The work was led by the National Cancer Centre Singapore in collaboration with the A*STAR Institute of Molecular and Cell Biology, Singapore General Hospital and the Yong Loo Lin School of Medicine at the National University of Singapore. The peer-reviewed paper, “Identification of novel paracrine therapeutic targets in peritoneal metastases through integrative proteotranscriptomic analysis of ascites,” was published in Molecular Cancer on 11 September 2026.

Peritoneal metastases occur when cancers originating in abdominal organs spread to the peritoneum, the lining of the abdominal cavity. The source identifies gastric, colorectal and ovarian cancers as common primary sources, giving reported ranges of 20–25%, 20–45% and 60–75%, respectively. The condition is often diagnosed at Stage IV, has a median survival of under nine months and is frequently managed with palliative intent. Surgery, regionalised chemotherapy and systemic chemotherapy have limited curative ability.

Many patients with peritoneal metastases develop ascites, an accumulation of fluid in the abdomen. The 2026 study investigated the molecular pathways in this fluid that may support tumour development. It built on an earlier 2022 National Cancer Centre Singapore study, which had established that targeting ascites could inhibit tumour growth; the new analysis sought to characterise biological pathways and therapeutic vulnerabilities more comprehensively.

Using multiomic profiling, the team analysed proteins and biological pathways in a 397-sample ascites dataset spanning 21 primary cancer types. The collection is described by the researchers as one of the world’s largest ascites datasets. The analysis identified several paracrine factors implicated in tumour development, with PAI-1 and S100A10 emerging as key drivers in the ascites microenvironment.

The team selected PAI-1 as a therapeutic target because of its clinical and prognostic relevance in cancer progression. Instead of developing a conventional monoclonal antibody, the researchers created E8, a smaller antibody-derived drug known as a nanobody. The source notes that nanobodies have low immunogenicity and a reduced toxicity risk, and that these properties could support better tolerability than some existing treatments, although this was not established in patients in the reported experiments.

Investigators assessed E8 in three preclinical systems: in vitro cell assays, in vivo mouse models and ex vivo human tissue explants in which tumour tissue was grown in patients’ ascites. Across all three systems, E8 suppressed tumour growth with minimal toxicity. These findings support further investigation of E8 as a treatment directed at the ascites microenvironment, but the experiments do not demonstrate clinical efficacy, safety or patient benefit.

Associate Professor Johnny Ong, the study’s senior author and head of the Department of Sarcoma, Peritoneal and Rare Tumours at Singapore General Hospital and NCCS, said the work creates the possibility of a new class of treatments for peritoneal metastases that targets the tumour microenvironment through ascites rather than tumour cells directly. He said the pathway may have implications for cancers that result in ascites, including ovarian, gastrointestinal, liver, breast and lung cancers.

Dr Joe Yeong, co-author and principal investigator at A*STAR’s Institute of Molecular and Cell Biology, described E8 as a therapeutic candidate with potential to improve outcomes for patients who have limited treatment options. Professor Jimmy So of NUS Medicine said the research could help pave the way for effective therapy for peritoneal metastasis. These statements describe translational potential; the study itself was experimental and preclinical.

The researchers are working to advance E8 toward clinical-grade development and a first-in-human clinical trial. In parallel, they are developing a companion biomarker test to identify patients with elevated PAI-1 levels who may be most likely to benefit from E8. These efforts are part of the Melyorax Initiative, a newly established biologics research programme led by Ong, with E8 as its lead asset. Whether the suppression observed in laboratory and animal models will translate into clinical benefit remains to be established.

Subject of Research: Preclinical identification of ascites targets and testing of the E8 nanobody in peritoneal metastases

Article Title: SG research team develops novel nanobody targeting tumor-fueling fluid to treat late-stage cancer of the abdominal lining

Article References: SG research team develops novel nanobody targeting tumor-fueling fluid to treat late-stage cancer of the abdominal lining. Original publication (Bibliographic details unavailable.)

Image Credits: AI Generated

DOI: Not provided

Keywords: peritoneal metastases, ascites, PAI-1, S100A10, E8 nanobody, tumour microenvironment, preclinical research, cancer therapeutics, Singapore, research, team, develops

Peritoneal metastasis occurs when cancer cells detach from a primary tumor, most commonly of gastrointestinal, gynecological or appendiceal origin, and implant on the peritoneum, the membrane lining the abdominal wall and covering the organs within. As the disease progresses, many patients develop ascites, an abnormal buildup of protein-rich fluid that causes distension, discomfort and breathing difficulties. Clinically, the appearance of malignant ascites portends poor survival, yet the molecular contents of that fluid and their biological significance have remained surprisingly underexplored. Because ascites sits in direct contact with peritoneal tumor deposits, any signaling molecules it contains are, by definition, delivered to the tumor microenvironment, making it an ideal starting point for hunting paracrine factors, the secreted ligands that cells use to influence their neighbors.

To interrogate that signaling milieu systematically, the team assembled an extraordinary cohort: 397 samples of malignant cell-free ascites representing 21 distinct histological origins of peritoneal metastasis. Cell-free ascites, the liquid fraction left after cells and debris are removed, concentrates the soluble protein complement of the peritoneal cavity. The researchers applied quantitative mass spectrometry, using tandem mass tag labeling to measure protein abundance across large batches of samples with high reproducibility, complemented by rigorous quality control based on metrics such as coefficients of variation. This proteomic backbone was then integrated with transcriptomic data, allowing the team to confirm that the proteins they detected were not passive leakage from dying cells but were actively produced and secreted by defined cell populations within the tumor ecosystem.

The mass spectrometry analysis revealed multiple coherent biological themes woven through the ascites proteome. Chief among them was the coagulation cascade and its counterbalancing system, fibrinolysis, the enzymatic machinery that dissolves blood clots. The team identified both previously recognized paracrine factors and a set of novel candidates that appear amenable to therapeutic perturbation, including proteins such as S100A10, macrophage migration inhibitory factor, peroxiredoxin 1, phosphoglycerate kinase 1 and lactate dehydrogenase A. Each of these has known roles in tumor biology, from plasminogen activation and extracellular matrix remodeling to inflammatory signaling and metabolic adaptation, and their enrichment in ascites points to a peritoneal microenvironment actively shaped by tumor-driven secretion.

To determine which cells were responsible for producing the key ligands, the researchers turned to single-cell transcriptomic sequencing of cells recovered from peritoneal metastasis ascites. This analysis delivered one of the study’s most conceptually important findings: two functionally antagonistic regulators of fibrinolysis, S100A10 and plasminogen activator inhibitor-1, or PAI-1, are secreted by distinct cell populations within the tumor microenvironment. S100A10, which promotes plasmin generation and matrix degradation, and PAI-1, which inhibits it, sit on opposite sides of the fibrinolytic balance. Their co-elevation, attributed to separate cellular sources, indicates that upregulation of these ligands reflects active, coordinated secretion rather than passive homeostatic drift. In other words, the tumors and their stromal partners are not merely accumulating proteins; they are engineering the extracellular fluid to favor their own agenda.

Why would a cancer care about fibrinolysis? The answer lies in the mechanics of peritoneal spread. Plasmin activity remodels the extracellular matrix, clears fibrin barriers and facilitates cell motility, all processes that help disseminated tumor cells implant, invade and expand on peritoneal surfaces. By mapping the components of this system, including tissue-type and urokinase plasminogen activators, annexin 2, and their inhibitors, the study frames the ascites compartment as a regulated enzymatic workspace where coagulation and dissolution are choreographed to tumor advantage. That framing immediately suggests intervention points: if specific ligands sustain the peritoneal ecosystem, neutralizing them could degrade the environment on which the metastases depend.

The team went beyond cataloguing. They selected PAI-1, a well-studied but therapeutically elusive target, and engineered a nanobody, a single-domain antibody fragment derived from camelid heavy-chain antibodies, designed to bind and neutralize it. Nanobodies are roughly a tenth the size of conventional antibodies, offering deep tissue penetration, high stability and straightforward manufacturing, properties well suited to an intraperitoneal disease. Using surface plasmon resonance the researchers confirmed the nanobody bound PAI-1 with high affinity, and they then challenged it across a battery of preclinical models: in vitro assays with cancer and stromal cells, in vivo animal models of peritoneal metastasis, and ex vivo human explant systems that preserve fragments of actual patient tumor in their native architecture.

The results provided proof of concept that intercepting PAI-1 signaling can constrain peritoneal tumor growth with minimal toxicity. In the experimental systems, the anti-PAI-1 nanobody suppressed the fibrinolytic and pro-metastatic programs the multiomic analysis had flagged, while sparing normal physiology, an encouraging safety signal for a target embedded in the hemostatic system. Because PAI-1 is secreted into ascites and acts on neighboring cells, antibody-based neutralization is a rational modality: the drug circulates in the same fluid compartment where the ligand operates. The strategy also generalizes, since the same proteotranscriptomic pipeline that surfaced PAI-1 can be reapplied to the other paracrine factors the study identified, each now carrying a mechanistic rationale for drug development.

The broader significance of the work lies in its scale and integration. By pairing deep proteomics across nearly four hundred ascites samples and twenty-one tumor types with single-cell resolution attribution of ligand sources and functional validation in vitro, in vivo and ex vivo, the study demonstrates a template for target discovery in fluid-biopsied tumor microenvironments. It reframes malignant ascites from a prognostic liability into an analytical asset, a lens through which the signaling economy of peritoneal metastasis becomes visible. For a disease whose treatment options remain limited largely to cytoreductive surgery with hyperthermic intraperitoneal chemotherapy and systemic regimens of modest efficacy, the identification of actionable paracrine targets opens a genuinely new therapeutic front.

Considerable work remains before patients benefit. The nanobody must advance through formal preclinical development and clinical trials, biomarkers must be established to select patients whose tumors depend on PAI-1, and the logistics of intraperitoneal delivery will need clinical refinement. Yet the conceptual shift is already tangible. The study shows that the fluid bathing peritoneal tumors can be read, almost like a molecular transcript of the tumor’s intentions, and that reading it carefully reveals not just how the disease progresses but precisely where it is vulnerable. In turning the fluid of a poor prognosis into a map of therapeutic opportunity, the Singapore-led consortium has offered one of the more hopeful reframings of advanced abdominal cancer in recent years.

The scale of the ascites cohort deserves particular emphasis. Studies of malignant fluid have often been limited to a few dozen samples drawn from a single tumor type, making it difficult to distinguish cancer-specific signals from generic inflammatory noise. By sampling 397 specimens across 21 histological origins, the investigators could identify protein themes that recur regardless of whether the primary tumor arose in the stomach, ovary, colon or appendix, strengthening the case that the observed ligands reflect shared biology of the peritoneal metastatic niche rather than the idiosyncrasies of any one malignancy.

The choice of PAI-1 as the validation target also carries scientific weight. PAI-1, formally known as serpin E1, is a member of the serine protease inhibitor family and functions as the principal physiological brake on both tissue-type and urokinase plasminogen activators. Although elevated PAI-1 levels have been associated with poor outcomes in numerous cancers, the protein has historically been considered difficult to drug, partly because it interacts with multiple partners, including vitronectin, and because its role in hemostasis raises concerns about systemic interference. A nanobody format, with its compact size and high affinity confirmed here by surface plasmon resonance, offers one route around these obstacles, particularly when delivered locally into the peritoneal cavity where the ligand is concentrated.

The single-cell finding that S100A10 and PAI-1 originate from separate cell populations adds a layer of ecological interpretation to the fibrinolytic data. It implies a division of labor within the tumor microenvironment, in which different cellular compartments contribute opposing components of the same pathway, potentially allowing the tumor to tune plasmin activity spatially and temporally. Spatially resolved methods such as Stereo-seq, referenced in the study’s methodology, are increasingly enabling this kind of anatomical attribution of secreted signals.

Finally, the open-access release of the proteomic dataset itself represents a resource for the field. Researchers studying individual tumor types can query the ascites proteome for their own candidate ligands, extending the paracrine target list well beyond the factors validated in this initial report.

Subject of Research: Proteotranscriptomic identification of paracrine therapeutic targets in malignant ascites from peritoneal metastases

Article Title: Identification of novel paracrine therapeutic targets in peritoneal metastases through integrative proteotranscriptomic analysis of ascites

Article References: Liu, Y., Chong, C. Y. L., Wang, Y., Xu, L., Lim, H. J., Tan, Q. X., Hendrikson, J., Tan, J. W.-S., Ng, G., Guo, W., Wang, Y., Huang, L., Ge, W., Rethineswaran, V. K., Ong, J. Z. L., Quek, Y. E., Ang, A. J. Y., Teo, F., Chen, O. W., … Singapore Peritoneal Oncology Study (SPOS) Group and Singapore Gastric Cancer Consortium (SGCC) (2026). Identification of novel paracrine therapeutic targets in peritoneal metastases through integrative proteotranscriptomic analysis of ascites. Molecular Cancer. https://doi.org/10.1186/s12943-026-02787-3

Image Credits: AI Generated

DOI: 10.1186/s12943-026-02787-3

Keywords: peritoneal metastases, ascites, proteomics, PAI-1, S100A10, fibrinolysis, nanobody, paracrine signaling, single-cell sequencing, coagulation cascade, drug targets, Molecular Cancer