Comparing swallowing function in oral tongue squamous cell carcinoma after upfront brachytherapy followed by IMRT vs. surgery

Purpose

The oral tongue is the most common site of oral cavity cancer, as it accounts for 22-39% of all oral cancer cases [1-3]. Most malignancies of the oral tongue are squamous cell carcinomas. Cigarette smoking and alcohol consumption are risk factors that contribute to the increases in oral tongue squamous cell carcinoma (OTSCC) worldwide. Various treatments are available for OTSCC. Surgical options, including wide local resection, hemiglossectomy, and total glossectomy, are the most well-established treatments that offer effective local control. However, the resection of excessive tongue tissue to achieve clear tumour margins, especially in advanced lesions, may impair tongue function. External beam radiotherapy (EBRT), such as 3D conformal radiotherapy and intensity-modulated radiation therapy (IMRT), are other treatment options for OTSCC [4]. IMRT has the potential to deliver a more targeted dose to the tumour while sparing the surrounding normal structures such as tissue involved in the swallowing function. Brachytherapy can be used as part of a non-operative organ preservation approach to increase the local control rate of the primary site in the treatment of OTSCC [5]. Based on the National Comprehensive Cancer Network guidelines Version 3.2024 for oral tongue carcinoma, early-stage tumours (T1, T2 N0) can be treated with a single modality, either surgery or definitive radiation therapy (EBRT or brachytherapy). For advanced-stage tumours (T3, T4a, N1-N3), the recommended treatment is a combined modality approach, involving surgery along with radiation therapy, with or without chemotherapy. However, not all patients are suitable for surgery and some patients may have refused surgical options. At the Advanced Medical and Dental Institute (AMDI), the upfront high-dose-rate interstitial brachytherapy followed by IMRT (HyBIRT) technique, which integrates interstitial brachytherapy, followed by definitive concurrent chemoradiation with intensity-modulated radiotherapy (IMRT) and weekly cisplatin, is offered to patients who either refused surgery or were deemed ineligible for surgery due to medical comorbidities. However, the impact of this treatment combination on their swallowing function has been uncertain, particularly when compared to the surgical methods. We therefore conducted this study to assess the swallowing outcomes of patients who received the combined brachytherapy and IMRT (HyBIRT technique) compared to those who were treated with surgical methods, with or without adjuvant chemoradiation. At the AMDI, all OTSCC cases are treated with the HyBIRT technique to improve the LC rate in the tongue, and hence an IMRT-alone arm was not included in this study for comparison because of insufficient patient numbers. The HyBIRT technique for treating the oral tongue SCC is effective, with a local control (LC) rate of 94% at both 1 and 3 years [6, 7]. Given these positive oncologic outcomes, we decided to assess the impact of this treatment on patients’ quality of life, specifically their swallowing function.

The objective of this study was to delineate the differential impact of surgery versus brachytherapy on the swallowing outcomes of patients with OTSCC. We aimed to elucidate and compare the effects of these two treatment approaches specifically on swallowing function to provide clinicians with insights regarding the comparative outcomes.

Material and methods

This cross-sectional study focused on evaluating the swallowing outcomes in OTSCC patients who had completed their treatment. The study included 30 patients from three tertiary hospitals, who were enrolled between June 2022 and October 2023. All cases were discussed in the multidisciplinary team/tumour board and before commencing treatment, and all cases underwent an intra-department peer review process. Fourteen of these patients were treated with brachytherapy followed by IMRT with concurrent chemotherapy for definitive management, and the remaining 16 underwent surgical procedures with or without adjuvant chemoradiation. The inclusion criteria were patients aged 18 years and older with histopathologically proven OTSCC who had completed their treatment regimes. Additionally, the participants needed to be at least six months post-treatment and to have been confirmed clinically and radiologically as free from tumours. Patients with metastatic disease and those with neurological disorders that might impair swallowing were excluded from the study. The study received ethical approval from both Universiti Sains Malaysia [code USM/JEPeM/22060371] and the National Medical Research Register [code NMRR ID-22-02363-9BL]. A notable limitation of this study is the selection criteria, which restricted the sample to patients who completed their treatments. This inherently excluded individuals who discontinued treatment due to adverse events, complications, or disease progression. Patients who developed recurrence after completing treatment were also excluded, as recurrence could impair swallowing function and confound post-treatment assessments. In addition, patients who died from treatment-related toxicity or disease were not included, as their swallowing outcomes could not be evaluated. While these exclusions were necessary to ensure that the analysis reflects the functional outcomes of completed treatment courses, they may also limit the generalizability of the finding.

Each patient first completed the Malay Eating Assessment Tool-10 (EAT-10) questionnaire and then the Swallowing Capacity Scale, to assess their subjective swallowing difficulties. The Malay EAT-10 is a validated, self-administered questionnaire used to assess subjective swallowing difficulties in patients. It consists of 10 items, each scored on a 5-point Likert scale (0 = no problem to 4 = severe problem), with a total score ranging from 0 to 40. Higher scores indicate greater swallowing impairment. The Malay EAT-10 has demonstrated good psychometric properties, including high internal consistency, excellent test-retest reliability, and strong construct validity. It has been culturally adapted and validated for use in the Malaysian population, making it a reliable tool for evaluating dysphagia in this context. They subsequently underwent a fibreoptic endoscopic evaluation of swallowing (FEES) test, which was performed by a trained medical officer in accordance with the Langmore protocol [8]. The endoscope was passed through the most patent nostril, with the tip of the flexible endoscope advanced into the hypopharynx. Thereafter, the patient received 5 ml (1 teaspoon) of three different food consistencies. The first food consistency was solid, the second a semisolid bolus, and the third liquid. To evaluate swallowing, the endoscope was placed in a high position above the epiglottis. After swallowing, the endoscope was advanced for about 10 seconds to a low position just above the vocal folds to evaluate the Penetration-Aspiration Scale. The collected data were analysed using the Mann-Whitney U test and independent t-test via SPSS version 29.

The study is cross-sectional in nature, as it evaluates swallowing outcomes at a specific point in time after patients have completed their treatment. The retrospective aspect pertains to the enrolment of patients who had already completed their treatment between June 2022 and October 2023, but the actual assessment of swallowing function was conducted cross-sectionally at a post-treatment time point.

The eligibility criteria for the HyBIRT regimen in the treatment of oral tongue squamous cell carcinoma (OTSCC) include histopathologically confirmed SCC and TNM7 stages T1–T4a, N1-N3, M0. Only patients who had completed the full course of treatment, including brachytherapy, IMRT, and chemotherapy were included. The HyBIRT treatment technique and disease control outcomes are described in detail in another publication [6, 7] which was presented during ESTRO 2025. The dose prescriptions used were: HDRIBT 20 Gy in 5 fractions to primary gross tumour volume (GTV), followed by IMRT of 69.96 Gy to GTV node, 61.71 Gy to high-risk region, 59.4 Gy to pre-treated HDRIBT field and 56.1 Gy to low-risk regions in 33 fractions. The first GEC-ESTRO ACROP update on head and neck brachytherapy in squamous cell carcinomas emphasizes improvements in treatment planning through cross-sectional imaging and the use of stepping source technology. The American Brachytherapy Society (ABS) has also published recommendations for high-dose-rate brachytherapy for head-and-neck carcinoma. While the dose prescription and constraints used in the current study conform to the recommendations in both the GEC-ESTRO and ABS, the guidelines agree that there are limited data on the use of HDRIBT in oral cavity despite the excellent local control seen in published studies. Our study differs from the ABS and GEC-ESTRO recommendations, as based on current available literature, HDRIBT is recommended as a boost after EBRT. However, the lack of recommendation for upfront HDRIBT is due to the lack of published data. Our current study is novel in that it represents the first documented use of upfront high-dose-rate interstitial brachytherapy (HDRIBT) followed by intensity-modulated radiation therapy (IMRT) in the treatment of head and neck cancers. The rationale for the upfront treatment approach was presented at ESTRO 2023 [6, 7]. Administering brachytherapy prior to chemoradiotherapy allows for easier identification of tumour margins and facilitates optimal applicator placement to ensure optimal dose coverage. In a post-chemoradiotherapy setting, there is a theoretical risk of microscopic disease being missed in the area affected by the primary tumour.

Results

Demographic details of the patients in the brachytherapy and SG are presented in Table 1. The median age was higher in the SG at 61.5 years (IQR 14.25) compared to the HyBIRT, which had a median age of 55.5 years (IQR 22.75). The HyBIRT included more men, while the SG had a higher proportion of women. The follow-up duration was longer for the surgical group. In the T staging distribution, most patients in the HyBIRT were classified as T3, while most patients in the SG were classified as T2. In terms of N staging distribution, more than half of the cases in both the HyBIRT and SG groups were classified as N+.

The perioperative details of the surgical patients are shown in Table 2. For surgical management, the type of resection was determined by the size and extent of the tumour, with the primary goal being complete tumour removal with a 2 cm safety margin. In cases of smaller tumours, a wide local excision or hemiglossectomy may suffice. However, for larger tumours, more extensive resections such as subtotal or even total glossectomy might be required to achieve adequate oncological control. The choice of reconstruction methods depended on the extent of tongue tissue resection, incorporating techniques such as skin grafts, buccinator flaps, or pectoralis major flaps. In the surgical group of the study, 10 out of 16 patients (62.5%) required postoperative radiotherapy, of which 3 patients (18.75%) also necessitated adjuvant chemotherapy. Indications of adjuvant chemoradiation are margin involvement and extra-nodal extension.

The patients’ swallowing outcomes after brachytherapy and surgery are presented in Table 3. The HyBIRT showed significantly better outcomes in the Penetration-Aspiration Scale for both the liquid and semisolid boluses (p = 0.02), Swallowing Capacity Scale (p < 0.01), and Malay EAT-10 score (p < 0.01) compared to the surgical group. No significant difference was noted between the groups in the Penetration-Aspiration Scale for the solid bolus (p = 0.94).

Table 4 shows a comparison of the adverse swallowing outcomes between the HyBIRT and SG. Five patients (31.5%) in the SG experienced solid bolus arrest in the oral phase, so no pharyngeal phase occurred to score penetration/aspiration. The reduction in the number of patients who participated in this segment may have contributed to the lack of significant difference between the groups for the solid bolus with the Penetration-Aspiration Scale. The absence of solid bolus arrest in the oral phase among the patients in the HyBIRT suggests a better oral transit of the solid bolus in HyBIRT compared to the SG. Two patients from the SG required nasogastric tube feeding.

Discussion

Difficulty swallowing is a frequent and challenging issue faced by patients with OTSCC before and after treatment. Both surgery and brachytherapy disrupt the normal mechanism of swallowing in different ways. This issue leads to severe problems, such as a compromised nutritional status, dehydration, and an increased risk of aspiration pneumonia, which profoundly impacts patients’ quality of life. It was therefore the critical focus of investigation in this study.

In the current study, we integrated both subjective and objective evaluations to provide a comprehensive assessment of swallowing function in patients recovering from OTSCC treatment. The Penetration-Aspiration Scale using the FEES test was chosen for the objective evaluation because it was readily available in the clinic and is a routine swallowing assessment method used in our centres. This test allows clinicians to examine the swallowing process and identify any abnormalities that could lead to unsafe swallowing and potential aspiration pneumonia [9]. The FEES test is a validated and reliable method that provides real-time visualisation of swallowing mechanics, which is crucial for detecting the key indicators of dysphagia severity, namely, aspiration and penetration [10, 11]. However, while objective testing is crucial for verifying the safety of swallowing to prevent aspiration pneumonia, it does not always capture a patient’s experience of dysphagia and its impacts on their quality of life [12, 13]. We therefore opted to include subjective evaluations in our study. Subjective evaluations are based on patient-reported experiences and involve the use of detailed questionnaires to gauge the broader impacts of swallowing difficulties on daily life. In this study, the Malay EAT-10 was selected for use as the subjective swallowing assessment instrument because the questionnaire has previously been translated into Malay and validated, which makes it appropriate for assessing the swallowing function within a Malaysian patient population [14].

Based on the Penetration-Aspiration Scale and the corresponding binary classifications of swallowing safety status, a Penetration-Aspiration Scale score of less than 3 represents safe swallowing [10]. In the present study, both the HyBIRT and SG had a median Penetration-Aspiration Scale score of less than 3 for all three types of boluses, which indicated that safe swallowing was achieved with both types of OTSCC treatments. However, the Penetration-Aspiration Scale scores for the liquid and semisolid boluses were significantly better in the HyBIRT. No significant difference was evident between the two groups for the solid bolus, because a significant proportion (31.3%) of the patients from the SG experienced swallowing arrest in the oral phase. To obtain a Penetration-Aspiration Scale score, patients must reach the pharyngeal phase of swallowing. The aforementioned patients in the SG did not reach the pharyngeal phase and were therefore not included in the Penetration-Aspiration Scale scoring for the solid bolus.

Before patients with OTSCC receive treatment, swallowing difficulties can arise due to the presence of a tumour in the tongue, which impacts the oral phase of swallowing. In the oral propel phase, the bolus is normally propelled backward to the oropharynx by the sequential anterior-to-posterior elevation of the tongue. The tongue tip rises to touch the hard palate, while the base of the tongue drops to receive the bolus. The disruption of this bolus propelling mechanism by a tongue tumour can lead to bolus arrest occurring in the oral cavity. Normally, once the bolus crosses the oropharyngeal sphincter and comes into contact with the palatoglossal and palatopharyngeal arches, which stimulate the glossopharyngeal and vagus nerves, the involuntary swallowing reflex is initiated (i.e. the pharyngeal phase of swallowing). This phase propels the food bolus from the oropharynx into the oesophagus. Typically, the upward movement of the tongue base presses the bolus against the pharyngeal wall to create positive pressure that acts as a piston pump to propel the bolus towards the oesophagus. In the case of OTSCC, particularly with a posterior extension, the propulsive mechanism of the tongue base may be affected, which leads to food bolus retention in the oropharynx and an increased risk of aspiration pneumonia. This altered swallowing physiology highlights the importance of optimising the swallowing function in the treatment for OTSCC, be it via surgery or radiotherapy.

After treatment for OTSCC, difficulty with swallowing may occur for a variety of reasons. External radiation therapy in the head and neck region has the potential to damage the healthy tissue surrounding the treatment area, including the pharyngeal constrictor muscle. This damage can result in fibrosis, which disrupts the sequential contraction from the superior to inferior constrictors that are essential for creating the necessary positive pressure to propel the bolus into the oesophagus. Additionally, external radiation can lead to fibrosis of the suprahyoid muscle, with a subsequent impact on hyolaryngeal excursion. The elevation of the larynx by the suprahyoid muscle plays an important role in generating the traction force needed to open the upper oesophagus sphincter and create the negative pressure that assists with swallowing – a process known as hypopharyngeal suction. Laryngeal oedema, a common complication of neck radiation, can further complicate swallowing and increase the risk of aspiration. In our study, poorer swallowing outcomes were observed in the SG compared to the HyBIRT, which can be attributed to the fact that a significant majority (62.5%) of the surgical patients underwent not only surgical resection but also external radiotherapy. While radiotherapy can enhance oncological outcomes, particularly in higher T and N stages, its side effects, which include mucositis, xerostomia, fibrosis of the oropharyngeal musculature, and osteoradionecrosis of the mandible bone, can add complexity to postoperative recovery and often worsen swallowing difficulties [15-17]. These complications may impede swallowing by increasing the oral transit time due to xerostomia and reducing pharyngeal contraction due to radiation-induced fibrosis [16, 17]. Evidence has consistently shown that postoperative radiotherapy results in more pronounced swallowing impairments than surgery alone in the treatment of tongue cancer. For example, a study by Finlay et al. showed that patients undergoing combined surgical and radiotherapy approaches exhibited a significant reduction in the ability to return to solid food intake compared to those who received surgery alone. In addition, the studies of both Shin et al. and Pauloski et al. demonstrated that in patients with tongue cancer who have undergone glossectomy, postoperative radiotherapy significantly impairs swallowing, as evidenced by both objective assessments via videofluoroscopy and subjective assessments using the MD Anderson dysphagia inventory [17, 18]. The decline in swallowing function is due to the significant shrinkage of the reconstructed tissue and radiotherapy-related complications, such as fibrosis, which impede the coordinated movement of the residual tongue and oropharyngeal muscle. A systematic review of the FEES test and videofluoroscopy found that chemoradiation for head and neck cancer commonly leads to swallowing difficulties, such as decreased laryngeal movement, tongue base dysfunction, and weakened pharyngeal contraction, with silent aspiration occurring in 35-85% of cases [15].

Like radiotherapy, surgery can lead to swallowing difficulties. The surgical resection of an OTSCC can impact the swallowing process at both the oral and pharyngeal phases. The tongue plays a crucial role in manipulating food particles in the oral cavity to create a bolus that can be transferred to the pharynx and then swallowed via the oesophagus. Extensive resection of the tongue can result in the loss of tongue tissue and muscle, which reduces its ability to effectively shape and compress the food bolus and impedes its propulsion towards the oropharynx and oesophagus during the swallowing process. This alteration in the swallowing function could explain why patients in the SG in our study experienced solid bolus arrest at the oral phase while none in the HyBIRT were affected in this way. The bulkiness of the reconstruction, such as the pectoralis major flap used in the surgical group, can affect the movement of the residual tongue muscle, which impacts the driving force that the residual tongue muscle provides for the piston pump action that propels the bolus towards the oesophagus during swallowing. The choice of reconstruction technique following surgical treatment for tongue cancer can influence swallowing outcomes. Postsurgical reconstruction techniques for tongue cancer typically encompass primary closure, a skin graft, pedicled flaps, and microvascular free flaps. Evidence suggests that patients whose surgical sites are closed directly or with skin grafts tend to exhibit superior swallowing function postoperatively compared to those who have reconstruction using pedicled or free flaps [17, 19-24]. Pedicled flaps, such as the pectoralis major flap, have been associated with several challenges, including compromised tongue mobility, prolonged oral transit times, more significant post-swallow residue, and overall decreased swallowing efficiency. This is often attributed to the bulkiness of these flaps, which can impede the movement of the remaining tongue tissue. On the other hand, microvascular free flaps, such as the radial forearm free flap, are valued for their thin yet flexible nature, which allows the defect to close without imposing undue bulk that could impede the mobility of the remaining tongue tissue [25, 26]. Studies have shown that microvascular free flap reconstruction is superior to pedicled pectoralis major flaps in terms of swallowing function [27]. In our study, the inferior swallowing outcomes in the SG could be attributed to the use of the pectoralis major flap, which constituted 50% of the reconstructions. The ideal flap for OTSCC should be a microvascular free flap, as it offers better functional outcomes. However, the choice of the pedicled pectoralis major flap in this study was due to limited resources in our centres, which constrained the availability of microvascular free flaps. The complexities associated with surgical resection and reconstruction in the treatment of OTSCC underline the importance of rehabilitation and continued support to assist patients in adjusting to functional changes, optimising the swallowing function, and improving patients’ quality of life following surgery.

It is crucial to acknowledge that the success of tongue cancer treatment is not measured solely by the eradication of the disease but also by the preservation of the tongue’s essential functions. Optimal outcomes from cancer treatment are significantly compromised if they lead to considerable functional impairment. Conversely, efforts to preserve function hold limited value if they fail to provide adequate oncological control. Hence, a nuanced approach to the management of such cancers is required to ensure that the patient’s post-treatment quality of life is considered equally important as the effectiveness of the cancer treatment itself. Brachytherapy, an organ preservation approach, has been gaining attention among clinicians as a valuable treatment for OTSCC, as it offers a balance between effective cancer control and the preservation of tongue function. Evidence has increasingly shown that brachytherapy offers low-toxicity oncological results that are on par with those of surgical approaches [28-31]. A study by Budrukkar et al. suggested that brachytherapy not only maintains oral competence but also supports the preservation of speech and swallowing abilities while reducing the risk of aspiration [31]. Urashima et al. observed that among a large cohort that included 493 OTSCC patients treated with brachytherapy with or without external radiotherapy, nearly all were able to maintain a normal diet following treatment, although it is noteworthy that this group consisted predominantly of early-stage disease cases [32]. Furthermore, Yoshimura et al. reported that the overall quality of life of oral cancer patients who received brachytherapy was good [33]. The findings from the literature, alongside the results of the current study, reveal a superior swallowing outcome following treatment for OTSCC with brachytherapy. This indicates an advantageous equilibrium between treatment efficacy and functional preservation, which underscores brachytherapy as a modality that effectively balances oncological success with the maintenance of swallowing function.

The shorter follow-up and smaller sample size for the HyBIRT group and the non-stratified enrolment are limitations of this study.

Conclusions

The organ-preserving HyBIRT technique is a promising alternative for managing OTSCC, particularly in preserving swallowing function, and should be further explored in larger, stratified studies with longer follow-up periods to validate these findings.