The problem TAKO addresses is a familiar one in the genomics community. Large-scale single-cell perturbation screens, such as Perturb-seq, can in principle map the function of thousands of genes at single-cell resolution, but in practice genome-editing efficiencies are imperfect, experimental cycles are long, and ethical and safety constraints limit what can be tested. Meanwhile single-cell measurements are noisy and sparse, complicating the detection of true differential responses. For most biological systems, researchers therefore have only wild-type data, and they must guess which genes will change when a candidate gene is removed. Existing in silico perturbation tools fall into two broad camps: graph-centric methods that operate directly on wild-type-derived gene regulatory networks, and representation-learning approaches such as scGen and GenKI that train latent generative models and simulate interventions in the learned space. Both camps have drawbacks. Graph methods inherit every error in the inferred network, while generative models demand substantial training, careful tuning of hyperparameters, and often yield predictions that are hard to interpret mechanistically.

TAKO takes a deliberately minimalist path between these approaches. The workflow starts with a wild-type expression matrix of cells by genes. The authors normalise and log-transform the data, select highly variable genes, and then construct a directed gene regulatory graph using a principal component regression scheme: they compute a low-rank singular value decomposition of the expression data, fit ridge regression in the reduced space, and back-project the coefficients to obtain directed gene-gene interactions. The strongest off-diagonal entries are retained, converted to non-negative edge weights, and row-normalised, producing a stochastic transition matrix that defines a Markov chain over genes. In other words, the regulatory network becomes something a signal can random-walk across, much in the way PageRank crawls the web.

The knockout itself is modelled with what the authors call a strict cut-in-and-out intervention. For a candidate gene, TAKO removes every transition into that gene, removes every transition out of it, assigns the node a pure self-loop, and re-normalises the affected rows so the matrix remains a valid Markov chain. The gene is thereby isolated from the network in a single algebraic operation. To quantify the damage, TAKO runs personalised PageRank-style propagation on both the original and the perturbed transition matrices, using the same uniform restart distribution over all non-knockout genes. The signed difference between the two stationary propagation profiles becomes the knockout impact profile, and sorting genes by that score yields a ranked list of putative knockout-responsive targets. Because each knockout requires only sparse matrix iterations, the configured scans of dozens of candidate knockouts completed within seconds on standard hardware, whereas GenKI required model training and scTenifoldKnk was substantially slower in the authors’ implementation.

To test the method, the team turned to SERGIO, a simulator that generates realistic single-cell expression data from a known directed gene regulatory network, allowing controlled evaluation against ground-truth edges. They benchmarked networks of 100, 400 and 1,200 genes, defining candidate knockouts by out-degree and treating the simulator’s outgoing regulatory edges as the positive labels for each knockout. Performance was measured with area under the receiver operating characteristic curve and average precision, against a random ranking, scTenifoldKnk, GenKI, signed Pearson correlation and absolute Pearson correlation. The results were competitive but sobering. In the 400-gene network, TAKO achieved a mean AUROC of 0.853 versus 0.816 for absolute correlation, a paired difference of 0.036 with a bootstrap confidence interval excluding zero. In the 1,200-gene network, however, absolute correlation was slightly higher in both AUROC and average precision, and in the 100-gene network, where only six eligible knockouts produced wide confidence intervals, the two methods were statistically indistinguishable. TAKO recovered substantial structural signal, but it did not show a consistent global advantage over the naive correlation baseline.

That outcome, the authors argue, is less a failure of TAKO than a revelation about the benchmark itself. The simulator’s edge labels record whether a regulator controls a target, but not whether the target should rise or fall after the knockout. Signed correlation therefore ranks strongly negatively correlated true targets near the bottom, dragging its AUROC down to barely above random in the larger networks, while unsigned absolute correlation ranks both strongly positive and strongly negative associations at the top, matching the unsigned structural task almost perfectly. A sensitivity analysis with Spearman correlation reproduced the same signed-versus-absolute pattern, showing the effect is not an artifact of Pearson’s linear scale. In effect, absolute correlation is nearly an ideal answer to the specific question the benchmark asks, which makes it an unusually tough baseline for any network method to beat.

Where TAKO showed more promise was at the head of the rankings and in biological coherence. Applied to four real single-cell datasets covering Trem2 knockout in mouse microglia, Nkx2-1 knockout in mouse alveolar type 1 epithelial cells, virtual knockout of STAT1 in human airway epithelium from COVID-19 samples, and Mecp2 deficiency in mouse neurons, the TAKO rankings fed into preranked gene set enrichment analysis produced pathway signatures that aligned with known biology in every case. The Trem2 rankings enriched for lysosomal organisation, phagosome processes and complement activation, matching established consequences of losing this lipid-sensing microglial receptor. The Nkx2-1 rankings highlighted extracellular matrix organisation, cell adhesion and epithelial differentiation; the STAT1 virtual knockout surfaced interferon signalling and antigen presentation; and Mecp2 profiles enriched synaptic vesicle cycling and cytoskeletal processes, consistent with roles in neuronal maturation and Rett syndrome.

Direct evaluation against measured knockout responses, however, yielded mixed discrimination. For Trem2, with 765 wild-type and 885 knockout microglial cells and 201 reference-positive genes among 2,999 evaluated, absolute correlation achieved higher global discrimination, an AUROC of 0.636 versus 0.571, yet TAKO placed far more differentially expressed genes near the top of its list, with precision at ten of 0.20 versus 0.00 for correlation. For Nkx2-1, both methods were close to random by AUROC, around 0.50 to 0.51, but TAKO again showed markedly better precision at the top, 0.50 at ten positions versus 0.10. With only two knockout targets available and cell-level labels that cannot provide replicate-aware inference across animals, the authors are careful to describe these top-ranked patterns as hypothesis-generating rather than proof of a general advantage.

The study is notable as much for its candour as for its method. The authors enumerate their limitations explicitly: TAKO inherits errors and directional uncertainty from the inferred network; converting to non-negative transition weights discards the sign of regulatory interactions, so the scores cannot predict the direction of expression change; the cut-in-and-out intervention is a network surrogate, not a mechanistic model of transcriptional compensation or incomplete knockouts; and the current implementation handles one knockout at a time without combinatorial perturbations. They also caution that absolute correlation, being computationally simpler than every network method, cannot be used to support a speed advantage for TAKO, and that the high positive prevalence in the smallest simulated network limits cross-scale interpretation.

Within those boundaries, the authors position TAKO as what it demonstrably is: a fast, auditable, training-free screening tool for the increasingly common situation in which only wild-type data exist. Rather than replacing experiments or claiming superiority over either generative models or plain correlation, it offers a topology-derived candidate list that can be laid alongside a correlation ranking, letting investigators identify consensus and complementary candidates before committing laboratory resources. The authors suggest that future work should evaluate this usage in multi-target Perturb-seq datasets with independent biological replication and prespecified top-ranked endpoints, and should test whether signed gene regulatory networks, which preserve activation and inhibition, deliver reproducible gains over simple co-variation. For a field drowning in perturbation data it cannot yet afford to collect, a transparent method that tells you where to look first, and is honest about the limits of what it knows, may prove to be exactly the kind of tool the pipeline needs.

Subject of Research: In silico prediction of single-gene knockout effects from wild-type single-cell gene regulatory networks

Article Title: Topology-adjusted knockout (TAKO): interpretable in silico single-gene knockout ranking on wild-type single-cell gene regulatory networks

Article References: Wang, H., & Oba, K. (2026). Topology-adjusted knockout (TAKO): interpretable in silico single-gene knockout ranking on wild-type single-cell gene regulatory networks. Frontiers in Bioinformatics, 6, Article 1924703. https://doi.org/10.3389/fbinf.2026.1924703

Image Credits: AI Generated

DOI: 10.3389/fbinf.2026.1924703

Keywords: gene regulatory networks, in silico knockout, single-cell RNA-seq, network propagation, personalised PageRank, gene knockout ranking, bioinformatics, CRISPR perturbation, SERGIO simulation, Trem2, Nkx2-1, computational biology

The core of the argument is that neurodegeneration is fundamentally different from neurodevelopment. Many neurodevelopmental disorders arise from intrinsic defects in neural stem cells or early neurons, so neuron-focused organoids can be genuinely informative. Neurodegenerative diseases, by contrast, unfold over decades of aging and emerge from a combination of intrinsic neuronal defects and dynamic, changing interactions among neurons, glia, vascular cells, immune cells, and their surrounding microenvironment. Neuroinflammation and vascular dysfunction are hallmarks of most neurodegenerative diseases, and Liu and Ng insist that immune and vascular cells should therefore be treated as fundamental components of organoids intended to model neurodegeneration, not as optional extras bolted on for realism.

A growing body of evidence supports this broader view. Although specific subtypes of neurons are the cells that ultimately die, neuronal death is unlikely to mark the beginning of the disease process. Studies cited in the Perspective show that astrocytes, oligodendrocytes, microglia, and endothelial cells all undergo disease-associated changes before overt neurodegeneration becomes visible. In mouse models of juvenile neuronal ceroid lipofuscinosis, low-level glial activation precedes late-onset neurodegeneration; in mice carrying a frontotemporal dementia-causing CHMP2B mutation, early microgliosis appears before neuronal loss and behavioral impairment. Genetic analyses of complex brain traits have also pointed to oligodendrocyte involvement in Parkinson’s disease etiology, while blood-brain barrier breakdown and endothelial dysfunction have been documented in Alzheimer’s disease and related disorders.

This reframing carries a profound implication for drug development. Neurodegeneration is increasingly understood as the consequence of both cell-autonomous and non-cell-autonomous processes, in which pathological changes in one cell type can initiate or amplify dysfunction in another. If that is true, it may help explain why therapeutic strategies focused primarily on protecting neurons have yielded limited clinical success despite decades of intensive research. Protecting neurons in isolation may be insufficient if the disease is initiated or sustained by interactions between multiple cell types. The authors argue that nonneuronal cells and pathological cell-cell interactions should be considered legitimate therapeutic targets in their own right, rather than background noise in a neuron-centric narrative.

Among the nonneuronal cells implicated in neurodegeneration, microglia receive particular attention in the Perspective. Microglia are the resident immune cells of the central nervous system, responding to changes in the extracellular environment by adopting a continuum of phenotypes that ranges from pro-inflammatory to anti-inflammatory. They are highly heterogeneous, displaying diverse molecular and functional states that emerge in response to disease, injury, or shifts in the local microenvironment. For many years, microglia were regarded primarily as responders to neuronal injury, with activated microglia considered a consequence of neurodegeneration rather than a primary contributor. That view, the authors note, has changed substantially, and the change has been driven largely by human genetics.

The genetic evidence is compelling. Mutations in TREM2 were first identified as genetic risk factors for Alzheimer’s disease in 2012. Within the brain, TREM2 is expressed primarily in microglia, where it functions as a cell surface receptor that senses cellular damage and drives phagocytosis. Studies have now convincingly shown that replacing Trem2-mutant microglia with Trem2 wild-type microglia reduces disease phenotypes in a mouse model of Alzheimer’s disease, a result that suggests microglial dysfunction actively influences the initiation and progression of neuronal disease phenotypes. The logical conclusion is stark: if microglial dysfunction can shape neurodegenerative disease progression, then an organoid that excludes microglia cannot replicate a key cellular process it is supposed to model.

These findings have direct consequences for therapeutic discovery pipelines. Most phenotypic screening in neurodegeneration currently relies on neuronal mono-cultures, with compounds evaluated against relatively simple endpoints such as neuronal survival, neurite morphology, or protein aggregation. Such assays are amenable to high-throughput screening, but they rest on the assumption that the therapeutic target resides within the neuron. What if it does not? A drug candidate that protects neurons in isolation may fail in vivo because it never corrects the pathological neural microenvironment. If diseases such as Alzheimer’s and motor neuron disease are driven at least in part by early dysfunction of nonneuronal cells like microglia, then screening paradigms built on neuronal mono-cultures may require fundamental re-evaluation, the authors warn.

The Perspective does not simply call for bigger, more complicated organoids. The next frontier, Liu and Ng argue, is not complexity for its own sake but disease relevance. If neuroinflammation is a fundamental component of pathogenesis, immune cells must be incorporated; if vascular dysfunction is central, endothelial cells should be too. The objective is not to recreate every component of the human brain but to reconstruct the cellular interactions necessary to model the disease process. To that end, the authors propose explicit benchmarks: at a minimum, organoids intended to model neurodegenerative diseases should contain relevant neuronal and astrocytic populations plus a functional neuroimmune component including microglia. Crucially, the mere presence of microglia should not be considered sufficient. These cells must exhibit appropriate maturation, spatial localization, and molecular states in response to the neural environment. Where vascular dysfunction is central to a particular disease, vascular components should similarly be included.

As organoid systems grow more sophisticated, the technologies needed to deconvolute multicellular interactions become equally important. Single-cell and spatial profiling can determine whether cellular states within organoids resemble those observed in human disease, while genetic and pharmacological perturbation can establish whether specific cell types or signaling pathways are causal rather than merely correlated with pathology. Whole-organoid imaging, automated phenotypic screening, and computational modeling can enable longitudinal interrogation of how disease-associated states emerge, interact, and propagate across the tissue. The authors envision scalable small-molecule or CRISPR perturbation screens coupled with spatial and single-cell multi-omics and whole-organoid imaging, identifying perturbations that reverse disease-associated phenotypes and uncovering therapeutic targets for prioritization and in vivo validation. Recent work on vascularized neuroimmune organoids, in which extracts from Alzheimer’s disease patient brains induced multiple pathologies, hints at what such platforms could achieve for disease modeling and drug discovery.

Ultimately, the authors conclude, the value of next-generation organoids will not be measured by how closely they resemble the human brain anatomically, but by whether they enable reconstruction of the cellular interactions that drive disease. For neurodegeneration, microglia are central to that equation. An organoid without microglia might still be a useful model for some questions, but for diseases in which neuro-immune dysfunction is integral to pathogenesis, it should no longer be considered a complete model of neurodegeneration. No disease model can fully recapitulate human disease, the authors acknowledge, and the benchmark should always be a clear definition of the biological questions a model is designed to address, together with honest acknowledgment of its limitations. But as the genetics, pathology, and therapeutic failures of neurodegeneration converge on the same message, the era of the neuron-only brain organoid may be drawing to a close, and with it, perhaps, a new chapter in the search for effective treatments.

Subject of Research: Cellular diversity requirements for human neural organoid models of neurodegenerative disease

Article Title: Cellular diversity in neural organoids is critical for neurodegenerative disease modeling

Article References: Liu, S., & Ng, S.-Y. (2026). Cellular diversity in neural organoids is critical for neurodegenerative disease modeling. PLOS Biology, 24(10), e3004074. https://doi.org/10.1371/journal.pbio.3004074

Image Credits: AI Generated

DOI: 10.1371/journal.pbio.3004074

Keywords: brain organoids, neurodegeneration, microglia, neuroinflammation, Alzheimer's disease, iPSC, TREM2, vascular dysfunction, drug screening, glial cells, PLOS Biology, disease modeling

The scale of the resource is what sets it apart. Most previous single-cell RNA sequencing studies of human liver pooled relatively small numbers of cells from limited patient cohorts, which made it hard to resolve subtle transcriptional states relevant to disease. The new atlas integrates seven published studies into a single reference containing 649,295 cells drawn from 77 donors—42 with healthy livers and 35 with chronic liver disease, balanced across 38 males and 39 females. After rigorous quality control, iterative subclustering and manual annotation, the researchers organized the data into three hierarchical levels: 16 broad cell lineages, 34 intermediate cell types and, at the finest resolution, 125 transcriptionally distinct cellular states. The atlas even captured rare residents of the liver, including neurons, mast cells, basophils, neutrophils and hematopoietic progenitor cells, which individual studies had rarely described. An interactive web browser allows any researcher to query the gene expression profiles.

The analytical centerpiece concerns macrophages, the scavenger immune cells that both fuel and resolve fibrosis. Within 81,577 nonproliferating mononuclear phagocytes, the team distinguished tissue monocytes, conventional dendritic cells, resident Kupffer cells and, crucially, two separate populations of so-called scar-associated macrophages, or SAMacs—cells marked by the proteins TREM2, CD9, CD63 and SPP1 that previous work had implicated in activating the scar-producing myofibroblasts of the liver. Differential abundance testing with the Milo algorithm showed that both SAMac populations expand in diseased livers, but with a telling difference: SAMac2 expanded consistently across every etiology of chronic liver disease, whereas SAMac1 expansion was most pronounced in the biliary disorders primary biliary cholangitis and primary sclerosing cholangitis.

Why does splitting SAMacs into two matter? Because the field has been stuck in a contradiction. Some studies painted TREM2-positive scar macrophages as fibrosis drivers; others found that TREM2 and the SAMac molecule SPP1 actually carry anti-inflammatory and antifibrotic properties, promoting scar resolution. The new data dissolve the paradox by showing that the two SAMac populations are transcriptionally almost opposite. Differential expression analysis, run with a statistical hurdle model and Bonferroni correction, flagged 903 genes enriched in SAMac1 and 1,176 in SAMac2. SAMac1 expressed GPNMB along with a suite of reparative factors—MMP9, IL10, GDF15, CD36 and PPARG—resembling the lipid-associated macrophages described in mouse and human fatty tissue. SAMac2, by contrast, lit up with pro-inflammatory and pro-fibrotic mediators: IL1B, TNF, IL18, the inflammasome component NLRP3, AREG and PDGFB, plus the scavenger receptors OLR1 and CLEC5A. Flow cytometry confirmed expansion of TREM2-positive, OLR1-positive macrophages in diseased human livers, and immunofluorescence placed OLR1-positive CD68-positive cells precisely within fibrotic tissue.

Spatial context came from a 36-plex antibody panel run on the PhenoCycler-Fusion platform across 46 patient liver samples, spanning healthy tissue and explants or biopsies from metabolic dysfunction-associated steatotic liver disease (MASLD), alcohol-related liver disease, PBC and PSC. Computational neighborhood analysis identified three cellular niches, one of them a fibrotic niche packed with fibroinflammatory cells and dense fibrillar collagen deposition that expanded dramatically in disease. Both OLR1-positive and GPNMB-positive macrophages preferentially accumulated inside that niche, yet they remained largely distinct from one another—occasional double-positive cells existed, but the two markers mostly labeled separate cells, mirroring the single-cell data and confirming genuine macrophage heterogeneity at the site of scarring.

The clinical stakes became clear when the team turned to outcome data. In liver needle biopsies from MASLD patients, SAMacs expanded and upregulated OLR1, which stayed macrophage-specific. Then, mining the SteatoSITE multimodal database of 632 histologically staged MASLD cases, the researchers found that hepatic OLR1 expression correlated positively with fibrosis stage and with other fibroinflammatory genes, and negatively with markers of homeostatic Kupffer cells and sinusoidal endothelial cells. Deconvolution of bulk RNA-sequencing data using the single-cell reference showed both SAMac populations rising with fibrosis stage, a pattern replicated in independent cohorts of MASLD, PSC, ARLD and PBC patients. Most strikingly, in the longitudinal SteatoSITE cohort, high OLR1 expression and high estimated SAMac2 abundance were both associated with increased all-cause mortality and faster progression to hepatic decompensation—the point at which a scarred liver stops coping.

Cross-species work strengthened the case. In mouse models of fibrosis induced by carbon tetrachloride, bile duct ligation and a choline-deficient high-fat diet, the team found the same two Trem2-positive SAMac subpopulations, with the Olr1-positive counterpart again enriched for inflammatory genes including Il1b and Nlrp3. Both populations expanded during active injury and returned to baseline once injury ceased, tracking fibrogenic activity in time as well as space. Flow cytometry showed that Olr1-positive SAMacs produced more interleukin-1 beta than their CD319-positive SAMac1 counterparts. Bone-marrow chimera experiments established that both subpopulations arise from circulating monocytes rather than resident Kupffer cells. Single-cell ATAC-sequencing added a chromatin dimension: Olr1-positive SAMac2 carried open chromatin enriched for pro-inflammatory transcription factor motifs—Fos, Jun, NF-kB and Rel family members—while SAMac1 showed motifs linked to interferon responses and regulatory factors, indicating the two states are epigenetically primed for different jobs.

Ligand-receptor modeling with the LIANA and Tensor-cell2cell frameworks then predicted how each SAMac population talks to the mesenchymal cells that manufacture collagen. SAMac1’s outgoing signal repertoire contained recognized antifibrotic factors such as HGF, IGF1, MMP9 and GDF15. SAMac2’s contained TNF, TGFB1, IL1B, IL18, complement C3 and PDGFB—mediators associated with promoting fibrogenesis. Functional tests followed. Conditioned media from human monocyte-derived macrophages with OLR1 knocked down by siRNA were less able to induce collagen gene expression in LX2 stellate cells, and treatment with MEDI6570, a clinically tested OLR1-blocking antibody, similarly dampened fibrillar collagen expression. In a three-dimensional multilineage spheroid model of MASH—primary hepatocytes, macrophages and stellate cells challenged with lipotoxic free fatty acids—OLR1 knockdown reduced fibrogenic activity and abolished fibrillar collagen protein production. The team also showed that a chronic inflammatory stimulus cocktail (TNF, prostaglandin E2 and Pam3CSK4) pushed macrophages toward the SAMac2 program, whereas fatty acids and glucocorticoids pushed toward the SAMac1-like state.

The therapeutic implication is unusually concrete, because OLR1—also known as LOX-1, a class E scavenger receptor—already has a drug-development history in cardiovascular medicine, where it has been implicated in atherosclerosis, cardiac fibrosis and cancer biology. MEDI6570 has been administered to patients in phase 1 and phase 2 trials, raising the prospect of repurposing it, or developing small-molecule OLR1 inhibitors, for liver disease. The authors caution that much remains to be learned: the activating ligands of OLR1 in the injured liver, whether macrophages can switch between the SAMac1 and SAMac2 states, and whether OLR1 can serve as a predictive biomarker all require further study. They also note that hepatocytes and stellate cells are better captured by single-nucleus approaches, and that candidate targets from single-cell studies must be validated in independent, histologically staged biopsy cohorts—as OLR1-positive SAMacs were here. Still, by resolving the heterogeneity that had made scar macrophages such a confusing target, the atlas converts a paradox into a plan: spare the reparative macrophages, silence the inflammatory ones, and perhaps finally slow the scarring that ends so many livers.

Subject of Research: Single-cell transcriptomic identification of OLR1-positive scar-associated macrophages as a therapeutic target in chronic liver disease fibrosis

Article Title: A human single-cell atlas identifies OLR1+ scar-associated macrophages as a potential therapeutic target for chronic liver disease

Article References: Papachristoforou, E., Kong, K., Colella, F., Tam, J., Luft, J., Parhar, R., Liang, Y., McCafferty-Brown, A., Finney, G., Rao, K., Ravichandar, J. D., Sutherland, E. F., Veizades, S., Grzelka, M., Qiu, P. K., Asif, A., Battle, I., Hammer, M., Kuschnereit, T., … Ramachandran, P. (2026). A human single-cell atlas identifies OLR1+ scar-associated macrophages as a potential therapeutic target for chronic liver disease. Nature Genetics, 58(10), 2550-2565. https://doi.org/10.1038/s41588-026-02774-w

Image Credits: AI Generated

DOI: 10.1038/s41588-026-02774-w

Keywords: single-cell RNA sequencing, liver fibrosis, macrophages, OLR1, chronic liver disease, scar-associated macrophages, TREM2, MASLD, spatial proteomics, antifibrotic therapy, Nature Genetics, liver atlas

The receptor in question is TREM2, short for triggering receptor expressed on myeloid cells 2. Over the past several years, TREM2 has attracted intense interest in tumor immunology because it is preferentially expressed by myeloid cells within tumors and appears to reinforce their suppressive, tissue-repair-like behavior. In the new work, the team led by co-first authors Jassaud Marius and Ziane-Chaouche Lydia, with Michel Salzet and Marie Duhamel as co-senior authors, began by mining breast cancer transcriptomic datasets. They found that TREM2 is enriched in breast tumors relative to healthy tissue and that its expression correlates with myeloid gene signatures and immunoregulatory networks, the molecular circuits that dampen antitumor immune responses.

That correlation set the stage for the central experiment. The researchers engineered chimeric antigen receptor macrophages, or CAR-M, macrophages fitted with a synthetic receptor that directs them to recognize and engulf cells expressing HER2, a growth factor receptor amplified in a substantial subset of breast cancers. CAR macrophages are an emerging class of engineered immune cells, distinct from the CAR T cells that have transformed treatment of blood cancers. Because macrophages naturally infiltrate solid tumors and physically devour malignant cells, they are attractive candidates for solid tumor therapy, but their tendency to adopt anti-inflammatory, tumor-promoting states inside the hostile tumor microenvironment has limited their effectiveness.

To build their CAR-M, the team used the THP-1 human monocytic cell line, a widely employed model for generating macrophage-like cells in the laboratory. A key observation emerged early: when the cells were stimulated under anti-inflammatory conditions, the very conditions that resemble the immunosuppressive milieu of a tumor, TREM2 expression rose. In other words, the receptor appeared to be part of the machinery that macrophages deploy when they shift into their pacified, repair-oriented mode. That finding made TREM2 an obvious suspect in the failure of CAR-M to maintain an aggressive, tumor-killing phenotype.

The researchers then tested what happens when TREM2 is removed from the equation, using two complementary strategies. The first was genetic: interfering with the receptor’s expression directly. The second was pharmacological: blocking the receptor’s activity with inhibitory compounds. Both approaches produced the same directional result, reducing the anti-inflammatory polarization signatures that normally accumulate in the engineered macrophages. The consistency between genetic and pharmacological inhibition strengthens the conclusion that TREM2 itself, rather than an off-target effect of any single intervention, functions as a checkpoint holding macrophages in their suppressive state.

To understand the reprogramming at molecular depth, the team turned to data-independent acquisition proteomics, a mass spectrometry technique that quantifies thousands of proteins in a single experiment. The proteomic landscape of TREM2-blockaded CAR macrophages told a coherent story. Programs associated with the anti-inflammatory cytokines IL-4, IL-13 and IL-10, the classic drivers of suppressive macrophage polarization, were attenuated. In their place, the researchers observed enrichment of inflammatory signatures, interferon-related programs, cytoskeletal components, and metabolism-related pathways. Each of these categories makes functional sense for a cell being asked to hunt and consume cancer cells: interferon signaling primes antimicrobial and antitumor activity, cytoskeletal remodeling powers the physical machinery of engulfment, and metabolic reprogramming fuels the energetic demands of phagocytosis and inflammation.

The functional consequences were measured directly. TREM2 inhibition enhanced HER2-dependent phagocytosis, the ability of the CAR macrophages to recognize, bind and ingest HER2-expressing breast cancer cells, in experiments conducted on breast cancer cell lines. Critically, the team did not stop at simplified two-dimensional cultures. They extended their findings to patient-derived tumoroid models, three-dimensional clusters of tumor cells grown from patient samples that more faithfully recapitulate the architecture and immunosuppressive chemistry of real tumors. The phagocytosis advantage held in these more demanding models, an important benchmark for any cellular therapy claim.

One of the most striking details of the study is the role of interferon-gamma. The strongest enhancement of phagocytosis was observed when the TREM2-blockaded CAR macrophages were stimulated with IFN-γ, a potent inflammatory cytokine produced by activated T cells and natural killer cells. This suggests a potential combination strategy: pairing TREM2-targeted CAR macrophages with interventions that raise interferon-gamma levels in the tumor microenvironment, such as checkpoint inhibitor antibodies or other immunostimulatory agents. The finding also hints that the reprogrammed macrophages are not merely less suppressive but genuinely more responsive to inflammatory activation, a qualitative shift in their functional identity.

The concept of a macrophage-intrinsic checkpoint is the study’s most consequential idea. Immune checkpoint blockade, exemplified by antibodies against PD-1 and CTLA-4, works by releasing T cells from inhibitory signaling. TREM2 blockade would apply the same logic to a different arm of the immune system, releasing macrophages from their own inhibitory receptor. Because TREM2 is largely restricted to myeloid cells, targeting it could in principle reprogram the suppressive tumor-associated macrophage population itself, converting an obstacle to immunotherapy into an instrument of it. The Lille team’s data support TREM2 as a candidate target for exactly this purpose in macrophage-based therapies for breast cancer.

As with any preclinical study, important caveats apply. The CAR macrophages were generated from a cell line rather than primary human cells, and the tumoroid models, while more realistic than flat cultures, still fall short of intact tumors with their full vascular, stromal and immune complexity. Whether pharmacological TREM2 blockade can be delivered safely in patients, and whether the reprogrammed macrophages retain their inflammatory phenotype in vivo over time, remain open questions that will require animal studies and eventually clinical trials. Nevertheless, the study, which was supported by the French National Cancer Institute, the Ligue Contre le Cancer and regional funding, and approved by the ethics committees of the Oscar Lambret Cancer Center with written informed patient consent, provides a mechanistically grounded and proteomically detailed case that TREM2 is a druggable lever for flipping macrophages from tumor guardians into tumor executioners. For a field searching for ways to make solid tumor cellular immunotherapy work, that lever is one worth pulling.

Subject of Research: TREM2 blockade to enhance HER2 CAR macrophage immunotherapy in breast cancer

Article Title: TREM2 blockade reprograms HER2 CAR macrophages toward inflammatory phagocytes in breast cancer models

Article References: Marius, J., Lydia, Z.-C., Antonella, R.-R., Michel, S., & Marie, D. (2026). TREM2 blockade reprograms HER2 CAR macrophages toward inflammatory phagocytes in breast cancer models. Cancer Immunology, Immunotherapy. https://doi.org/10.1007/s00262-026-04576-9

Image Credits: AI Generated

DOI: 10.1007/s00262-026-04576-9

Keywords: TREM2, CAR macrophages, breast cancer, immunotherapy, phagocytosis, tumor-associated macrophages, HER2, proteomics, interferon-gamma, tumor microenvironment, cellular therapy, myeloid cells

The challenge the researchers set out to address is well known to immunotherapy researchers. Checkpoint-blocking drugs have transformed outcomes for some patients with melanoma, lung cancer and other malignancies, but many solid tumours remain resistant because their surroundings actively suppress immune activity. Tumour-associated macrophages, or TAMs, are abundant white blood cells found inside tumours that, in their tumour-associated state, help the cancer evade destruction. Professor Zhao explained the central problem succinctly: the immune system may be fully capable of attacking a tumour, but the tumour environment can stop those immune cells from doing their job. Her team’s answer was to change that environment from within, delivering treatment precisely to the cells that maintain the immunosuppressive order and persuading them to switch sides.

The delivery vehicle at the heart of the study is the lipid nanoparticle, the same class of fatty droplet that carries mRNA in COVID-19 vaccines. But these particles were given a targeting upgrade: their surface is coated with an antibody that recognises TREM2, a protein expressed at high levels on tumour-associated macrophages. This molecular address label guides the nanoparticles to the very cells responsible for tumour immune suppression. Once inside the macrophages, the particles release two payloads with complementary functions. The first is an mRNA molecule carrying instructions for the macrophage to produce CXCL9, a chemical signalling protein. The second is Resiquimod, a small-molecule compound that pushes macrophages away from their immune-suppressing behaviour and toward a more inflammatory, immune-supportive state.

The two payloads work in concert in a way that illustrates the elegance of the design. CXCL9 acts as a chemical beacon, drawing cytotoxic CD8+ T cells, the immune system’s primary tumour-killing soldiers, into the tumour mass, where they are often excluded or rendered inactive. Meanwhile, Resiquimod shifts the local macrophage population away from suppression, easing the hostile conditions that would normally exhaust or repel those T cells. Rather than depleting the macrophages outright, an approach that can carry inflammatory side effects, the researchers effectively re-educated them, converting a tumour-protective population into an accomplice of the anti-cancer immune response.

In experiments with mouse models, the approach produced measurable shifts in the tumour immune landscape. Treatment reduced the proportion of immune-suppressing macrophages by more than 60 percent, a substantial remodelling of the tumour’s defensive cellular makeup. Levels of CXCL9 within the tumours rose fourfold, confirming that the delivered mRNA was being translated into functional chemical beacon by the targeted cells. The researchers also documented greater numbers and heightened activity of cancer-fighting T cells inside the treated tumours, along with a moderate reduction in tumour growth. While the growth slowdown alone was not dramatic, the immunological changes suggest a tumour environment becoming markedly more permeable and hospitable to immune attack.

The team then tested whether their nanoparticle therapy could amplify the effects of existing immunotherapies. When combined with immune checkpoint-blocking antibodies targeting PD-L1 and CTLA-4, two of the most widely used targets in clinical oncology, the treatment produced further increases in cytotoxic T cells and, notably, the generation of central memory T cells. These long-lived memory cells could give the immune system the ability to recognise and respond to cancer if it returns, a property highly valued in cancer therapy because it hints at durable protection rather than transient tumour shrinkage. Interestingly, however, the combination did not yield additional tumour-growth inhibition in this particular mouse model, a nuance the researchers report candidly and one that will guide future experimental design.

Professor Zhao framed the findings as an important proof of concept that mRNA and nanoparticle technology can reprogramme the immune environment of a tumour. She emphasised that significant work remains before the approach could be considered for patients, but described the results as an encouraging foundation for developing more targeted cancer immunotherapies. That caution reflects the well-known gap between promising mouse studies and clinical reality: nanoparticle manufacturing, dosing, safety profiling and the variability of human tumour environments all present substantial hurdles. Still, the strategy addresses a specific failure mode of current immunotherapy, immune exclusion, and does so with a precision that conventional drugs have struggled to achieve.

The broader significance of the work lies in the expanding repertoire of mRNA medicine. The pandemic demonstrated that lipid nanoparticles can deliver genetic instructions safely and at scale; researchers worldwide are now exploring whether the same platform can carry therapeutic instructions for enzymes, antibodies, cytokines and tumour antigens. This study adds a subtle new use case: not delivering a drug or an antigen, but delivering the blueprint for a signalling molecule that changes the behaviour of the cells receiving it. By targeting TREM2-positive macrophages, the Adelaide team also taps into a growing body of research on macrophage reprogramming, an area increasingly seen as fertile ground for solid tumour therapy where T-cell-focused approaches alone often fall short.

The research was led by Adelaide University scientists in collaboration with SA Pathology and the Royal Adelaide Hospital, and is published under the title ‘Targeting tumor-associated macrophages using mRNA lipid nanoparticles for cytotoxic T lymphocyte–mediated cancer immunotherapy’. For patients whose tumours shut out immunotherapy, the study offers a vision of treatment that does not fight the tumour’s fortress head-on but instead quietly converts its guards, arming them with the instructions to raise a signal flare that guides the immune system’s most lethal cells inside. If subsequent studies can translate those beacon-lit results into durable clinical benefit, mRNA nanoparticles may find a second act in oncology as architects of the tumour microenvironment rather than mere couriers of vaccines.

The biology underlying the study helps explain why macrophages have become such a sought-after target. Macrophages are not inherently tumour-friendly; in healthy tissue they patrol, clear debris and coordinate inflammatory responses. Tumours, however, gradually reshape the macrophages they recruit, coaxing them into a state that suppresses cytotoxic T cells, promotes new blood vessel growth and remodels the fibrous matrix surrounding the cancer. High densities of these suppressive macrophages have been associated in many cancer types with poorer responses to checkpoint inhibitors, which is why strategies that convert rather than eliminate them have attracted growing interest.

The choice of TREM2 as a targeting marker reflects recent advances in understanding macrophage identity within tumours. TREM2 is a receptor expressed on a subset of macrophages that accumulates in tumours and is linked to immunosuppressive function, and blocking or depleting TREM2-positive cells has emerged as an active area of preclinical investigation. Using an antibody against TREM2 as a homing device, rather than as a therapeutic agent itself, is a distinctive feature of the Adelaide approach: the antibody serves as an address label that concentrates the therapeutic payload where it is most needed, potentially limiting off-target effects in healthy tissue.

The signalling components of the formulation also draw on established immunology. Resiquimod is a synthetic agonist of toll-like receptors, pathways that act as alarm bells for the innate immune system and have previously been explored as topical treatments and vaccine adjuvants. Encapsulating it alongside mRNA allows the two stimuli to act within the same cell, pairing a behavioural switch with a genetic instruction. CXCL9, meanwhile, belongs to a family of chemokines long known for their role in recruiting CD8+ T cells to sites of inflammation, and low CXCL9 expression has been linked to immune-excluded tumours in human studies, making it a rational choice for restoring T-cell infiltration.

The appearance of central memory T cells in the combination experiments deserves particular attention. Memory T cells persist long after an initial immune response subsides and can mount faster, stronger reactions upon re-encounter with their target. In cancer, inducing such cells is a goal of therapeutic approaches such as cancer vaccines and intratumoural therapies, because tumour recurrence remains a leading cause of treatment failure even when initial therapy is successful. The fact that this memory phenotype emerged despite the absence of added tumour shrinkage suggests that immunological benefit and immediate tumour control may follow different timelines, a distinction that longer animal studies would need to resolve.

The candid reporting of the combination result also illustrates a broader principle in immunotherapy research: immune activation and tumour regression are not always tightly coupled in early experiments. Factors such as the mouse model used, the timing of treatment relative to tumour establishment, and the dose and schedule of each component can all influence whether enhanced T-cell activity translates into measurable growth control. The authors’ decision to publish these nuances alongside the positive findings provides a transparent baseline for other laboratories seeking to refine the formulation, test it across additional tumour types, and determine which patient populations might ultimately benefit most from a macrophage-reprogramming strategy.

Subject of Research: mRNA lipid nanoparticle reprogramming of tumour-associated macrophages for cancer immunotherapy.

Article Title: New ‘smart’ nanoparticles help the immune system better attack tumours

Article References: New ‘smart’ nanoparticles help the immune system better attack tumours. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: lipid nanoparticles, mRNA, tumour-associated macrophages, cancer immunotherapy, CXCL9, TREM2, CD8+ T cells, immune checkpoint inhibitors, tumour microenvironment, Resiquimod, Science Advances, Adelaide University