Institutional experience with tumor budding in invasive breast carcinoma: A histopathological study of 25 cases.

INTRODUCTION

Tumor budding is a histopathological feature defined by the presence of detached individual or small clusters of tumor cells at the invasive front or edge of the tumor with surrounding normal tissue. It is commonly linked to aggressive tumor behavior and poor prognosis. Tumor cells detach, lose their epithelial characteristics of cell adhesion, and gain mesenchymal properties (Epithelial-Mesenchymal transition), which in turn help the tumor cells migrate through the extracellular matrix, invade the lymphatic and/or vascular channels, leading to metastasis, thus resulting in poor overall survival.^[1]

In 1949, Japanese researcher Imai first described tumor budding in relation to gastric cancer, initially referring to it as "sprouting" at the invasive edge of carcinomas.^[1,2] Subsequently, other researchers also discovered an association between tumor budding and prognosis in several cancers, including tongue, larynx, esophagus, and colon. Tumor budding is now regarded as an important histopathological feature for assessing the aggressiveness of cancers, especially colorectal cancer, where it serves as an independent adverse prognostic factor.^[3] The 2016 International Tumor Budding Consensus Conference (ITBCC) sought to standardize the pathological classification of tumor budding, defining it as isolated cancer cells or small clusters of up to four tumor cells at the invasive front.^[4] This is validated and now included as a recommended element in the College of American Pathologists (CAP) protocol reporting of colorectal cancers.

Globally, breast cancer is the most common cancer in women. In 2020, there were over 2.3 million new cases of breast cancer and 685,000 related deaths.^[5] Prognosis and treatment of breast cancer vary based on tumor subtype and grade, tumor size, lymph node status, and expression of hormone receptors. Patients with similar prognostic characteristics may still experience different clinical outcomes. Therefore, significant research efforts are needed to investigate new prognostic factors like tumor budding.

Interest in tumor budding and its clinical implications has increased significantly in recent years, with several recent studies demonstrating that high tumor budding in breast carcinoma is linked to unfavorable clinicopathological features, including tumor size, tumor differentiation, lymph node metastasis, and lymphatic or vascular invasion.^[6,7] However, the absence of a validated histological tumor budding scoring system and the lack of clinical data supporting its practical application limit the broader incorporation of tumor budding into routine histopathology reporting of breast cancers.

The objective of the study was to assess the clinicopathologic significance of tumor budding in breast carcinoma and explore its correlation with various histopathological parameters.

MATERIAL AND METHODS

This observational retrospective study was conducted in the department of histopathology over a period of 1 year, from January 2024 to December 2024.

Tumor budding

Tumor buds were defined as either a single tumor cell or a cluster of up to four cells located at the invasive front of the tumor.^[8] Microscopic visualization and tumor bud assessment were done by two pathologists independently using a standard bright field microscope (Olympus BX43, Olympus corporation, Tokyo, Japan) equipped with 4x, 10x, and 40x objectives. To assess tumor budding on H&E-stained slides, the invasive front of the tumor (hotspot) was located under a 4x objective of the microscope [Figure 1]. Within this region, tumor buds were then counted in 10 consecutive high-power fields (HPF, 40x objective, area 0.950 mm2) and the mean count per 10 high power field (HPF) was calculated in each case.

Close

Figure 1:

Hematoxylin and eosin (H and E) stain, 10× magnification. Low-power image showing the invasive tumor front with tumor cells infiltrating the surrounding stroma. Tumor buds are identified and counted at the invasive front.

Export to PPT

Cells displaying a similar morphology to the tumor were selected for counting. Structures that resembled tumor buds, such as peri-tumoral inflammatory cells, endothelial cells, and fibroblasts, were excluded after careful examination under high power [Figure 2]. Tumor buds bordering the ductal carcinoma in situ or confined within the retraction spaces were not counted as they did not represent true stromal invasion [Figure 3]. Tumor buds adjacent to areas showing crushing or cautery artefact and overlapped/folded tissue were excluded. Only the clearest available fields were considered. International Tumor Budding Consensus Conference (ITBCC) recommends counting tumor buds in a single hotspot using a 20x objective (field area 0.785mm2). Since our microscope lacked a 20x objective, tumor bud count was performed using the 40x objective (Field area 0.950mm2). To ensure comparability with the ITBCC reference field size, tumor bud counts obtained at 40x magnification were normalised using the ITBCC normalization table, mentioned in the CAP protocol of colorectal carcinoma.

Close

Figure 2:

Hematoxylin and eosin (H and E) stain, 40× magnification. High-power image of the invasive tumor front showing tumor buds (arrow; defined as clusters of fewer than four tumor cells). Also visible are inflammatory cells, endothelial cells, and fibroblasts, which can mimic the morphology of tumor buds.

Export to PPT

Close

Figure 3:

H and E (40x), High power image of invasive tumor front displaying tumor cells within retraction spaces. Such foci were not considered for counting tumor buds. H and E: Hematoxyline and eosin

Export to PPT

Cases were subsequently categorized into low-grade tumor buds (mean tumor buds < 4/10 HPF) and high-grade tumor buds (mean tumor buds ≥ 4 /10 HPF).^[9,10] For interobserver consistency, two pathologists independently assessed the tumor budding score. Discrepant cases were jointly reviewed to reach consensus. In cases where there was a discrepancy in the tumor bud grades between the two pathologists, the slides were reviewed, and a consensus grading was given (Cohen’s Kappa 0.655).

RESULTS

In this study, a total of 25 cases of diagnosed invasive breast carcinoma, of no special type, were studied for tumour budding.

Female patients had a wide age distribution ranging from 33 to 73 years, with the median age being 53 years. The majority of the patients were in the age range of 41-60 years. 40% of the patients were <50 years, while 60% were > 50 years of age.

In 60% (15/25) of cases, the tumor was present on the left side of the breast, and in 40% (10/25) of cases, the tumor was on the right side of the breast.

The majority of the cases 72% (18/25), had the greatest tumour dimension of 2–5 cm (T2 stage). 28% (7/25) cases had tumor size > 5cm. There were no cases with a tumor size < 2cm.

Most tumors were grade 2 and grade 3, i.e., 44% each, while 12% were grade 1 tumors. 68% (17/25) of the tumors showed necrosis. The majority of cases showed lymphovascular invasion (72%) and had positive axillary lymph nodes (68%). Amongst the patients with axillary lymph node involvement, most were in the N2 stage (28%), followed by the N1 (24%) and the N3 (16%).

The majority were ER positive (68%), PR positive (64%), HER2 negative (92%), with a Ki67 index >15% (68%), characteristic of luminal B molecular subtype (56%). 16% of the cases were Luminal A subtype, followed by triple negative (20%) and HER-2 enriched (8%). Table 1 shows the distribution of clinicopathological parameters.

DISCUSSION

In this retrospective observational study, we investigated the clinicopathological significance of tumor budding in invasive breast carcinoma of no special type. Our findings indicate that high tumor budding is significantly associated with adverse prognostic factors such as lymphovascular invasion, tumor necrosis, advanced nodal stage, and a high Ki-67 proliferation index. These results reinforce the growing body of evidence supporting tumor budding as a potential histopathological marker of tumor aggressiveness in breast cancer.

Tumor budding, initially described in gastric cancer by Imai in 1949 ^[2] was later formalized as a prognostic tool in colorectal carcinoma by the ITBCC in 2016.^[4] It has since gained recognition as a morphological correlate of epithelialmesenchymal transition (EMT), with budding cells showing reduced adhesion and enhanced motility, favoring invasion and metastasis.^[1] While routinely assessed in colorectal cancer, its application in breast cancer is still emerging due to limited standardization and validation studies.

In our study, 68% of cases exhibited high tumor budding, which was significantly correlated with lymphovascular invasion (p = 0.008). This finding is consistent with previous reports by Liang et al. and Salhia et al., who showed that tumor budding at the invasive margin was associated with vascular and lymphatic spread in breast cancer.^[6,7] EMT, characterized by loss of E-cadherin and gain of vimentin expression, underpins this biological behavior and may explain the aggressive phenotype associated with budding cells.^[1]

We also observed a strong association between high tumor budding and tumor necrosis (p = 0.001). Necrosis is commonly associated with high-grade, rapidly proliferating tumors and hypoxic microenvironments. Hypoxia-inducible factors such as HIF-1α have been implicated in driving EMT under such conditions, which could further promote tumor budding.^[11]

Therefore, necrosis and budding may represent converging morphological indicators of tumor hypoxia and aggressiveness.

Another critical association was observed between tumor budding and advanced pathological nodal stage (N2 and N3) (p = 0.029). Lymph node metastasis remains a pivotal factor in breast cancer prognosis and treatment decisions. Previous studies have demonstrated that tumor budding is an early event in the metastatic cascade and may predict nodal involvement even before overt metastasis is histologically evident.^[9] This suggests that tumor budding could be a valuable adjunct in risk stratification, particularly in clinically node-negative patients.

Moreover, a high Ki-67 index (>15%) was significantly associated with high tumor budding (p = 0.024). Ki-67 is a well-established marker of cellular proliferation and poor prognosis, especially in hormone receptor-positive breast cancers.^[12] The combined presence of high budding and a high Ki-67 index may signify a particularly aggressive tumor subset that warrants closer surveillance and possibly more intensive therapy.

Interestingly, while over half of the patients with high tumor budding had grade 3 tumors and larger tumor sizes (>5 cm), these associations were not statistically significant. Tumor grade, as per the Nottingham system, evaluates tubular formation, nuclear pleomorphism, and mitotic count. However, tumor budding may represent a distinct biological phenomenon not fully captured by conventional grading.^[13] This supports the view that budding should be evaluated as a separate parameter with independent prognostic significance.

No significant associations were found between tumor budding and hormone receptor status (ER/PR) or HER2 expression. A 2023 meta-analysis of 1,763 breast cancer patients showed that low-grade budding correlated significantly with the triple-negative subtype. Our findings did not mirror this subtype association, possibly due to the limited sample size or population-specific differences.^[14]

Salhia et al. found that high peripheral tumor budding in breast cancer, assessed using pan-cytokeratin staining, significantly correlated with lymph node metastasis and lymphatic invasion, even in ER-positive, low proliferative tumors.^[7] Their findings support our results and suggest tumor budding's utility as an early marker of metastatic potential and as a useful addition to preoperative risk assessment.

Gabal et al. demonstrated that tumor budding significantly correlated with MMP-2 expression in invasive breast carcinoma, particularly in tumors with ill-defined borders and lymph node metastasis.^[15] The co-expression of tumor budding and MMP-2 highlights a possible interplay between cellular detachment, proteolysis, and ER signaling in promoting tumor invasion.

A recent study by Manimaran et al. further validated tumor budding's prognostic value, showing its association with advanced tumor and nodal stages, as well as systemic metastasis.^[16] Notably, grade 3 tumors with high tumor budding exhibited poorer event-free survival, reinforcing budding as a simple but potent morphological predictor of aggressive behavior in breast cancer.

From a technical perspective, our method of assessing tumor budding on H and E-stained sections was straightforward and reproducible. Nonetheless, budding can be underestimated in areas with dense stroma or inflammatory infiltrates. Studies suggest that immunohistochemical staining using pancytokeratin can enhance accuracy and interobserver agreement, especially in low-grade cases or those with subtle budding.^[17]

Several limitations must be acknowledged. Our study was monocentric, retrospective, and involved a modest sample size, all of which may limit the generalizability of our results. A sample size of 25 cases was determined pragmatically based on the number of eligible mastectomies during the study period. This is because the primary aim was hypothesis-generating rather than confirmatory.

Additionally, the absence of survival endpoints, disease-free survival/overall survival, precludes direct assessment of the prognostic value of tumor budding. Thus, its relevance remains inferential via associations with adverse pathological features. Some patients were lost to follow-up, precluding meaningful survival analysis. Larger, outcomes-based studies are needed to confirm its clinical utility.

The inclusion of mastectomy-only specimens introduces a potential sampling bias as smaller tumors managed by lumpectomy were excluded. The absence of a standardised consensus-based scoring system for tumor budding in breast cancer, unlike the well-established ITBCC criteria in colorectal cancer, poses a methodological limitation. Furthermore, possible preanalytical variability related to tissue fixation and processing could have influenced the microscopic assessment of tumor budding.

Nevertheless, the study adds valuable data to the limited literature on tumor budding in breast cancer, particularly from South Asian cohorts.

A key strength of this study lies in its use of a uniform histological subtype (invasive breast carcinoma, no special type), which reduced variability and allowed for focused analysis. Comprehensive assessment of histopathological and immunohistochemical parameters further enhanced the robustness of our correlations.

CONCLUSION

In summary, this retrospective, single-centre study adds to the emerging evidence supporting the role of tumor budding as a potential histopathological marker of tumor aggressiveness in breast cancer. Its significant associations with lymphovascular invasion, tumor necrosis, high Ki-67 index, and advanced nodal stage underscore its prognostic potential. However, given the limited cohort size and lack of correlation with survival outcomes, these findings should be interpreted as hypothesis-generating. Larger multicentric prospective studies are needed to validate the prognostic value before routine implementation in histopathological reporting.