Unilateral radiotherapy for tonsillar cancer with multiple ipsilateral neck lymph nodes

Article information

Radiat Oncol J. 2024;42(3):192-199

Publication date (electronic) : 2024 September 25

doi : https://doi.org/10.3857/roj.2024.00164

Tae Hyun Kim^1,2, Hong-Gyun Wu¹, Soon-Hyun Ahn³, Woo-Jin Jeong⁴, Wonjae Cha⁴, Keun-Yong Eom^,5

¹Department of Radiation Oncology, Seoul National University College of Medicine, Seoul, Republic of Korea

²Department of Human Systems Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

³Department of Otorhinolaryngology, Seoul National University College of Medicine, Seoul, Republic of Korea

⁴Department of Otorhinolaryngology, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Republic of Korea

⁵Department of Radiation Oncology, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Republic of Korea

Correspondence: Keun-Yong Eom Department of Radiation Oncology, Seoul National University Bundang Hospital, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 13620, Republic of Korea. Tel: +82-31-787-7653 E-mail: 978sarang@daum.net

Received 2024 March 12; Revised 2024 April 20; Accepted 2024 May 7.

Abstract

Introduction

The lateralized nature of tonsillar cancer, especially at early stage tumors, has been a subject of long-standing debate over unilateral radiotherapy (UNRT). The common practice is to apply UNRT for node negative or single ipsilateral node positive tumors (N0-N2a), and to mandatorily apply bilateral radiotherapy (BNRT) for tumors with bilateral or contralateral node involvement or with nodes larger than 6 cm (N2c-N3) [1,2]. However, there is no definite evidence to support one over the other regarding patients with multiple ipsilateral node involvement (N2b). All consensus guidelines that are available to date consider BNRT as a conventional approach for patients with multiple ipsilateral node involvement; American College of Radiology recommendation in 2011 and American Radium Society appropriate use criteria [3,4]. Also, the American Society for Radiation Oncology Evidence-Based Clinical Practice Guideline does not include tumors with multiple ipsilateral node involvement in the indication of UNRT [2].

Nonetheless, some prior studies have questioned the implication of contralateral neck treatment considering the fact that contralateral neck failure is rare, and even unobservable in some cases [5–9]. These studies have demonstrated contralateral neck failure rates as low as 0% to 2%–3%. On the other hand, there have been studies that asserted the necessity of contralateral neck treatment based on a relatively high rates of contralateral neck failure rates from their own data [10,11].

On the other side of the issue lies the radiation toxicity. UNRT has been shown to be associated with significant reduction in toxicities including xerostomia, mucositis, edema, and fibrosis [12]. The extent of reduction was quite significant, up to from 2 to 3 times at maximum. Moreover, one study has demonstrated that patient-reported quality of life (QoL), as measured by the M.D. Anderson Dysphagia Inventory, exhibited improvement across all time periods up to 27 months after receiving UNRT compared to BNRT [13]. This improvement was observed not only in overall QoL, but also in specific symptom-related QoL, particularly concerning xerostomia.

In the light of aforementioned controversy in addition to the merits and demerits of UNRT, this study aimed to elucidate the safety and efficacy of UNRT for tonsillar cancer with multiple ipsilateral neck lymph nodes. The data was based on the treatment results of two tertiary referral hospitals that share same treatment policy, and thus may provide a more coherent conclusion on the research question.

Materials and Methods

2. Diagnosis and treatment

Patients were diagnosed via biopsy of primary tumor +/- suspicious lymph nodes. Imaging modalities including CT scan, MRI scan, and/or positron-emission tomography/CT (PET/CT) were performed for staging work up. Staging was established according to TNM classification of malignant tumors (TNM) 7th edition, 2009 considering the unavailability of p16 data in some patients.

Patients underwent treatment with one of the following options or a combination of them: induction chemotherapy (ICT), surgery, RT, and concurrent chemoradiotherapy (CCRT) (Table 1). Surgery followed by postoperative RT or CCRT was performed in 48.3% of the patients. ICT was given in about half of patients who received radical RT or CCRT. The most frequent ICT regimen was 5-fluorouracil/cisplatin (n = 3, 10.3%) followed by docetaxel/cisplatin (n = 2, 6.9%). Concurrent chemotherapy regimen was mainly weekly cisplatin (n = 17, 58.7%). ICT was administered to patients at high risk of distant metastasis, those with extremely bulky primary tumors, or individuals intended for RT volume and/or dose de-escalation. CCRT was strongly considered for patients presenting with extranodal extension, perineural or lymphatic invasion, and primary tumors categorized as T3 or higher. However, individual factors such as renal and hepatic function, as well as the overall general condition was considered as well. The treatment regimen for each patient was selected under multi-disciplinary team’s discretion.

Table 1.

Treatment scheme statistics (n = 29)

Regarding the radiation treatment, all patients were simulated with a CT scan using thermoplastic mask as a fixator. For definitive RT, high-risk clinical target volume (CTV) encompassed primary tumor and gross lymph nodes with 3–5 mm margin. Intermediate-risk CTV included involved lymph node level and subsequent level of the involved neck level. Low-risk CTV was established when deemed necessary. The prescribed radiation doses for the High-risk CTV ranged from 67.5–70 Gy delivered at a fraction size of 2.0–2.25 Gy. The intermediate-risk CTV received a prescription of 54–60 Gy in 1.8–2.0 Gy per fraction, while the low-risk CTV was prescribed a dose of 48 Gy in 1.6 Gy per fraction. For post-operative radiotherapy, prescription doses of 57–60 Gy, 54 Gy, and 48 Gy were administered accordingly. A 3-mm expansion from CTV was applied to create planning target volume. Treatment delivery was performed with either a 6-MV or 10-MV photon beam on a linear accelerator. The treatment delivery was verified using image-guided RT technique with on board imager or cone-beam CT, daily or twice weekly as necessary.

Results

1. Patient characteristics

The patient characteristics of 29 UNRT patients are outlined below (Table 2). The median follow-up time of this study was 68 months (range, 6 to 165). All patients were male, and the median age of the patients was 58 years (range, 45 to 82). Twenty-seven out of 29 patients were diagnosed as early T-stage, with one patient each for clinical T3 and T4. Regarding p16 and HPV status, information was missing for patients treated before 2010’s. But among the 17 patients who underwent HPV examination, majority was positive for HPV (n = 14). About a quarter of the patients were current smokers, and another quarter were former smokers. Three-dimensional conformal RT (3D-CRT) were delivered in 51.7% (n = 15), and intensity-modulated RT (IMRT) in 48.3% (n = 14).

Table 2.

Baseline patient characteristics and treatment information (n = 29)

2. Patterns of failure and survival after UNRT

After UNRT, there were no contralateral locoregional failure in this study (Fig. 1). There were six cases of ipsilateral failure in total: two in the primary mass, two in ipsilateral neck level II, one in ipsilateral neck level II and III, one in ipsilateral retropharyngeal node (Table 3). Two patients developed distant metastasis. One patient was found with a lung nodule on regular follow up at 15.4 months post-treatment, and the other was found with a metastasis in the suprasternal area at 10.7 months post-treatment with ipsilateral regional failure in level II–III neck. All patients received a maximum radiation dose of 66 Gy or above, except for one post-operative case patient who received a maximum dose of 57.5 Gy.

Fig. 1.

Patterns of failure according to subsites after unilateral radiotherapy.

Table 3.

Detailed information on all patients who developed any type of failure after unilateral radiotherapy

The 3- and 5-year actuarial overall LRFS after UNRT was 89.5% and 85.6% respectively (Fig. 2A). For DMFS, those values were 91.8% and 91.8% (Fig. 2B). For OS, those were 92.7% and 92.7% for the same time span (Fig. 2C). All three deaths that occurred during the follow-up was associated with disease progression. For those three patients, time to recurrence was 4.3, 4.8, and 10.3 months, and time to death was 5.9, 18.4, and 21.8 months.

Fig. 2.

Kaplan–Meier curves for (A) locoregional recurrence-free survival (LRFS), (B) distant metastasis-free survival (DMFS), and (C) overall survival (OS) after unilateral radiotherapy.

3. Toxicity comparison after UNRT and BNRT

A propensity score-matched cohort was constructed in order to analyze the acute and chronic toxicity of UNRT compared to BNRT (Table 4). The matched cohorts consisted of 29 patients in each group, exhibiting no discernible differences in patient demographics, tumor characteristics, or treatment specifics.

Table 4.

Baseline variables before and after propensity score matching, stratified by radiation field

Within the UNRT group, 27.5% of patients experienced grade 3–4 acute mucositis, while 48.2% encountered grade 1–2 mucositis during this phase (Table 5). The incidence of dermatitis during the acute phase was notably low, with no instances of grade 3 or higher dermatitis reported. 31% of patients experienced grade 1–2 dermatitis. Xerostomia was observed in 55.1% of patients at grades 1–2, with no occurrences of grade 3 or higher xerostomia during the acute phase.

Table 5.

Comparison of acute and chronic toxicity after unilateral and bilateral radiotherapy

In the chronic phase, only one case of chronic grade 3 dysphagia was recorded, while 17.2% of patients experienced chronic grade 1–2 dysphagia. Additionally, two patients exhibited chronic grade 1 dysgeusia. The prevalence of xerostomia increased from 55.1% during the acute phase to 68.9% in the chronic phase within the UNRT group.

When comparing the toxicity between UNRT and BNRT groups, statistically significant differences were noted only in acute xerostomia. 55.1% of UNRT patients experienced xerostomia compared to 82.7% in the BNRT group (p = 0.002). While neither group reported cases of grade 3 or higher xerostomia, acute grade 2 xerostomia occurred in 10.3% of UNRT patients versus 51.7% in the BNRT group.

The incidence of chronic xerostomia did not exhibit statistical variation between the UNRT and BNRT groups. However, a persistent trend of xerostomia was observed in both groups, as evidenced by the increase from acute to chronic phase: UNRT showed an increase from 55.1% to 68.9%, while BNRT showed a shift from 82.7% to 74.3% at grades 1–2 xerostomia.

Discussion and Conclusion

The results of our study demonstrate the safety and efficacy of UNRT for tonsillar cancer with multiple ipsilateral neck lymph nodes. The most important finding of this study is that there was no contralateral neck failure. This implies that we may be reassured about the concern regarding the omission of contralateral neck irradiation when performing UNRT, at least in select patients. Regarding radiation toxicity, only one patient experienced grade 4 toxicity, and less than a quarter of patients experienced grade 3 toxicity. Compared to the BNRT group, there were significantly less acute xerostomia.

The outcomes of this study are comparable to the results of prior researches. Regarding contralateral neck failure, more than half of the published studies also confirmed absence of contralateral recurrence in N2b subgroup [1,5–7,9]. The largest data on N2b population of 181 patients corroborated a 3% of contralateral neck failure [14].

Two of the preceding studies showed contralateral neck failure rate 10% and 17.4%, which are quite high [10,11]. One is a brief report reviewing the outcomes of UNRT in patients diagnosed with well-lateralized (not invading the tongue base or extending more than 1 cm into the soft palate), T0-3N0-2bM0 squamous cell carcinoma of tonsil at a single medical center [10]. Among 46 patients in the cohort, 23 had N2b, and all these received chemotherapy either concomitant, induction or both. With median follow-up of 41 months, four patients experienced contralateral neck recurrence (4/23, 17.4%). However, the definition of well-laterality of the study was ambiguous, and detailed information regarding patients’ characteristics, diagnostic protocols, and treatment scheme were not shown. In another retrospective study, the outcomes of a single-center treatment were also retrospectively examined [11]. Patients with well-lateralized squamous cell carcinoma, not extending to ≤1 cm from the midline, and staged as T1-2N0-2bM0, were chosen for UNRT. Primary tumors were required not to invade the base of the tongue, posterior pharyngeal wall, or more than one third of the soft palate. In this study, staging predominantly relied on MRI or CT scans, while PET scans were employed only in more challenging cases. All patients underwent radiotherapy using a hypofractionated 3D-CRT technique. Among 28 patients with N2b, four had contralateral neck failure (4/28, 14.3%). All with contralateral recurrence were HPV positive, and three out of four patients had involvement in ipsilateral lower neck regions (level 3 and/or 4). While there are no exact numbers on how many patients received PET scans during the initial evaluation, the authors acknowledged that PET scans may have been underutilized at this stage and unintentionally included N2c patients in the cohort.

In two preceding articles [7,8], extensive data on radiation-induced toxicity were presented. The acute mucositis rate, ranging from 79% to 85%, and chronic dysphagia rate, varying between 13% and 18%, observed in the UNRT group align closely with the findings of the present study. However, the chronic xerostomia rate of 70% in the UNRT group stands notably higher compared to previously reported rates of 36% to 45%. The underlying causes for these discrepancies remain unclear. One plausible explanation could be linked to the infrequent use of ICT, a factor not widely employed in the aforementioned studies. Although conclusive evidence is still lacking, certain research indicates reduced stimulated and unstimulated salivary gland functions subsequent to chemotherapy in solid tumors [15,16]. The variance in acute xerostomia incidence between the UNRT and BNRT groups might be attributable to the RT regimen. Notably, the mean radiation doses to the ipsilateral and contralateral parotid glands were 17.2 Gy and 7.7 Gy, respectively, within the UNRT group. Nonetheless, these mean doses do not significantly surpass conventionally reported doses to the parotid glands, thereby warranting caution when interpreting the relationship between radiation dosage and resultant xerostomia. An intriguing observation from this study was the escalation in xerostomia incidence from the acute to chronic phase in the UNRT group. This phenomenon could potentially be explained by the masking or underestimation of symptoms during the acute phase, leading to an apparent rise in reported xerostomia cases during the chronic phase.

The main reason behind the absence of contralateral neck failure in this study can be summarized as follows. First is the careful evaluation of tumor laterality. As in many other previous studies, well-lateralized tumors were defined as being at least 1 cm away from the midline of the neck and was meticulously reviewed in all available imaging modalities. Any tumors in close proximity to midline were excluded from receiving UNRT. Midline involvement of the primary mass has been shown to be the most important prognostic factor of contralateral neck failure in meta-analysis [17]. This is specified by the result of another study which demonstrated that all contralateral neck failure after UNRT in T1-T4, N0-N3 tonsillar cancer occurred due to primary mass invading the 1 cm margin from midline or contralateral second primary tumor [6].

Second is the active employment of PET scan during the diagnosis. In this study, 25 out of 29 patients received PET scan before receiving RT, which may have helped discover undetected lymph nodes in contralateral neck. The specificity of PET scan has been well established in head and neck cancers, and PET scan is generally considered a more sensitive modality compared to CT or MRI [18]. The importance of initial workup using PET scan is reinforced by the fact that occult contralateral metastasis is not uncommon. A meta-analysis that looked into clinically contralateral lymph node negative oropharyngeal cancer concluded that the occult contralateral lymph node metastasis rate was 15.4% after bilateral neck dissection, for clinically ipsilateral lymph node positive cancers [19].

A possible limitation to this study may be the retrospective nature of the research design, and the small number of patients enrolled. Also, p16 status was not available for 41.4% of the patients, raising concerns about the integrity of the data. Considering that most of the patients with unknown p16 status would be p16 positive, the applicability of UNRT might need investigation in the p16 negative population as well. Furthermore, patients treated within a 20 years of time span were incorporated in the analysis, and thus some unarticulated changes in the medical environment need to be taken into account when interpreting the findings. The mainstay of radiation treatment technique has changed from 3D-CRT to IMRT, the imaging technique and diagnostic scheme have seen a big progress, and the decision makings have changed especially regarding multidisciplinary approaches for complicated cases, as well as surgery or chemotherapy standards. Therefore, a further study in prospective setting with full data regarding p16 status is warranted to draw a more definitive conclusion. Also, a treatment strategy for p16 negative population still remains to be answered, although small in numbers.

Nonetheless, this study, to best of our knowledge, is the first to present the clinical outcomes and toxicity data of pure N2b tonsillar cancer after UNRT. Moreover, all of the treatment plans were designed by two physicians belonging to two affiliated tertiary referral hospitals that share the same treatment policy. Thus the coherency of the data behind this study serves as a strong foundation for advocating for UNRT in tonsillar cancer with multiple ipsilateral neck lymph nodes.

Article information Continued