Metastatic neck involvement in oral squamous cell carcinoma: A retrospective clinicopathological study

Abstract

Oral squamous cell carcinoma (OSCC) is the most common oral malignancy, with neck metastasis significantly affecting prognosis and treatment. This study examines clinicopathological factors associated with neck metastasis in a Middle Eastern cohort of OSCC patients. A retrospective analysis of 84 OSCC patients treated at King Hussein Cancer Center (2018-2024) was conducted. Clinicopathological factors, including tumor subsite, T stage, depth of invasion, differentiation, perineural invasion, and lymphovascular invasion were evaluated for their association with lymph node metastasis. Pathological neck involvement was observed in 47.6% of cases. Late T stage (T3-T4) and LVI positivity were identified as independent predictors of nodal metastasis (aOR = 4.25, p = 0.009; aOR = 4.39, p = 0.027, respectively). These findings highlight the importance of comprehensive pathological assessment in OSCC.

Introduction

Oral squamous cell carcinoma (OSCC), a subtype of head and neck cancer, is the sixth most common malignancy worldwide, with approximately 300,000 new cases diagnosed annually, predominantly in Asian countries (Tanaka et al. 2011). Research on OSCC incidence in the Middle East is limited, though the high prevalence of its risk factors in the region might be linked to high incidence rates in the future.

The primary risk factors for OSCC include tobacco and alcohol consumption, infectious agents such as Candida and viruses, and prolonged sun exposure (particularly in lip squamous cell carcinoma) (Tanaka et al. 2011).

This association is particularly relevant in Jordan, which the WHO global report on trends in prevalence of tobacco use 2000–2030 identifies as one of six countries worldwide where tobacco use is still rising. According to this report’s 2022 estimates, 45.7% of Jordanian males aged 15 years and older are current cigarette smokers – an age-standardised prevalence, underscoring the population’s elevated risk of OSCC (WHO 2024).

Tobacco and alcohol consumption remain the main risk factors for OSCC in Europe. Betel quid chewing is also an established risk in Southeast Asia countries (Nokovitch et al. 2023).

Neck metastasis is a key factor in OSCC prognosis and treatment planning, as it significantly impacts staging and survival outcomes. Early detection and appropriate management of cervical lymph node (CLN) involvement are crucial for improving prognosis and reducing disease recurrence (Pantvaidya et al. 2020; Mishra et al. 2016; Al-Rajhi et al. 2000). Various imaging modalities (Ultrasonography (US), Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and Ultrasound-guided Fine Needle Aspiration Biopsy (US-FNAB)) assist in assessing nodal involvement before treatment initiation (Khunteta et al. 2022).

While elective neck dissection (END) remains a topic of ongoing discussion in specific clinical scenarios, its role is becoming more defined within evidence-based clinical guidelines (D'Cruz et al. 2009). Many organizations, such as American society of clinical oncology (ASCO) consider prophylactic neck dissection the gold standard in the management of the neck in patients with squamous cell carcinoma of the oral cavity (Koyfman et al. 2019). It has also been stated that elective neck dissection not only improves regional control in patients with adequate primary resection margins, but also determines and guides adjuvant therapy decisions (Lin et al. 2011).

Elective neck dissection had proved to result in higher rates of overall and disease-free survival than did therapeutic neck dissection approach as cervical lymph nodes remain the most important site of relapse in patients in whom neck dissection is not performed at the time of primary surgery (D’Cruz et al. 2015a).

The anatomical subsite of OSCC influences tumour behaviour and metastatic patterns. The tongue is the most common site, accounting for approximately 40% of cases, followed by the floor of the mouth, while other sites such as the gingiva, buccal mucosa, labial mucosa, and hard palate are less frequently involved (Tanaka et al. 2011; Pantvaidya et al. 2020; Mishra et al. 2016; Al-Rajhi et al. 2000; Khunteta et al. 2022; Arain et al. 2020). Tumour size, depth of invasion (DOI), and histological differentiation are critical predictors of lymphatic spread, with larger and poorly differentiated tumours demonstrating a higher risk of metastasis (Tanaka et al. 2011; Khunteta et al. 2022; Arain et al. 2020; Massey et al. 2019; Shear et al. 1976). In oral tongue SCC, a tumour thickness greater than 4 mm is associated with an increased risk of nodal metastasis (Arain et al. 2020).

Additionally, perineural invasion (PNI) and lymphovascular invasion (LVI) are independent risk factors for cervical nodal metastasis (Fagan et al. 1998); Huang et al. 2021a). The presence of extranodal extension (ENE) further reflects aggressive tumour behaviour and increased likelihood of distant metastasis (Huang et al. 2021b).

This study aimed to analyse the patterns of neck metastasis (pN +) in OSCC and their correlation with primary tumour characteristics, including subsite, T stage, DOI, histological differentiation, PNI, and LVI. While these factors are well-studied in Western and Asian populations, there are significant gaps in their applicability to Middle Eastern cohorts (Al-Jaber et al. 2016; Kujan et al. 2017).

This includes – but is not limited to—regional variations in risk profiles, diagnostic delays, and potential differences in metastatic patterns. By addressing this gap, this research provides valuable insight that can guide clinical decision-making and improve OSCC management in the region.

Methodology

This study acquired KHCC institutional review board (IRB) approval. Study IRB number: 24 KHCC 64.

We conducted a retrospective study of all OSCC patients treated at King Hussein Cancer Center (KHCC) (Amman, Jordan) between 2018 and 2024 who met the following criteria: (1) surgically managed OSCC cases that underwent neck dissection as part of their treatment, and (2) complete clinicopathological records. Cases were included for analysis with no limitations on age, gender, or medical status. Patients assessed as ‘beyond surgery’, who were referred to either chemo/radiotherapy or palliative care were excluded from this study.

Data collection focused on: Demographics (age and gender), tumour characteristics (primary tumour subsite, T stage), histopathology (Depth of invasion (DOI), extra-nodal extension (ENE) histological differentiation, perineural invasion (PNI), lymphovascular invasion (LVI)) and nodal assessment (Number of lymph nodes removed, anatomical distribution of metastases (for pN + cases)).

Primary outcomes were the rates of pathological nodal involvement (pN +) and their correlations with primary tumor subsite and histopathological features. Secondary outcomes included the false-negative rate of clinical neck staging (cN0 vs. pN +). All statistical analyses were conducted using RStudio (Version 2024.12.0.467, Posit Software, PBC) (R Core Team 2024).

Results

Study population and demographics

114 patients diagnosed with OSCC patients were assessed for this study (Table 1). A total of 84 cases met all the inclusion criteria and none of the exclusion criteria. The mean age of participants was 54.8 years (SD: 14.5), with one paediatric case (11 years old). The majority of patients were male (59 cases, 70.24%). Among the 84 patients, 46 were clinically neck-positive (cN +), while 38 were clinically neck-negative (cN0). However, 7 of the 38 clinically negative cases (18.4%) were found to have pN +, indicating a false-negative rate of 18.4%.

Table 1.

Pathological neck involvement (pN +) in OCSCC based on anatomical subsite, tumour stage, neck level, differentiation, PNI, and LVI

Factor		Number of cases	Number of pN + cases	Percentage of pN + cases
Anatomical subsite	Maxilla	12	3	0.25
	Buccal	8	4	0.50
	Tongue	37	17	0.459
	Retromolar	8	5	0.625
	Mandible/Alveoli	8	2	0.25
	Floor of the mouth	6	3	0.50
	Lips	5	1	0.20
Stage	Early (T1, T2)	31	7	0.226
	Late (T3, T4)	53	33	0.623
Neck level	1A	-	8	0.20
	1B	-	27	0.675
	2A	-	26	0.65
	2B	-	5	0.125
	3	-	13	0.325
	4	-	2	0.05
	5	-	1	0.025
Differentiation	Well	7	0	0.00
	Moderate	64	30	0.469
	Poor	13	10	0.769
PNI	PNI +	20	14	0.70
	PNI-	64	26	0.406
LVI	LVI +	17	13	0.765
	LVI-	67	27	0.403

Tumour characteristics

The tongue was the most common primary tumour site (38 cases, 45.24%), followed by the maxilla (12 cases, 14.29%), while the lips were the least common (5 cases, 5.95%). Most cases presented at an advanced stage, with 53 patients (63.1%) diagnosed at T3 or T4, whereas only 31 patients (36.9%) had early-stage disease (T1 or T2).

Analysis of neck metastasis (pN +)

Pathological neck involvement (pN +) was observed in 40 of the 84 cases (47.62%). To assess associations between clinicopathological variables and neck involvement status statistical analyses included Fisher's exact tests for categorical variables (T stage, tumour subsite, differentiation grade, PNI, LVI), Welch's t-tests for continuous variables (DOI), and multivariate logistic regression to control for confounders. All tests were two-tailed with α = 0.05. Analyses were performed using RStudio (version 4.4.2).

T Stage

pN + was more frequent in late-stage (T3-T4) tumours (33/53, 62.3%) compared to early-stage (T1-T2) tumours (7/31, 22.6%). This shows that advanced T stage can be associated with a three more times higher rate of neck metastasis compared to early-stage tumours. Fisher’s exact test confirmed a statistically significant difference in pN + rates between early and late-stage cases (p < 0.001, 95% CI: 0.06–0.53). The odds of neck involvement were significantly lower in early-stage cancer (OR = 0.18, 95% CI: 0.06–0.53). [Fig. 1: Predicted Probability of Nodal metastasis based on T stage].

Fig. 1.

Predicted probability of nodal metastasis based on T stage

Tumour subsite

Neck metastasis was most common in retromolar OCSCC (87.5%) and least common in maxillary tumours (25.0%). Nodal metastasis was observed in 3 of 6 (50%) floor-of-mouth tumors. Fisher’s exact test confirmed a significant association between primary tumour subsite and nodal involvement (p = 0.035).

Neck levels

Among pN+ patients, the most frequently involved levels were 1B (67.5%) and 2A (65.0%), while level 5 was the least affected (2.5%). Only 5 cases (5.95%) had metastasis to level 2B, originating from the tongue (2), buccal mucosa (2), and mandible (1).

Differentiation grade

Among the 13 poorly differentiated tumours, 10 cases (76.92%) had nodal metastasis, whereas 30 of the 64 moderately differentiated tumours (46.88%) were pN +. None of the well-differentiated tumours exhibited nodal involvement. Tumour differentiation can be a strong predictor of nodal involvement. Studying the difference between moderately and poorly differentiated tumours using Fisher’s exact test shows that poorly differentiated tumours had 3.72 higher odds of neck metastasis than moderately differentiated tumours (p = 0.068, 95% CI: 0.85–22.97). The wide range of confidence interval could be attributed to the small sample size (13 cases of poorly differentiated tumours) and p value > 0.05 suggests a not statistically significant result, but a strong trend towards higher chance of metastasis in poorly differentiated OSCC.

PNI and LVI

PNI and LVI were both significantly associated with nodal metastasis. Neck involvement was observed in 70.0% of PNI-positive cases and 76.47% of LVI-positive cases.

Univariate analysis of PNI and LVI shows both factors being significantly associated with nodal metastasis. PNI + tumours have a 3.41 times higher chance of neck metastasis compared to PNI- tumours (Fisher’s exact test, p = 0.039, 95% CI: 1.15–10.03). LVI + tumours have a 4.81 times higher chance of neck metastasis compared to LVI- tumours (Fisher’s exact test, p = 0.013, 95% CI: 1.42–16.35).

Multivariate analysis adjusting for both factors shows that LVI and PNI are independent factors (Pr( >|z|) = 0.019 and 0.045, consecutively). LVI remained a stronger independent predictor than PNI after adjustment.

DOI and ENE

The mean depth of invasion was greater in pN + cases (14.05 mm, SD: 5.03) compared to pN- cases (10.90 mm, SD: 6.83) with statistical analysis, as a continuous variable, confirming a significant difference between the two means (Welch two-sample t-test (p = 0.038, 95% CI: 0.19–6.10)) [Fig. 2: DOI by Nodal Metastasis Status].

Fig. 2.

DOI by nodal metastasis status

ENE, indicative of aggressive tumour behaviour, was present in 16 of the 40 pN + cases (40%).

To account for potential confounders, a multivariate logistic regression model was performed with pN + status as the dependent variable and PNI, LVI, T stage (early: T1 and T2 vs late: T3 and T4) and differentiation (well/moderate vs. poor) as independent variables.

The analysis identified late T stage (T3 and T4) (aOR = 4.25, 95% CI: 1.43–12.67; p = 0.009) and LVI + (aOR = 4.39, 95% CI: 1.18–16.35; p = 0.027) as independent predictors of nodal metastasis. Well/moderately differentiated tumours showed nonsignificant trend towards lower risk compared to poorly differentiated tumours (aOR = 0.26, p = 0.074).

Well/moderate differentiation shows a protective trend (74% lower odds) but needs larger samples to confirm.

Discussion

This study assessed the clinicopathological factors associated with neck metastasis in OSCC within a Middle Eastern population, particularly in Jordan, where smokeless tobacco and cigarette smoking remain highly prevalent risk factors (Al-Jaber et al. 2016). Our findings align with prior research, reinforcing the significant burden of nodal metastasis in OSCC and emphasizing the importance of identifying high-risk features for optimal disease management.

Tumour location played a key role in predicting neck involvement, with the tongue and retromolar trigone demonstrating higher metastatic potential. These findings are consistent with existing literature highlighting the aggressive behaviour of tumours arising in these regions (Vidya et al. 2022; Demir et al. 2020). The retro molar trigone, in particular, is known for its high rates of nodal metastasis, likely due to its rich lymphatic drainage and close proximity to multiple lymph node levels (Faisal et al. 2017).

Patterns of lymph node involvement in our cohort further support prior studies emphasizing the predominance of metastasis in levels 1 and 2 (Zhang et al. 2021). Notably, metastasis to level 2B was infrequent, suggesting that routine dissection of this level may be unnecessary; however, subgroup analysis by clinical nodal status was limited. This aligns with findings from Elsheikh et al. (7.3%) and Chakraborty et al. (3.2%), reinforcing the idea that selective dissection of level 2B should be considered only in cases with advanced nodal disease or ENE, particularly in tumours originating from high-risk subsites such as the tongue and floor of the mouth (Elsheikh et al. 2008; Chakraborty et al. 2019).

Our study revealed a strong association between advanced T-stage and neck metastasis. However, evidence suggests that elective neck dissection improves survival even in early-stage OSCC due to the high risk of occult metastasis (D’Cruz et al. 2015b). Histopathological features, particularly LVI and PNI, were associated with nodal metastasis. Poorly differentiated tumours exhibited a trend toward higher metastatic potential compared to well-differentiated counterparts, supporting prior findings that tumour grade may be an important prognostic factor influencing recurrence and overall survival (Alqutub et al. 2024; Thomas et al. 2014). Michcik et al. found that increasing tumour grade correlates with a higher likelihood of ENE, LVI, PNI, nodal metastases, and recurrence (Michcik et al. 2023). The presence of LVI and PNI further underscores their role as critical indicators of disease aggressiveness, facilitating both regional and distant spread (Huang et al. 2023; Ting et al. 2021).

In this study, a significant association was observed between DOI and nodal metastasis, reinforcing its role in risk stratification and treatment planning. However, it was not included in the multivariate regression model due to sample size limitations and to avoid model overfitting. According to Edri et al., assessing DOI as a continuous variable may improve prognosis accuracy and guide better surgical and adjuvant therapy decisions (Edri et al. 2025) ENE is also a recognized marker of tumour aggressiveness, linked to poor outcomes and incorporated into the latest AJCC staging for risk stratification and treatment planning (Ghorbanpour et al. 2024). Its varying degrees may carry distinct prognostic significance, emphasizing the need for standardized assessment. The growing role of radiologic ENE (rENE) in preoperative evaluation further highlights the importance of consistent diagnostic criteria (Huang et al. 2021b).

In our cohort, 18.4% of clinically node-negative (cN0) patients were found to have occult nodal metastases, underscoring the limitations of current imaging in detecting subclinical disease. This aligns with findings from D’Cruz et al., who reported a 29.6% rate of occult metastasis in early-stage OSCC and identified depth of invasion as a key predictor (D’Cruz et al. 2015b). Their randomized trial demonstrated a significant survival benefit with elective neck dissection over therapeutic dissection, with 3-year overall and disease-free survival rates of 80.0% vs. 67.5% and 69.5% vs. 45.9%, respectively. Given that CT, MRI, and PET may miss micrometastases (Thoenissen et al. 2023), especially in high-risk early-stage tumours, our findings support elective neck dissection in cN0 patients with adverse features such as DOI > 4 mm, poor differentiation, LVI, or PNI. This also highlights the need for improved imaging criteria and adjunctive tools like sentinel node biopsy to reduce false-negative rates (Hiraki et al. 2016).

Additionally, the UK-based SEND randomized trial confirmed that patients with T1/T2 N0 oral cancers undergoing upfront END had improved disease-free survival, even in those with small tumours. The study found a DFS hazard ratio of 0.66 (p = 0.04) on intention-to-treat analysis and 0.56 (p = 0.007) per protocol (Hutchison et al. 2019). Furthermore, a meta-analysis by Fasunla et al. including four randomized trials with 283 patients found that END significantly reduced disease-specific death in oral cancer patients with a clinical N0 neck. The relative risk of disease-specific death was 0.57 (p = 0.014) using a fixed-effects model, and 0.59 (p = 0.034) using a random-effects model, further supporting the role of END in this population (Fasunla et al. 2011). This growing body of high-level evidence strengthens the rationale for recommending elective neck dissection even in early-stage tumours with clinically negative necks.

This study has certain limitations, including its retrospective design and relatively small sample size, which may affect its generalizability. However, its well-defined inclusion criteria and meticulous data collection enhance the reliability of the findings. The results contribute to the growing body of evidence on OSCC management, particularly in identifying high-risk factors for nodal metastasis. Future research should focus on larger, prospective cohorts to refine risk stratification models and explore personalized treatment approaches aimed at reducing nodal metastases and improving patient outcomes.

Conclusion

This study highlights key predictors of neck metastasis in OSCC, emphasizing the role of tumour location, T stage, and histopathological features such as poor differentiation, ENE, LVI, and PNI. These findings reinforce the need for thorough pathological assessment to guide optimal treatment planning. Further research with larger cohorts and prospective studies is essential to validate these findings and develop targeted interventions to reduce the burden of nodal metastases in OSCC.

Author contribution

Medyan Al Rousan: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing. Ra'ad Alsunna: Conceptualization, Data curation. Lamees Al-Khalili: Data curation, Formal analysis, Visualization, Writing – original draft, Writing – review and editing. Jowan Al-Nusair: Validation, Visualization, Writing – original draft, Writing – review and editing. Raya Manasrah: Investigation, Resources. Sumaiah H. AlNaimat: Formal analysis. Raed Dwaik: Data curation. Lina Alelaumi: Writing – review and editing. Jareer Shatnawi: Investigation.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data that support the findings of this study are available on request from the corresponding author, Al-Khalili, Lamees. The data are not publicly available due to their containing information that could compromise the privacy of research participants.

Declarations

Ethics approval

This is an observational study. It is confirmed that no ethical approval is required.

Consent for publication

Informed consent for publication was not needed.

Consent to participate

This is an observational study. No consent to participate was needed to be taken.

Competing interest

The authors have no competing interests to declare that are relevant to the content of this article.