Salivary gland tumors manifest themselves in various entities, localizations and morphologies. They contribute 3 – 10% of all head and neck tumors, making them relatively rare. The estimated incidence ranges from 0.4 to 13.5 cases per 100,000 annually in the United Kingdom (1). Given their low incidence, comprehensive studies with substantial numbers of cases are rare. The World Health Organization (WHO) lists 39 salivary gland pathologies in the current 5th WHO Classification of Tumors. These are categorized into four groups: mesenchymal tumors specific to the salivary glands, malignant epithelial tumors, benign epithelial tumors and non-neoplastic epithelial lesions (2). While most tumors in major and minor salivary glands are benign, up to 35% of the lesions prove malignant, exhibiting varied distributions among the different salivary glands (1, 3). The most prevalent malignant entities are the mucoepidermoid carcinoma (MEC) and the adenoid cystic carcinoma (ACC) (4).

Grading plays a pivotal role in determining subsequent therapeutic interventions in the contemporary management of major salivary gland carcinoma (MSGC) (5, 6). Due to the diverse histogenetic, biological, prognostical and phenotypical properties of these carcinomas, a uniform grading system is lacking. Consequently, two grading strategies have emerged. One group of malignant tumors is classified due to its entity-related characteristics as low- or high-grade, respectively. For instance, acinic cell carcinomas and basal cell adenocarcinomas are consistently classified as low-grade, while lymphoepithelial carcinomas and salivary duct carcinomas fall into the high-grade category by definition. The other strategy involves grading entities like mucoepidermoid carcinoma, adenocarcinoma and adenoid cystic carcinoma based on their cytological and histological appearance (2, 7–11).

Tumor budding has gained interest across various medical disciplines in the current discourse on prognostic biomarkers of tumor entities. First described by Imai et al. in the 1950s (12), tumor budding is now well-established in colorectal cancer as an independent prognostic biomarker. Tumor buds are defined as single-cell nests of cancer cells at the invasive tumor front, with up to four cells in total (13). Tumor buds are part of the tumor microenvironment and are associated with processes of epithelial-mesenchymal transition (EMT) (14, 15). The tumor budding described at the invasive front must be distinguished from intratumoral budding (16). While there are indications of a correlation between increased tumor buds and tumor progression in head and neck squamous cell carcinoma (HNSCC) (17), the impact of tumor buds in MSGC has not been systemically explored. This study aims to analyze the influence of tumor budding in low-grade MSGC on metastasis, tumor size, recurrences, overall survival and disease-free survival. Insights into survival and recurrences patterns are clinically significant and could potentially alter therapy regimens for patients.

This retrospective exploratory study is the first to explicitly evaluate the influence of tumor budding on 5-YSR, 5-YDFSR, mean OS and DFS in MSGC. As an emerging prognostic factor in tumor diagnosis and therapy, tumor budding is gaining attention in several solid tumors. The prognostic function of tumor budding in HNSCC (head and neck squamous cell cancer) has been discussed since the 2010s (20). However, there has been no published study analyzing tumor budding as a prognostic factor for MSGC until now.

The ongoing debate on less extended surgical procedures for MSGCs presents both benefits and risks for patients (21, 22). Undoubtedly aggressive surgical procedures such as total or radical parotidectomy, neck dissections and resection of the facial nerve, may result in impairment of quality of life and a higher morbidity (21). In this study 15 (23.4%) of low-grade MSGC patients experienced recurrence, emphasizing the need for identifying risk factors to tailor specific therapy regimens. These findings suggest that there is an unmet need to improve locoregional control of MSGC. Tumor budding might serve as the prognostic factor for locoregional recurrences to improve the therapy of MSGC.

Gaining a better understanding of the locoregional recurrence of low-grade MSGC presupposes a better understanding of the multistep process of carcinogenesis. It involves invasion and metastasis, clinically important hallmarks in cancer prognosis and treatment. Tumor budding, characterized by the loss of cell-cell contact and increased mobility, is thought to play a pivotal role in these processes across various cancer types (13, 14, 23). Higher tumor budding is correlated with advanced T- and N-classifications and poorer OS and DFS in colorectal carcinoma (19), ductal carcinoma of the breast (24), cholangiocellular carcinoma (25), nasopharyngeal carcinoma (26) and oral squamous cell carcinoma (27). While tumor budding serves as an additional prognostic marker in colorectal carcinoma, its implementation in head and neck cancer, including MSGC, has not yet become established.

The underlying mechanisms of tumor budding remain incompletely understood, with recent suspicions pointing toward epithelial-mesenchymal transition (EMT) as the potentially underlying mechanism. EMT, a developmental program, is associated with increased motility, resistance to apoptosis and higher invasiveness (13). Tumor budding seems to be a dynamic process in which the ‘hybrid’ EMT phenotype of buds with downregulation of epithelial (e.g. cytokeratin) and upregulation of mesenchymal (e.g. vimentin) molecules causes further invasion into tissue (28). The mechanism of EMT is the subject of current research in salivary gland carcinomas. There is evidence that EMT is associated with higher invasiveness and therefore a worse prognosis (29, 30). Nevertheless, there is no evidence for a link between EMT and tumor budding in salivary gland cancer yet. The assumption of a more aggressive cancer and therefore higher metastasis and more frequent recurrences due to high tumor budding is currently under debate in many solid cancers.

In this study, the evaluation and reporting of tumor budding was adapted to the current recommendations of the ITBCC (19). Due to the small number of patients included in our study, the existing tumor budding score was modified into a two-group score: low tumor budding (0–4 buds/0.785 mm²) and high tumor budding (>4 buds/0.785 mm²). The analysis focused on peritumoral tumor buds (PTB) at the invasion front, aligning with classical reporting for tumor buds as recommended in colorectal carcinoma (19). Nevertheless, evidence suggests that intratumoral tumor buds (ITB) may correlate with higher T- and N-stages in colorectal carcinoma and intrahepatic cholangiocellular carcinoma, making them useful in biopsy specimens (16, 25).

The results of our study indicate that there might be a need for further locoregional control in the subgroup of low-grade MSGC with a high tumor budding score, as patients in this group showed a lower 5-YDFS (61.1% vs. 87%, p=0.021) and mean DFS. However, the current American Society of Clinical Oncology (ASCO) guideline recommends a limited local surgical approach for T1/2 low-grade parotid gland cancers, reserving total or subtotal parotidectomy for high-grade cancers or advanced local spread (T3/4) (31). Controversies also exist regarding neck dissection in cN0 patients (21, 22). These guidelines show a trend toward surgical de-escalation of MSGC treatment. In order to avoid undertreatment, identifying potential risk factors beyond the established ones becomes crucial for further individualization of oncological management.

The parotid gland emerged as the most common location for MSGC in the current study, consistent with prior reports (25). Furthermore, our analysis demonstrated a significant decrease in OS for higher UICC stages, aligning with findings of other research groups (32–35). Nodal metastasis significantly lowers 5-year survival, as reported by Meyer and colleagues (35). This correlates with our data, showing mean OS of 45.5 and 132.4 months in nodal positive vs. nodal negative patients. At the current state, poor outcome of patients with MSGC is strongly associated with tumor stage, grading, lymph node metastasis and age (36). However, there is still a need for identification of further risk factors to achieve a better prognosis and differentiation in such patients.

Tumor grading has already been established as a prognostic factor and therefore directly affects the therapy regimen of MSGC. High-grade MSGC typically entails surgical resection, elective neck dissection and adjuvant radiation therapy (RT). Conversely, low-grade MSGC without nodal metastasis often achieves excellent OS and DFS with resection alone (31, 33). Elective neck dissection seems to be favorable for patients with T3/4 tumors and all high-grade MSGC (31, 34). However, our data indicate that there are also low-grade MSGCs with high tumor budding that lack locoregional control, as evidenced by the lower 5-YDFSR. The publication by Armstrong et al. demonstrated an increased occult lymph node metastasis in high-grade carcinoma, leading to the need for elective neck dissection to improve OS and DFS (37). The rate of occult metastasis differs in the literature. The Ketterer at al. study from 2019 did not find any occult metastases in cN0 staged MSGC (38). Concordant with decision-making for elective neck dissection, adjuvant RT of the ipsilateral neck is not routinely performed in low-grade MSGC (39–41). The data of both Ali et al. and Park et al. indicate a need of further local control in low-grade MSGC patients with locally advanced carcinomas, pathological risk factors and nodal metastasis (34, 42).

The findings in the current study show a significantly higher rate of recurrence and more lymph node metastasis in low-grade MSGC patients with high tumor budding. Therefore high tumor budding might add another level besides grading, tumor size and locoregional metastasis to take into consideration while planning the therapy of MSGC. Recurrences occurred in 19.6% of the low tumor budding group compared to 33.3% of the high tumor budding group. Furthermore, Kaplan-Meier analysis of mean DFS revealed 100.2 months in the low tumor budding score group and 58.7 months in the high tumor budding group (log-rank test p=0.032). Tumor budding correlated with UICC-stage (p=0.021), N-stage (p=0.019), lymphangioinvasion (p=0.022) and perineural invasion (p=0.03). These observations suggest a potentially more aggressive tumor infiltration, adding another feature to the considerations for patients with MSGC.

In 2019 Nakaguro et al. proposed a novel histologic risk stratification for salivary duct carcinoma. The risk stratification includes pathological factors such as prominent nuclear pleomorphism, mitosis ≥30/10 HPF, vascular invasion, high tumor budding and poorly differentiated clusters. The retrospective analysis demonstrated a significant impact of tumor budding on OS. DFS was not examined in their study (43). The current study, however, did not find a significant difference in 5-YSR or mean OS for patients with high tumor budding scores. The difference in findings may be attributed to variations in the number of patients (151 vs. 64) and the specific entity of salivary gland carcinoma included in the study by Nakaguro. Furthermore Nakaguro et al. used a different cut-off with 10 tumor buds/0.785 mm², not the 4 tumor buds/0.785 mm² that were used in our study. However, our findings indicate that tumor budding may be a factor of high interest in the treatment strategy for low-grade MSGC. It is associated with a worse 5-YDFSR, shorter mean DFS and correlates with higher N-stages, UICC-stages, lymphangioinvasion and perineural invasion. It might serve as a pathological risk factor, influencing decisions about neck dissection and adjuvant RT.

The presented study has limitations including the retrospective character, which generates a potential selection bias and the relatively low number of included patients. Being a single-center study, the results are noted to be vulnerable to the specific surgical and medical skills at that particular clinic. The lack of differentiation between entities of MSGC is acknowledged as a limitation, potentially leading to false implications for some specific entities. Despite these limitations, the explorative nature of the study is emphasized, as it is the first to analyze the impact of tumor budding on MSGC in terms of 5-YSR, 5-YDFSR mean OS, mean DFS, tumor size, nodal status, and recurrence. The authors stress the need for further studies in larger patient collectives with differentiation of tumor entities to fully understand the influence of tumor budding and its potential as a prognostic factor in the treatment of MSGC.

Current trends in oncological treatment of MSGC favor a less invasive surgical approach in terms of primary resection of mostly parotideal carcinomas. Elective neck dissection is not carried out in cN0 staged patients unless risk factors are detected during staging or histopathological examination. These risk factors include locally advanced tumors (T3/4), high-grade entities and lymphangioinvasion. Recognizing these risk factors leads to further personalized oncological treatment and is crucial to prevent undertreatment and recurrences. The findings support the hypothesis of tumor budding as a prognostic factor in MSGC, urging further exploration for enhanced therapy of patients to prevent recurrences and improve overall survival.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving humans were approved by Ethik-Kommission der Albert-Ludwigs-Universität Freiburg. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

VB: Conceptualization, Data curation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. GK: Conceptualization, Investigation, Methodology, Writing – original draft, Writing – review & editing. TV: Conceptualization, Data curation, Investigation, Writing – original draft, Writing – review & editing. CB: Conceptualization, Investigation, Supervision, Writing – original draft, Writing – review & editing.