Tumor budding in invasive breast carcinoma and its association with clinicopathological parameters: an experience from a tertiary care center in India
Article information
Abstract
Purpose
Breast cancer is one of the most common cancers globally, with an increasing incidence rate. It is a heterogeneous disease, and early metastasis remains a challenge. Tumor budding, defined as single tumor cells or small clusters at the invasive front, has been suggested as a prognostic marker in various cancers, including breast cancer. This study aims to evaluate tumor budding in invasive breast carcinoma using the International Tumor Budding Consensus Conference (ITBCC) scoring system and explore its association with pathological characteristics and prognosis.
Methods
A retrospective study was conducted on 100 mastectomy specimens of histopathologically confirmed invasive breast carcinoma, excluding cases that underwent chemotherapy or radiotherapy. Tumor budding was classified as low, intermediate, or high based on the ITBCC scoring method, and associations with clinicopathological features were analyzed using appropriate statistical tests.
Results
Tumor budding was classified as high in 52% of cases. A significant association was found between high tumor budding and higher tumor grade (P<0.001), negative estrogen receptor and progesterone receptor status (P<0.001), positive HER2neu status (P=0.003), and high Ki-67 levels (P<0.001). High tumor budding was also linked to higher T stage, and dermal lymphovascular invasion (P=0.001). Our findings support previous studies showing that high tumor budding is associated with poor prognostic factors such as higher tumor grade, negative hormone receptor status, and higher T stage.
Conclusion
Tumor budding is a potential prognostic marker in breast cancer, associated with more aggressive tumor characteristics.
INTRODUCTION
Breast cancer is one of the most commonly diagnosed cancers in the world. It accounts for nearly one-third of all new cancer diagnoses in women and is a leading cause of cancer mortality. Female breast cancer incidence rates have slowly been increasing by about 0.5% per year since the mid-2000s [1].
Breast cancer is a heterogeneous disease and has been classified into various pathological subtypes, each of which has a different prognosis and shows different responses to therapy [2]. Poor prognosis is associated with early invasion and metastasis. Currently used pathological characteristics do not fully represent early metastases such as tumor differentiation or lymphovascular infiltration [3]. This requires the need for an alternative histopathological characteristic which can be used to predict metastasis and prognosis in breast cancer cases.
Tumor budding is defined as a single tumor cell or a cluster of tumor cells (up to 4) at the invasive front of tumor or hot spot region [4]. Tumor buds represent cells in a partial epithelial-mesenchymal transition (EMT) state. Budding cells that were doubly stained have shown co-expression of mesenchymal marker vimentin and epithelial marker cytokeratin. This proved that the cells are undergoing EMT, indicating their invasive and migratory potential [4,5]. Tumor budding has been observed in many cancers, including colorectal, esophageal, gastric, lung, breast, and cervical cancer. It has now become a well-established prognostic factor in colorectal cancer. An attempt was made to standardize the assessment and scoring of tumor budding by the International Tumor Budding Consensus Conference (ITBCC) in 2016 [4]. The ITBCC group recommended a three-tier system to classify tumor budding on hematoxylin and eosin (H&E)-stained sections [4]. In colorectal cancer, tumor budding was seen in 20% to 40% of cases and was associated with lymph node metastasis, lymphovascular invasion (LVI), and an overall poorer prognosis [5]. Recently, few researchers attempted to study tumor budding associations in breast carcinoma and they observed its association with LVI, and other poor prognostic markers [3].
This study aimed to assess tumor budding in cases of invasive breast cancer with no special subtype. The ITBCC scoring system was used to count and classify tumor budding in cases of invasive breast cancer. The tumor budding score was compared with tumor grade, pathological TNM stage, and surrogate molecular classification. Its association was also established with LVI, dermal LVI, and perineural invasion.
Tumor budding in breast cancer has not been studied in the North Indian set-up. Our study will add to the limited data, and the associations found will help support the idea that tumor budding can become an independent prognostic parameter.
METHODS
This retrospective study included 100 resected specimens of histopathologically proven cases of invasive breast carcinoma, no special type, over one and a half years (from January 2023 to June 2024). Patients on chemotherapy/radiotherapy were excluded. Sarcomas, metastatic lesions, and in situ carcinomas of the breast were also not included. The study was approved by the Ethical Committee at Vardhman Mahavir Medical College and Safdarjung Hospital, New Delhi (IEC approval number is IEC/VMMC/SJH/Thesis/Project2023-10/CC-388). The written patient consent was waived.
Collection and preparation of material
Mastectomy specimens of histopathologically proven cases of invasive breast carcinoma received in the department of pathology were stained with H&E staining as per the standardized procedure. Tumor sections were evaluated for grade, pathological staging (pTNM) and receptor status. Each stained section was evaluated for histological grading of the tumor according to the modified Bloom-Richardson grading system [6]. Pathological staging (pTNM) was assessed as per the 8th edition of the American Joint Committee on Cancer staging system [6], and immunohistochemistry was performed for estrogen receptor (ER), progesterone receptor (PR), HER-2neu, and Ki-67. H&E-stained tumor sections with hotspot regions were used for the evaluation of tumor budding.
Counting and grading of tumor budding
Each H&E-stained mastectomy specimen was first viewed under the ×10 objective lens to identify the invasive front (hot spot) of the tumor, where the maximum number of tumor buds were located. Ten separate fields were scanned before selecting the hotspot. Followed by which the tumor buds were counted under the ×40 objective lens and were categorized as low, intermediate and high based on the ITBCC group as; 0–4 buds: low budding; 5–9 buds: intermediate budding; 10 or more buds: high budding (Fig. 1). The procedure is summarized in Fig. 2. However, no immunohistochemical staining was applied on the tissue section with the most invasive front to aid in better visualization of tumor buds.
Hematoxylin and eosin-stained section (×40) illustrating tumor budding in invasive breast carcinoma. Tumor budding, defined as the presence of isolated single tumor cells or small clusters of ≤4 cells at the invasive front, is marked with arrows. The image highlights tumor buds infiltrating the surrounding stroma, a feature associated with poor prognosis in breast carcinoma. (A) High tumor budding is observed, indicating aggressiveness (marked with arrows). (B) Intermediate tumor budding is observed (marked with arrows). (C) Low tumor budding is observed (marked with arrow).
Flowchart showing the methodology followed for counting tumor buds.
Statistical analysis
The presentation of the categorical variables was done in the form of numbers and percentages (%). On the other hand, the quantitative data with normal distribution were presented as the mean±standard deviation and the data with non-normal distribution as median with 25th and 75th percentiles (interquartile range). The data normality was checked by using the Shapiro-Wilk test. In the cases in which the data was not normal, we used non-parametric tests. The following statistical tests were applied to the results: (1) The association of the variables which were quantitative and not normally distributed in nature was analyzed using Kruskal-Wallis test and variables which were quantitative and normally distributed in nature were analyzed using analysis of variance; (2) The association of the variables which were qualitative in nature were analyzed using the chi-square test. If any cell had an expected value of less than 5 then Fisher exact test was used.
The data entry was done in the Microsoft EXCEL spreadsheet and the final analysis was done with the use of Statistical Package for Social Sciences (SPSS) ver. 25.0 software (IBM Corp.). For statistical significance, P-value of less than 0.05 was considered statistically significant.
RESULTS
The study involved 100 female patients with a mean age of 50.8 years. The age distribution showed that the majority of patients were in the 51–60 (29%) and 41–50 (28%) age groups. In the 21–30 age group, low budding accounted for 3.70%, intermediate for 0%, and high budding for 5.77%. In the 31–40 age group, low budding was 18.52%, intermediate 28.57%, and high 15.38%. For the 41–50, 51–60, 61–70, and 71–80 age groups, the percentages for low, intermediate, and high budding were as follows: 33.33%, 23.81%, 26.92%; 22.22%, 14.29%, 38.46%; 11.11%, 23.81%, 13.46%; and 11.11%, 9.52%, 0%, respectively. Age distribution was comparable with tumor budding categories; however, no significant association was found between age and tumor budding (P=0.144). Tumor grading, based on the modified Bloom-Richardson score, classified 51% of cases as grade 3, 47% as grade 2, and 2% as grade 1. Regarding receptor status, 54% were ER positive, and 50% were PR positive. HER2neu status was negative in 70% of cases. Ki-67 levels were greater than 14% in 80% of cases. Molecular classification revealed that 35% of cases were luminal B, 23% were HER2neu-enriched, 22% were triple negative, and 20% were luminal A (Table 1). Thirty-eight percent of cases were T2, while 47% had no nodal involvement (N0). LVI was present in 54% of cases, while perineural invasion occurred in 11%. Tumor budding was classified as high in 52% of cases. Dermal LVI was absent in 80% of cases. Tumor grade varied significantly across tumor budding categories. Grade 1 tumors were primarily associated with low budding (7.41%), while grade 3 tumors were predominantly seen in the high budding group (96.15%). The difference was statistically significant (P<0.001). Positive ER and PR status were significantly higher in low and intermediate budding groups, while negative status was predominant in the high budding group (P<0.001). Negative HER2neu status was more common in low budding (92.59%), while positive status was higher in the high budding group (44.23%) (P=0.003). Patients with high Ki-67 levels (>14%) were mostly in the high budding group (98.08%), while low Ki-67 levels were more frequent in the low budding group (P<0.001) (Table 1). In surrogate molecular classification, luminal A was most prevalent in the low budding group (59.26%), whereas triple-negative breast cancer (TNBC) and HER2neu-enriched subtypes showed predominantly higher tumor buds (38.46% and 44.23%, respectively, P<0.001). When tumor stage was compared with tumor budding, a higher stage (T3 & T4) was found to be significantly associated with high tumor budding (26.92% for both, P=0.011). However, no significant association was observed between tumor budding and N stage and LVI. A high tumor budding was also observed in cases with dermal LVI (P=0.001) (Table 1).
Association of low, intermediate, and high TB in invasive breast carcinoma with established clinicopathological parameters
DISCUSSION
This study aimed to quantify tumor budding (according to the ITBCC system) in invasive breast carcinoma and evaluate its association with tumor grade, pathological TNM stage, surrogate molecular classification, LVI, dermal LVI, and perineural invasion. Our findings highlight the prognostic significance of tumor budding, particularly in relation to tumor aggressiveness and molecular subtypes. Few studies have been done on tumor budding in breast cancer. They have classified tumor budding as high grade or low grade, according to the number of tumor buds counted. The quantification method widely varies across studies. For this reason, we chose to classify tumor budding as low, intermediate, or high according to the ITBCC scoring method, as it is the recommended method for colon cancer [4].
ITBCC group counted tumor buds under ×20 objective. However, some studies conducted on breast cancer used the ×40 objective lens to count tumor buds as it would be challenging to differentiate tumor buds from other mimickers such as macrophages, giant cells, fibroblasts, endothelial cells, smooth muscle cells and other artefacts on ×20 [7–10]. Therefore, we counted tumor buds at ×40 objective in our study as well. Our study found a significant association between higher tumor grade and high tumor budding. The proportion of patients with grade 3 scores was significantly higher in the high budding group compared to both the low and intermediate budding groups (P<0.001) [11]. On the other hand, the grade 2 score was significantly higher in the low and intermediate budding groups. This aligns with the observations of Renuka et al. [7] who found a significant association between high intratumoral budding and high tumor grade. Salhia et al. [10] also showed a signification correlation between high tumor budding with tumor grade.
Tumor budding was compared with T and N staging. The proportion of patients with Tis and T1 stage was significantly higher in the low budding group compared to both intermediate and high tumor budding categories. The proportion of patients with T2 stage was significantly higher in the intermediate budding group. Conversely, the proportion of patients with T3 and T4 stages was significantly higher in the high budding group compared to both the low and intermediate budding groups (P=0.011).
It can be concluded from these results that a higher T stage is associated with high tumor budding. However, no significant association was found with N stage. Our study found that negative ER status and negative PR status were both significantly associated with high tumor budding (P<0.001). Positive HER2neu status was seen in high tumor budding (P=0.003). The proportion of patients with Ki-67 levels above 14% was also significantly higher in the high budding group (P<0.001).
Patients classified as luminal A were significantly higher in low tumor budding, and luminal B were significantly higher in intermediate budding. Triple negative and HER2neu-enriched patients were seen in high tumor budding (P<0.001).
A systematic review and meta-analysis of seven studies by Lloyd et al. [8] found an increased likelihood of high tumor budding in ER− and PR-positive cases. They also found that TNBC showed a reduced incidence in high tumor budding. They concluded that high tumor budding is seen in hormone receptor-positive breast cancer and is associated with known adverse prognostic variables.
While our study aligns with prior research in linking high tumor budding with advanced tumor grade and aggressive behavior, discrepancies exist regarding its association with molecular subtypes. Lloyd et al. [8] reported an increased likelihood of high tumor budding in ER− and PR-positive cases, while our study found a significant association with ER− and PR-negative tumors (P<0.001). Additionally, we observed a higher frequency of tumor budding in HER2neu-enriched and TNBC (44.23% and 38.46%, respectively, P<0.001).
One possible explanation for these differences is methodological variation in tumor budding assessment. The counting method and magnification used (×20 vs. ×40) may influence the detection rate, as previous studies employed different approaches. Furthermore, tumor budding may exhibit molecular heterogeneity, behaving differently across racial and ethnic populations or in response to specific microenvironmental factors. These variations highlight the need for standardized scoring systems tailored specifically to breast cancer. Salhia et al. [10] also found high intratumoral budding to be significantly associated with ER-positive and −negative HER2neu status. Gujam et al. [12] found that high tumor budding was associated with ER-positive status.
The contrasting nature of associations found shows the need for further studies to be conducted on the correlation between tumor budding and molecular subtypes. Our study may have been limited due to a small sample size.
ER−, PR-negative tumors generally show a worse prognosis since drugs targeting hormone receptor-positive cases now have increased survival rates [13–15]. High tumor budding in ER−, PR− negative tumors indicate a correlation with worse prognosis. Thus, it could potentially be used as an additional prognostic factor to detect high-risk cases with low survival rates.
A key driver of tumor budding is EMT, a biological process that enables tumor cells to detach from the primary tumor, invade surrounding tissues, and ultimately metastasize. EMT is characterized by the loss of epithelial markers (e.g., E-cadherin) and gain of mesenchymal markers (e.g., vimentin and N-cadherin), leading to enhanced motility and invasiveness.
Triple-negative and HER2neu-enriched subtypes, which exhibited the highest tumor budding in our study, are known to have increased EMT activity. HER2neu overexpression has been linked to EMT induction via pathways such as PI3K/AKT (phosphatidylinositol-3-kinase/protein kinase B) and Ras/MAPK (Ras/mitogen-activated protein kinase), contributing to heightened invasiveness. Similarly, TNBC is enriched with mesenchymal-like cells and has been shown to undergo EMT at higher rates than luminal subtypes. This could explain the greater frequency of tumor budding observed in these subtypes.
Additionally, the significant correlation between tumor budding and high Ki-67 levels (P<0.001) supports the hypothesis that proliferative activity and EMT-driven invasion are closely linked. These findings suggest that tumor budding may not only serve as a prognostic marker but also indicate tumors that may respond poorly to conventional therapies and require EMT-targeted interventions.
No significant association was found with LVI and tumor budding. When considering the presence of LVI, the percentages were 44.44% for low budding, 42.86% for intermediate budding, and 63.46% for high budding (P=0.141).
Dermal LVI was significantly higher in the high tumor budding group (P=0.001). Perineural invasion showed a borderline significant difference in distribution among the groups (P=0.05). When perineural invasion was present, the percentages were 0% for low budding, 9.52% for intermediate budding, and 17.31% for high budding.
Renuka et al. [7] and Lloyd et al. [8] found that high tumor budding was significantly associated with LVI and lymph node metastasis. Masilamani et al. [9] also correlated it with LVI. LVI, along with positive lymph node status, was associated with high tumor budding by Salhia et al. [10] and Gujam et al. [12]. Tumor budding is emerging as a crucial histopathological marker associated with tumor invasiveness and poor prognosis. Our study demonstrates a significant association between high tumor budding and aggressive tumor characteristics, including high histologic grade (96.15% in grade 3 tumors, P<0.001), advanced T stage (P=0.011), and increased Ki-67 proliferation index (P<0.001). These findings suggest that tumor budding could serve as an additional prognostic parameter, potentially guiding treatment decisions.
From a clinical standpoint, patients with high tumor budding may require more aggressive treatment strategies, similar to those recommended for high-grade tumors. The significant correlation between high budding and increased dermal LVI (P=0.001) further underscores its role in metastatic potential. The early identification of tumor budding in routine pathological evaluation could enable risk stratification, facilitating timely intervention and improving patient outcomes.
Limitations of the study
Pan-cytokeratin immunohistochemistry can be additionally applied on the tumor section with the most invasive front to aid in the easy identification and confirmation of tumor buds, which could not be done in the present study due to resource constraints. Application of immunohistochemistry may further improve the accuracy of tumor buds calculation as other cells in the tumor microenvironment including macrophages, fibroblasts can be misinterpreted as epithelial cells. Furthermore, the study is restricted by its limited sample size and being a single-center study. Moreover, another difference in the methodology of the tumor budding assessment magnification employed in the present study, where tumor buds were counted at ×40 magnification, instead of the ITBCC’s recommendation of ×20, could also have an impact on the outcomes obtained.
Conclusion
Despite the limitations of the present study, it provides valuable insights into the assessment of tumor budding and its association with other prognostic parameters in invasive breast carcinoma. Our findings indicate that high tumor budding is linked to a more aggressive tumor phenotype, reinforcing its potential as a prognostic marker. Notably, tumor budding can be evaluated using routine pathological examination and standard staining techniques, highlighting its practical applicability in clinical settings. Tumor budding holds significant promise as a prognostic tool in breast cancer, with the potential to refine risk stratification, guide treatment decisions, and improve patient outcomes. Early detection of tumor budding could reduce mortality by identifying metastatic potential at an earlier stage, allowing for timely and more targeted interventions. However, before it can be integrated into routine clinical practice, further research is needed to standardize its quantification and establish universally accepted scoring criteria. The limitations of our study, including a small sample size, underscore the need for larger-scale, prospective studies to validate these associations and explore additional prognostic implications. Future research should also investigate the molecular mechanisms driving tumor budding, particularly its interactions with the tumor microenvironment and EMT pathways. Additionally, exploring potential therapeutic targets related to tumor budding may open new avenues for treatment. As our understanding of tumor budding evolves, its incorporation into breast cancer prognostication could significantly enhance personalized treatment strategies, ultimately improving patient survival and clinical outcomes.
Notes
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.
Funding
None.