Lab-Grown Tumors From Dogs’ Noses Open a New Window on Hard-to-Treat Cancer

Researchers in Japan have grown miniature tumors in a dish from the nasal carcinomas of pet dogs, creating the first organoid culture system for one of the most stubborn cancers in veterinary medicine. The work, published in Frontiers in Veterinary Science, offers an experimental platform for a disease whose treatment options remain limited.

Canine nasal carcinoma is rare but brutal. It accounts for roughly one to two percent of malignant tumors in dogs, yet around three-quarters of these tumors behave malignantly, and nearly half of affected dogs — 46 percent — show metastasis to the lungs or regional lymph nodes. Dogs typically arrive at the clinic with nosebleeds, purulent nasal discharge, and sneezing, often accompanied by destruction of the nasal turbinate or septum. Exposure to environmental chemicals, including cigarette smoke, has been suggested as a risk factor. Nasal adenocarcinoma is the most common subtype, while squamous cell carcinoma carries a poorer prognosis.

Treatment today leans heavily on surgery combined with radiation therapy or chemotherapy. Radiation is often considered the standard of care, with reported median survival times of eight to 23 months. But the approach carries real drawbacks: treatment-related toxicities, limited therapeutic efficacy, high cost, and anesthetic risks for the elderly dogs who most often develop the disease. The authors argue that these limitations make the search for alternative drug therapies — and for laboratory models in which to test them — an urgent need.

Organoids are three-dimensional structures grown from epithelial cells embedded in Matrigel, a matrix that supports stem-like properties. Unlike flat two-dimensional cell cultures, they can reproduce structural and biological characteristics of their tissue of origin, offering a more realistic representation of tumor behavior in vivo, despite greater cost and technical demands. The research group had previously established organoids from canine bladder cancer, lung cancer, malignant mesothelioma, and apocrine adenocarcinoma, but no culture method had been reported for nasal carcinoma in dogs.

To build the new model, the team collected tumor tissue during surgery from dogs with nasal carcinoma at clinics across Japan. The tissue was minced, enzymatically digested, filtered, and mixed with Matrigel in culture plates. Within one to two weeks, the cancer cells formed spherical organoid structures, whereas normal nasal mucosal cells collected from healthy dogs proliferated slowly and could not sustain long-term culture. Transmission electron microscopy of the organoids revealed well-developed microvilli, undifferentiated cells, goblet cells packed with mucus granules, and squamous or intermediate cells containing tonofilaments — a pattern of structural differentiation mirroring the native tumor tissue.

The organoids also proved tumorigenic. When cells from passages two to three were implanted subcutaneously into immunodeficient SCID mice, they formed tumors within three weeks that histopathologically resembled the original cancers, showing solid structures, variable nuclei, and mitotic activity. Immunohistochemistry confirmed positive staining for cytokeratin 19, E-cadherin, and Ki67, markers of epithelial identity and proliferation, with expression patterns corresponding to those in the original tissues. Notably, the xenograft tissue did not metastasize to any other region during the study period.

A key test was whether the organoids could reveal differences in drug response among individual patients. Organoids from six cases were treated with three anti-cancer drugs — carboplatin, doxorubicin, and toceranib — at concentration ranges selected from previously reported clinical dosing and pharmacokinetic studies in dogs. The lines showed variable sensitivities, demonstrating inter-individual differences in drug responses. The authors caution, however, that this study did not compare organoid drug responses with actual clinical treatment outcomes, so the platform’s predictive value for therapeutic selection in living patients remains to be established.

To probe the molecular biology, the researchers performed RNA sequencing on organoids from six carcinoma cases and normal nasal cells from four healthy dogs. Principal component analysis cleanly separated the two groups, and the comparison yielded 5,707 differentially expressed genes, with 3,286 higher in the carcinomas and 2,521 higher in normal tissue. Gene set enrichment analysis pointed to pathways related to MYC targets, oxidative phosphorylation, and protein secretion as enriched in the tumor organoids. The authors emphasize an important caveat: because the normal cells were collected by cotton swab and could not be maintained in long-term culture, differences in cellular composition, culture conditions, and proliferative state — not just tumor status — may have contributed to the transcriptomic differences, so the identified genes cannot be interpreted exclusively as cancer-associated alterations.

Among the differentially expressed genes, S100A2 stood out for its markedly higher expression in the tumor organoids, a pattern confirmed in original tumor tissues. S100A2 is one of 20 members of the S100 family of calcium-binding secreted proteins, and its role in cancer is context-dependent: it has been reported as downregulated in tumors such as oral and liver cancers, yet other studies link it to proliferation, metabolism, invasion, and metastasis through mechanisms involving GLUT1 and PI3K/AKT/mTOR-related signaling. This is the first report examining its relationship with canine nasal carcinoma.

Functionally, the team transfected organoid cells with three distinct siRNAs against S100A2 and confirmed knockdown by western blotting. Silencing the gene significantly suppressed both cell proliferation and invasiveness in a Matrigel-coated invasion assay. However, no statistically significant change in c-MYC mRNA expression was detected after knockdown; with only four samples, the authors note this result should be interpreted cautiously and does not exclude a relationship between S100A2 and MYC signaling. They also stress that the functional experiments were performed in a single patient-derived organoid line, that downstream signaling mechanisms were not directly investigated, and that comprehensive molecular fidelity between parental tumors, organoids, and xenografts was not evaluated — validations they say future work must address across multiple independent lines.

Beyond veterinary medicine, the model may hold relevance for humans. Nasal cavity tumors in people are uncommon, accounting for roughly 0.5 to one percent of all tumors, but more than three-quarters are malignant, the five-year survival rate is approximately 30 percent, and high local invasiveness near the skull often makes complete surgical removal impossible. No organoid models focused on human nasal carcinoma currently exist. Because canine and human nasal carcinoma share similarities in disease progression and histological characteristics, the authors suggest canine organoids could provide a basis for future comparative studies — while noting that direct cross-species molecular comparisons will be required to determine whether therapeutic vulnerabilities are actually conserved between the two species.

Subject of Research: Development of patient-derived organoids for canine nasal carcinoma and functional characterization of S100A2

Article Title: Establishment of canine nasal carcinoma organoids for drug sensitivity testing and functional characterization of S100A2

Article References: Yamamoto, H., Elbadawy, M., Abugomaa, A., Tsunedomi, R., Kaneda, M., Usui, T., & Sasaki, K. (2026). Establishment of canine nasal carcinoma organoids for drug sensitivity testing and functional characterization of S100A2. Frontiers in Veterinary Science, 13, Article 1912335. https://doi.org/10.3389/fvets.2026.1912335

Image Credits: AI Generated

DOI: 10.3389/fvets.2026.1912335

Keywords: canine nasal carcinoma, organoids, drug sensitivity, S100A2, RNA sequencing, veterinary oncology, xenograft, siRNA knockdown, transcriptomics, dog cancer, MYC signaling, cancer invasion