The platform is built on precision-cut tumor slices, or PCTS, a technique borrowed in spirit from lung and liver research but adapted here for urothelial cancer. Fresh bladder cancer specimens, obtained from patients undergoing surgery at Skåne University Hospital in Malmö, were cut into thin, uniform slices and placed in optimized culture conditions designed to maintain both tissue architecture and cellular viability. Unlike single-cell suspensions or organoids, which strip away much of the native structure of a tumor, these slices retain the complete three-dimensional organization of the tissue, including the spatial relationships between cancer cells, stromal cells, blood vessels, and the diverse immune populations that infiltrate the tumor microenvironment.

Preserving that architecture matters because immunotherapy is fundamentally a spatial therapy. Immune checkpoint inhibitors such as nivolumab, which blocks the programmed cell death protein 1, or PD-1, do not kill cancer cells directly. Instead, they release the brakes on T cells that must physically locate and engage their targets within the tumor. A model that scatters the cells into a flat culture dish or grows tumor cells in isolation cannot reproduce that choreography. The Lund team, led by corresponding author Kristina Lundberg of the Department of Immunotechnology, set out to determine whether slices of bladder tumors could sustain the relevant cellular players long enough, and in a functional enough state, to serve as a meaningful testing ground.

The answer, according to their comprehensive characterization, is yes. Using flow cytometry, the researchers performed detailed phenotyping of both the tumor cells and the immune populations within the slices, confirming that a broad repertoire of immune cells survived in culture. Critically, the slices maintained the viability of both cancer cells and diverse immune populations for up to six days, a window long enough to run meaningful drug experiments. The team also demonstrated that the platform could address intra-tumor heterogeneity, the well-known phenomenon in which different regions of the same tumor harbor different cell types and molecular characteristics, while maintaining experimental reproducibility across patient samples, a balance that is notoriously difficult to strike in ex vivo systems.

Viability alone, however, would not be enough. The immune cells inside a preserved tumor slice might simply sit dormant, present but inert, offering no useful signal when a drug is applied. To rule this out, the researchers validated immune functionality directly. They stimulated the T cells within the slices using antibodies against CD3 and CD28, two proteins on the T cell surface whose engagement mimics the activation signals a T cell would normally receive. In response, the cells raised their expression of well-established activation markers, including CD69, CD137, and HLA-DR, and secreted significantly increased amounts of inflammatory mediators such as interferon gamma, or IFNγ, and the chemokine CXCL10, compared with untreated control samples. Those molecules are hallmarks of a genuine, functional immune response, not merely the passive presence of immune cells.

With the platform validated, the team moved to the question that matters most clinically: could the slices predict how real patients respond to real drugs? Nine patient samples were treated with nivolumab, a PD-1 checkpoint inhibitor that is a mainstay of treatment for bladder cancer yet benefits only a minority of patients. In the PCTS model, the responses were heterogeneous, mirroring the clinical reality. Two of the nine samples, roughly 22 percent, demonstrated dose-dependent immune activation after nivolumab treatment, a figure consistent with the response rates published for this drug in bladder cancer patients. For a preclinical model, reproducing that clinical distribution is a striking result, suggesting the slices may capture the biological determinants of response and resistance that simpler models miss.

The platform was also designed to evaluate more than immunotherapy alone. Using multiplex immunofluorescence microscopy on fixed slices, the researchers showed they could visualize the spatial distribution of cells and detect the direct cytotoxic effects of chemotherapy agents on cancer cells. This dual capability, capturing both immune-mediated responses and direct tumor cell killing within the same preparation, means the system can support the evaluation of combination therapies, an increasingly important area as oncology moves toward pairing checkpoint inhibitors with chemotherapy, targeted agents, and other immunomodulators. The technique also preserves the spatial information that pathologists value, allowing researchers to see not just whether a drug works but where in the tumor microenvironment its effects are concentrated.

Bladder cancer presents a particularly compelling case for such a model. The disease encompasses a spectrum from non-muscle-invasive tumors, which are typically managed with intravesical therapies, to muscle-invasive bladder cancer, which often requires aggressive systemic treatment. Molecular subtypes, including the basal/squamous, genomically unstable, and urothelial-like classifications referenced in the study, respond differently to therapy, and the immune landscape varies accordingly. A platform that preserves the native tumor microenvironment of each individual patient’s tumor offers a way to study these differences directly, rather than inferring them from cell lines that have spent decades adapting to plastic or from mouse models engrafted with human tumors.

The broader implications extend to how immunotherapy drugs are developed in the first place. The authors position PCTS among what the field calls new approach methodologies, a class of techniques intended to reduce reliance on animal models while improving translational relevance. Animal models, particularly humanized mice, remain expensive, slow, and imperfect predictors of human immune responses. An ex vivo system that uses a patient’s own tumor, keeps its immune system intact, and yields results within days could accelerate preclinical screening, help prioritize which drug combinations advance to clinical trials, and reduce the number of therapies that fail late in development after showing promise in less faithful models.

The Lund team is careful about what comes next. The current study establishes and validates the technical foundation, but the authors note that future prospective studies will be needed to correlate ex vivo responses in the slices with actual patient clinical outcomes. Only then could the platform move from a research tool toward genuine personalized treatment selection, in which a patient’s tumor is sliced, tested against a panel of therapies, and used to guide clinical decisions. The ethical framework is already in place: the study was approved by the ethics committee at Lund University, conducted in accordance with the Declaration of Helsinki, and all patients provided informed consent. If those prospective correlations hold, the humble tumor slice, a technology conceptually simple enough to be described in a sentence, could become a bridge between the immune complexity of real cancers and the controlled world of the laboratory, bringing precision immunotherapy a measurable step closer to the clinic.

Subject of Research: A precision-cut tumor slice platform preserving the functional immune microenvironment of bladder cancer for ex vivo immunotherapy and drug evaluation

Article Title: Bladder cancer precision-cut tumor slices (PCTS) preserve functional immune microenvironment enabling ex vivo drug evaluation

Article References: Richtmann, S., Radak, M., Sincic, V., Hugoson, J., Werner, N., Skoryk, V., Wagner, D., Liedberg, F., & Lundberg, K. (2026). Bladder cancer precision-cut tumor slices (PCTS) preserve functional immune microenvironment enabling ex vivo drug evaluation. BMC Medicine, 24(1), Article 468. https://doi.org/10.1186/s12916-026-05173-4

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05173-4

Keywords: bladder cancer, precision-cut tumor slices, tumor microenvironment, immunotherapy, nivolumab, PD-1, ex vivo models, drug development, flow cytometry, multiplex immunofluorescence, T cell activation, personalized medicine

The research team, led by Jackson Emanuel and Marcel A. Müller with senior collaborators including Christian Drosten and Jan Felix Drexler, set out to overcome a persistent limitation of wildlife virology. Most virus discovery efforts depend on detecting viral nucleotide sequences in tissues or swabs, an approach that is highly specific but easily missed. Viral RNA concentrations vary enormously over the course of an infection, viruses concentrate in particular organs rather than in routinely sampled secretions, field samples degrade under tropical conditions, and many infections are transient. Antibodies, by contrast, persist in blood long after a virus has been cleared, providing a retrospective record of exposure. Serology therefore captures a wider window of infection than molecular detection alone.

To exploit this advantage, the investigators deployed a multiplex immunofluorescence assay capable of screening a single specimen for reactive antibodies against 14 different human-pathogenic RNA viruses spanning seven families: Coronaviridae, Flaviviridae, Hantaviridae, Paramyxoviridae, Phenuiviridae, Pneumoviridae, and Togaviridae. The assay uses virus-infected cells fixed on slides, allowing sera to be tested in parallel against multiple antigens while preserving the ability to distinguish genuine virus-specific reactivity from nonspecific background staining. This multiplexed design makes efficient use of precious wildlife samples, which are difficult and expensive to collect across remote field sites.

The scale of the survey is one of its defining strengths. The team retrospectively analyzed 1,135 bat specimens representing 16 species and 454 rodent specimens representing 15 species. The samples were collected at 63 distinct sites across seven countries, an unusually broad geographic and taxonomic footprint for a single serological study. This breadth matters because patterns of virus circulation in wildlife are shaped by local ecology, and conclusions drawn from one region or one host species may not generalize elsewhere.

The headline finding concerns tropical bats, which exhibited strikingly high rates of seropositivity against certain virus families. Overall, 42.7 percent of the tropical bats in the dataset carried antibodies reactive with members of the family Flaviviridae, and 67.7 percent showed reactivity against Paramyxoviridae, while 7.5 percent were seropositive for Togaviridae. These figures substantially exceed the prevalence typically observed in molecular screening studies, reinforcing the argument that antibody-based surveillance reveals enzootic virus circulation that sequence-based methods systematically underestimate. In practical terms, the results indicate that flaviviruses and paramyxoviruses are not occasional visitors to African fruit bat populations but established, continuously circulating residents.

Both virus families carry substantial public health relevance. The Flaviviridae include dengue, Zika, yellow fever, and West Nile viruses, while the Paramyxoviridae encompass measles, Nipah, and Hendra viruses as well as mumps and several rodent- and bat-borne relatives. Previous work has detected paramyxovirus sequences in African Pteropodidae, the family of Old World fruit bats that includes the straw-coloured fruit bat Eidolon helvum and the Egyptian rousette Rousettus aegyptiacus. The new serological data suggest that exposure to these viruses is far more common than sequence detection alone would imply, consistent with infections that are transient in individual animals but persistent at the population level.

Beyond mapping seroprevalence, the study probed the ecological determinants of infection. The authors are careful to note that confounding factors and sparse data still complicate the identification of reliable correlates of RNA virus infection in wildlife, and the models they present are explicitly provisional. Nevertheless, multivariable logistic regression analyses of the dataset pointed to one host characteristic in particular: colony size. Species that form large colonies, exemplified by Eidolon helvum and Rousettus aegyptiacus, were more likely to show seropositivity against orthoflaviviruses and orthorubulaviruses than species with smaller or more solitary roosting habits.

This association fits a well-established epidemiological principle. In large, dense, communally roosting colonies, viruses transmitted directly between individuals or through contaminated roost environments encounter a steady supply of susceptible hosts, particularly during annual pulses of newborn pups that lack maternal antibodies. High host density and repeated mixing of individuals from across vast foraging ranges create conditions favorable to sustained viral transmission, allowing bat-adapted viruses to circulate endemically rather than burning out after localized outbreaks. Eidolon helvum, which forms colonies numbering in the millions across equatorial Africa, and Rousettus aegyptiacus, which roosts in caves in dense aggregations, are both known to undertake long-distance seasonal movements that could further disperse viruses across the continent.

The findings carry implications for how viral surveillance in wildlife should be conducted and interpreted. Because molecular and serological methods capture different phases of the infection process, the authors argue implicitly that the two approaches are complementary: sequence detection pinpoints which virus is present and enables phylogenetic reconstruction, while serology quantifies how widely hosts are exposed over time. Integrating both into long-term monitoring programs, particularly at sites where large fruit bat colonies intersect with human activities such as bushmeat hunting, agriculture, and cave tourism, could sharpen risk assessments for zoonotic emergence before outbreaks occur, rather than after.

The study also underscores the remaining uncertainties in the field. Antibody reactivity against a virus family does not always identify the exact virus species involved, since closely related viruses can elicit cross-reactive responses, and sample quality, host age, sex, and seasonal effects can all influence serological results. The authors themselves emphasize that the regression findings should be treated as provisional within this dataset rather than as definitive laws of bat virology. Even so, by combining an unusually large and geographically diverse sample set with a multiplexed assay and formal statistical modeling, the work provides a template for future reservoir studies and strengthens the evidence that African Pteropodidae, especially species forming vast communal colonies, constitute a major and ongoing reservoir of flavi- and paramyxoviruses with potential relevance to human health.

Subject of Research: Serological surveillance of RNA virus antibodies in African bats and rodents

Article Title: A multiregional serosurvey of RNA virus-reactive antibodies in bats and rodents reveals the enzootic presence of flavi- and paramyxoviruses in African Pteropodidae

Article References: A multiregional serosurvey of RNA virus-reactive antibodies in bats and rodents reveals the enzootic presence of flavi- and paramyxoviruses in African Pteropodidae. (n.d.). https://doi.org/10.1371/journal.pntd.0013689

Image Credits: AI Generated

DOI: 10.1371/journal.pntd.0013689

Keywords: bats, rodents, serosurvey, flaviviridae, paramyxoviridae, Pteropodidae, Eidolon helvum, Rousettus aegyptiacus, zoonosis, multiplex immunofluorescence, Africa, virus reservoirs

The gene in question is GNG4, which encodes the gamma-4 subunit of heterotrimeric G proteins, the signaling molecules best known for relaying messages from G protein-coupled receptors at the cell surface. Rather than studying GNG4 in isolation, a research team led by Xiangchou Yang, Yi Xu, and Chenwei Pan, working across institutions including Wenzhou Medical University and Shanghai Jiao Tong University School of Medicine, took an unusually comprehensive approach. They combined single-cell RNA sequencing, which profiles gene expression in thousands of individual cells, with spatial transcriptomics, which maps where those genes are active within intact tissue, and multiplex immunofluorescence, which visualizes multiple proteins simultaneously under the microscope. This triangulation allowed them to move beyond correlation and ask where, precisely, in the three-dimensional architecture of a liver tumor the immune-exclusion program operates.

The first finding to emerge was epidemiological in character. GNG4 turned out to be enriched in malignant hepatocytes, the cancerous liver cells themselves, and high expression of the gene was strongly associated with poor patient prognosis and elevated scores on a single-cell-based risk stratification model the team had constructed. In other words, patients whose tumors lit up for GNG4 fared measurably worse than those whose tumors did not. More striking still, when the researchers examined tumors from patients who had received neoadjuvant immunotherapy, treatment given before surgery to prime the immune system, they found that GNG4 expression was spatially concentrated in the regions of the tumor that had failed to respond. The gene was not merely a passive marker of aggressive disease; its geography within the tumor mapped directly onto therapeutic failure.

To test whether GNG4 was causally involved rather than a bystander, the researchers used short hairpin RNA, or shRNA, a technique that silences a specific gene by triggering its degradation, to knock down GNG4 in hepatocellular carcinoma cells. The effects were immediate and broad. Deprived of GNG4, the cancer cells proliferated more slowly, migrated less efficiently, and lost much of their invasive capacity in laboratory cultures. When the team implanted these GNG4-deficient cells into animal models, tumor growth was significantly suppressed. What makes this result particularly interesting is that the suppression was not purely a cell-autonomous phenomenon. The tumors lacking GNG4 were not just weaker; they were also more visible to the immune system.

That second, immunological mechanism is where the study becomes genuinely novel. In the animal models, GNG4-deficient tumors showed markedly increased infiltration by CD8-positive T cells, the cytotoxic lymphocytes responsible for directly killing infected and malignant cells. Not only were more of these cells present, but they were functionally hotter: the infiltrating T cells expressed elevated levels of interferon-gamma, the signature cytokine of activated cellular immunity, and granzyme B, the protease enzyme that T cells use to punch lethal holes in target cells. The authors describe this as a dual mechanism, in which removing GNG4 simultaneously restrains the tumor cells’ own growth and hands the surrounding microenvironment back to the immune system. Both arms of the effect converged on tumor restriction.

The mechanistic thread connecting the tumor cell to the T cell turned out to run through an unexpected molecule called MID1, short for Midline-1, a protein whose expression state in neighboring CD8-positive T cells changed in lockstep with GNG4 expression in adjacent malignant cells. Computational models of cellular communication, which infer signaling interactions between cell types from their co-localization and receptor-ligand expression patterns, together with spatial transcriptomics, revealed that regions of GNG4-high tumor tissue were consistently accompanied by an altered MID1 expression state in the T cells hovering at the tumor’s edge. The implication is that malignant cells broadcasting GNG4-associated signals somehow reach across the microenvironment and reprogram the identity of the immune cells nearest to them.

To establish that this relationship was functional rather than merely correlational, the researchers ran an in vitro rescue experiment using primary murine CD8-positive T cells, freshly isolated rather than immortalized. When T cells were exposed to conditioned medium, the nutrient broth in which malignant cells had been growing, from GNG4-silenced cancer cells, the T cells recovered function. Critically, that functional restoration was tied to the recovery of Mid1 expression within the T cells themselves. This is the kind of experimental loop that elevates a spatial observation into a mechanistic claim: perturb the tumor cell, watch the T cell change, and then show that the change in the T cell depends on a specific molecular marker reverting to its healthy state.

The final layer of validation came from human tissue. Using multiplex immunofluorescence on human hepatocellular carcinoma samples, the team stained for GNG4 in malignant cells and for the combination of CD8 and MID1 in infiltrating T cells, then quantified their spatial relationships across the tissue sections. The result confirmed the animal and single-cell findings: GNG4 expression in malignant cells was inversely correlated with the local density of CD8-positive MID1-positive T cells. Where the gene was loud, the killer T cells were scarce; where the gene was quiet, they clustered. In human disease, as in the models, the GNG4 axis appears to define an immune-excluded architecture, a tumor state in which immune cells are present at the periphery but barred from penetrating the malignant core.

The translational implications are considerable. Immune exclusion is one of the recognized reasons checkpoint inhibitors fail, because a drug that releases the brakes on T cells cannot help cells that never arrive at the tumor in the first place. If GNG4-high tumors represent a distinct, identifiable subset of hepatocellular carcinoma characterized by this exclusion phenotype, then GNG4 itself, or the signaling pathways it feeds into, becomes a candidate target for combination therapies designed to tear down the exclusion barrier before or alongside checkpoint blockade. The gene’s association with prognosis also suggests it could serve as a biomarker, helping clinicians identify up front which patients are unlikely to benefit from immunotherapy alone and might need a different strategy.

As with any study that moves from computational atlases to mouse models to human tissue, important questions remain before these findings reach the clinic. The precise signaling route by which GNG4 in a malignant hepatocyte influences MID1 in a neighboring T cell has not yet been fully mapped, and whether pharmacological inhibition of GNG4 in established tumors can reverse immune exclusion in patients is a question only clinical trials can answer. What the study delivers now is a clearly articulated hypothesis with unusually strong spatial evidence behind it: liver cancer is not just a disease of malignant cells, but of the conversations those cells conduct with their surroundings, and GNG4 appears to be one of the words in that vocabulary that tells the immune system to stay outside.

Subject of Research: The role of GNG4 in tumor immune exclusion and CD8+ T cell infiltration in hepatocellular carcinoma

Article Title: GNG4 associated with tumor immune exclusion and compromised CD8+ T cell infiltration via microenvironmental crosstalk in hepatocellular carcinoma

Article References: Yang, X., Zhou, Y., Jin, X., Ying, Z., Chen, Q., Chen, J., Ahmed, A.-A. W. H., Shan, Y., Xu, Y., & Pan, C. (2026). GNG4 associated with tumor immune exclusion and compromised CD8+ T cell infiltration via microenvironmental crosstalk in hepatocellular carcinoma. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08967-5

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08967-5

Keywords: hepatocellular carcinoma, GNG4, CD8+ T cells, tumor microenvironment, immune exclusion, MID1, single-cell RNA sequencing, spatial transcriptomics, multiplex immunofluorescence, immunotherapy resistance, liver cancer, tumor immunology