Vaccine Adjuvant Injected Straight Into a Dog’s Tumor Sparks a Hidden Immune Assault

In a development that could reshape how veterinarians and oncologists think about treating solid tumors, researchers have reported that injecting an experimental immune-stimulating compound directly into a canine soft tissue sarcoma triggered a striking influx of immune cells into what is normally an immunologically quiet tumor. The case report, published in the journal Veterinary Oncology, describes the preliminary immunologic observations of EmT4™, a synthetic agonist of Toll-like receptor 4 (TLR4), administered intratumorally to an eight-year-old Boston terrier before surgical removal of a three-centimeter mass on her right forelimb. While the findings come from a single patient and must be interpreted with caution, the cellular changes observed inside the tumor are consistent with the kind of in situ immune activation that has produced remarkable results in early human trials of related compounds.

The compound at the center of the study belongs to a family of molecules derived from one of immunology’s most storied discoveries: lipopolysaccharide, or LPS, the toxic component of the outer membrane of gram-negative bacteria. LPS was the first identified agonist of TLR4, a transmembrane signaling receptor expressed on macrophages and dendritic cells that serves as an alarm bell for the innate immune system. Raw LPS, however, provokes dangerous, unregulated inflammation. Scientists therefore developed mimetics of monophosphoryl lipid A (MPLA), the immunostimulatory core of LPS, which retain the receptor-activating power while shedding most of the toxicity. These MPLA-based adjuvants are already embedded in approved human vaccines, including GSK’s AS01 and AS04 formulations, and have been delivered to millions of people.

EmT4™ takes this concept a step further. Rather than a hydrolyzed mixture of acylated lipid species derived from Salmonella bacteria, it contains a single, fully defined synthetic molecule called 3D (6-acyl)-PHAD, the hexaacyl form known to be the active congener in humans. The agonist is incorporated into a squalene-based oil-in-water emulsion stabilized with glycerol and polysorbate 80, with particles of roughly 100 nanometers. Produced under good manufacturing practice conditions in an ISO7 cleanroom, the formulation is functionally analogous to GLA-SE and AS02, adjuvants that have themselves been tested intratumorally in human cancer patients. In mouse studies, EmT4™ proved well tolerated even with daily dosing, boosted the breadth and strength of mRNA vaccine responses against SARS-CoV-2 spike protein, and generated robust protective Th1-type CD4 T cell immunity as an adjuvant for a tuberculosis vaccine.

The canine patient in the report was an eight-year-old, thirteen-kilogram spayed female Boston terrier with a soft, fluctuant, broad-based mass over her right elbow that had been slowly enlarging for approximately five months. Two cytological aspirates had yielded only adipose tissue, leading her veterinarians to initially suspect a benign lipoma. Because of the tumor’s size and location, complete surgical excision seemed unlikely, and incomplete removal of a soft tissue sarcoma typically condemns a dog to costly follow-up radiation. The owner, who happened to have long professional familiarity with TLR4 adjuvants, requested intratumoral EmT4™ injections before surgery in the hope of stimulating an anti-tumor immune response in place.

Two injections were given two weeks apart, each delivering 65 micrograms of EmT4™, equivalent to 5 micrograms per kilogram, a dose identical to that used in human trials of the related TLR4 agonist G100 in Merkel cell carcinoma and follicular lymphoma, and well below the 100-microgram-per-kilogram single dose of MPLA that research dogs tolerated without toxicity in earlier safety studies. Within hours of the first injection, the tumor swelled from roughly three to five centimeters, the overlying skin became taut, warm, and painful, and the dog became mildly lethargic, though she remained alert, responsive, and continued eating and drinking normally. The signs corresponded to a grade 1 adverse event under veterinary oncology criteria and resolved within a day. Notably, the second injection produced neither swelling nor discomfort, a pattern the authors attribute to the autoregulation that typically follows strong initial innate immune activation.

When the tumor was surgically excised four weeks after the second injection, histopathology confirmed a grade 2 soft tissue sarcoma, but with an unexpected feature: dense, tightly packed cuffs of immune cells encircled approximately half of the tumor’s blood vessels. Immunohistochemistry performed on the excised tissue revealed a heterogeneous immune infiltrate within these perivascular clusters, with CD20-positive B lymphocytes constituting the largest population at 57 percent, followed by CD3-positive T lymphocytes at 25 percent, Iba-1-positive macrophages and dendritic cells at 36 percent, FOXP3-positive regulatory T cells at just 8 percent, and CD204-positive macrophages at 5 percent. Two archived, non-injected soft tissue sarcomas of comparable grade and histology showed no CD3, CD20, or FOXP3 staining at all, and roughly threefold fewer Iba-1-positive cells, offering an informal point of comparison.

To probe the functional state of these infiltrating cells, the team employed RNAscope in situ hybridization, a technique that detects specific RNA transcripts within intact tissue sections. The results were striking: approximately 29 percent of cells in the lymphocyte clusters carried CD4 transcripts and 24 percent carried transcripts for tumor necrosis factor-alpha, the proinflammatory cytokine central to TLR4-driven inflammation, while 3.2 percent expressed the CD8 cytotoxic T cell marker and 1.3 percent carried interferon-gamma transcripts. The authors interpret the combination of clustered CD4 cells and abundant TNF-alpha as the signature of a Th1-biased immune response, the same profile observed in human soft tissue sarcoma patients treated with intratumoral GLA-SE and radiation. Given the short half-life of TNF-alpha protein and messenger RNA, its detection four weeks after the final injection suggests that immune activation was sustained up to the time of surgery.

The investigators are careful to emphasize the limitations of their observations. No pretreatment biopsy was available, because the mass had been presumed benign and was only sampled by non-diagnostic aspirates, so the team cannot prove that EmT4™ caused the immune infiltration rather than revealing a pre-existing one. Two archived tumors serve only as illustrative examples, not true controls. Canine soft tissue sarcomas are generally considered immunologically cold, owing to low mutational burdens and sparse neoantigen expression, and fibrosarcoma-type tumors in particular typically harbor few lymphocytes, which lends circumstantial weight to the idea that the agonist recruited the infiltrate. The authors also note that the immune cells remained concentrated in perivascular niches near the tumor periphery rather than penetrating its core, mirroring patterns seen in murine models where hypoxia and immunosuppressive factors in tumor interiors may block immune cell migration, and raising questions about whether differences in endothelial adhesion molecules dictate which vessels serve as entry points.

The clinical course of the patient offers a tantalizing coda. Although histology showed tumor cells extending to the surgical margins, the site was treated with two rounds of electrochemotherapy, and approximately thirty months later there has been no recurrence. The authors frame the case as foundational rather than conclusive, calling for controlled trials with pretreatment biopsies to establish appropriate dosing, toxicity profiles, and repeatability. Looking forward, they envision intratumoral EmT4™ as a component of combination regimens, potentially paired with immunogenic-cell-death-inducing radiation, low-dose cyclophosphamide to deplete regulatory T cells, or the veterinary immune checkpoint inhibitors now entering clinical use. In human medicine, intratumoral TLR4 agonists have already produced durable complete and partial remissions in Merkel cell carcinoma and abscopal tumor regressions in follicular lymphoma, with matching T cell clones detected in both tumors and circulating blood, evidence that local injection can seed systemic immunity. If controlled veterinary studies confirm what this single Boston terrier’s tumor suggested, the therapy could one day turn an ordinary vaccine adjuvant into a personalized, in situ cancer vaccine for dogs, and perhaps, by extension, a bridge to new approaches for human patients.

Subject of Research: Immunologic effects of intratumoral TLR4 agonist EmT4™ injection in a canine soft tissue sarcoma

Article Title: Preliminary immunologic observations of intratumoral EmT4™, a TLR4 agonist, in a canine soft tissue sarcoma

Article References: Helfand, S. C., Carossino, M., Kim, J., McMonigle Jr, D. R., Carter, D., & Gray, S. A. (2026). Preliminary immunologic observations of intratumoral EmT4™, a TLR4 agonist, in a canine soft tissue sarcoma. Veterinary Oncology, 3(1), Article 17. https://doi.org/10.1186/s44356-026-00063-6

Image Credits: AI Generated

DOI: 10.1186/s44356-026-00063-6

Keywords: TLR4 agonist, EmT4, monophosphoryl lipid A, intratumoral immunotherapy, canine soft tissue sarcoma, tumor microenvironment, TNF-alpha, veterinary oncology, perivascular immune niches, in situ vaccination, vaccine adjuvant, Boston terrier