The core of the argument rests on three epidemiological realities. First, roughly half of all canine mammary tumors are benign, meaning that any staging protocol applied indiscriminately before a diagnosis of malignancy is confirmed will expose a large share of patients to unnecessary testing. Second, a clinically relevant proportion of malignant mammary tumors are low grade and highly unlikely to metastasize, so even among dogs with cancer, many derive little benefit from systemic imaging. Third, and perhaps most strikingly, distant metastasis at the time of initial presentation appears to be genuinely rare. In data drawn from the PennVet Shelter Canine Mammary Tumor Program, only 13 of 249 dogs screened for enrollment had detectable distant metastasis, and the median tumor size in that subgroup was 6.7 centimeters, suggesting that metastasis at staging is largely a phenomenon of advanced local disease.

Despite these numbers, recent recommendations have shifted toward more elaborate pre-surgical workups, including thoracic computed tomography, abdominal ultrasound, and sentinel lymph node mapping. The authors contend that this trend is driven partly by the availability of more sensitive imaging technology and partly by a fear of missing occult metastatic lesions, a fear that has not been critically examined against actual detection rates. The consequences are not trivial. Advanced imaging adds substantial cost, requires additional anesthesia or heavy sedation, and delays definitive surgical treatment. In some cases, the accumulated expense of staging may prevent owners from pursuing the surgery that would actually cure their dog, a paradox the authors describe as directly contrary to the goal of cancer care.

To build their case, the researchers looked closely at how human breast cancer, long the conceptual model for canine mammary tumor management, handles the same question. Human breast cancer staging begins with mammography and biopsy, and only after malignancy is confirmed do clinicians proceed to sentinel lymph node biopsy, molecular typing, and, selectively, systemic imaging. Notably, guidelines from the National Comprehensive Cancer Network have evolved over the past decade in the opposite direction from veterinary practice: women with early-stage, low-risk disease based on receptor profile and sentinel lymph node status now forgo systemic imaging altogether. The rationale is that metastasis is rare in early-stage asymptomatic patients, and false-positive imaging findings automatically trigger cascades of additional diagnostics, cost, and anxiety. The canine situation, the authors argue, shares these same mathematical and psychological dynamics, yet veterinary recommendations have moved toward more testing rather than less.

Sentinel lymph node biopsy receives particularly detailed scrutiny. In human breast cancer, the technique replaced radical axillary dissection because it accurately identifies the first lymph node to receive drainage from the tumor with far less morbidity. In dogs, the anatomy is simpler: the axillary lymphocenter typically contains only one or two nodes, and the predictable drainage pattern sends the two cranial mammary glands to the axillary node and the two caudal glands to the superficial inguinal node, with the third gland draining in either direction. Reviews of lymphographic studies show that this normal pattern persists as the predominant route in most dogs with tumors, with deviations being rare and possibly linked to high tumor grade, which promotes lymphangiogenesis, inflammation, and vascular remodeling within the tumor microenvironment.

Yet the authors identify practical obstacles that make routine sentinel node biopsy in dogs problematic. Dogs commonly present with multiple independent mammary tumors, with studies reporting that 60 to 70 percent have tumors in different glands, each with its own histology and biological behavior. The largest tumor is often, but not always, the most aggressive, so selecting a single tumor for mapping risks sampling the wrong lesion, while injecting multiple tumors yields multiple biopsied nodes and larger, more painful surgeries with higher complication rates. Furthermore, because sentinel node techniques require peritumoral or subareolar contrast injection before complete resection, dogs without a confirmed diagnosis of malignancy are routinely subjected to the procedure even when their tumors turn out to be benign or low risk. The authors also note that in dogs, unlike in women, lymph node status has weak implications for chemotherapy decisions, since dogs with high-grade tumors are likely to receive systemic therapy regardless of nodal findings, making the procedure’s main human indication largely moot.

Thoracic imaging comes under similar examination. Three-view thoracic radiographs remain a reasonably sensitive, widely available, and cost-effective method that can be performed under light sedation. Computed tomography detects more small pulmonary nodules, but the clinical significance of that sensitivity is questionable in this population. In human early-stage breast cancer, the false-positive rate for chest CT, about 10 percent, actually exceeded the overall yield of metastasis detection, roughly 2 percent, a cautionary statistic for anyone considering withholding surgery based on small CT findings. Classic necropsy work by Fidler and Brodey established that pulmonary metastasis was present in 81 percent of dogs with metastatic disease, compared with only 5 to 16 percent for abdominal organs, meaning the lungs are the primary metastatic site and abdominal spread is unlikely when thoracic imaging is negative. The authors’ own prospective trial data found that every dog later confirmed at necropsy to have pulmonary metastasis had preceding three-view radiographs documenting it.

Abdominal ultrasound fares similarly in the analysis. Because the lungs dominate the metastatic pattern, the risk of metastasis confined to abdominal organs in the face of negative thoracic radiographs is low. Ultrasound also cannot reliably distinguish benign from malignant hepatic or splenic nodules, generating false positives that invite further invasive diagnostics. A large prospective screening study in healthy middle-aged and older dogs found that physical examination plus cytology of superficial masses detected the great majority of cancers, undercutting the argument that abdominal imaging is needed to catch unrelated occult tumors. Follow-up data from the authors’ earlier cohort showed that non-mammary tumors detected during monitoring emerged a median of 500 to 600 days after the original surgery, indicating they were new developments rather than lesions that staging had missed.

Fine needle aspiration of the tumors themselves is likewise judged unreliable as a screening tool. Reported sensitivity for distinguishing benign from malignant mammary lesions ranges from 25 to 95.2 percent across studies, and even modern cytological grading systems show poor concordance with histology for low-grade tumors, precisely the category where misclassification matters most for staging decisions. Because cytologists evaluate cellular atypia without the architectural context available to surgical pathologists, and because even pathologists disagree about the benign-malignant boundary in low-grade lesions across institutions and countries, the authors decline to recommend routine cytology as a gatekeeper for staging. They do concede it retains value for identifying non-mammary masses such as lipomas or mast cell tumors.

The alternative framework the authors propose is elegantly simple: let tumor size and clinical risk factors decide. For tumors under 1 centimeter, or under 3 centimeters in WHO T1 category, no pre-surgical staging is recommended; post-operative thoracic radiographs are applied by grade, skipped for grade 1, optional for grade 2, and recommended for grade 3. Tumors larger than 3 centimeters warrant pre-surgical three-view thoracic radiographs, as do any tumors with cytologically positive local nodes or obvious malignant features such as rapid growth, ulceration, or deep fixation. Routine fine needle aspiration of palpably normal regional nodes is deemed unnecessary for small tumors, since inguinal nodes are typically removed anyway with the fifth gland, but firm or enlarged nodes should always be aspirated before surgery to adjust the operative plan. Patent blue dye is endorsed for identifying axillary nodes during removal of larger cranial tumors. Abdominal ultrasound is reserved for dogs with positive inguinal nodes, to assess iliac chains, and thoracic CT is judged not indicated in most cases. The authors acknowledge that a small number of small tumors with metastatic nodes could slip through the pre-surgical net, but argue that post-operative grade assessment catches grade 2 and 3 cases. Their closing message extends beyond mammary tumors: staging recommendations across veterinary oncology should be judged by whether results change prognosis or treatment, because diagnostic intensity that offers minimal clinical benefit while imposing financial and emotional costs may ultimately harm the very patients it aims to protect.

Subject of Research: Evidence-based revision of pre- and post-surgical staging practices for canine mammary tumors

Article Title: Considerations for revised staging practices in canine mammary tumors

Article References: Sorenmo, K. U., Cofone, M. A., Donnelly, L., Duda, L., & Stefanovski, D. (2026). Considerations for revised staging practices in canine mammary tumors. Veterinary Oncology, 3(1), Article 9. https://doi.org/10.1186/s44356-026-00061-8

Image Credits: AI Generated

DOI: 10.1186/s44356-026-00061-8

Keywords: canine mammary tumors, cancer staging, veterinary oncology, sentinel lymph node biopsy, thoracic radiographs, thoracic CT, abdominal ultrasound, fine needle aspiration, tumor grade, metastasis, risk-based staging, dog health