Tumor-stromal ratio as a predictive marker for lymph node metastasis in patients with oral cavity squamous cell carcinoma: a cross-sectional study

Abstract

Objective

The prognostic potential of the tumor-stromal ratio (TSR) in esophageal, colon, breast, cervical, and oral squamous cell carcinomas (OSCC) has been well documented. Lymph node involvement is a critical prognostic factor in oral cavity cancers, and understanding its association with histopathological features may improve prognostic assessments. This study aimed to evaluate the association between TSR and lymph node metastasis (LNM) in patients with oral cavity squamous cell carcinoma.

Methods

A cross-sectional study was conducted on 55 patients diagnosed with oral cavity squamous cell carcinoma who underwent upfront surgical treatment. Tumor specimens were assessed for TSR, defined as the proportion of tumor cells relative to the total tumor area (tumor cells plus stroma), using routine hematoxylin and eosin (H&E)-stained slides. A TSR ≥ 50% was classified as the tumor-rich (high TSR) group, and TSR < 50% as the stroma-rich (low TSR) group. The association between TSR and lymph node metastasis was analyzed using Pearson’s chi-square test and Fisher’s exact test where appropriate.

Results

A significant association was observed between high TSR and the presence of lymph node metastasis (p < 0.001). In the high TSR group (n = 26), 20 patients (76.9%) had lymph node metastasis, compared with 7 patients (24.1%) in the low TSR group (n = 29). The positive predictive value (PPV) was 74.04% and the negative predictive value (NPV) was 78.5%. TSR was also significantly associated with tumor grade (p < 0.001), lymphovascular space invasion (LVSI) (p < 0.001), and tumor size (p = 0.002).

Conclusion

TSR, assessed on routine H&E-stained slides, is significantly associated with lymph node metastasis in oral cavity squamous cell carcinoma. It represents a simple, cost-effective histopathological parameter that may aid in predicting nodal spread and informing treatment planning. Further prospective studies with larger cohorts are warranted to validate these findings.

Introduction

Oral cavity cancers constitute a major health burden in India and represent a multifactorial disease associated with significant morbidity and mortality.

Approximately 60% of cases present at a locally advanced or advanced stage, which is associated with poor prognosis [1]. Lymph node metastasis (LNM) is a well-established adverse prognostic factor in oral cavity squamous cell carcinoma (OSCC) [2].

The tumor microenvironment of oral cavity cancers comprises both tumor cells and stromal elements. Current staging and grading systems consider only tumor-related factors and do not account for stromal characteristics, which may independently influence disease behavior.

Tumor-stromal ratio (TSR), defined as the proportion of tumor cells relative to the total tumor area comprising tumor cells and surrounding stroma, has been introduced as a valuable prognostic factor across several solid tumors [3]. Its prognostic significance has been demonstrated in esophageal, breast, colon, and cervical cancers. Despite this, the role of TSR in predicting lymph node metastasis specifically in oral cavity cancer remains incompletely characterized. This study aimed to assess the association between TSR and LNM in patients with OSCC.

Methodology

Study design and patient selection

A cross-sectional study was conducted at the Department of Surgical Oncology, Madurai Medical College, from August 2023 to August 2024. The study included 55 patients diagnosed with oral cavity squamous cell carcinoma who underwent upfront surgical treatment with informed consent. Patients with synchronous cancers, recurrent tumors, or those who received preoperative therapy (neoadjuvant chemotherapy or radiation) were excluded. The sample size of 55 was determined using a non-convenience sampling method.

Histopathological assessment of TSR

Tumor-stromal ratio (TSR) was assessed on routine hematoxylin and eosin (H&E)-stained sections of the primary resection specimen. TSR is defined as the proportion of tumor cells relative to the total tumor area (tumor cells plus stroma), expressed as a percentage. All slides were independently evaluated by two experienced pathologists, both of whom were blinded to the clinical and nodal status of the patients. In cases of discordance, a consensus was reached by joint re-examination of the slide.

The invasive tumor front of the entire section was first identified under a 4× objective lens. Fields where both tumor cells and stromal tissue were simultaneously visible were then selected and scored under a 10× objective lens. Areas where tumor cells were observed on all four sides of the microscopic field were specifically chosen for evaluation, in accordance with previously published methodology.

A TSR of ≥ 50% was classified as the tumor-rich (high TSR) group, indicating that tumor cells occupied at least half of the assessed area. A TSR of < 50% was classified as the stroma-rich (low TSR) group, indicating stromal predominance. This 50% threshold was adopted from previously published studies in colon and breast cancers, where it demonstrated maximum discriminative power [4].No digital image analysis software was used; TSR was assessed by direct microscopic examination.

Statistical analysis

Data were analyzed using IBM SPSS Statistics Version 29. Categorical variables were compared between the high TSR and low TSR groups using Pearson’s chi-square test. Fisher’s exact test was applied where expected cell counts were fewer than five. The positive predictive value (PPV) and negative predictive value (NPV) of TSR for lymph node metastasis were calculated from the 2 × 2 contingency table. A p-value of < 0.05 was considered statistically significant. No multivariate analysis was performed given the sample size; findings are therefore reported as associations.

Ethical considerations

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Ethical approval was obtained from the Institutional Ethics Committee of Madurai Medical College prior to commencement. Written informed consent was obtained from all participants.

Results

Patient demographics

The study cohort comprised 55 patients, with a male predominance of 72.7% (n = 40) and 27.3% (n = 15) females. The age of participants ranged from 27 to 85 years, with a mean age of approximately 57.6 years.

TSR distribution and association with lymph node metastasis

Based on histopathological assessment, 26 patients (47.3%) were classified as tumor-rich (high TSR ≥ 50%) and 29 patients (52.7%) as stroma-rich (low TSR < 50%). Lymph node metastasis was identified in 27 patients (49.1%), while 28 patients (50.9%) had node-negative disease.

Among the high TSR group, 20 of 26 patients (76.9%) had lymph node metastasis compared with only 7 of 29 patients (24.1%) in the low TSR group. This difference was statistically significant (p < 0.001). The PPV of high TSR for lymph node metastasis was 74.04% and the NPV was 78.5% (Table 1; Fig. 1).

Table 1.

Association of TSR with clinicopathological variables (n = 55)

Variable	High TSR (n = 26)	Low TSR (n = 29)	Total (n = 55)	p-value
Sex
Male	19 (73.1%)	21 (72.4%)	40 (72.7%)
Female	7 (26.9%)	8 (27.6%)	15 (27.3%)
Lymph Node Metastasis
Present	20 (76.9%)	7 (24.1%)	27 (49.1%)	< 0.001
Absent	6 (23.1%)	22 (75.9%)	28 (50.9%)
Lymphovascular Space Invasion
Present	16 (61.5%)	2 (6.9%)	18 (32.7%)	< 0.001
Absent	10 (38.5%)	27 (93.1%)	37 (67.3%)
Tumor Grade
Grade 1 (Well differentiated)	5 (19.2%)	21 (72.4%)	26 (47.3%)	< 0.001
Grade 2 (Moderately differentiated)	19 (73.1%)	8 (27.6%)	27 (49.1%)
Grade 3 (Poorly differentiated)	2 (7.7%)	0 (0%)	2 (3.6%)
Tumor Size
> 4 cm	14 (53.8%)	4 (13.8%)	18 (32.7%)	0.002
≤ 4 cm	12 (46.2%)	25 (86.2%)	37 (67.3%)
Mean Depth of Invasion (cm)	1.35 cm	1.63 cm	—	—

 p-values derived from Pearson’s chi-square test or Fisher’s exact test as appropriate

TSR tumor-stromal ratio, LNM lymph node metastasis, LVSI lymphovascular space invasion, DOI depth of invasion

Fig. 1.

TSR proportion. Bar chart showing the distribution of lymph node metastasis status within the high TSR (tumor-rich, n = 26) and low TSR (stroma-rich, n = 29) groups. p < 0.001 (Pearson’s chi-square test)

Association of TSR with other clinicopathological variables

TSR showed a significant association with several clinicopathological variables (Table 1). LVSI was present in 18 patients overall; 16 of these (88.9%) were in the high TSR group (p < 0.001). Regarding tumor grade, poorer differentiation was significantly associated with high TSR (p < 0.001): 19.2% of high TSR patients had grade 1 tumors, compared with 72.4% of low TSR patients. Tumors larger than 4 cm were more frequently associated with high TSR (53.8% vs. 13.8%; p = 0.002). The difference in mean depth of invasion (DOI) between the high TSR group (1.35 cm) and low TSR group (1.63 cm) did not reach statistical significance.

Subsite analysis

Subgroup analysis by tumor subsite revealed a stronger association between TSR and lymph node metastasis in the tongue (p = 0.003) and buccal mucosa (p = 0.012) compared with the alveolus and floor of mouth, where the association did not reach statistical significance (Table 2; Figs. 2 and 3). These findings should be interpreted with caution given the small numbers within each subsite group.

Table 2.

Association of TSR with lymph node metastasis by tumor subsite

Subsite	Total (n)	High TSR with LNM	Low TSR with LNM	Total LNM	p-value
Tongue	18	10	2	12 (66.7%)	0.003
Buccal Mucosa	16	6	2	8 (50.0%)	0.012
Alveolus	12	3	2	5 (41.7%)	0.210
Floor of Mouth	9	1	1	2 (22.2%)	0.480
Total	55	20	7	27 (49.1%)	< 0.001

p-values derived from Pearson’s chi-square test or Fisher’s exact test as appropriate

LNM lymph node metastasis, TSR tumor-stromal ratio

Fig. 2.

Comparison of TSR with Lymph node metastasis. Bar chart illustrating the association between TSR groups (high vs. low) and lymph node metastasis across tumor subsites (tongue, buccal mucosa, alveolus, floor of mouth). p-values for each subsite are presented in Table 2

Fig. 3.

Tumor site distribution with TSR

Discussion

The tumor microenvironment has garnered increasing attention for its role in tumor initiation, progression, and metastasis [5]. As Paget’s “seed and soil” hypothesis suggests, metastasis is not merely a result of chance; tumor cells preferentially colonize microenvironments that support their growth, including the lymphatic system [5].

TSR, as conceptualized by Mesker et al., quantifies the balance between tumor cells and surrounding stroma within the invasive tumor front, and has been shown to reflect the biological aggressiveness of several solid tumors [6]. In our study, 47.3% of patients were classified as tumor-rich (high TSR), and 76.9% of this group had lymph node metastasis, compared with 24.1% in the stroma-rich (low TSR) group, with a highly significant p-value of < 0.001. These findings are consistent with published data in head and neck cancers. Niranjan et al. demonstrated TSR as an adjunct prognostic factor for overall survival (OS) and disease-free survival (DFS) in OSCC. A systematic review and meta-analysis by Alhadi et al. confirmed the prognostic value of TSR in oral cavity cancers, correlating higher stromal content with poorer outcomes [7]. Shaui Wang et al. further reported associations between TSR and perineural invasion (PNI), pattern of invasion, nodal involvement, and worst pattern of invasion (WPOI), observations that align with our findings [8].

In contrast to some prior studies that defined a high stromal proportion as the adverse phenotype, our study defines TSR as the proportion of tumor cells, such that a high TSR (≥ 50% tumor cells) is the unfavorable group. This reflects the conceptual framework that a greater density of tumor cells at the invasive front is associated with more aggressive behavior and greater nodal spread. The terminology and directionality of TSR classification vary across the literature, and readers are directed to the specific definition used in each study when making comparisons.

In addition to lymph node metastasis, high TSR was significantly associated with poorer tumor differentiation, presence of LVSI, and larger tumor size in our cohort. These associations are biologically plausible: a tumor-rich microenvironment likely reflects greater proliferative activity and a diminished stromal barrier to lymphovascular invasion. The absence of a statistically significant difference in DOI between the two groups warrants further investigation; the mean DOI values (1.35 cm in the high TSR group and 1.63 cm in the low TSR group) did not differ significantly, suggesting that DOI and TSR may capture complementary rather than redundant information about tumor biology.

Subsite analysis revealed a stronger association between TSR and LNM in the tongue and buccal mucosa compared with the alveolus and floor of mouth. This may reflect the greater proportion of soft tissue in the tongue and buccal mucosa, facilitating more reliable TSR assessment at the invasive front. In contrast, subsites adjacent to bone (mandible and floor of mouth) may present technical challenges in identifying unambiguous invasive front fields. However, the small number of cases in each subsite group limits interpretation, and these findings should be regarded as exploratory.

The present study has several important limitations. First, this is a single-center, cross-sectional study with a relatively small sample size (n = 55), which limits statistical power, particularly for subsite analyses. Second, no multivariate analysis was performed to adjust for confounders; the associations reported here do not imply independent predictive value and should be interpreted accordingly. Third, interobserver agreement between the two pathologists was not formally quantified using a kappa statistic. Fourth, no long-term survival data were available, precluding assessment of the prognostic significance of TSR for DFS or OS in this cohort. Future prospective studies with larger, multi-center cohorts and formal interobserver reliability analysis are needed to validate these findings and to determine whether TSR retains its association with LNM after adjustment for established prognostic variables.

Conclusion

This study demonstrates a significant association between a tumor-rich TSR (≥ 50% tumor cells) and lymph node metastasis in oral cavity squamous cell carcinoma. TSR assessment on routine H&E-stained slides is simple, reproducible, and does not require additional staining or specialized equipment, making it readily applicable in standard pathological practice. These findings suggest that TSR is a potentially valuable histopathological parameter that may complement existing staging tools and inform treatment planning, particularly regarding the management of the cervical lymph nodes.

Further prospective studies involving larger patient cohorts, multivariate analysis, and extended follow-up are needed to clarify the independent contribution of TSR and to elucidate the biological mechanisms underlying the association between tumor cell predominance and lymphatic spread.

Abbreviations

TSR

Tumor-stromal ratio

OSCC

Oral squamous cell carcinoma

LNM

Lymph node metastasis

DOI

Depth of invasion

PNI

Perineural invasion

LVSI

Lymphovascular space invasion

DFS

Disease-free survival

OS

Overall survival

PPV

Positive predictive value

NPV

Negative predictive value

H&E

Hematoxylin and eosin

WPOI

Worst pattern of invasion

Authors’ contributions

All authors have their contribution in study which is mentioned already.

Funding

No financial support or funding was received for this study.

Authors’ contributions: All authors contributed to the study conception, data collection, analysis, and manuscript preparation.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Institutional Ethics Committee of Madurai Medical College. Written informed consent was obtained from all participants prior to enrolment. This study was conducted in accordance with the Declaration of Helsinki.

Competing interests

The authors declare no competing interests.