Neck Dissection for Salivary Gland Malignancies

18 Neck Dissection for Salivary Gland Malignancies


Vincent Vander Poorten


Abstract


A minority of patients with salivary gland carcinomas presents with clinically or radiologically obvious cervical lymph node metastasis. These patients’ prognosis following treatment is significantly worse than that of patients with a clinically uninvolved neck. Adequate management of this aspect of the disease is thus of vital importance. There is little controversy on the management of clinically evident metastases. Primary surgery aiming at removal of all obvious suspected nodal disease will almost invariably result in pathological confirmation of the preoperative suspicion, and then entail postoperative radiotherapy, which, for all sites, improves locoregional control and survival. The optimal treatment of the clinically and radiologically uninvolved neck does continue to provoke discussion. It is not the question of whether or not to treat the neck that is the problem, as there is wide agreement on the tumor and the patient factors that make it more likely that the lymph nodes harbor occult metastases. When these factors are present, the neck should be electively treated. The major discussion, however, revolves around the best treatment strategy to use once the decision to proceed with neck treatment is made. On the one hand, there are the clinicians who prefer to treat those patients with an elective neck dissection, followed by postoperative radiotherapy on indication. On the other hand, others favor radiotherapy over elective neck dissection in this scenario. The rationale behind these two strategies is summarized.


Keywords: salivary gland carcinoma, clinically negative neck, elective neck dissection, elective radiotherapy, clinically positive neck, therapeutic neck dissection, neck metastasis


18.1 Introduction


Globally, the incidence of salivary gland carcinomas is 0.4 to 13.5 cases per 100,000 people per year; in the United States, the incidence rate is 1 per 100,000 people per year. The European incidences appear to be lower, with Belgium, the Netherlands, the United Kingdom, and Finland reporting around 0.6 to 0.7 cases per 100,000 people per year. A Danish population-based study reported a crude incidence of 1.1 per 100,000 people per year.1


Around 70% of these carcinomas arise in the largest gland, the parotid,2 and 10 to 25% of salivary carcinomas arise in the minor salivary glands.3 The rest are submandibular carcinomas, with sublingual carcinomas being very rare.


Prognostic indicators explaining the observed variability in chance of cure following treatment include patient, tumor, and treatment characteristics. Tumor characteristics with prognostic impact include anatomic site of the affected salivary gland, histotype of the tumor (22 malignant types in the most recent 2017 WHO classification),4 TNM stage,5 and specific growth characteristics (perineural/intraneural invasion, lymphatic invasion, surgical margins, extraglandular extension), which can be observed histopathologically following resection. One of the tumor-related factors with a strong independent prognostic impact is regional metastasis, as reflected in the clinical and pathological N-classification.6


Evaluation of the neck is mandatory whenever a salivary malignancy is suspected on clinical grounds. Available options, both for the primary and the neck, are ultrasound, which allows for ultrasound-guided fine-needle aspiration, CT scanning (image Fig. 18.1), which is superior for bone detail, and MR imaging, which is strongly advised when tumor mobility is impaired, and which has a superior soft-tissue detail, including visualization of perineural extension (image Fig. 18.2).


In clinically suspected or fine-needle aspiration cytology (FNAC) proven malignancy, before embarking on locoregional therapy, positron emission tomography (PET) with or without CT co-localization (PET-CT) is mainly important for detecting disease recurrence and to exclude (gross) distant disease. It is important to note that this modality fails in differentiating benign from malignant disease, as Warthin’s tumors and pleomorphic adenomas show an increased uptake (high false-positive rate), and not infrequently, malignant tumors are not fluorodeoxyglucose-avid (high false-negative rate).1,3,7


18.2 Treatment of the Neck According to the Gland of Origin of the Primary Tumor


18.2.1 Parotid Carcinoma The Clinically Positive Neck


Regional metastasis is clinically and/or radiologically evident at presentation (clinically positive [cN + ] disease) in 14 to 29% of patients.6,8 This percentage increases in high-grade tumors and advanced T-status tumors.9,10 The lymph node levels most frequently involved are levels II, III, and IV.8,11 Parotid cancer–related cN + disease requires a (modified) radical neck dissection, removing levels I to V.12 This therapeutic comprehensive neck dissection implies radicality toward nonlymphatic structures (nerve XI, jugular vein, or sternocleidomastoid muscle) depending on proximity of or involvement by lymph node metastases13 (image Fig. 18.3a, image Fig. 18.4). Recent studies confirmed this “old knowledge”; rates of pathologically positive (pN +) involvement in a recent study from Memorial Sloan Kettering Cancer Center were 52% in level I, 77% in level II, 73% in level III, 53% in level IV, and 40% in level V.14 In a comparable study from Korea, rates of pN + involvement were 43% in level I, 90% in level II, 40% in level III, 57% in level IV, and still 43% in level V.15




As such, clinical neck disease implies a well-accepted negative prognostic value,6,16 but recent reports revealed the independent negative prognostic impact of an increasing “lymph-node density,” which is the ratio of the number of metastatic nodes to the total number of lymph nodes removed.17,18


A sometimes-overlooked problem is the deep lobe intraparotid lymph nodes. A significant proportion (53–65%) of patients with pN + disease on neck dissection will also have metastatic deposits in the “first echelon” intraparotid lymph nodes.8,11 When a neck dissection is needed for removal of cN + disease, it seems logical and consequent that a deep lobe parotidectomy is performed to address this problem. There is usually no discussion on performing this type of “total parotidectomy” for large tumors, deep lobe tumors, or tumors that have already caused a seventh cranial nerve (CN VII) paralysis (image Fig. 18.4), but the controversy surrounds the early-stage tumors with normal facial nerve function that need a therapeutic neck dissection19 (image Fig. 18.3a). Authors from Mayo Clinic recommend performing a deep-lobe parotidectomy in high-grade tumors, especially if an intraparotid node in the specimen of the initially performed superficial parotidectomy is positive on frozen section20 (image Fig. 18.3 a, b). The practical problem here is that preoperative grading (based on FNAC) is infrequently available for parotid cancer patients. The Köln group published a series of 142 patients in 2008, where, in their total parotidectomy specimens, 1 to 11 parotid lymph nodes were retrieved. Eighty percent of these parotid nodes were involved in cN0/pN + patients.11 While there is no direct evidence that resection of these nodes increases locoregional control in salivary gland cancer, this evidence is available in skin cancer, metastatic to the parotid. In this disease, a 20% local recurrence rate—the majority of which occurred in the parotid bed—was observed in patients treated with superficial parotidectomy (the deep lobe remaining in situ), despite being treated with postoperative radiotherapy.21,22


It is well accepted that pN + patients with salivary gland cancer need postoperative radiotherapy to the parotid bed and the ipsilateral neck. In this setting, adjuvant radiation not only doubles the rate of locoregional control, but also improves survival.23,24 Two recent reports documented the benefit of a postoperative, platinum-based, concomitant chemoradiation scheme for high-risk major salivary gland carcinomas, and the approach certainly merits further research.25



image

Fig. 18.3 (a) The same patient as in image Fig. 18.2, following a retrograde cranial nerve (CN) VII dissection. Intraoperatively, the facial nerve found to be displaced anterosuperiorly, but ultimately uninvolved. A total conservative parotidectomy was possible, including a deep lobe parotidectomy, in conjunction with a therapeutic neck dissection (ND I–V, sternocleidomastoid, CN XI).12 This picture was taken before removal of the last part of the deep lobe from underneath the main trunk of the facial nerve. (b) Another patient with total conservative parotidectomy in conjunction with a selective neck dissection of level II.


The Clinically Negative Neck

Observation, Elective Neck Dissection with or without Postoperative Radiation, or Elective Neck Irradiation

The rates of pN + disease, in patients who are defined as cN0 following clinical examination and high-quality imaging, are between 12 and 49%. This variation between series derives from the different tendency to regional metastasis of the plethora of salivary gland cancer histotypes, within which there even exist different grades, again determining metastatic behavior.8,10,11,26,27,28


For patients presenting with a cN0 parotid carcinoma, we usually try to estimate the theoretical risk of pN + disease, from the presence of established risk factors for occult neck disease. We then decide to treat the cN0 neck when the combined presence of different risk factors implies a probability that exceeds the threshold of 15 to 20%.


Among identified risk factors that predict micrometastatic disease in cN0 patients are clinical and histopathological factors. Clinical factors are age in the sixth decade or older, presence of pain, seventh nerve dysfunction, and locally advanced disease as reflected in the T-status. Histological factors include histotype and grade, extraglandular soft-tissue invasion, and lymphatic invasion.8,26,28,29,30


Histotypes that imply a high prevalence (> 50%) of occult nodal disease are salivary duct carcinoma (SDC), undifferentiated carcinoma (UC), adenocarcinoma not otherwise specified (ACNOS), high-grade mucoepidermoid carcinoma (HG-MEC), and squamous cell carcinoma (SCC).8,10 Textbook knowledge classically teaches that parotid adenoid cystic carcinoma (AdCC) has a low tendency to regional metastasis. Recently, however, a scrutinized analysis of the available literature revealed a 14.5% lymph node metastasis rate in parotid cN0 AdCC.31 The same authors identified a high-grade subgroup in parotid AdCC (AdCC-HGT) that implies a pN + rate of up to 57%.32 Acinic cell carcinoma (AcCC) and low-grade MEC are generally considered to have a low rate of pN + disease; nevertheless, authors routinely performing elective neck dissection (END) in patients with these subtypes also report higher-than-expected rates of occult nodal disease.10,11 Furthermore, also in AcCC, nowadays there is a high-grade subtype, which implies a higher risk.33 To complicate the matter, early-stage cancers and low-grade cancers can also present with cN0pN + disease.14,28,34 Stenner et al performed END in T1N0 and T2N0 patients and found a pN + rate of 21%.14,28,34


Feb 14, 2020 | Posted by in OTOLARYNGOLOGY | Comments Off on Neck Dissection for Salivary Gland Malignancies

Full access? Get Clinical Tree

Get Clinical Tree app for offline access