small molecules combinations, reducing MEK inhibitor concentrations, or utilizing YAP-induced naive hPSCs.^53,55,56 Chemical resetting approaches have been shown to enhance genomic stability during long-term culture.^55,56 Notably, the 4CL (four chemicals plus LIF) culture system achieved balanced genome-wide DNA demethylation, generated transcriptional profiles resembling those of the inner cell mass (ICM), and supported the maintenance of both hESCs and hiPSCs for at least 15 passages without evident karyotypic abnormalities.⁵⁶ The successful derivation of naïve pluripotency in nonhuman primates underscores the advantage of the 4CL system in sustaining long-term self-renewal while preserving genomic stability.^57,58 Moreover, the generation of chimeric monkeys with a high contribution from naïve hESCs cultured under 4CL conditions fulfills the gold standard of naïve pluripotency.⁵⁹

Physiological intracellular reactive oxygen species (ROS) levels are closely associated with DNA damage repair and genomic integrity in hESCs, with low ROS levels activating stress-response pathways and excessive ROS levels resulting in apoptosis.⁶⁰ For instance, the SNV mutation rate of the clinical primed hESC line Nutristem-MSHEF11 was reduced by 54% when cultured under hypoxic conditions (5% O₂) compared with normoxic conditions.⁶¹ Similarly, supplementation with the antioxidant supplementA1345 has been shown to exert a dose-dependent protective effect, with low concentrations (1-20 μM) effectively reducing DNA damage, whereas the apoptosis inhibitor Y27632 had no substantial impact.^62,63 Commonly used culture systems for primed hPSCs include MEF-conditioned medium (MEF-CM), Nutristem, mTeSR, E8, KSR/bFGF, and AIC.⁶⁴ The MEF-CM system maintains hESC pluripotency for over 130 passages, with chromosome duplications emerging during prolonged culture, whereas loss of heterozygosity(LOH) arises during derivation and is unlikely to be caused by the culture condition.⁶⁵ hPSCs maintained in the mTeSR and E8 systems maintain a normal karyotype up to passage 10.⁶⁶ In comparison to the KSR system, hPSCs (HuES9 cells) cultured in mTeSR and E8 exhibit elevated ROS levels as well as an increased frequency of INDELs and SNVs.^66,67 The AIC culture system enables single-cell passaging and suspension expansion of hPSCs, achieving superior survival and proliferation (∼25-fold in 4 days) with lower spontaneous differentiation than E8 or StemFlex, while maintaining pluripotency and genomic stability after long-term culture.⁶⁸

The morphology and proliferative dynamics of hPSCs differ markedly depending on the substrate or feeder layer.^69,70 Vitronectin has been shown to support hiPSC and hESC proliferation and pluripotency more effectively than Matrigel or other ECM coatings (eg, Laminin-511/521) in long-term cultures.⁷¹ Via distinct mechanisms, LN511E8 (myosin contraction) and LN211E8 (Wnt signaling) induce dense and mesenchymal-like hiPSC morphologies, respectively, with LN332E8 promoting epithelial-like forms.^72,73 Cytogenetic analyses in these studies revealed no genomic abnormalities. Some hESC lines, such as HS181 and SHEF-3, exhibit a higher incidence of genetic aberrations when cultured without a feeder layer. Upon transfer to feeder-free culture, HS181 cells acquired 89% trisomy 12 mosaicism by passage 30, versus stable karyotypes on feeders.⁷⁴ Several studies under feeder-free conditions have reported gains of 20q11.21 and 1q in hPSCs,^{8,11,12,15,33,47,75} underscoring the need for systematic evaluation of the impact of culture conditions on genomic stability.

Culture density also plays a critical role in genomic stability. High-density cultures (1.34 ± 0.30 × 10⁶ cells/well) are associated with elevated levels of genomic abnormalities, prolonged G1-phase stress, and S-phase arrest. In this context, subclones carrying recurrent 1q region duplications often gain a selective growth advantage.^33,75 Increasing the frequency of medium changes effectively reduces DNA damage and genetic instability.⁷⁶ Prolonged in vitro culture, elevated lactic acid levels, and low pH in high-density hESC cultures hinder cell growth and affect pluripotency, potentially linked to DNA methylation loss and specific histone modifications.^77,78

Accumulation of genetic instability in hPSCs expansion

Expansion of hPSCs requires repeated cell passage and cryopreservation, both of which may influence genetic stability. Passage methods are generally classified as mechanical or enzymatic, and further subdivided into mass or single-cell approaches. Most studies on the genomic stability of hPSCs have focused on enzymatic passage, which employs agents such as collagenase, Accutase, traditional trypsin, or animal-free trypsin-like enzymes (eg, TrypLE) to dissociate cells. Evidence suggests that hPSCs can maintain a normal karyotype under various passage conditions, including enzymatic single-cell passage (Accutase, collagenase B, trypsin, or TrypLE), non-enzymatic single-cell passage (using dissociation solutions from Sigma, Gibco), and mass passage techniques.^79–81 Small‑block passaging with EDTA preserves genomic stability in >50 hPSCs lines cultured in feeder‑free media (TeSR, E8) for >50 generations.⁸¹ However, findings are not entirely consistent. One study reported that enzymatic dissociation with trypsin–EGTA better preserved karyotypic integrity than the EDTA-based method.⁸² In addition, although hESCs maintain normal karyotypes under mechanical passage, switching to TrypLE-based single-cell passage often induces abnormalities within five generations, including trisomies 12 and 20 and deletions on chromosome 7q.⁸³ Given the influence of multiple variables—including culture system, dissociation duration, and reagent concentration—further systematic studies are needed to clarify the relationship between passage methods and genomic instability.

Cryopreservation is widely used for the long-term preservation of cells. However, this process exposes cells to various stresses, including osmotic pressure changes and cold stress, leading to oxidative accumulation and freezing damage.^79,84 Current cryopreservation strategies for hPSCs mainly include slow freezing, ultra-slow programmable freezing, and vitrification.^84,85 To improve post-thaw survival, researchers have investigated combinations of cryoprotectants such as dimethyl sulfoxide (Me₂SO), hydroxyethyl starch (HES), and trehalose.^85,86 Nonetheless, some studies report that while protective agents or repeated freeze–thaw cycles do not affect genomic stability, they can negatively impact cell viability.^87,88 Despite maintaining a stable karyotype, hESCs exhibit high sensitivity to cryoinjury, with adherent cells being more vulnerable than suspension cells, and only a limited number of cells able to reform colonies after cryopreservation.⁸⁹ Notably, other reports describe karyotypic abnormalities, mitochondrial damage, and chromosomal aberrations following cryopreservation.⁹⁰ In human embryos, cryopreservation has been associated with increased mitochondrial DNA mutations, with vitrification causing significantly less DNA damage than slow freezing.⁹¹

Beyond cryopreservation, numerous studies have documented the frequent occurrence of chromosomal abnormalities and the accumulation of genetic variations during stem cell passage. Some hESC lines (eg, SA002, SA121) preserve overall karyotype integrity during extended culture, with only dynamic changes in CNVs and SNVs observed.^14,42,43,92 However, later passages typically show gains or losses of specific chromosomes—most of chromosomes 1, 12, 17, and 20—with the frequency of chromosomal aberrations reported to double relative to early passages.^14,93 As shown in Figure 2, some mosaic variants are selected during long-term passaging, allowing growth-dominant clones to become fixed and potentially interfering with stem cell differentiation. Notably, approximately 20% of hPSC lines acquire duplications of the 20q11.21 region, which confers a selective advantage by activating anti-apoptotic mechanisms through overexpression of the BCL2L1 gene. This leads to aberrant mitosis and impaired neuroectodermal differentiation.^93,94 Similarly, mutations in TP53 (located at 17p13.1) increase with passage number at a rate of ∼1.9-fold per generation, providing a strong selective advantage.⁹⁵ With increasing passage number, stem cells exhibit progressive decline in DNA damage repair capacity, coupled with dynamic alterations in DNA methylation.^96,97 Interestingly, when culture duration is held constant but passage frequency is reduced, hESCs are more likely to maintain a normal karyotype.^41,42,93,94 Although many molecular mechanisms underlying the enrichment of specific genetic abnormalities during prolonged culture have been elucidated, this remains an essentially unavoidable phenomenon.

Figure 2.

The impact of mutation accumulation on the differentiation potential of hPSCs. With increasing passage number of hPSCs, subpopulations with mutation gradually expand and become fixed, leading to differentiation blockade or lineage-biased outcomes. Normal hPSCs typically possess the ability to differentiate into cells of the three germ layers, but mutation accumulation may enhance or block their differentiation into certain cell types. Gray arrows represent differentiation potential, larger yellow arrows represent biased differentiation, and gray T-bars indicate blocked differentiation.

Genomic stability assessment of hPSCs and cell line selection

The enrichment of genetic abnormalities in hPSCs is influenced by numerous environmental factors, underscoring the importance of monitoring genomic integrity. However, the dynamics and molecular mechanisms driving the emergence and elimination of mutant clones remain poorly understood. Notably, some high-risk mutations, as indicated by elevated Combined Annotation-Dependent Depletion (CADD) scores, appear to confer advantages for embryonic growth and functionality. A further challenge lies in linking non-coding region variants with gene expression, as only 20% to 50% of disease-associated genes overlap with expression quantitative trait loci (eQTL).⁹⁸ Thus, preserving genomic stability while avoiding aberrations during in vitro culture remains a critical challenge in stem cell research.

Over recent decades, several approaches have been developed to evaluate genomic stability. Commonly employed methods include fluorescence in situ hybridization (FISH), quantitative polymerase chain reaction (qPCR), Giemsa-banding karyotyping, chromosomal microarrays, and second-generation short-read sequencing (a form of next-generation sequencing, NGS), each with distinct advantages and limitations. FISH and qPCR are primarily used to target specific chromosomal regions, with qPCR capable of identifying variations at a resolution of ≥100 bp and FISH at a resolution of ≥100 kb. In contrast, karyotyping, chromosomal microarrays, and NGS allow genome-wide assessments, with NGS providing single-nucleotide mutation detection at exceptionally high resolution.⁹⁹ Nevertheless, point mutations often arise in only small subpopulations of cells. At the standard sequencing depth of 30× in bulk NGS, reliable mutation detection typically requires 30%-40% of cells to carry the variant.¹⁰⁰ A strategy combining whole-genome sequencing (WGS) and RNA sequencing has been shown to increase mutation detection positivity.¹⁰¹ High-depth whole-exome sequencing (WES) complemented by FISH or qPCR enhances risk variant confirmation and detection accuracy. However, large-scale studies of cell population mutations remain constrained by high costs.^99,102

Single-cell sequencing enables the detection of mutations at the individual-cell level and, when integrated with transcriptional profiles, reveals their functional impact. In the stem cell field, single-cell sequencing has been widely applied to dissect heterogeneity within stem cell populations and, when integrated with multi-omics approaches, to resolve CNVs in mosaic embryos.^103,104 Like bulk RNA‑seq, scRNA‑seq detects only transcribed variants, limiting coverage and sensitivity and potentially masking mutation effects in specific cell states. Single-cell genomics, by contrast, offers higher fidelity and accuracy in detecting SNVs, as exemplified by multiple displacement amplification (MDA), albeit at the cost of reduced coverage uniformity.¹⁰⁵ Additionally, while short‑read NGS excels at detecting SNVs and indels, it falls short in characterizing large structural variants (SVs). In recent years, third-generation/real-time single-molecule sequencing (TGS) overcomes short-read limitations by enabling full-length cDNA sequencing and more accurate detection of complex SVs.¹⁰⁶ Single-molecule sequencing (SMS) achieves >98% allele‑specific SNP accuracy in single‑cell full-length transcriptomics.¹⁰⁷ SMS-based single-cell genome analysis (eg, dMDA) enables high-fidelity variant detection and >16-fold improved SV detection.¹⁰⁸ SMS-based single-cell epigenome captures kilobase-scale regions (eg, LINEs, LTRs) missed by conventional short-read sequencing, and increases sensitivity for allele-specific chromatin accessibility peaks by three- to ten-fold.¹⁰⁹ Given the complexity of mutation data and current technical limitations, method choice should be tailored to research goals and cost.

The International Society for Stem Cell Research (ISSCR) and relevant health authorities have formulated unified standards governing current human stem cell research.¹¹⁰ Standards mandate legally sourced cells, sterility testing, genomic stability, and functional validation (marker expression, teratoma, tri-lineage differentiation). Guidelines also recommend monitoring culture-acquired genetic changes of a cell culture overtime.¹¹¹ In 2022, Florian et al. analyzed 141 clinical-grade hESC lines, further expanding existing guidelines. They advocated the use of stem cell lines with extensively characterized genomes, minimal derivation limitations, and an absence of disease-associated mutations, including those linked to developmental disorders.⁸ This analysis outlines the trajectory of hESC genetic stability research. Established genomic integrity standards for hESC lines provide a framework for clinical translation. Future work should elucidate mechanisms driving mutant clone advantage and reduce mutation burden via optimized culture. Addressing these challenges will enhance hPSC safety, inform developmental disorders, and accelerate personalized regenerative therapies.

Conclusion

Accumulating evidence indicates that stem cell–based therapies have demonstrated substantial therapeutic potential across approximately 15 diseases previously considered difficult to treat, integrating findings from 23 leading laboratories worldwide.¹¹² For example, clinical studies of hESC-derived pancreatic progenitors in type 1 diabetes have shown that 11 out of 12 patients were able to markedly reduce or discontinue insulin administration.¹¹³ To date, the United States has led most early-phase clinical trials under a well-established FDA regulatory framework, Japan has advanced hiPSC-based applications in ophthalmic and neurological disorders, and China has made notable progress in diabetes, cardiovascular repair and reproductive medicine.¹¹⁴ Although differentiated derivatives of human pluripotent stem cells (hPSCs), such as hematopoietic stem cells (HSCs) and mesenchymal stem cells (MSCs), have been approved for clinical use in certain countries, most hPSC products remain in clinical trials. At present, commonly used clinical-grade female hESC lines, including WA09, UCLA9 and MShef08, as well as male lines such as MShef11, HUES63 and UCLA10, have been demonstrated to maintain genomic stability beyond 10 passages compared with other stem cell lines, providing an important reference for safety concerns related to the parental origin of hPSCs.⁸

Amid the balance between technological innovation and ethical governance, regenerative medicine is approaching a new translational era. However, accumulating studies have revealed that genetic and epigenetic variations are difficult to fully avoid during hPSC derivation and expansion, making the pursuit of an “perfect” cell line unrealistic. The establishment of unified and broadly applicable good manufacturing practices and quality assessment frameworks is essential to ensure the reproducibility, homogeneity and long-term safety of stem cell–based products. This study on culture‑induced hPSC variants provides a reference for optimizing systems and ensuring clinical safety. Besides, we advocate for the establishment of an organization dedicated to dynamic oversight that monitors research advances and maintains hPSCs quality. Such efforts will promote standardization, industrialization, and robust clinical translation of stem cell products.

Author contributions

Congge Li and Manqing Long contributed equally to this work. Congge Li conceived the project and wrote the manuscript with input from all authors. Manqing Long contributed to material collection and figure design. Yu Yin, Zongyong Ai, and Lifeng Xiang critically reviewed and edited the manuscript. All authors discussed and approved the final version.

Funding

This work was supported by the National Natural Science Foundation of China (Grant No. 82192874, 32360178 and No. 32360177), Yunnan Fundamental Research Projects (202401CF070089), the Joint Special Funds for the Department of Science and Technology of Yunnan Province‑Kunming Medical University (202301AY070001-298), Yunnan Provincial Young and Middle-aged Academic and Technical Leaders Reserve Talents Program (202305AC160022).

Conflicts of interest

All authors declare no potential conflicts of interest.

Data availability

No new data were generated or analyzed in support of this research.