The review, led by Tao Huang and Qiaoying Li of Southwest Medical University in Sichuan, China, together with colleagues, takes an unusually rigorous approach to a field notorious for loose claims. Rather than accepting every published association between fibroblasts and drug failure, the authors graded the evidence according to strict criteria: whether a defined drug was actually being administered, whether treatment-related outcomes were measured, and—critically—whether blocking the fibroblast pathway functionally restored drug sensitivity in experiments. Only mechanisms passing these tests earn the label of true resistance mechanisms. The distinction matters, because fibroblasts drive tumor progression in many ways, and progression is not the same thing as resistance to therapy.

Cancer-associated fibroblasts arise largely from hepatic stellate cells, the liver’s resident vitamin A-storing cells that normally lie dormant until injury wakes them. In chronic liver disease they become the workhorses of fibrosis, laying down scar tissue. When a tumor takes root in this scarred landscape, the activated fibroblasts multiply and diversify into subtypes with very different personalities: inflammatory fibroblasts that pump out signaling molecules, myofibroblastic fibroblasts that build dense extracellular matrix, and even antigen-presenting fibroblasts that can interact directly with immune cells. Single-cell and spatial transcriptomic studies have mapped these populations in hepatocellular carcinoma in increasing detail, and the picture that emerges is of a tumor wrapped in a living, communicating stroma.

The most convincing resistance story in the review concerns a fibroblast-secreted protein called SPP1, also known as osteopontin. When patients receive sorafenib or lenvatinib, fibroblasts respond by secreting SPP1, which binds integrins on the surface of tumor cells and activates a bypass signaling route through protein kinase C alpha. In effect, the drug blocks the main growth pathway while the fibroblasts quietly open a side door. Experiments in which SPP1 was knocked down genetically or blocked pharmacologically restored the tumor cells’ sensitivity to the kinase inhibitors, satisfying the review’s strictest evidentiary standard. This is no longer a correlation; it is a mechanism with a demonstrated rescue experiment behind it.

A second well-supported pathway involves the chemokine CXCL12. Fibroblast-derived CXCL12 signals through its receptor CXCR4 on hepatocellular carcinoma cells, triggering upregulation of a molecule called FOLR1, the folate receptor 1. The result is a reduction in the apoptosis—the programmed cell death—that sorafenib is supposed to induce. Independent work using quantitative proteomics had already implicated FOLR1 in sorafenib resistance through activation of autophagy, the cellular recycling program that helps stressed cancer cells survive. When researchers interfered with the CXCL12–CXCR4 axis, including with the CXCR4 antagonist plerixafor, sorafenib’s killing power returned.

The third mechanistically resolved pathway is perhaps the most elegant. Fibroblasts deposit fibronectin bearing an extra domain A, an embryonic splice variant of the matrix protein that is largely absent from healthy adult liver but abundant in tumors and fibrotic tissue. This oncofetal form of fibronectin, long recognized as a molecular marker of hepatocellular carcinoma, appears to signal through the innate immune receptor Toll-like receptor 4 on tumor cells, activating the transcription factor NF-κB and the metabolic enzyme SHMT1. The consequence is a shift in redox metabolism: the cancer cells adapt to the oxidative stress that sorafenib generates and simply keep going. Blocking this axis restores drug sensitivity, completing the causal chain from fibroblast product to receptor to transcription factor to metabolic enzyme to survival.

When the review turns to immunotherapy, the evidence becomes more circumstantial, and the authors are careful to say so. Periostin-positive fibroblasts, a stromal subset identified in patient tumors, are associated with recruitment of macrophages through the interleukin-6 signaling pathway, elevated SPP1 expression within those macrophages, physical exclusion of immune cells from the tumor, and reduced responsiveness to immune checkpoint inhibitors. The proposed chain of events is coherent: fibroblasts build a stromal-immune barrier that keeps cytotoxic T cells at arm’s length, so the checkpoint drug has no targets to reactivate. But unlike the tyrosine kinase inhibitor stories, no experiment has yet shown that selectively reversing this specific periostin-positive fibroblast program restores immunotherapy responses in hepatocellular carcinoma. The association is strong; the causal proof is pending.

That gap reflects a broader and sobering lesson from the fibroblast field. In pancreatic cancer, the classic cautionary tale, researchers depleted carcinoma-associated fibroblasts in mouse models and expected tumors to shrink. Instead, the tumors became more aggressive and the animals died faster, revealing that some fibroblasts restrain cancer rather than feed it. The review emphasizes that fibroblast heterogeneity makes blanket depletion strategies dangerous: eliminating the wrong subtype could dismantle the stromal barriers that contain the tumor while sparing the very cells that drive resistance. The field’s central question has accordingly shifted from whether to target fibroblasts to which ones, when, and how.

Translational efforts are already adapting to this nuance. Nanoparticle systems have been engineered to deliver small interfering RNA directly to fibroblasts, silencing stromal targets such as microfibril-associated protein 5 in antistromal therapy for hepatocellular carcinoma. Other platforms exploit fibroblast biology for drug delivery: lipid nanoparticles that transform in size in response to fibroblast activation protein activity, releasing their payload deep within fibrotic tumor tissue, and dual-targeting nanoparticles that simultaneously remodel the stroma and reprogram immune cells. CXCR4 antagonist-loaded nanoparticles have been shown to reprogram the tumor microenvironment and enhance immunotherapy in preclinical models. Even diagnostic imaging is joining in, with fibroblast activation protein-targeted PET tracers now being tested to monitor liver fibrosis and tumor burden, potentially allowing clinicians to stratify patients by the stromal character of their tumors before choosing a therapy.

The review’s authors, funded by the National Natural Science Foundation of China, frame their conclusions as a call for discipline in a rapidly growing literature. The fibroblast field has generated an abundance of associations—extracellular vesicles carrying resistance-promoting microRNAs, exosomal signals that decrease sorafenib sensitivity, fibroblast-derived growth factors that sustain tumor cells through metabolic crosstalk—but only a handful of mechanisms in hepatocellular carcinoma have been validated with the combination of defined drug exposure, treatment outcomes, and functional rescue. SPP1, CXCL12–FOLR1, and fibronectin extra domain A now stand on the firmest ground, each with a plausible route to clinical intervention. For patients whose liver cancers stop responding to the current arsenal, the hope is that the next generation of treatments will not merely attack the tumor harder, but will disarm the cellular accomplices that have been quietly protecting it all along.

Subject of Research: Cancer-associated fibroblast mechanisms of therapeutic resistance in hepatocellular carcinoma

Article Title: Cancer-associated fibroblasts in therapeutic resistance of hepatocellular carcinoma: evidence, mechanisms, and translational implications

Article References: Huang, T., Li, Q., Yu, Z., Wu, Z., Cheng, H., Zhang, X., Yang, L., & Wang, S. (2026). Cancer-associated fibroblasts in therapeutic resistance of hepatocellular carcinoma: evidence, mechanisms, and translational implications. Medical Oncology, 43(11), Article 314. https://doi.org/10.1007/s12032-026-03435-y

Image Credits: AI Generated

DOI: 10.1007/s12032-026-03435-y

Keywords: hepatocellular carcinoma, cancer-associated fibroblasts, therapeutic resistance, tyrosine kinase inhibitors, immune checkpoint inhibitors, sorafenib, lenvatinib, SPP1, CXCL12, fibronectin extra domain A, tumor microenvironment, spatial transcriptomics

The motivation for the work stems from a persistent gap in regenerative biology. Chronic and acute pulmonary diseases, including chronic obstructive pulmonary disease, asthma, pneumonia and bronchiectasis, all involve damage to the bronchial epithelium, and impaired airway regeneration worsens disease progression. Scientists already knew that airway progenitors can mobilize to repair the alveoli, a one-way street of cross-compartmental help. In vitro studies had even suggested that human AT2 cells can become airway basal cells and bronchiolar secretory cells, and similar behavior had been observed in non-human primate injury models. But definitive in vivo evidence was missing, largely because the genetic tools available could not cleanly separate the cell populations involved.

The central technical obstacle was a marker problem. The standard lineage-tracing system for AT2 cells relies on the Sftpc-creER driver, which labels any cell expressing the Sftpc gene. Unfortunately, both genuine AT2 cells and bronchioalveolar stem cells (BASCs), a rare multipotent population sitting at the bronchioalveolar-duct junction that co-expresses Scgb1a1 and Sftpc, light up with this marker. Because BASCs are naphthalene-resistant and known to convert into club cells after injury, any experiment using Sftpc-creER alone could not tell whether newly formed airway cells came from AT2 cells or from BASCs. To break this ambiguity, the team built a series of five complementary dual-recombinase tracing systems that distinguish AT2 cells, BASCs and club cells from one another with unprecedented precision.

The first system, called AT2-tracer1, combined Sftpc-creER labelling with a newly generated Scgb1a1-DTR-GFP knock-in line. In triple transgenic mice, AT2 cells carried only the tdTomato reporter, while club cells and BASCs additionally carried GFP and a diphtheria toxin receptor. Administering diphtheria toxin selectively eliminated all Scgb1a1-expressing cells, wiping out the club cell lining of the bronchioles while leaving pure tdTomato-positive AT2 cells in the alveoli. Four weeks later, lineage tracing revealed that tdTomato-labelled AT2 cells had migrated into the bronchioles and generated tdTomato-positive, GFP-positive club cells. Quantification showed that 77.42 percent, give or take 4.71 percent, of the new club cells originated from AT2 cells, an astonishingly large contribution from a cell type previously thought to stay home.

The AT2-derived cells proved remarkably versatile. Within the bronchioles they differentiated into club cells, AT1 cells, ciliated cells and undifferentiated AT2 cells, all while remaining within the epithelial layer. Clonal analysis using a Confetti multicolor reporter confirmed that individually labelled AT2 cells possess multipotent transdifferentiation capacity at the single-cell level. The researchers also mapped two migratory routes: one through the bronchioalveolar-duct junction, which dominated, and a second through transient channels in the bronchiolar wall lined by AT2-derived AT1-like cells, which served as a supplementary pathway. To rule out artifacts of the toxin-ablation model, the team repeated the experiments with naphthalene injury using a nested dual-recombinase reporter system, AT2-tracer2, and again found that roughly 43 percent of regenerated club cells were of AT2 origin. A third system, AT2-tracer3, allowed simultaneous comparison of club cells, BASCs and AT2 cells in the same mouse, revealing that AT2 cells contributed the majority, about 52 percent, of regenerated club cells, outpacing both BASCs and resident club cells themselves.

Single-cell RNA sequencing then exposed the molecular choreography underlying the fate switch. Pseudotime analysis identified a transient intermediate state marked by expression of Cldn4, the gene encoding the tight-junction protein claudin-4. As AT2 cells progressed through this CLDN4-positive stage, the alveolar program was gradually silenced and the airway secretory program was switched on, with Sftpc expression falling and Scgb1a1 rising. The team built a dedicated genetic system to trace these transient cells directly and found that 62 percent of new club cells passed through the CLDN4-positive intermediate. Critically, deleting Cldn4 in AT2 cells slashed their contribution to club cells from about 66 percent to about 25 percent, demonstrating that this intermediate state is not merely a passive waypoint but a functional requirement for efficient transdifferentiation.

The next question was what instructs the migrating alveolar cells to adopt an airway identity. The answer, it turns out, is Notch signalling delivered by the cells that survive injury. Single-cell analysis showed that Notch pathway genes were upregulated in AT2-derived club cells, and ligand-receptor interaction mapping pointed to JAG1 and JAG2 on residual ciliated cells engaging NOTCH1 and NOTCH2 on the incoming alveolar cells. Functional tests were decisive: deleting the Notch signal transducer Rbpj in AT2 cells nearly abolished club cell conversion, dropping AT2-derived club output from about 38 percent to under 2 percent, while forcing Notch activation through NICD overexpression boosted conversion to about 45 percent. Ablating ciliated cells or deleting Jag1 specifically in ciliated cells, or deleting Notch2 in AT2-derived cells, each independently impaired the transition. Notch therefore acts as a bidirectional rheostat, and when it is disrupted, alveolar cells pile up in the bronchioles without differentiating, thickening airway walls and impairing ventilation.

Equally compelling is the mechanism that summons AT2 cells to the injury site in the first place. Spatial transcriptomics revealed that bronchiolar injury triggers a localized, roughly threefold increase in peribronchiolar immune infiltration, with neutrophils and monocytes secreting the chemotactic factor SPP1, also known as osteopontin. This secreted molecule forms a concentration gradient radiating outward from the damaged bronchioles, and AT2 cells sense it through the integrin subunit ITGB1 on their surface. The evidence chain is thorough: recombinant SPP1 accelerated wound closure and Transwell migration of AT2-like cells in vitro; an SPP1 inhibitor blocked the migratory effect of immune cells in co-culture; pharmacological SPP1 inhibition in living mice reduced bronchiolar coverage by AT2-derived cells; and genetic deletion of Itgb1 in AT2 cells significantly impaired their migration, thickened bronchiolar walls and worsened small-airway function. Matrix metalloproteinases such as MMP9 and MMP14, induced in immune and stromal cells, likely help clear a physical path through the extracellular matrix.

One of the most reassuring aspects of the study is what happens over the long term. At four weeks after injury, AT2-derived club cells still carried residual alveolar and PATS-like transcriptional signatures, raising the question of whether they were truly equivalent to native airway cells. By six months, however, single-cell multi-omics showed that these cells had converged with BASC-derived and club-cell-derived populations at both the transcriptomic and epigenetic levels. The AT2-specific chromatin accessibility signature declined rapidly and stayed low, while club-cell-specific secretory and immune gene programs rose progressively. AT2-derived club cells even retained progenitor capacity, proliferating and generating ciliated cells after a secondary naphthalene challenge. In other words, prolonged repair resolves lineage-specific differences through maturation to a common functional endpoint.

The implications reach well beyond basic biology. Current clinical strategies for airway repair focus on airway-resident stem cells such as basal and secretory cells, but this work suggests AT2 cells could serve as an alternative, and arguably more accessible, cellular reservoir for airway regeneration, particularly given their ability to mature fully into functional club cells. The identification of SPP1-ITGB1 signalling as the migration cue and Notch activation as the fate switch offers two concrete molecular targets for enhancing cross-compartmental repair. The authors are careful to note the caveats: mouse and human lungs differ structurally, and whether a conserved AT2-to-club program exists in humans remains unresolved, though in vitro evidence from human and primate systems is suggestive. Questions also remain about whether AT2 cells pass through a BASC-like state and whether other niche cells contribute. Still, the demonstration that the lung’s two great compartments are connected by a two-way regenerative dialogue rewrites a core assumption of pulmonary biology and opens a genuinely new chapter in the search for therapies that rebuild damaged airways.

Subject of Research: Cross-compartmental transdifferentiation of alveolar type 2 stem cells driving bronchiolar epithelial regeneration in the mouse lung

Article Title: Alveolar stem cells transdifferentiate to drive bronchiolar regeneration

Article References: Liu, K., Liu, Z., Meng, X., Tang, M., Di, F., Wang, Z., Li, C., Yang, S., Zhang, Y., Yang, X., Lv, Z., Li, X., Jin, H., Pu, W., Zhao, H., Li, F., Sui, P., & Zhou, B. (2026). Alveolar stem cells transdifferentiate to drive bronchiolar regeneration. Nature. https://doi.org/10.1038/s41586-026-11127-w

Image Credits: AI Generated

DOI: 10.1038/s41586-026-11127-w

Keywords: alveolar type 2 cells, bronchiolar regeneration, lineage tracing, Notch signalling, SPP1, ITGB1, club cells, CLDN4, lung stem cells, transdifferentiation, ciliated cells, lung injury repair

SPP1 is a secreted glycosylated phosphoprotein with a long-established resume in skeletal biology, where it participates in bone mineralization and the remodeling of the extracellular matrix. But in tumors, the same molecule takes on a darker role. The review synthesizes evidence showing that SPP1 is produced by multiple players within the tumor microenvironment, including tumor cells themselves, tumor-associated macrophages and cancer-associated fibroblasts. Once secreted, it signals through a family of cell-surface receptors, chiefly integrins that recognize its arginine-glycine-aspartic acid motif, as well as variants of the CD44 receptor. Through these interactions, SPP1 transmits instructions that reshape how immune cells behave, how they move through tissue, and whether they can reach the tumor at all.

One of the most consequential effects of SPP1 signaling is its influence on macrophages, the versatile immune cells that populate nearly every tumor. The review describes how SPP1 drives the polarization of tumor-associated macrophages toward a pro-tumor phenotype, a state in which these cells abandon their immune-surveillance duties and instead support tumor growth, angiogenesis and tissue remodeling. Single-cell RNA sequencing studies, which allow researchers to profile gene expression in individual cells, have repeatedly identified distinct populations of SPP1-positive macrophages in human tumors, and these cells are consistently associated with poor clinical outcomes. Rather than simply being bystanders, these macrophages appear to be active architects of an immunosuppressive environment, secreting SPP1 that reinforces their own suppressive identity and that of their neighbors.

The consequences extend well beyond macrophages. According to the review, SPP1 impairs the function of innate immune cells, the rapid-response arm of the immune system that includes natural killer cells and dendritic cells. It also induces the exclusion and exhaustion of T cells, the cytotoxic lymphocytes that immunotherapies such as immune checkpoint inhibitors are designed to unleash. T cell exhaustion is a state of progressive dysfunction in which T cells lose their ability to kill tumor cells, and SPP1 appears to promote this process while simultaneously building physical barriers that keep T cells away from tumor cells in the first place. By remodeling the extracellular matrix, the fibrous scaffold that permeates tumors, SPP1 helps create a form of immune exclusion in which anti-tumor lymphocytes linger at the tumor margin, unable to penetrate the tumor core.

This combination of effects helps explain why many patients fail to respond to immune checkpoint inhibitors, the blockbuster therapies that block inhibitory receptors such as PD-1 and have transformed the treatment of melanoma, lung cancer and other malignancies. Checkpoint inhibitors work best in tumors that are already infiltrated by T cells, so-called hot tumors. SPP1-rich tumors, by contrast, tend to be cold, lacking meaningful lymphocyte infiltration and exhibiting dense, suppressive stroma. The review positions the SPP1 axis as a promising therapeutic target precisely because it operates upstream of these failures: neutralizing SPP1 or blocking its receptors could convert a cold, excluded tumor into one that immune therapies can actually reach.

Encouragingly, the review catalogs a growing arsenal of strategies aimed at doing exactly that. Direct SPP1 neutralization, using antibodies or other decoy molecules, aims to mop up the protein before it reaches its receptors. Receptor blockade seeks to sever the downstream signaling that integrins and CD44 variants transmit into immune and stromal cells. A third approach targets the source: depleting or reprogramming the SPP1-positive tumor-associated macrophages that manufacture much of the tumor’s supply. Each strategy attacks a different node of the axis, and the review suggests that combining them with existing modalities may be where the greatest potential lies.

Those combination possibilities are particularly striking. The authors describe evidence supporting the pairing of SPP1 targeting with immune checkpoint inhibitors, with agonists of the STING pathway, a innate immune sensor that when activated can ignite anti-tumor inflammation, with cytokine-based therapies, and with conventional treatments such as radiotherapy and chemotherapy. The logic is complementary: SPP1 blockade dismantles the immunosuppressive and exclusionary architecture of the tumor, while the partner therapy supplies or amplifies the immune attack. Radiotherapy and chemotherapy, for their part, can release tumor antigens and induce immunogenic cell death, effects that SPP1 targeting may help convert into durable systemic immunity rather than transient local responses.

Drug delivery innovation is also entering the picture. The review highlights work on lipid nanoparticles, the same lipid-based vehicles that proved their worth in mRNA vaccines, engineered to deliver SPP1-targeting small interfering RNA into tumors. One described design uses mannosylated nanoparticles, decorated with sugar molecules that guide them toward macrophages, to deliver SPP1 siRNA in combination with interferon-gamma, a cytokine that can push macrophages back toward an anti-tumor state. Such approaches illustrate how the field is moving beyond simple antibody blockade toward precision tools that can reprogram specific cell populations within the tumor microenvironment, potentially resetting the balance between suppression and immunity at its source.

Yet the review is careful to temper enthusiasm with a sober accounting of the translational challenges ahead. SPP1 signaling is context-dependent: the same molecule can exert different, sometimes opposing, effects depending on the receptor engaged, the cell type targeted and the stage of tumor progression. Osteopontin has documented roles in normal wound healing and tissue homeostasis, raising the possibility that systemic blockade could interfere with beneficial biology. The authors also emphasize the need for biomarker development and patient stratification, since identifying which patients harbor SPP1-driven, immune-excluded tumors will be essential for selecting who benefits from SPP1-directed therapies and for designing clinical trials capable of demonstrating meaningful benefit.

The broader message of the review is that the tumor microenvironment is not a passive backdrop but an actively engineered landscape, and SPP1 is one of its chief engineers. By promoting pro-tumor macrophage polarization, disabling innate immunity, exhausting and excluding T cells, and hardening the extracellular matrix into a physical barrier, SPP1 sits at a convergence point of nearly every mechanism tumors use to hide from the immune system. The authors conclude that targeting the SPP1 axis offers a promising strategy for remodeling the tumor microenvironment and enhancing antitumor immunity, a conclusion that is likely to accelerate preclinical and clinical efforts. If those efforts succeed, a molecule named for its abundance in bone may become one of the most important levers for turning immunologically cold tumors into ones that modern immunotherapy can finally heat up.

Subject of Research: The role of secreted phosphoprotein 1 (SPP1) in regulating immunosuppression in the tumor microenvironment and therapeutic strategies targeting the SPP1 axis

Article Title: Reprogramming the tumor immune landscape through SPP1 targeting

Article References: Gao, L., Yang, J., & Teng, Y. (2026). Reprogramming the tumor immune landscape through SPP1 targeting. Journal of Experimental & Clinical Cancer Research. https://doi.org/10.1186/s13046-026-03844-x

Image Credits: AI Generated

DOI: 10.1186/s13046-026-03844-x

Keywords: SPP1, osteopontin, tumor microenvironment, tumor-associated macrophages, cancer-associated fibroblasts, immune evasion, T cell exclusion, immune checkpoint inhibitors, CD44, integrins, lipid nanoparticles, cancer immunotherapy

The research team, led by Bingxuan Zheng, Junbo Li, Qi He, and senior author Chenguang Ding, established a mouse model of kidney transplantation in which donor kidneys were transplanted into unrelated recipients, recreating the immunological conflict that occurs in human transplantation. Rather than sampling the graft at a single moment, the investigators collected immune cells at multiple post-transplant stages, capturing the transition from acute rejection, when the immune assault is at its most violent, to the chronic phase, when fibrosis and gradual functional decline take hold. This longitudinal design is critical, because chronic rejection is not a static condition but a dynamic process in which the cellular players and their relationships shift over time. By sampling across that timeline, the researchers could watch the immune landscape evolve rather than merely snapshot its endpoint.

The technological core of the study is the pairing of single-cell RNA sequencing with single-cell ATAC sequencing. Single-cell RNA sequencing, or scRNA-seq, measures which genes are actively being transcribed in each individual cell, revealing the identity and functional state of thousands of cells simultaneously. Single-cell ATAC sequencing, or scATAC-seq, measures which regions of chromatin are physically accessible in each cell’s genome, exposing the regulatory landscape that determines which genes a cell could potentially activate. By analyzing both layers of information in the same immune cell populations, the researchers gained not just a census of which cells were present, but insight into the transcription factor programs driving their behavior. The team supplemented these core assays with computational tools including CellChat for mapping communication between cell types, RNA velocity and Monocle3 for inferring developmental trajectories, UMAP for visualization, and Weighted Gene Co-expression Network Analysis for identifying gene modules correlated with disease states.

Out of this multi-omic atlas emerged a striking finding: a late-stage, terminal-like state of macrophages defined by expression of the gene SPP1, which encodes secreted phosphoprotein 1, also known as osteopontin. These SPP1-positive macrophages were markedly enriched in chronically rejecting allografts compared with earlier stages. Macrophages are the tissue-resident scavengers and sentinels of the immune system, capable of enormous functional plasticity, and the study’s trajectory analyses suggest that the SPP1-positive population represents a terminal differentiation state that macrophages in the graft progressively adopt as rejection moves from acute inflammation toward chronic fibrosis. In other words, these cells appear to be not transient visitors but an end-stage product of immune evolution within the failing graft.

Position matters in tissue immunology, and the SPP1-positive macrophages were not merely abundant; they were central. When the researchers constructed ligand-receptor interaction networks to map how different immune cell populations communicate with one another, the SPP1-positive macrophages occupied a hub position in a macrophage-centered communication network. This means that many of the signaling conversations shaping the graft’s immune environment either originated from or passed through these cells. A cell at the center of such a network has outsized influence: it can recruit other immune cells, shape their differentiation, and coordinate collective behaviors such as tissue remodeling. The implication is that SPP1-positive macrophages may act as orchestrators of the fibrotic microenvironment rather than passive bystanders in a deteriorating graft.

The chromatin accessibility data added a crucial mechanistic dimension. Motif enrichment analysis of the scATAC-seq data revealed that the open regulatory regions in SPP1-positive macrophages were characterized by binding sites for CEBPβ, a transcription factor known as CCAAT/enhancer-binding protein beta. This finding identifies a CEBPβ-centered regulatory program as a candidate driver of the SPP1-positive macrophage state. Even more tellingly, the genes under this regulatory program were associated with sustained activation of extracellular matrix remodeling pathways. The extracellular matrix is the protein scaffold that gives tissue its structure, and its excessive deposition and remodeling is the defining feature of fibrosis. A regulatory program that keeps matrix-remodeling genes switched on in macrophages provides a plausible molecular link between chronic inflammation and the scarring that ultimately destroys graft function.

Perhaps the most intriguing cell-cell interaction identified in the study was a regulatory axis connecting SPP1-positive macrophages with CD8-positive tissue-resident memory T cells, abbreviated Trm cells. Tissue-resident memory T cells are a specialized population of T lymphocytes that take up long-term residence in tissues rather than circulating through blood and lymph nodes. In transplanted organs, they are increasingly recognized as persistent local actors that can sustain immune attack independently of the systemic immune system. The integrative ligand-receptor and pathway analyses in this study suggest that SPP1-positive macrophages and CD8-positive Trm cells engage in a predicted interaction network that is associated with the fibrosis-related immune remodeling characteristic of chronic rejection. This macrophage-T cell axis offers a concrete cellular partnership around which future mechanistic studies can be designed.

The significance of this work lies partly in its methodology and partly in its framing. Most studies of transplant rejection have relied on bulk tissue analysis, which averages signals across millions of cells and can mask the critical contributions of rare or state-specific populations. By resolving the graft immune environment at single-cell resolution across time, and by integrating gene expression with chromatin accessibility, the researchers have produced what they describe as a high-resolution longitudinal immune atlas of kidney allograft rejection. Within that atlas, SPP1-positive macrophages and the CEBPβ regulatory program stand out as candidate contributors to the fibrotic process. If validated, these targets could open therapeutic avenues that current immunosuppression does not address, since standard drugs broadly dampen lymphocyte activation but are not designed to reprogram macrophage differentiation or block matrix remodeling programs.

The authors are careful to emphasize the limits of what their study establishes. The work was performed in a mouse transplantation model, and the findings are framed as candidate regulators and predicted interaction networks rather than proven causal mechanisms. Functional validation, in which the activity of SPP1 or CEBPβ is experimentally manipulated to test whether fibrosis is altered, remains necessary. Human translation is an additional hurdle, since the immune environment of human transplanted kidneys may differ in important ways from the mouse model, and sampling human graft tissue at multiple time points presents practical and ethical challenges. The study was conducted under approved animal protocols at Xi’an Jiaotong University, and the researchers state that their findings offer a framework for future mechanistic and therapeutic investigations rather than immediate clinical application.

Even with those caveats, the study adds an important piece to the puzzle of why transplanted kidneys fail over the long term. Chronic rejection has long been viewed as an intractable, slowly accumulating process, but single-cell multi-omics is beginning to reveal that it has an identifiable cellular logic: specific immune cell states, governed by specific transcription factors, communicating through specific ligand-receptor pairs, drive the tissue toward scarring. By naming SPP1-positive macrophages, CEBPβ, and the CD8-positive tissue-resident memory T cell axis as central players, this atlas gives transplant immunologists a concrete set of hypotheses to test and, potentially, a new generation of therapeutic targets to pursue. For the hundreds of thousands of transplant recipients worldwide whose grafts slowly fail despite the best available immunosuppression, that kind of molecular specificity is exactly what the field has been waiting for.

Subject of Research: Single-cell multi-omics of immune cell dynamics during chronic rejection of transplanted kidneys

Article Title: Single-cell multi-omics characterization of rejection in transplanted kidneys reveals key cell types and candidate regulators of chronic rejection

Article References: Zheng, B., Li, J., He, Q., Li, Z., Wei, T., Gao, Y., Zhu, Y., Deng, G., Yang, M., Tian, P., & Ding, C. (2026). Single-cell multi-omics characterization of rejection in transplanted kidneys reveals key cell types and candidate regulators of chronic rejection. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-09002-3

Image Credits: AI Generated

DOI: 10.1186/s12967-026-09002-3

Keywords: kidney transplantation, chronic rejection, single-cell RNA sequencing, single-cell ATAC sequencing, macrophages, SPP1, CEBPβ, CD8-positive tissue-resident memory T cells, fibrosis, extracellular matrix remodeling, transplant immunology, multi-omics

Hirschsprung disease arises when enteric neural crest cells fail to migrate and differentiate properly during development, leaving the distal intestine devoid of the myenteric and submucosal nerve plexuses that normally coordinate peristalsis. The result is abdominal distension and refractory constipation, typically treated by surgically removing the aganglionic segment and reconstructing the bowel. Yet a substantial proportion of patients continues to suffer long-term complications, including enterocolitis, anastomotic strictures, fecal soiling, persistent constipation, and the need for reoperation. These persistent problems have intensified the search for the molecular mechanisms underlying the disease, and the new work points squarely at the glial microenvironment.

Enteric glial cells are the most abundant non-neuronal component of the enteric nervous system, and they are far more than passive scaffolding. They maintain epithelial integrity, coordinate neuro-immune signaling, and support neuronal survival. Recent single-cell RNA sequencing studies have revealed that glia exist in multiple transcriptionally distinct states, some of which resemble reactive microglia in the brain. In tumor-associated macrophages, researchers have described a CXCL9-to-SPP1 polarization axis that separates anti-inflammatory from pro-disease functional states. Whether enteric glia undergo a comparable polarization program in Hirschsprung disease had never been established, and that question became the starting point for the investigation.

The team began with an integrated single-cell transcriptomic analysis of colonic tissue from a mouse model of Hirschsprung disease based on loss of the endothelin receptor B gene, Ednrb, alongside wild-type controls. Among the five transcriptionally distinct glial subclusters they identified, one stood out: a subcluster enriched for both TXNRD1 and SPP1 that expanded markedly in disease samples, while a subcluster marked by Apoe shrank. Enrichment analyses of this expanded population highlighted extracellular matrix organization, cell adhesion, focal adhesion, and PI3K-Akt and MAPK signaling programs. Computational modeling of ligand-receptor interactions further showed that this subcluster engaged in the strongest predicted communication with enteric neurons of any glial population, including a predicted VCAM1-NCAM1 signaling axis.

To probe the regulatory logic behind this glial remodeling, the researchers turned to network-based virtual perturbation using the scTenifoldNet framework, which simulates gene knockout or overexpression within an inferred gene regulatory network. Both virtual knockout and virtual overexpression of TXNRD1 converged on regulatory programs involving redox control, ERK/MAPK-associated immediate-early response genes, and stress-associated glial features. Virtual overexpression produced a more focused pathological shift, with SPP1 emerging as the most prominently perturbed gene. These in silico results positioned TXNRD1, a central regulator of intracellular redox homeostasis, as a plausible upstream driver of the disease-associated glial state.

Human tissue evidence followed. Proteomic profiling of paired proximal dilated and stenotic colonic segments from children undergoing pull-through surgery revealed marked TXNRD1 upregulation in the stenotic, aganglionic tissue, together with significant enrichment of the MAPK signaling pathway. Validation in eight paired patient samples using quantitative PCR, western blotting, and immunofluorescence confirmed elevated TXNRD1 alongside increased phosphorylation of Raf, MEK, and ERK, while the stress kinases p38 and JNK showed no significant changes, indicating selective activation of the ERK branch of the MAPK cascade. Immunofluorescence staining also showed that SPP1 was increased and CXCL9 reduced within S100β-positive enteric glia in the stenotic segments, mirroring the single-cell findings in human tissue.

The mechanistic experiments in cultured rat enteric glial cells tied these observations together. When the researchers overexpressed TXNRD1, the cells selectively increased ERK phosphorylation, elevated their production of SPP1, suppressed CXCL9, accumulated reactive oxygen species, migrated less effectively, and secreted less glial cell line-derived neurotrophic factor, or GDNF. Treatment with the ERK inhibitor SCH772984 partially restored CXCL9 expression and blunted SPP1 upregulation, while the antioxidant N-acetyl-L-cysteine reduced ROS accumulation, dampened ERK activation, and reversed the SPP1-high/CXCL9-low polarization. Notably, ERK inhibition did not normalize ROS levels, suggesting that redox dysregulation acts upstream of ERK/MAPK signaling rather than as a downstream consequence.

The consequences for neurons were striking. In co-culture experiments with primary dorsal root ganglion neurons, TXNRD1-overexpressing glia increased neuronal apoptosis, impaired neuronal migration, reduced phosphorylation of the RET receptor and expression of its co-receptor GFRα1, downregulated the synaptic proteins PSD-95 and synaptophysin, and disrupted intracellular calcium homeostasis. Levels of the neurotrophic factors NGF and BDNF also fell in the co-culture system. Crucially, both ERK inhibition and supplementation with exogenous GDNF partially rescued these defects, implicating suppression of the GDNF/GFRα1/RET signaling axis as a key route by which the remodeled glia damage neuronal development and function.

The in vivo experiments provided the most compelling translational signal. The team generated Ednrb conditional knockout mice, which develop Hirschsprung-like phenotypes, and delivered an adeno-associated virus carrying TXNRD1-targeting shRNA by intraperitoneal injection at postnatal day 5. Knockdown of TXNRD1 markedly ameliorated the disease features, reducing abdominal distension, proximal colonic dilatation, distal stenosis, and fecal retention. Treated mice survived significantly longer, with four animals reaching the study endpoint, and fecal water content rose toward control levels, indicating improved intestinal function. At the molecular level, TXNRD1 suppression decreased SPP1, restored CXCL9, increased the neuronal marker TUJ1 and the neural crest marker p75, and upregulated the communication mediators NCAM1 and VCAM1 in the bowel.

The authors are careful to acknowledge the limitations of their work. The proteomic screen relied on only two paired patient samples, the validation cohorts were small, and the intraperitoneal viral delivery was not restricted to enteric glia, complicating cell-type-specific interpretation. Primary dorsal root ganglion neurons served as a surrogate for enteric neurons, and the temporal origins of the SPP1-high glial state remain unresolved, with lineage tracing and spatial transcriptomics needed to clarify its development. Even so, the study establishes a coherent mechanistic framework linking redox dysregulation, ERK/MAPK activation, pathological glial polarization, and neuronal dysfunction in Hirschsprung disease. By showing that enteric glia are active participants in the disease rather than passive bystanders, and that silencing TXNRD1 alleviates HSCR-like phenotypes in mice, the work nominates TXNRD1 and the glial polarization program it controls as promising targets for future therapies aimed at improving outcomes for children with this challenging disorder.

Subject of Research: The role of TXNRD1-driven enteric glial remodeling in neuronal dysfunction in Hirschsprung disease

Article Title: TXNRD1-driven enteric glial remodeling promotes neuronal dysfunction in Hirschsprung disease

Article References: Li, Y., Wang, Y., Jin, Z., Tang, C., Xia, X., Gong, Y., Du, Q., Huang, L., Li, Z., Liao, Y., He, S., Wang, B., Liu, Y., & Zheng, Z. (2026). TXNRD1-driven enteric glial remodeling promotes neuronal dysfunction in Hirschsprung disease. iScience, 29(10), Article 117636. https://doi.org/10.1016/j.isci.2026.117636

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117636

Keywords: Hirschsprung disease, enteric glial cells, TXNRD1, SPP1, CXCL9, ERK/MAPK signaling, redox dysregulation, GDNF, single-cell RNA sequencing, enteric nervous system, Ednrb knockout mice, neurotrophic signaling

The central difficulty the researchers set out to solve is a familiar one in genomics. Bulk RNA sequencing, which measures the average gene activity across millions of cells in a tissue sample, is cheap and widely available but blurs the contributions of individual cell types. Single-cell RNA sequencing, by contrast, resolves gene expression cell by cell, revealing the dazzling diversity of immune cells infiltrating a tumour, but it is expensive and applied to relatively few samples. The team, led by Ye Wang, Zhicheng Zhuang and Tengyi Peng, who contributed equally to the work, downloaded large volumes of both bulk and single-cell CRC data from multiple public databases and set about integrating the two scales of information into a single, coherent picture of the colorectal tumour microenvironment.

The linchpin of their integrative strategy is an algorithm called Scissor. Rather than treating bulk and single-cell datasets as separate worlds, Scissor correlates the two, using bulk tumour samples annotated with clinical outcomes and genomic data to tag individual cells in the single-cell data as either associated with a poor phenotype or a favourable one. Cells linked to adverse outcomes are designated Scissor-positive, while those linked to better outcomes are Scissor-negative. When the researchers applied this framework to macrophages infiltrating colorectal tumours, a striking pattern emerged: the Scissor-positive subgroup of TAMs, the macrophages whose expression patterns mirrored the molecular profiles of tumours from patients with poor outcomes, was strongly associated with worse prognosis. In other words, not all macrophages are equal, and a specific transcriptional state of these cells carries a measurable survival penalty.

Having split TAMs into these prognostically distinct subtypes, the team moved to identify the characteristic genes that define the dangerous macrophage state. By comparing the gene expression profiles of Scissor-positive and Scissor-negative macrophages, and cross-referencing with clinical follow-up data, they pinpointed a set of TAM subset genes closely tied to patient survival. Among these, one gene stood out: SPP1, which encodes osteopontin, a secreted protein with well-documented roles in cell adhesion, migration and immune modulation. To probe what SPP1 actually does in colorectal cancer cells, the researchers turned to the laboratory bench, performing colony formation assays to measure proliferative capacity, scratch assays to gauge cell migration, and invasion assays to test the cells’ ability to penetrate surrounding matrix. These functional experiments provided direct evidence that SPP1 is not merely a passive biomarker but an active participant in the malignant behaviour of colorectal cancer cells.

The next phase of the study tackled a problem that has increasingly dogged the field of computational oncology: with dozens of machine learning algorithms available, each with its own biases and assumptions, how do you decide which model best captures the biology? The team deployed multiple machine learning techniques in parallel and compared their outputs, ultimately converging on nine TAM signature genes associated with colorectal cancer progression. This consensus-driven approach, letting several algorithms vote rather than trusting a single method, is designed to produce gene sets that are robust rather than artefacts of one particular modelling framework. The nine-gene signature then served as the backbone for a risk stratification scheme, in which each patient with colorectal cancer could be assigned a risk score based on the expression levels of these macrophage-derived genes in their tumour.

The clinical payoff of this risk score is potentially substantial. Colorectal cancer management currently leans heavily on tumour stage, nodal status and a handful of established markers, but outcomes among patients with apparently similar disease can diverge dramatically. A macrophage-based signature offers a window into the immunological character of an individual’s tumour, information that conventional staging cannot capture. Patients flagged as high risk by the TAM signature might, in principle, be candidates for more intensive surveillance or adjuvant therapy, while those in the low-risk group could potentially be spared unnecessary treatment. The study’s authors suggest that this TAM-related prognostic model could become an important tool in the management of patients with colorectal cancer, complementing rather than replacing existing clinical criteria.

To make the signature usable at the bedside, the researchers went one step further and linked the TAM signature genes to clinicopathological features, the standard clinical variables such as patient characteristics and disease measures that oncologists already collect. From this integration they constructed a nomogram, a graphical calculating tool long favoured in oncology for translating complex multivariable models into a single, individualized probability estimate. By plotting a given patient’s values on the nomogram’s axes and summing the corresponding points, a clinician can obtain a personalized assessment of prognosis that blends the molecular intelligence of the macrophage signature with the familiar anchors of clinical practice. It is a concrete step toward turning single-cell atlas data, which can feel abstract and remote, into a decision-support instrument.

The study also speaks to a broader shift in how cancer immunology is being conducted. The tumour microenvironment has moved from the margins of cancer research to its centre, driven in large part by the success of immunotherapies that unleash immune cells against tumours. But immunotherapy responses are notoriously uneven, and macrophages are increasingly implicated as one reason why: in many tumours, TAMs actively suppress T-cell activity and promote tissue remodeling that favours invasion and metastasis. By dissecting TAM heterogeneity at single-cell resolution and tying specific macrophage states to patient survival, the Fujian team’s work adds to a growing body of evidence that the immune contexture of a tumour, not just its genetic mutations, shapes its clinical trajectory. The identification of SPP1 as a functional driver further suggests that macrophage-related pathways could be druggable targets in their own right.

Caveats remain, as they do for any study built primarily on computational analysis of public datasets. Gene expression signatures derived from retrospective cohorts must be validated in independent, prospectively collected patient populations before they can inform treatment decisions, and the biological mechanisms linking the nine-gene signature to tumour progression will require deeper experimental dissection beyond the SPP1 functional assays reported here. The authors themselves frame the work as an analysis of the potential association between TAMs and colorectal cancer prognosis, a foundation on which future studies can build. Still, the methodological template, fusing bulk and single-cell transcriptomics through the Scissor algorithm, stress-testing candidate gene sets with multiple machine learning approaches, and validating key players in the wet lab, represents a rigorous and reproducible pipeline that other cancer types could readily adopt.

For patients with colorectal cancer, the promise is a future in which a routine genomic readout of their tumour reveals not only which mutations it carries but what kind of immune neighbourhood it inhabits, and in which that information translates into a personalized risk estimate and a tailored treatment plan. For researchers, the study delivers a curated set of macrophage signature genes and a functional lead in SPP1 to interrogate further. The work was supported by the National Natural Science Foundation of China, the Fujian Province Finance Project and talent programs from the First Affiliated Hospital of Fujian Medical University, and it appears as an open-access publication, making the full technical detail available to clinicians and computational biologists alike. As single-cell technologies mature and machine learning methods are tamed by consensus approaches, the macrophages lurking inside colorectal tumours are giving up their secrets, one transcript at a time.

Subject of Research: Tumour-associated macrophage gene signatures and prognostic modelling in colorectal cancer using bulk and single-cell RNA sequencing and machine learning

Article Title: Bulk and single-cell RNA-sequencing analyses along with machine learning methods identify macrophage signatures in colorectal cancer

Article References: Wang, Y., Zhuang, Z., Peng, T., Pan, Z., Zhang, Q., Zheng, S., Liu, X., Zhuang, J., Guan, G., Li, S., & Chen, B. (2026). Bulk and single-cell RNA-sequencing analyses along with machine learning methods identify macrophage signatures in colorectal cancer. Cancer Cell International. https://doi.org/10.1186/s12935-026-04445-6

Image Credits: AI Generated

DOI: 10.1186/s12935-026-04445-6

Keywords: colorectal cancer, tumour-associated macrophages, single-cell RNA sequencing, bulk RNA sequencing, machine learning, Scissor algorithm, tumour microenvironment, SPP1, prognostic signature, nomogram, gene expression profiling, transcriptomics

The work, published in Molecular Cancer, addresses a long-standing frustration in tumor biology. Cancer-associated fibroblasts, or CAFs, have been recognized for decades as central players in stromal remodeling, immune exclusion, and drug resistance, but most classifications of these cells have been static snapshots. Pathologists and single-cell biologists could label fibroblasts as myofibroblastic, inflammatory, or antigen-presenting subtypes, yet the field lacked a causal account of how tumor cells instruct fibroblasts to adopt these pathogenic identities in the first place. The new study set out to answer precisely that question: does the tumor merely coexist with certain fibroblast states, or does it actively manufacture them?

To map the landscape, the team performed integrative single-cell RNA sequencing on lung adenocarcinoma specimens, constructing a detailed stromal atlas of the disease and tracing fibroblast lineage trajectories. Rather than falling into discrete categories, fibroblasts in these tumors followed a continuous differentiation path that converged on a myofibroblast-like endpoint marked by MMP11 expression. This MMP11-positive subset carried a transcriptional program dominated by extracellular matrix remodeling and collagen organization, and its gene-expression signature overlapped strongly with markers of immune exclusion. Across multiple independent patient cohorts, the abundance of this fibroblast state tracked with worse clinical outcomes, suggesting it is not a bystander but an active participant in disease progression.

The next question was mechanistic: what turns ordinary fibroblasts into MMP11-producing matrix builders? The researchers co-cultured tumor cells with fibroblasts and found that different lung adenocarcinoma cell lines varied dramatically in their ability to trigger the program. When they separated the tumor cells’ influence into soluble factors and extracellular vesicles, the vesicle fraction carried most of the programming activity. Extracellular vesicles are nanoscale lipid-enclosed particles that cells release to shuttle proteins, lipids, and nucleic acids to recipient cells, functioning as a kind of biological postal system. Using isolation and functional fractionation techniques, the team demonstrated that removing vesicles from tumor-conditioned media largely abolished the induction of the MMP11-positive state, while purified vesicles restored it.

To identify the cargo responsible, the researchers turned to quantitative proteomics of the vesicles themselves, comparing vesicles with high CAF-programming activity against those with low activity using both data-dependent and data-independent acquisition mass spectrometry. One protein stood out: secreted phosphoprotein 1, better known as SPP1 or osteopontin, a secreted glycoprotein previously implicated in metastasis and immune regulation. SPP1 was selectively enriched in the vesicles that most potently drove fibroblast reprogramming, indicating that the tumor does not simply dump a generic mix of molecules into its vesicles but actively sorts specific signaling proteins into them.

The mechanistic chain was then traced to its receptor. Vesicle-associated SPP1, the study found, engages primarily the CD44 receptor on the surface of fibroblasts. This binding activates the PI3K-AKT signaling pathway, a canonical pro-survival and pro-growth cascade, and sustains the transcriptional changes that push fibroblasts into the MMP11-positive phenotype. Genetic and pharmacologic perturbations confirmed the pathway’s causality: blocking SPP1, CD44, or PI3K-AKT signaling prevented the fibroblast reprogramming in co-culture systems. In other words, the tumor-to-stroma communication is not diffuse background noise but a specific, druggable ligand-receptor axis.

The clinical implications emerged clearly from multiplex immunofluorescence staining of lung adenocarcinoma specimens. Tumors rich in MMP11-positive fibroblasts showed dense collagenous matrix deposition, physical exclusion of T cells from the tumor nests, and reduced intratumoral immune infiltration. In cohorts of patients treated with chemo-immunotherapy, the presence of this fibroblast state predicted poorer responses, consistent with the idea that a matrix-rich, immune-excluded microenvironment undermines the ability of checkpoint inhibitors to mobilize anti-tumor T cells. The findings thus provide a mechanistic bridge between a specific stromal cell state and the well-known clinical problem of immunotherapy resistance in lung cancer.

Crucially, the study went beyond correlation and tested whether interrupting the EV-SPP1 axis could reverse the process in living organisms. In immunocompetent mouse models of lung tumor growth and metastasis, disrupting the pathway reduced the development of MMP11-positive CAFs, loosened the dense stromal architecture, and revived anti-tumor T-cell immunity. Most strikingly, combining EV-SPP1 pathway disruption with PD-1 immune checkpoint blockade made tumors substantially more sensitive to the immunotherapy, transforming a resistant microenvironment into one that immune cells could penetrate and attack. This positions the vesicle cargo protein not just as a biomarker but as a candidate therapeutic target in its own right.

The broader significance of the work lies in reframing how scientists think about the tumor microenvironment. Rather than treating fibroblast diversity as an intrinsic property of stromal cells, the study shows that tumor cells can actively author the identity of their neighbors through targeted vesicle-mediated delivery of signaling proteins. This tumor-instructed model of stromal reprogramming suggests that effective immunotherapy strategies in lung adenocarcinoma may need to target the communication lines between cancer cells and fibroblasts, not only the immune checkpoints themselves. If the EV-SPP1-CD44-PI3K-AKT circuit can be safely intercepted in patients, it could convert a large fraction of immune-excluded, therapy-resistant lung tumors into ones that respond to existing treatments, offering a rational path to combination regimens built on the biology of intercellular vesicle trafficking.

Subject of Research: Tumor-derived extracellular vesicle SPP1 programming of MMP11-positive cancer-associated fibroblasts driving immune exclusion in lung adenocarcinoma

Article Title: Extracellular vesicle–enriched SPP1 programs MMP11⁺ fibroblast states to drive immune exclusion in lung adenocarcinoma

Article References: He, X., Qian, L., Zhao, D., Wang, G., Zhang, X., Ye, Y., Li, L., Wen, Y., & Zhang, L. (2026). Extracellular vesicle–enriched SPP1 programs MMP11⁺ fibroblast states to drive immune exclusion in lung adenocarcinoma. Molecular Cancer. https://doi.org/10.1186/s12943-026-02785-5

Image Credits: AI Generated

DOI: 10.1186/s12943-026-02785-5

Keywords: lung adenocarcinoma, cancer-associated fibroblasts, extracellular vesicles, SPP1, MMP11, immune exclusion, tumor microenvironment, immunotherapy resistance, PD-1 blockade, PI3K-AKT signaling, CD44, single-cell RNA sequencing