Efficacy of adjuvant radiation in early-stage cancers of buccal mucosa: an institutional experience

The purpose of this study is to determine if depth of infiltration is the only risk factor that will determine the outcomes in early-stage buccal mucosa patients or do the other minor risk factors have an impact too. It is a retrospective analysis of 226 patients with early-stage buccal mucosa cancer who were treated with curative intent from 2010 to 2020. These patients were grouped in two arms, surgery alone (n = 111) and surgery followed by adjuvant radiotherapy (n = 115). Patients were followed up and local and regional recurrences and distant metastasis were recorded. Our results show that addition of radiation to the standard surgery arm improves overall survival and disease-free survival, though the improvement in overall survival was not statistically significant. This improvement was more pronounced in the depth of infiltration > 5 mm and in 5 mm or less depth of infiltration this benefit was not statistically significant. Other factors like perineural invasion, lymphovascular invasion, tumour size, node positive, margin positive were considered for univariate analysis. Although there was a trend towards improvement of OS and DFS, it was not statistically significant as far as these factors are concerned. The role of adjuvant radiation in early-stage cancers of buccal mucosa is a crucial tool with definitive DFS benefit and requires more prospective trials to answer the OS benefit.


Introduction
Although one of the most common cancers in men in our country, the outcomes of cancers of the oral cavity have a huge scope for improvement.
Surgery in the form of gross tumour resection with adequate margins and reconstruction forms the standard treatment approach for squamous cell cancer of the buccal mucosa with or without radiation. The role of radiation was well defined in the era of 7th AJCC TNM with a clear demarcation of major and minor risk factors and as to which subset of patients might benefit with the same. However, with the incorporation of depth of infiltration as a major predictor of prognosis in oral cavity cancers, there has been a paradigm shift in determining when and whom to offer adjuvant radiation, since the depth of infiltration does not always correlates with other risk factors.
With this background, we formulated a hypothesis-is there a subset of early-stage cancers of oral cavity that might benefit from radiation if so what are the factors that define this risk group, is it only depth of infiltration that will determine the outcomes in such patients or do the other minor risk factors have an impact too.

Materials and methods
We retrospectively, identified a total of 226 patients who were treated with curative intent having cancers of the buccal mucosa from 2010 to 2020 registered in the department of Radiation Oncology, Vydehi Cancer Centre, Bangalore, Karnataka, India. Ours was a primary data collected in a quantitative method from Electronic Medical Records available at our institute. These patients were grouped in two arms, surgery alone (n = 111) and surgery followed by adjuvant radiotherapy (n = 115).
All these patients had undergone total excision of tumour and neck dissection as well as reconstruction, if it was indicated. Patients in the radiotherapy arm were treated with a dose of 60-66 Gy in 30-33 fractions to the tumour bed and a dose of 50-54 Gy to elective nodal regions, via 3DCRT or IMRT technique over a duration of 6 weeks. Patients were followed up every 3 months for the first 2 years, every 6 months for the next 3 years and yearly thereafter with history and clinical examination and imaging, if clinically indicated. On each follow-up, complaints and toxicities according to CTCAE guidelines were recorded. Local and regional recurrences and distant metastasis were recorded and patients were further managed accordingly. All the patients were called for follow-up physically and telephonic follow-up was done when physical presence was not feasible.

Statistical analysis
Survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test. Cox proportional hazards models were used to estimate hazard ratios and 95% confidence intervals for the association between adjuvant radiation therapy and survival outcomes. We performed univariate analysis to identify potential predictors of overall survival and disease-free survival. The variables considered in the univariate analysis included age, sex, depth of invasion, margin status, lymph node status. All statistical analyses were performed using R software.
Survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test. Cox proportional hazards models were used to estimate hazard ratios and 95% confidence intervals for the association between adjuvant radiation therapy and survival outcomes. We performed univariate analysis to identify potential predictors of overall survival and disease-free survival. The variables considered in the univariate analysis included age, sex, depth of invasion, margin status, lymph node status. All statistical analyses were performed using R software.

Study population
Group A consisted of patients who underwent both surgery followed by radiation and Group B surgery followed by observation. Both the arms were well balanced in terms of patient characteristics mean age at diagnosis was 50.11 (± 11.65) for group A and 53.12 (± 11.94) in group B, all the patients were restaged using the AJCC-TNM 8th edition where ever necessary, majority of the patients were T2 in both the arms 65.2% vs 58.2% and node negative 56.5 vs 92.7, all the patients were divided based on the depth of infiltration into two main categories </= 0.5 cm (cm) and 0.6-1 cm. Other parameters used are, pathological size of the tumour, skin involvement, margin status, bone involvement, peri-neural invasion (PNI), lymphoma-vascular invasion (LVSI), whether concurrent chemotherapy was administered or not.
Detailed description is given in Table 1.

Survival outcomes
Our analyses showed the following results. Median followup period for Group A was 73 months and for Group B was

Overall survival
When depth of infiltration was taken into account, for patients with depth of infiltration less than 5 mm, there was no significant difference in the hazard of death between the two groups over time (p = 0.2) (Fig. 1).
For patients with depth of infiltration of tumour 6-10 mm, the overall survival of Group A was higher than that of Group B at all points, however, the difference was not statistically significant (0.2) (Fig. 2).

Disease free survival
For patients with depth of infiltration of tumour less than 5 mm, the difference in the disease-free survival between the two groups was not statistically significant (0.4) (Fig. 3).
For depth of infiltration of 6-10 mm, at all time points, the disease-free survival probability for Group A was consistently higher than that of Group B for patients with tumour size between 6 and 10 mm. The Cox hazard ratio of 2.1 (with 95% CI of 0.98-4.77) and a p-value of 0.05 suggests that the hazard was higher for patients in Group B compared to Group A, although there may be some uncertainty around this estimate (Fig. 4).  On univariate analysis this benefit was more pronounced in patients who had depth of infiltration more than 5 mm. Amongst the other factors the T size, LVSI, node positive status and ENE have shown to have poorer survival in both the groups, however, this was not statistically significant (Tables 2, 3).

Discussion
Our results show that addition of radiation to the management of early-stage tumours of buccal mucosa shows statistically significant disease-free survival benefit. Literature search revealed that most of the studies which are done till date, that look into the depth of invasion of tumour as a prognostic factor includes all the subsites of oral cavity. Subsite-specific studies are mostly focussed on carcinoma of tongue. Hence, with this study, we aim to correlate depth of invasion in early buccal mucosa cancers with outcomes of treatment. With the introduction of depth of invasion as a main parameter in the current TNM staging and oral cavity being the second most common cancer in India with buccal mucosa cancers being the major contributor, unlike western population where lip and tongue cancers predominate, it is imperative to have such studies (Sung et al. 2021).
The role of radiation in advanced stage buccal mucosa is well established. Dinshaw et al. (2005) conducted a retrospective analysis of head and neck squamous cell cancers, majority of these patients were oral cavity primaries this study concluded that oral cavity subsite and presence of peri nodal extension were the two major prognostic factors that had a positive impact with addition of post-operative radiotherapy (Dinshaw et al. 2005). Murthy et al. (2010) studied prognostic factors in 1180 patients of oral cavity  -Total cancers treated with both definitive or adjuvant radiation. The result showed superior outcomes in 810 gingivo-alveolar-buccal complex patients treated with post-operative radiation particularly in advanced stages and this study concluded that there is scope for improving outcomes in buccal mucosa patients by adopting treatment intensification strategies (Murthy et al. 2010). Mishra et al. (1996) conducted a prospective randomised trial in which locally advanced cancers of buccal mucus were randomly assigned to surgery alone and surgery with post-operative radiation. This study concluded that addition of radiation has clearly shown a disease-free survival benefit which is statistically significant, however, overall survival was not evaluated by this group (Mishra et al. 1996). Kligerman et al. (2022) evaluated the impact of risk factors like worst pattern of invasion, tumour infiltrating lymphocytes, tumour budding and predominant pattern of invasion as variables to determine risk of recurrence in early-stage oral cavity cancers. It proved that most of them were not statistically significant risk factors for recurrence. However, this study did not include depth of invasion as one of the risk factors (Kligerman et al. 2022). Huang et al. (2009) studied the tumour thickness in patients with squamous cell carcinoma of the oral cavity and its predictive value for cervical nodal involvement. The study concluded that there was an association between tumour thickness and cervical nodal involvement and the author also stated that tumour thickness was the strongest predictor of outcome and a cut-off of 4 mm for considering neck management (Huang et al. 2009). Pentenero et al. (2005) conducted a review of the studies analysing tumour thickness/depth of invasion in predicting regional metastases and survival. In this review, most of the studies agreed that tumour thickness is a significant predictor for nodal metastasis and survival but the cut-off thickness was quite variable ranging from 1.5 to 10 mm (Pentenero et al. 2005 Dec). One of the studies in this review (Woolgar and Scott 1995) showed different cut-off thickness values for different tumour sites (Woolgar and Scott 1995). Ebrahimi et al. (2014) in the retrospective analysis of 3149 patients with oral squamous cell carcinoma patients showed that DOI was a significant predictor of diseasespecific survival and they concluded that the T staging of oral cavity should be modified with incorporation of DOI and can be used in clinical practice after external validation. Matos et al. (2017) have provided an external validation for the 8th edition AJCC staging by analysing a cohort of 298 patients. Their results showed that 22.8% had been upstaged when depth of invasion was included into the T staging and these patients had worse DFS and OS. This study concluded that incorporation of depth of invasion into pT classification helps in identifying a worse disease-free survival and overall survival (Matos et al. 2017).
After a thorough literature search, ours is one of the few studies that has validated depth of invasion in early-stage squamous cell carcinoma of buccal mucosa. Our results show that addition of Radiation to the standard surgery arm improves overall survival and disease-free survival, though the improvement in overall survival was not statistically significant. This improvement was more pronounced in the depth of infiltration > 5 mm and in 5 mm or less depth of infiltration this benefit was not statistically significant. Other factors like perineural invasion, lymphovascular invasion, tumour size, node positive, margin positive were considered for univariate analysis. Although there was a trend towards improvement of OS and DFS, it was not statistically significant as far as these factors are concerned.
As this is a retrospective study conducted at a single institute, although very relevant in the 8th AJCC era, we could not achieve statistically significant benefit in patients with < 5 mm depth of infiltration with addition of adjuvant radiotherapy (Amin et al. 2023).
Although ours is a retrospective analysis and single institutional experience, the large sample size, well-balanced treatment arms, buccal mucosa alone as a subsite and incorporating DoI and other risk factors which are potential confounders help in determining the benefit of radiation in adjuvant setting in cancers of buccal mucosa.
Ongoing studies like AREST which is a prospective multicenter trial with longer and more robust follow-up might be able to address these issues better (Adjuvant Radiotherapy in Early Stage Oral Cancers (AREST) 2023).

Conclusion
The role of adjuvant radiation in early-stage cancers of buccal mucosa is a crucial tool with definitive DFS benefit and requires more prospective trials to answer the OS benefit.