As the body of knowledge on feline mammary tumors (FMTs) continues to grow, their histological classification and grading systems have undergone revisions and updates to better reflect the biological behavior of these tumors. In this review, the historical evolution of these frameworks is traced and later revisited in the context of their prognostic relevance. Numerous studies have investigated clinicopathological prognostic factors in feline mammary carcinomas (FMCs); however, the heterogeneity in assessment methods, inclusion criteria for survival analysis, and the clinical endpoints considered can often complicate direct comparisons across different studies and may contribute to seemingly conflicting results. Furthermore, the small cohort size of many studies limits the robustness and transferability of their findings. This paper provides an updated overview of the epidemiological, clinical, and pathological prognostic factors of these tumors, while also highlighting current challenges, methodological limitations, and areas for future improvement.

The tumor microenvironment (TME) comprises neoplastic and stromal cells, and extracellular matrix elements, all engaging in a complex interplay that ultimately dictates tumorigenesis, cancer progression, and therapeutic response. While extensive research on the TME has been conducted in human oncology, data on its veterinary counterpart, particularly in feline mammary tumors (FMTs), are still scarce. In this review, we explore current understanding of feline mammary carcinoma (FMC) microenvironment, focusing on tumor necrosis, fibrosis, angiogenesis, adipose tissue tumor-associated inflammation, extracellular vesicles, and epithelial–mesenchymal transition (EMT) and their prognostic implications. In FMC, remodeling of collagen fibers, cancer-associated fibroblasts (CAFs), regulatory T cells (Tregs) and elevated serum leptin have been associated with poor prognosis, whereas stromal cytotoxic T cells correlate with more favorable outcomes. By contrast, findings on necrosis and pro-angiogenic factors remain inconsistent, and research on extracellular vesicles (EVs) is still in its early stages. This review presents insights from human breast cancer (HBC) that further support and elucidate the potential relevance of these TME components. As FMCs are highly aggressive tumors, a deeper understanding of their microenvironment could not only improve prognostic accuracy but also uncover novel therapeutic targets. Furthermore, due to their similarities, FMCs offer a potential valuable spontaneous model for HBC, particularly for the aggressive triple-negative phenotypes.

Necrosis is a common finding in human and animal neoplasms. The percentage of tumor necrosis is included in tumor grading schemes in veterinary oncology; however, evaluation methods are often overlooked. Different studies have assessed the prognostic value of tumor necrosis in feline mammary tumors with contradictory results, which could be related to methodologic variability. In this study, a comprehensive evaluation of tumor necrosis in feline mammary tumors (FMTs) was conducted, by applying a semi-quantitative and a quantitative methodology for assessing necrosis. The interobserver agreement, the relationship with clinicopathological characteristics and the prognostic value of necrosis were analyzed in 154 FMT cases. Although subjectivity in the assessment of necrosis existed, an almost perfect agreement (weighted quadratic k = 0.851) between two observers was obtained. Furthermore, there was a significant positive correlation between the semi-quantitative and quantitative methods. Necrosis was more common and more extensive in malignant tumors than in their benign counterparts. Despite the non-significant results in the survival analysis, extensive necrosis was significantly associated with aggressive clinicopathological features, such as higher histological grade, high mitotic count and lymphovascular invasion. Our results support the potential relevance of necrosis in FMT.

Based on the longstanding Nottingham (EE) histological grading system - originally for grading of human breast carcinomas - a mitotic-modified (MMEE) and a revised (REE) grading systems were proposed to better accommodate feline mammary carcinoma (FMC) histopathological features. Additionally, a novel grading system (MGS) has also been proposed. This study aims to directly compare the EE grading with the proposed frameworks using the originally proposed mitotic count thresholds, and to evaluate their association with clinical outcomes. Additionally, we aim to calculate the veterinary Nottingham Prognostic Index (vet-NPI) and assess its prognostic value. Cohen's kappa agreement analysis revealed poor, slight and fair agreement between EE grading and MMEE, REE and MGS grading systems, respectively. Although all histological grading frameworks were associated with overall survival in the univariate analysis, only the EE and REE systems remained significant prognostic factors after adjustment to other covariates. Furthermore, REE and MGS gradings were associated with tumor-specific survival, but the latter lost its significance on multivariate analysis. The MGS system was the only one that was significantly linked with disease-free interval in both univariate and multivariate analyses. While the inclusion of LVI and nuclear form represent good steps toward improving the prognostic value of histological grade, the mitotic count thresholds appeared inadequate for our cohort, raising concerns about its applicability across different settings. Our findings support the inclusion of the vet-NPI in routine diagnostic histopathology reports of FMC to improve TSS prognostic assessment when using the EE and MMEE grading systems.

Regional lymph node status is a key prognostic factor in neoplastic lesions in cats and humans alike. The presence and clinical relevance of micrometastases and isolated tumour cells (ITCs) have been studied extensively in human breast cancer. However, comparable data in feline mammary carcinomas (FMCs) remain scarce; a single recent study addressed the presence of occult metastases. In this study, a multilevel sectioning protocol was applied to lymph nodes from 11 node-negative FMCs at first evaluation to search for occult metastases. The tissue in each paraffin-wax block was completely serially sectioned at 50 μm intervals, with one slide prepared at each tissue section and additional sections collected at every 150 μm interval for immunohistochemistry (IHC). ITCs were detected in two of the 11 cases on haematoxylin and eosin-stained tissue sections and confirmed by IHC for cytokeratin. No additional metastases were identified following IHC. These findings suggest that the assessment of multiple lymph node tissue levels may improve detection of metastases. Larger studies are necessary to determine the prevalence and clinical relevance of occult metastases in FMCs and their potential implications on treatment decisions.