Single-Tracer Methylene Blue-Guided Sentinel Lymph Node Biopsy in Early-Stage Squamous Cell Carcinoma of the Buccal Mucosa: A Prospective Study

Abstract

Background

Subsite and tumor stage-related heterogeneity in studies on optical tracer-guided sentinel lymph node biopsy (SLNB) in oral squamous cell carcinoma (SCC) has led to inconsistent results, limiting the applicability and wider adoption of this technique.

Methodology

This single-center, prospective validation study conducted in January and February 2022, included 29 consecutive patients with early-stage, node-negative SCC of the buccal mucosa undergoing methylene blue-guided SLNB followed by modified radical neck dissection (considered the reference standard) to determine the identification rate and test validity of SLNB, and secondarily, the clinicopathological factors associated with node-positivity using binary logistic regression.

Results

SLN identification rate with methylene blue was 93.1% with 66 SLNs retrieved in 27 patients, at a median 2 (IQR 2–3) nodes per procedure. Additionally, 644 nodes were grossed in the neck dissection specimens of 29 patients (median yield 24; IQR 17.5–26 per procedure). Overall accuracy of SLNB was 96.3% (95%CI 81.03–99.91%), with sensitivity and negative predictive value of 83.3% (95%CI 35.88–99.58%) and 95.5% (95%CI 77.16–99.88%), respectively. The only factor independently predictive of occult nodal involvement was pathological depth of infiltration (hazard ratio 3.312; 95%CI 1.040–10.546, p = 0.043) which at 6.5 mm was 100% sensitive and 91.3% specific (area under curve 0.975; 95%CI 0.925–1.000).

Conclusion

Methylene blue-guided SLNB may be considered a viable alternative to evaluate the neck in early-stage buccal mucosal SCC with acceptable test validity and reliability. [Registered with the Clinical Trials Registry of India (CTRI/2022/01/039523 dated 18th January 2022)].

Introduction

Elective neck dissection (END) has been established as an integral component of the surgical management of early-stage, clinically node-negative oral squamous cell carcinoma (OSCC), conferring an overall and cancer-specific survival benefit apart from being both therapeutic and prognostic [1]. However, only the 20–30% of patients who are eventually found to harbor occult nodal metastases will derive this benefit from END, inferring that the majority are overtreated with standard practice [1, 2]. Moreover, the complications associated with this procedure are formidable and adversely affect quality of life [3].

Sentinel lymph node biopsy (SLNB) was introduced as a less-invasive alternative for staging the neck in clinically node-negative OSCC with most studies using radiocolloid as tracer [4]. However, the utility of radiotracers in the Indian setting is limited by the availability of nuclear medicine facilities [5]. An optical tracer, methylene blue is widely available and economical with a negligibly low incidence of adverse events [6] and has been evaluated in early-stage OSCC for SLN detection. These studies, apart from being small trials conducted over prolonged periods of time, enrolled heterogenous patient cohorts inclusive of all tumor stages and subsites of the oral cavity failing to account for anatomical site-specific variations in lymphatic drainage pattern, hence leading to conflicting results and limiting the applicability of the procedure [7–9]. In the Indian subcontinent, population-wide trends in smokeless tobacco consumption have led to striking variances in the epidemiology of OSCC such that today, the buccal mucosa represents the single most common subsite of the oral cavity involved, and may anatomically be more conducive to optical tracer-guided SLNB on account of ease of access and a predictable pattern of lymphatic drainage [10].

This study sought to determine the incidence of occult nodal metastases, and the identification rate and test validity of SLNB in clinically early-stage, node-negative SCC of the buccal mucosa, secondarily identifying the clinicopathological factors associated with node involvement in this cohort.

Materials and Methods

Patient Selection

This prospective validation study was conducted in the Oral Oncology and Head and Neck Oncology services of a tertiary cancer center over 2 months (January and February 2022). Consecutive patients aged 18 years or older and presenting with early-stage, biopsy-proven, upfront-operable invasive SCC of the buccal mucosa, with no clinically or radiologically significant cervical lymphadenopathy, were included in the study.

The buccal mucosa was defined as the subsite of the oral cavity anatomically bound by the retromolar trigone posteriorly, 1 cm from the anterior commissure of the lip anteriorly, and craniocaudally, 1 cm lateral to the superior and inferior alveolar margins, inclusive of the respective gingivo-buccal sulci. Early-stage disease comprised tumors of largest dimension 4 cm and clinical depth of infiltration (DOI) up to 10 mm corresponding to T1 or T2 lesions [11], and radiologically non-pathological cervical nodes measuring up to 10 mm in the short-axis [12] with no morphological features suggestive of metastatic involvement (rounded shape, irregular borders, distortion of internal architecture, or enhancement) as reviewed by two radiologists.

Patients with a history of having undergone any intervention that could potentially alter the pattern of lymphatic drainage (including prior head and neck surgery, recurrent oral cancers, or receipt of neoadjuvant therapy) were excluded, as were those with tumors extending beyond the above predefined boundaries and involving the facial skin, cortical bone and/or sinuses, or associated with palpable lymphadenopathy (suggestive of locally-advanced disease), and subjects not consenting to the procedure.

Evaluation consisted of a detailed history followed by physical examination of the primary and the neck by two surgeons for concurrence on the clinical stage and the absence of palpable nodal disease. Tumor dimensions and extent with reference to the retromolar trigone, anterior commissure, and gingivo-buccal sulci were recorded. DOI was estimated clinically and stratified as < 5 mm and 5–10 mm. As part of locoregional staging and metastatic workup, CECT of the head, neck, and thorax was performed in all patients and correlated with the findings on physical examination, following which clinical stage was assigned [11].

Technique of SLNB

All patients underwent upfront surgery after multidisciplinary evaluation. A single-tracer technique using methylene blue was used to map the sentinel node(s) after excluding hypersensitivity to the dye by intradermal sensitivity testing. Intraoperatively, following induction of anesthesia and patient positioning, four-quadrant peritumoral injection of 0.5 ml 1% methylene blue dye (totaling 2 ml) was given using a 24 G needle. The dye was administered by a single surgeon (JM) in all cases and the subsequent nodal dissection performed by one of three surgeons (RBC, KR, or JM), all of whom were experienced in the technique of SLNB. A cervical incision (corresponding to the horizontal limb of the Schobinger incision) was made to permit exposure to nodal stations 1 and 2 (Fig. 1) also taking into account the tumor location, the extent of resection, and if applicable, the reconstruction planned. The superior subplatysmal flap was raised and the passage of dye along the lymphatic channels traced across the lower border of the mandible toward the first-echelon draining LN(s). The node(s) taking up dye within 10–15 min of injection were considered the sentinel node(s) and dissected separately, grossed on-table and sent individually for histopathological evaluation (HPE). Successful detection of the SLN(s), number of nodes stained, their respective diameters in the short-axis, and the parent nodal station were documented. On-table pathological assessment of the SLN(s) was not performed. Subsequently, all patients proceeded to undergo resection of the primary and an elective modified radical neck dissection (MRND), regarded as the reference standard. The latter involved completion of the Schobinger incision and sequential clearance of levels 1 to 5, preserving the sternocleidomastoid, internal jugular vein, and spinal accessory nerve. The dissected nodal stations were numbered and sent separately for HPE.

Fig. 1.

Marking of the incision for optical tracer-guided SLNB, extending from the symphysis menti to the mastoid process (corresponding to the horizontal limb of the Schobinger incision) to facilitate elevation of the flap and dissection of stations 1 and 2 (A). In this case, the level 1B nodes (B) taking up methylene blue within 15 min of injection were considered the sentinel nodes, dissected separately and grossed on-table (C). The vertical limb of the incision was placed before performing a modified radical neck dissection clearing levels 1 to 5 (D)

Pathological Assessment

Following fixation in 10% neutral buffered formalin for 24 hours, harvested SLNs were embedded in paraffin wax and, when macroscopically negative, subjected to serial sectioning, each node sliced transversely into 2-mm sections followed by hematoxylin and eosin staining and microscopic examination. In SLNs appearing grossly positive, one or more routine sections were obtained for tumor demonstration. Regardless of the size of deposit encountered (isolated tumor cells, micrometastases, or macrometastases), presence of disease in a resected node was considered unequivocal evidence of involvement. Microscopically-negative nodes were not subjected to further diagnostic testing.

Pathological primary tumor size, focality, margin status, grade (according to the World Health Organization classification based on Broders’ criteria) [13], worst pattern of invasion (WPOI), DOI, presence of lymphovascular (LVI) or perineural invasion (PNI), and stromal inflammatory cell infiltration were noted. LN yield and status were also recorded. Positive nodes in either the MRND or SLNB specimens were further characterized by the number, size (in the short-axis) and level, after which the pathological stage was assigned [11]. Depending on the status of the SLN, the results were determined to be a true-positive or a true-negative, when the SLN status was reflective of the nodal status of the neck, or a false-negative (wherein the SLN was found to be negative but with non-sentinel nodes harboring metastatic disease).

Sample Size Calculation

From a review of available literature, the incidence of nodal involvement in early-stage OSCC was estimated to range from 26 to 30% [1, 2]. Extrapolating data from the acceptable false-negative rates (FNR) for SLNB performed in other malignancies and equivalent to the incidence of neck recurrence in pathologically N0 patients post-END, 10% was set as the estimated FNR with maximum permissible limit not greater than 20% (corresponding to the cervical recurrence rate following END for a pathological node-positive neck) [14]. At 80% confidence, a sample size of 36 patients with early-stage node-negative buccal mucosal SCC was required to reject the null hypothesis and demonstrate that the false-negative rate of SLNB using single optical tracer did not exceed 20%. However, only 29 patients could be accrued during the study period for analysis.

Statistics

Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS) for Windows version 23.0 (IBM Corp, Armonk, NY, USA). Continuous variables, expressed as mean and standard deviation, were compared using the independent Student’s t-test for two groups. Associations between categorical variables (presented as counts and percentages) were analyzed using the chi-square test and when expected cell counts were less than 5, the Fisher’s exact test. Binary logistic regression, selecting variables with p value < 0.1 on univariate analysis for multivariate modeling, was performed to establish the relationship between the dependent variable (node-positivity) and potential independent variables after testing for multicollinearity using tolerance and variance inflation factors (VIF). For all calculations, p value < 0.05 was considered statistically significant.

This study complied with the Standards for Reporting Diagnostic accuracy studies (STARD) guidelines.

Results

During the study period, 72 patients with histologically-proven SCC of the buccal mucosa were identified of which 29 were included in the final cohort (Fig. 2). Baseline clinicodemographic features and tumor characteristics have been summarized in Table 1.

Fig. 2.

Flow diagram illustrating the study sample inclusion and exclusion criteria, and the outcomes associated with methylene blue-directed SLNB compared to the reference standard of a modified radical neck dissection

Table 1.

Clinicodemographic features, tumor characteristics, surgical management, SLNB-related outcomes, and pathological findings of patients in the study cohort (n = 29)

Characteristic						N	Value
Age (mean ± SD) in years						29	56.3 ± 13.03
Gender			Male			6	20.7%
			Female			23	79.3%
Comorbidities			No			22	75.9%
			Yes	Hypertension		4	13.8%
				Diabetes mellitus		2	6.9%
				COPD		2	6.9%
Tobacco consumption			Smokeless tobacco alone			26	89.7%
			Tobacco chewing + smoking			3	10.3%
Duration of tobacco consumption (median, IQR) in years						29	20.0 (11.0–27.5)
Complaints at presentation			Oral ulcer			17	58.6%
			Pain			7	24.1%
			Poor mouth opening			3	10.3%
			Swelling			2	6.9%
Duration of symptoms (median, IQR) in months						29	3.00 (1.25–4.00)
Mouth opening (mean ± SD) in mm						29	30.2 ± 7.62
Oral submucous fibrosis			Present			4	13.8%
			Absent			25	86.2%
Clinical evaluation	Tumor size in largest dimension (mean ± SD) in mm					29	26.9 ± 9.68
	Clinical DOI		< 5 mm			10	34.5%
			5–10 mm			19	65.5%
	Clinical T stagea		T1			6	20.7%
			T2			23	79.3%
Grade of SCC			Grade 1			23	79.3%
			Grade 2			6	20.7%
Surgery for the primary			WLE			9	31.0%
			WLE + hemimandibulectomy			15	51.7%
			WLE + bite resection			3	10.3%
			WLE + upper alveolectomy			2	6.9%
SLN detected			Yes			27	93.1%
			No			2	6.9%
Number of SLNs detected (median, IQR)						27	2 (2–3)
Parent nodal station of the SLN(s)			1B			15	55.6%
			1B + 2A			7	25.9%
			1A + 1B			3	11.1%
			2A			2	7.4%
SLN size in short axis (mean ± SD) in mm						27	8.5 ± 1.48
Pathological assessment		Tumor size in largest dimension (mean ± SD) in mm				29	23.0 ± 10.07
		Pathological DOI (median, IQR) in mm				29	5.0 (3.5–7.5)
		Lymphovascular invasion			Absent	21	72.4%
					Present	8	27.6%
		Perineural invasion			Absent	20	69.0%
					Present	9	31.0%
		Worst pattern of invasion			1–4	23	79.3%
					5	6	20.7%
		Stromal infiltration			Mild	10	38.5%
					Moderate	10	38.5%
					Dense	6	23.1%
MRND nodal yield (mean ± SD)						29	20.1 ± 6.84
Overall pathological stagea		pT stage			pT1	9	31.0%
					pT2	17	58.6%
					pT3	3	10.3%
		pN stage			pN0	23	79.3%
					pN1	2	6.9%
					pN2a (ENE +)	1	3.4%
					pN2b	3	10.3%

COPD, chronic obstructive pulmonary disease; DOI, depth of invasion; ENE, extranodal extension; IQR, interquartile range; MRND, modified radical neck dissection; pN, pathological N stage; pT, pathological T stage; SCC, squamous cell carcinoma; SD, standard deviation; SLN, sentinel lymph node; WLE, wide local excision

^aAs per AJCC Cancer Staging System, 8th Edition [11]

The SLN identification rate with methylene blue was 93.1%. In 17 patients (63%), the nodes were found in a single station, the most common being level 1B (in 15). Overall, 66 SLNs were retrieved in 27 patients, at a median of 2 nodes (IQR 2–3) per procedure. No adverse events were noted with methylene blue.

All patients underwent unilateral neck dissection. Exclusive of the number of SLNs harvested, an additional 644 nodes were grossed in the neck dissection specimens of 29 patients at a median nodal yield of 24 (IQR 17.5–26) per procedure. Overall, seven patients with clinically early-stage SCC were upstaged to pathological stage III or greater on account of occult cervical nodal metastases in six patients (20.7%) and T3 disease in three (10.4%). The correlation between clinical and pathological T stage was noted to be relatively strong (Cramer’s V 0.579; p = 0.008). Further analysis revealed a statistically significant correlation between clinically and pathologically determined DOI (Pearson’s point-biserial r = 0.437; p = 0.018) and primary tumor size (Spearman’s r = 0.928; p < 0.0001).

Of the six patients identified to have nodal metastases, five were SLN-positive (considered a true-positive test result). Of these, the SLN was the sole site of cervical nodal metastases in two patients; in the others, the involved cervical levels were adjacent to the sentinel nodal station(s). Skip metastases or noncontiguous nodal spread were not observed. Overall, the FNR of methylene blue-directed SLNB was 16.7%, on account of one patient in whom a positive level 1A node was detected on MRND in the absence of SLN (1B) involvement. For an occult node-positive disease prevalence of 22.2% (95% CI 8.62–42.26%), the accuracy of SLNB was 96.3% (95% CI 81.03–99.91%), with sensitivity of 83.3% (95% CI 35.88–99.58%) and negative predictive value (NPV) of 95.5% (95% CI 77.16–99.88%).

Additionally, variables that could potentially predict the presence of occult cervical nodal metastases in early-stage buccal mucosal SCC were evaluated using binary logistic regression. On univariate analysis, pathological DOI, LVI, PNI, and tumor grade reached statistical significance and were considered for multivariate regression after ensuring that the assumption of multicollinearity was met. The multivariate model was found to be significant [χ² (df = 4, N = 29) = 19.895, p = 0.001] and a good fit of the data [Hosmer–Lemeshow goodness-of-fit χ² (df = 7, N = 29) = 5.915, p = 0.550], correctly classifying 89.7% of patients in the cohort (from 79.3%) and explaining between 49.6% (Cox and Snell R-square) and 77.7% (Nagelkerke R-square) of the variance in the outcome variable. Among the covariates selected, pathological DOI of the primary was the only factor significantly and independently associated with node-positivity (Table 2).

Table 2.

Comparison of clinicopathological variables in patients with node-positive and node-negative disease, with sequential univariate and multivariate analyses of these factors (and the corresponding collinearity statistics) to predict the presence of clinically occult nodal involvement in early-stage buccal mucosal SCC (n = 29)

		Univariate analysis						Multivariate analyses				Collinearity
		Overall nodal status		Hazard ratio	95% CI		p value	Hazard ratio	95% CI		p value	Tolerance	VIF
		Negative (n = 23)	Positive (n = 6)
					Lower	Upper			Lower	Upper
Age (mean ± SD) in years		56.2 ± 13.12	56.7 ± 13.94	1.003	0.935	1.076	0.939
Gender	Male	4 (13.8%)	2 (6.9%)
	Female	19 (65.5%)	4 (13.8%)	0.421	0.056	3.145	0.399
Primary tumor size (mean ± SD) mm		21.5 ± 10.11	29.0 ± 8.00	1.085	0.980	1.202	0.115
Pathological DOI (mean ± SD) mm		4.6 ± 1.88	10.2 ± 2.14	3.170	1.209	8.315	0.019*	3.312	1.040	10.546	0.043*	0.548	1.825
Worst pattern of invasion	1–4	18 (62.1%)	5 (17.2%)
	5	5 (17.2%)	1 (3.4%)	0.720	0.068	7.661	0.785
Lymphovascular invasion	Absent	19 (65.5%)	2 (6.9%)
	Present	4 (13.8%)	4 (13.8%)	9.500	1.272	70.964	0.028*	1.670	0.046	60.295	0.779	0.640	1.563
Perineural invasion	Absent	18 (62.1%)	2 (6.9%)
	Present	5 (17.2%)	4 (13.8%)	7.200	1.009	51.392	0.049*	0.608	0.006	65.498	0.835	0.696	1.437
Tumor grade	Grade 1	20 (69.0%)	3 (10.3%)
	Grade 2	3 (10.3%)	3 (10.3%)	6.667	0.895	49.674	0.064*	0.580	0.011	30.429	0.788	0.709	1.410

CI, confidence intervals; DOI, depth of invasion; SD, standard deviation; VIF, variance inflation factors

*Statistically significant

The independent samples T-test was performed to determine the relationship between DOI and nodal status after satisfying the assumption of homogeneity-of-variance (p = 0.554 Levene’s test). Mean DOI in pathological node-positive patients was 10.2 ± 2.14 mm (compared to 4.6 ± 1.88 mm in pN0 patients; p < 0.001). Additionally, a receiver operating characteristic curve plotted to determine the corresponding validity parameters for a given DOI revealed that a pathological DOI of 6.5 mm was 100% sensitive and 91.3% specific for the presence of occult cervical nodal involvement (area under the curve 0.975, 95%CI 0.925–1.000) (Fig. 3).

Fig. 3.

Receiver operating characteristic curve describing the accuracy and validity parameters associated with pathological DOI for predicting the presence of occult cervical nodal involvement across the range of given thresholds [AUC, area under the curve; CI, confidence intervals; DOI, pathological depth of infiltration]

Discussion

Although versatile and feasible without additional equipment, optical tracer-guided SLNB has yielded inconsistent results in literature limiting its applicability and wider adoption. In contrast, radiotracers have demonstrated superior technical outcomes primarily on account of the feasibility of transcutaneous visualization and the identification of aberrant lymphatic pathways, facilitating the detection of contralateral nodal involvement and skip metastases. For these reasons, current consensus guidelines recommend lymphoscintigraphy (either alone or in combination with optical tracers as adjuncts) for SLN detection in early-stage OSCC [15].

However, tracer type alone does not predict procedure-related outcomes. The range in literature-reported rates of SLN identification and diagnostic efficacy associated with SLNB also reflect the influence of other study-related factors including the heterogeneity of subsites, operator experience, and technique of pathological assessment.

Sentinel node studies using dyes have included all subsites of the oral cavity, resulting in conflicting reports on efficacy [7–9]. Yet the buccal mucosa, as an anatomically well-lateralized subsite with a predictable pattern of lymphatic drainage and a low likelihood of contralateral nodal disease, may be more conducive to optical tracer-guided SLNB. Although clinically occult contralateral nodal metastases have been reported in up to 12% of lateralized tumors, this subsite has been underrepresented in trials on SLNB [2]. Also, proximity of buccal mucosal primaries to the first-echelon draining nodes may give rise to the shine-through effect with radiotracer-guided SLNB, adversely affecting procedure-related success rates [16].

Our results suggest that optical dyes may be equivalent to radiotracers with respect to buccal primaries: although SLN identification rate was lower compared to dual tracer-guided studies (approaching 100%), it was comparable to studies using single optical tracers (ranging from 83.3 to 92.3%). Moreover, efficacy parameters in terms of sensitivity, NPV, and FNR were not different from studies using radiotracers with or without optical tracers for SLN detection [2, 4] and may be attributed to the subsite selected, experienced operators following a standardized technique (minimizing intervention-associated heterogeneity), and use of serial sectioning for nodal assessment.

As a validation study, intraoperative pathological assessment of SLN(s) was not part of protocol. Despite being vital in real-time decision-making when contemplating a single-sitting completion END, procedure-associated limitations have been known to compromise their reliability and utility in this setting: the sensitivity of frozen section analysis in detecting tumor deposits < 2 mm is under 50% [17] and although immunohistochemical evaluation of SLN(s) undoubtedly improves sensitivity and NPV of SLNB by detecting ITCs and micrometastases, additional costs, prolonged theater time, and limited availability restrict the wider utilization of this technique especially in resource-constrained settings [18, 19]. Nonetheless, the absence of facilities for on-table pathological assessment should not be a factor against the performance of SLNB as an interval completion neck dissection (performed 2–3 weeks following index surgery) may be considered in patients found to have nodal metastases [15].

A finding that has been reported in other oral cavity subsites as well [1, 20], DOI was found to be an independent risk factor for lymphatic dissemination and predictor of node-positivity, highlighting the importance of accurate preoperative evaluation when selecting cases for SLNB. In this study, the degree of agreement between clinical and pathological T stage estimation was strong, the marginal disparity solely on account of DOI. Additionally, a cutoff pathological DOI of 6.5 mm was found to be highly sensitive and specific in predicting cervical nodal involvement. In view of the risk of failing in the neck, such patients may also be considered for interval END, although larger trials would be required to determine subsite-specific cutoffs for a given population.

Limitations

Despite the rarity of early-stage node-negative OSCC and the short study period, our failure to reach the calculated sample size within the stipulated timeframe should be considered a significant limitation. Also, on-table pathological evaluation of SLN(s) was not performed, the objective of the study being to compare SLNB to the reference standard of an END, using routine microscopy as the common metric. As a validation study, procedure-associated postoperative morbidity and long-term outcomes of optical tracer-directed neck dissection including failure rates in the contralateral neck and survival were not assessed.

With regard to the technique, possibly the most important limitation remains the inevitability of a neck incision to permit exposure to stations 1 and 2. Nonetheless, this would facilitate the visualization of dye-stained lymphatics up to the draining SLN(s), thereby avoiding a full-fledged level 1 and 2A dissection and a potentially-morbid level 2B dissection, favoring functionality over cosmesis [21]. SLNB using methylene blue may be ideal for small lesions of the retromolar trigone or gingivo-buccal sulcus which by virtue of its location and proximity to bone merit some form of mandibular resection and hence, a neck incision. Additionally, patients requiring vascular access for flap reconstruction may also represent a subset that is feasible for SLNB using dyes alone.

A significant caveat to the generalizability of these results is the associated learning curve: as a small molecule, methylene blue has a rapid washout, limiting the timeframe for node visualization and increasing the margin for error. Lastly, these findings cannot be extrapolated to other subsites of the oral cavity where lymphoscintigraphy remains the standard.

Future Implications

Being widely available and requiring no specialized procedural equipment, methylene blue may be a suitable option to evaluate the status of the neck in patients with buccal mucosal primaries, more so in resource-constrained settings. Taking into account the incidence of OSCC in India, this may represent a possible solution to screen the low-risk node-negative population at the community healthcare level, referring only those cases with proven nodal metastases to tertiary care institutions for interval END. However, it should be recognized that the procedure has a learning curve and that proper patient selection is vital to balance oncological and functional outcomes.

Conclusion

SLNB with methylene blue as a single optical tracer may be considered a viable option with acceptable test validity and reliability to assess the status of the neck in early-stage, clinically node-negative buccal mucosal SCC. However, larger studies with long-term follow-up will be required to determine failure rates in the neck as a surrogate marker for efficacy.

Author Contribution

Joseph Mathew: conceptualization, methodology, formal analysis and investigation, data curation, writing—original draft, writing—review and editing, visualization, resources, supervision, project administration. BC Rajani: conceptualization, methodology, writing—review and editing, resources, supervision. Deeksha Thakur: conceptualization, methodology, data curation, writing—review and editing. R Krishnappa: conceptualization, methodology, writing—review and editing. KS Sabitha: conceptualization, writing—review and editing, resources. Rajshekar Halkud: conceptualization, methodology, writing—review and editing, resources.

Data Availability

The data supporting the findings of this study is available from the corresponding author upon reasonable request. It is not publicly accessible as it contains information that could compromise the privacy of research participants.

Declarations

Ethics Approval and Consent to Participate

This study was approved by the Institutional Review Board (KMIO/MEC/025/17.February.2021) and performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Informed written consent was obtained from all participants enrolled in the study.

Consent for Publication

Not applicable.

Competing Interests

The authors declare no competing interests.