Tumor budding in gastric adenocarcinoma: prognostic value and association with clinicopathological markers.

INTRODUCTION
Gastric cancer (GC) is the fifth most common cancer worldwide and the fourth leading cause of cancer-related deaths
9
. In Tunisia, it has a five-year prevalence estimated at 7 cases per 100,000 inhabitants and a cumulative risk of mortality estimated at 0.34
6,8
. Despite advancements in diagnostic tools and treatment options, GC is frequently diagnosed at advanced stages, resulting in a poor prognosis and a five-year survival rate of less than 30%
20
.
The role of the pathologist is crucial in managing GC by identifying histo-prognostic factors that contribute to treatment decisions. Among these factors, markers of epithelial-mesenchymal transition, an important process in tumor progression, represent a new challenge for pathologists. One specific marker, known as tumor budding (TB), is characterized by the presence of isolated cells or clusters containing fewer than five carcinoma cells at the invasive front of the tumor. The analysis of TB and its prognostic value has been the subject of several studies, focusing on various cancers, particularly esophageal and colorectal cancers
17,22,27,26
.
In most of these studies, TB was associated with a poor prognosis, significantly reducing both overall and relapse-free survival. However, the results are inconsistent in GC, and this parameter is not included in the prognostic classifications, nor is it present in standardized pathological reports
10,14
.
In this study, we aimed to evaluate TB in gastric adenocarcinoma (GA) and to assess its prognostic value through survival analysis.
The objective is to explore the associations between TB and commonly recognized clinicopathological markers that are considered prognostic factors for this type of cancer.

METHODS
This retrospective study comprised a period of ten years, from January 2008 to December 2017. All patients who underwent surgery for GA at the Mongi Slim Hospital, in Tunisia, were included. The focus of the study lies on various clinicopathological factors, endoscopic features, tumor staging, and treatment details. Patients whose tumor sample slides were lost or whose paraffin blocks were depleted were excluded from the study. The pathological characteristics of the tumors were assessed using the Lauren classification and the WHO classifications from 2010 and 2019
15,16
. The pathological stages (pTNM) were determined according to the 2017 criteria set by the Union for International Cancer Control (UICC) and the American Joint Committee on Cancer (AJCC)
11,15
. Additionally, the presence of perineural invasion, vascular emboli, and the patient’s response to chemotherapy, evaluated through the Mandard Tumor Regression Grade (TRG), were documented
19
.

Tumor budding assessment
For TB assessment, histological slides from patients with GA were reexamined by a single pathologist (DB), focusing on the invasive fronts of the tumors. These slides were stained with hematoxylin and eosin (HE).
TB was defined as the presence of isolated cells or small clusters of fewer than five cells at the invasive front of the tumor, as commonly used in the scientific literature
30
. The evaluation followed the 2016 consensus guidelines for colorectal cancer, grading TB as follows: Bd1 (0–4 buds), Bd2 (5–9 buds), and Bd3 (≥10 buds). Furthermore, a two-grade classification was employed, categorizing budding as low-grade (<10 buds) and high-grade (≥10 buds)
18
.

Statistical analysis
For the statistical analysis, the Statistical Package for the Social Sciences (SPSS) statistical software was used. The confidence interval was set at 95% (95%CI). Continuous quantitative variables were represented by mean and standard deviation, while qualitative variables were expressed as percentages. To compare means, the Student’s t-test was employed. Associations between TB and specific clinicopathological factors were evaluated using Pearson’s χ² test. A correction was applied using Fisher’s exact test when the sample size was less than five. To identify risk factors that were independently associated with TB, a multivariate analysis using Cox regression, following a stepwise descending approach, was carried out. This analysis included all factors that had a significance level of p< 0.2 in the univariate analysis. For follow-up and survival analysis, survival curves for overall survival (OS) and recurrence-free survival (RFS) were created using the Kaplan-Meier method, and these curves were compared using the log-rank test. The study was approved by the Ethics Research Committee of the Institution (number 24/082).

RESULTS
Our research involved 68 patients who had undergone surgery for GA. The median age of the patients was 61.34 years (±13.58), with the highest incidence occurring between the ages of 61 and 70 years. The cohort consisted of 49 men (72%) and 19 women (28%), resulting in a male-to-female ratio of 2.57. The postoperative course was uneventful for 60 patients (88%), while eight cases (12%) required revisional surgery.
The clinicopathological features and patient management details are summarized in Table 1. The mean tumor size was 58.43±24.20 mm, with sizes ranging from 10 to 130 mm. In terms of TNM staging, 31 patients (45%) were classified as pT3, and 22 patients (32%), as pT4. Furthermore, 42 patients had lymph node metastasis, and 11 patients (16%) presented with synchronous distant metastasis.
The TB was assessed as follows: Bd1 in 35 patients (51%) (Figures 1 and 2), Bd2 in 12 patients (18%), and Bd3 in 21 patients (31%).
For the two-grade classification, tumors were classified as low-grade in 69% of cases and high-grade in 31% of cases. Local recurrence was observed in 12 patients (18%) within an average of 17.17±14.87 months, while metastatic recurrences occurred in 15 patients (22%) within an average of 21.93±12.2 months. In our study, 32 patients (47%) died during the follow-up period, with a time frame ranging from 0 to 48 months post-surgery.

Association between tumor budding and survival
The average OS for patients with high-grade TB was significantly lower than that of patients with low-grade TB (29.19 months vs 47.87 months; p=0.007, p<0.05) (Figure 3).
Similarly, the mean RFS for patients with high-grade TB was also lower than that for patients with low-grade TB, showing a significant difference (21.04 months vs. 35.42 months; p=0.004, p<0.05) (Figure 4).
According to the multivariate analysis, high-grade TB was an independent prognostic factor for RFS (p=0.005, p<0.05).

Association between tumor budding and clinicopathological features
The results of our univariate analysis are summarized in Table 2. In our study, we found a correlation between high-grade TB and several factors: poorly-differentiated tumors (p=0.035, p<0.05), vascular invasion (p=0.06, p>0.05), perineural invasion (p=0.038, p<0.05), deep parietal infiltration (pT3–pT4) (p=0.003, p<0.05), the occurrence of metastatic recurrence (p=0.033, p<0.05), nodal staging pN3 (p=0.005, p<0.05), synchronous distal metastasis (p=0.014, p<0.05), and stages III and IV (p=0.014, p<0.05). Additionally, high-grade TB was associated with metastatic recurrence (p=0.0033, p<0.05) and mortality (p=0.016, p<0.05). In the multivariate analysis, high-grade TB was significantly correlated with perineural invasion (odds ratio [OR] 2.98, 95%CI 1.04–8.53, p=0.004, p<0.05), advanced tumor stages (III–IV) (OR 4.04, 95%CI 1.27–12.83, p=0.01, p<0.05), and mortality (OR 3.65, 95%CI 1.24–10.74, p=0.02, p<0.05), as shown in Table 3.

DISCUSSION
According to our findings, OS and RFS were significantly lower in patients with high-grade TB compared to those with low-grade TB, with p-values of 0.007 and 0.004, respectively. High-grade TB was identified as an independent prognostic factor for RFS (p=0.005, p<0.05). Furthermore, high-grade TB were independently associated with poor histopathological factors, including the presence of perineural invasion (p=0.004, p<0.05), advanced tumor stages (III–IV) (p=0.01, p<0.05), and increased mortality (p=0.025, p<0.05).
These results led us to investigate the impact of TB in GA, drawing on existing literature.
The presence of TB has become a significant factor in predicting the progression of colorectal cancer (CRC), especially in stage II. Patients in this stage could benefit from new prognostic factors that help classify them as either low or high risk for recurrence, which has therapeutic implications
27
.
In 2016, the UICC recommended including TB as a criterion for adjuvant treatment
2,33
. Since then, authors of several studies have sought to demonstrate the prognostic value of TB in other types of cancer, including GA
31
. It has been the focus of various studies. The findings are summarized in Table 4.
Guo et al., in a meta-analysis, demonstrated the impact of high-grade TB on survival. This analysis involved seven stud ies with a total of 2,178 patients. The authors showed that high-grade TB predicted a poor 5-year OS with OR of 1.79 (95%CI 1.53–2.05, p<0.01) for patients with GA
27
. According to the combined OR values, high-grade tumor burden was significantly associated with several factors: tumor stage (OR 6.63, 95%CI 4.01–10.98, p<0.01), tumor differentiation (OR 3.74, 95%CI 2.68–5.22, p<0.01), lymphovascular invasion (OR 7.85, 95%CI 5.04–12.21, p<0.01), and lymph node metastasis (OR 5.75, 95%CI 3.20–10.32, p<0.01)
27
.
Furthermore, Ulase et al., in a study including 456 patients, verified that those with high-grade TB experienced significantly reduced 5-year OS compared to those with low-grade TB (p<0.001)
31
. Similar findings were reported in another study by Du et al., involving 621 patients (p<0.001)
4
.
Authors of several studies have identified an association between high-grade TB and poor tumor differentiation
1,3,13,25,28-30
. According to other findings, there are correlations with the presence of vascular and perineural invasion
5,12,23,24,29,31
, while some researchers reported associations with older age
25
and larger tumor size
28,32
. However, these latter findings were not observed in our study.
Variations in the literature results are primarily attributed to the significant differences in methods used to assess and quantify TB. Unlike CRC, there is currently no standardized method for evaluating this parameter in GA
4,23,25,30
. Several evaluation systems have been proposed, including one by Hase et al., which uses a binary classification based on the predominant morphological characteristics: either absent/minimal budding or moderate/marked budding
4
. Another system, developed by Nakamura et al., classifies TB into four scores based on the surface of the tumor’s invasive front
22
. In addition, quantitative systems, such as the one proposed by Ueno et al., focus on counting tumor buds in "hot spot" areas identified through low-magnification HE slides
30
. In our series, we employed the method validated in CRC, which distinguishes three grades based on the number of buds present.
It is essential to consider the reproducibility of these methods, as significant variability has been demonstrated among different evaluators, even when using the same technique
33
.
An immunohistochemistry (IHC) study using a pan-Cytokeratin antibody and image analysis (IA) technology can effectively identify carcinoma cells at the invasive front and aid in evaluating TB
25
.
However, it is worth noting that IHC involves additional costs in terms of both money and time. Assessing TB in GA can be particularly challenging in the presence of significant post-therapeutic fibrosis or when the distribution of this fibrosis is heterogeneous
25
. In such situations, distinguishing the invasive front between tumor tissue and adjacent healthy tissue from the junction of tumor remnants and post-therapeutic fibrosis may be difficult. In all cases, it is crucial to perform extensive sampling of the tumor along with adjacent tissue to analyze as many invasive-front areas as possible.
Our study has several limitations. It is a retrospective analysis based on available histological samples. During our investigation, we conducted a "prospective" re-evaluation of tumor slides to assess TB, as this parameter was not included in the routine histological reports. We chose not to use additional techniques, such as special staining, IHC, or IA. Instead, we adhered to the consensus for evaluating TB in CRC, which solely relies on histological examination of HE-stained slides.

CONCLUSIONS
We have shown that the prognostic and predictive value of TB in GA is significant, particularly regarding patient survival. Using a simple and cost-effective technique with relatively quick analysis, we can stratify patients with GA based on their prognosis. To further validate this parameter, larger prospective studies are needed to explore its potential therapeutic implications for increasingly personalized treatment approaches in GA. With these findings, we highlight the need for a standardized method to assess TB in GA, allowing for its inclusion in prognostic classifications and standardized reports of resected specimens of this cancer.