Tumors of the Salivary Glands

Published on 09/04/2015 by admin

Filed under Hematology, Oncology and Palliative Medicine

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 7844 times

32 Tumors of the Salivary Glands

Malignant salivary gland tumors are uncommon head and neck cancers. They share many characteristics, yet also have features that are uniquely attributable to their respective locations and subtypes. The two main categories are major and minor salivary gland tumors. They are presented separately to distinguish the presentations, natural histories, and therapeutic approaches between and within these groups.

Major Salivary Gland Tumors

Epidemiology

Tumors of the major salivary glands are uncommon and account for approximately 2% to 7% of all head and neck neoplasms. In adults, the annual incidence is approximately one to four new cases per 100,000 individuals. The parotid gland is overwhelmingly the most frequent gland affected; cancer in the parotid gland is much more common than submandibular and sublingual gland cancers by a factor of 10 and 100, respectively.

The ratio of benign to malignant tumors varies by site, and there is an inverse relationship between the size of the salivary gland and its incidence of malignancy: (1) parotid gland: 80% benign (primarily pleomorphic adenomas), 20% malignant; (2) submandibular gland: 50% benign, 50% malignant; (3) sublingual gland: less than 40% benign, more than 60% malignant. The average age for benign neoplasms is 40 years, whereas that for malignant tumors is 55 years. The incidence is equivalent for men and women.

The causal factors of these malignant tumors is unknown. The usual risk factors for squamous cell carcinomas of the head and neck—excessive use of tobacco and alcohol—are not clearly proven causal agents for malignant salivary gland tumors. Radiation-induced salivary gland malignancies have been reported in the literature in association with patients who had been irradiated to the head and neck region for benign conditions (e.g., acne, tinea capitis, infected tonsils) and in survivors of the atomic bomb in Hiroshima and Nagasaki. Although an association with female breast cancer has been reported, there is a lack of firm epidemiologic data to support this contention. Nutritional deficiency as a possible cause has been reported in the Arctic Inuit peoples who have a diet low in vitamins A and C. Infection has been cited as a possible cause in whites with the Epstein-Barr virus. Other possible epidemiologic causes include alcohol use, hair dye, and a higher education level.

Anatomy

The major salivary glands comprise the parotid, submandibular, and sublingual salivary glands (Fig. 32-1). Each are paired glands that have different secretory functions: (1) parotid gland—serous; (2) submandibular gland—seromucous; (3) sublingual gland—mucous.

image

FIGURE 32-1 • The anatomic locations of major salivary glands: parotid, submandibular, sublingual.

(From Spiro JD, Spiro RM: Salivary tumors. In Shah J (ed): Cancer of the head and neck, London, 2001, BC Decker, p 241.)

Parotid Gland

The parotid glands are the largest of the salivary glands (see Fig. 32-1 and Fig. 32-2). These paired glands are surrounded by a discrete capsule and have the following anatomic landmarks: (1) anterior—wraps around the ascending mandibular ramus anterior to the tragus and extends toward the anterior margin of the masseter muscle; (2) posterior—extends from the angle of the mandible under the earlobe toward the mastoid tip; (3) superior—extends to the inferior aspect of the zygoma at the temporomandibular joint level; (4) inferior—extends to the inferior aspect of the angle of the mandible; (5) medial—borders the parapharyngeal-base of skull; (6) lateral—located below the skin of the preauricular cheek–upper neck. The gland is divided into two lobes by the path of the facial nerve, which exits from the stylomastoid foramen (Fig. 32-3): (1) superficial lobe—accounts for 80% of the gland and is located just lateral to the facial nerve; (2) deep lobe—accounts for 20% of the gland and is located medial to the facial nerve and adjacent to the medial aspect of the angle of the mandible and is connected to the superficial lobe by the isthmus (Fig. 32-4). The deep lobe is in close anatomic proximity to the internal carotid artery; the internal jugular vein; the cervical sympathetic chain; and the cranial nerves IX, X, XI, and XII (Table 32-1).

image image

FIGURE 32-3 • A, Pathway of the facial nerve (cranial nerve VII) as it exits the stylomastoid foramen and then branches out into various anatomic regions. B, Note the presence of the deep lobe of the parotid gland just medial to the facial nerve. C, The facial nerve branches have several distribution variations.

(A from Leblanc A: The cranial nerves, ed 2, Berlin, 1995, Springer-Verlag, p 184. B from Crumley RL, Kelley TF: Rehabilitation of the patient with tumors of the salivary glands. In Thawley SE, Panje WR, Batsakis JG, et al (eds): Comprehensive management of head and neck tumor, vol 2, Philadelphia, 1987, WB Saunders, p 1202. C from Som PM, Curtin HD: Head and neck imaging, ed 3, St Louis, 1996, Mosby, p 829.)

Table 32-1 The Anatomic Relationship of the Major Salivary Glands and Cranial Nerves

Major Salivary Gland Adjacent Cranial Nerve Base of Skull Foramen
Parotid    
Superficial/deep lobes Facial nerve (CN VII) Stylomastoid foramen
Deep lobe Glossopharyngeal nerve (CN IX) Jugular foramen
  Vagus nerve (CN X) Jugular foramen
  Accessory nerve (CN XI) Jugular foramen
  Hypoglossal nerve (CN XII) Hypoglossal canal
Submandibular Lingual nerve (CNV3) Foramen ovale
  Facial nerve (CN VII): Stylomastoid foramen
  Mandibular/cervical branches  
  Hypoglossal nerve (CN XII) Hypoglossal canal
Sublingual Lingual nerve (CN V3) Foramen ovale

CN, Cranial nerve.

The parotid gland drains into the oral cavity through the Stensen duct (Fig. 32-5). This duct runs from the upper anterior third of the parotid gland and exits through the buccal mucosa by the second molar. The lymphatic drainage of the parotid gland progresses in an orderly fashion. Initially, it drains into the periparotid (Fig. 32-6) and intraparotid (Fig. 32-7) nodes that comprise two groups located within the fascia of the gland in two respective locations: between the gland and the superficial fascia and within the parenchyma. It then drains into the submandibular nodes (level IB), the upper (level II) and mid (level III) cervical nodes, and sometimes to the retropharyngeal nodes. Drainage to the contralateral nodes is very rare. However, it must be considered if the primary tumor extends across the midline or if there is massive ipsilateral cervical node involvement, which may disrupt the lymphatic pathways and cause spread to the opposite side.

image

FIGURE 32-6 • Periparotid lymph nodes are located on the surface of the superficial lobe but within the fascia.

(From Hoffman H, Funk G, Endres D: Evaluation and surgical treatment of tumors of the salivary glands. In Thawley SE, Panje WR, Batsakis JG, et al (eds): Comprehensive management of head and neck tumors, vol 2, Philadelphia, 1987, WB Saunders, p 1112.)

image

FIGURE 32-7 • Intraparotid nodes are located within the parenchyma of the gland.

(From Hoffman H, Funk G, Endres D: Evaluation and surgical treatment of tumors of the salivary glands. In Thawley SE, Panje WR, Batsakis JG, et al (eds): Comprehensive management of head and neck tumors, vol 2, Philadelphia, 1987, WB Saunders, p 1112.)

Adjacent to the superficial surface of the parotid gland fascia and anterior to the tragus are the preauricular nodes, which measure less than 3 mm in size and lie within the subcutaneous fat. These lymph nodes receive the lymphatic drainage from the dermis of the upper face, temple of the scalp, eye, nose, and ear. Their involvement is usually associated with skin carcinomas and melanomas in these regions as well as lymphomas. However, primary parotid malignancies do not commonly metastasize to these nodes. Secondary extension into the parotid gland and adjacent facial nerve can sometimes occur. These nodes can then drain into either the inferior parotid nodes (superficial lymph nodes) or into the jugular chain of nodes (level II).

Occasionally, there is an accessory parotid lobe (Fig. 32-8). This lobe has been reported in 21% of normal adult cadavers.1 However, in a review of 2261 patients with parotid lesions at Memorial Sloan-Kettering Cancer Center (MSKCC) between 1939 and 1978, only 23 patients (1%) had an accessory parotid gland.2 Accessory parotid glands are located either cephalad or lateral to the anterior aspect of the Stensen duct and lie over the anterior aspect of the masseter muscle but are separate from the superficial lobe of the parotid gland. Clinically, they are adjacent to the Stensen duct, which is located midway along a line drawn between the tragus of the ear to the lateral upper lip. There can be 1 to 10 tributary ducts from the accessory lobe that will empty into the Stensen duct. This lobe can be clinically significant in that it may be the primary site of a malignant tumor that presents as an asymptomatic swelling in the mid cheek region.

image

FIGURE 32-8 • A, The accessory parotid lobe is located in the anterior aspect of the Stensen duct. B, Normal parotid gland and the Stensen duct.

(From Johnson FE, Spiro RH: Tumors arising in accessory parotid tissue, Am J Surg 138:242, 1979.)

Submandibular Gland

The submandibular gland (see Fig. 32-1) is 25% of the size of the parotid gland and measures 3 to 4 cm. These paired glands are surrounded by a capsule and are located in the upper anterior triangle of the neck. They are medial to the proximal half of the mandible (Fig. 32-9). The majority of the gland is over the external surface of the mylohyoid muscle, which forms the muscular floor of mouth in the region between the insertion of the muscle and the mandible and which divides the gland into contiguous superficial and deep portions (Fig. 32-10). The posterior aspect is anterior to but near the lower anterior margin of the parotid gland. The inferior aspect can extend caudad a fair distance and approaches the level of the hyoid bone. Of clinical importance is the fact that the submandibular gland is lateral to and abuts the lingual nerve (cranial nerve V3) and hypoglossal nerve (cranial nerve XII) and is medial to the marginal mandibular and cervical branches of the facial nerve (cranial nerve VII) (see Table 32-1).

image

FIGURE 32-10 • Anatomic relationships between the submandibular gland and sublingual gland and the adjacent musculature and cranial nerves.

(A and B from Spiro JD, Spiro RM: Salivary tumors. In Shah J (ed): Cancer of the head and neck, London, 2001, BC Decker, p 241.)

The glands drain into the oral cavity through the Wharton duct. This 5-cm duct courses through the niche between the mylohyoid and hypoglossus muscles and emerges in the anterior floor of the mouth toward the midline.

The lymphatic drainage of the submandibular glands progresses in an orderly fashion. Initially it drains into the adjacent submandibular nodes (level IB) as there are no intraglandular parenchymal nodes. It then drains into the upper (level II) and mid (level III) cervical nodes. Drainage to the contralateral nodes is rare.

Sublingual Gland

The paired sublingual salivary glands (see Fig. 32-1) are the smallest of the major salivary glands and are approximately 10% of the size of parotid glands. These glands do not have a discrete capsule. They are located in the anterior floor of mouth adjacent to the medial aspect of the mandible and occupy a submucosal position. Sublingual salivary glands are adjacent and superior to the mylohyoid muscle in the area of the sublingual depression. They occupy a position just medial to the inner surface of the mandible near the mental symphysis (see Fig. 32-10). It is clinically important to note that the lingual nerve (cranial nerve V3) courses adjacent to the sublingual gland (see Table 32-1).

Primary malignant tumors of the sublingual glands are rare; however, any lesion in this area must be considered malignant until proven otherwise. It may be difficult to clinically rule out a primary squamous cell carcinoma of the anterior floor of mouth in this region.

These salivary glands drain into the oral cavity via 5 to 15 ducts (Rivinus ducts) that empty into the sublingual fold of the floor of mouth. Also, these ducts may anastomose to create a larger duct known as the Bartholin duct, which attaches to the submandibular duct.

The lymphatic drainage of the sublingual glands progresses in a predictable, orderly fashion, flowing initially to the submental (level IA) and submandibular nodes (level IB) and then to the deep jugular nodes (level II).

Pathologic Conditions

Malignant Tumors

Histopathologic Types

During the past several years, the World Health Organization has expanded its list of suggested histopathologic typing to include a very comprehensive and varied group of malignant tumors:

The relative incidence of the different histologic types according to gland of origin is detailed in Table 32-2.

Table 32-2 Distribution of Histologic Types of Major Salivary Gland Cancer

Type Percentage
Parotid (n = 1778 cases)  
Mucoepidermoid 32
Adenocarcinoma 16
Malignant mixed 14
Adenoid cystic 11
Acinic 11
Undifferentiated and squamous 16
Submandibular (n = 383 cases)  
Adenoid cystic 41
Acinic 17
Mucoepidermoid 12
Malignant mixed 10
Undifferentiated 9
Squamous 9
Adenocarcinoma 2

Data from Memorial Sloan-Kettering Cancer Center.1

Adenoid Cystic Carcinoma

This is the predominant malignant histologic type in submandibular and minor salivary gland tumors.3 The appearance (Fig. 32-12) can vary from a cribriform pattern (differentiated) to a mixed cribriform pattern and solid features (moderately differentiated) to solid features (undifferentiated). Some authors have observed that the solid variety can have a more malignant behavior. The natural history can be varied, ranging from a matter of months to 20 years or more. The first evidence of recurrence can be 20 years after diagnosis, making it very difficult to ever determine that an individual patient is cured. Lymph node spread is distinctly uncommon (<5%). Adenoid cystic tumors can cause perineural spread, which may track along the pathways of the cranial nerves to the base of skull. This is important in planning radiation treatment. Many patients (ultimately up to 40%) will develop pulmonary metastases. Because prolonged survival can occur (10 to 20 years) with pulmonary metastases, the primary site must be managed adequately despite the presence of metastatic disease. An example of the management of the primary site in a patient with adenoid cystic cancer and lung metastases is illustrated in Fig. 32-13.

Histologic Grade

The histologic grade or cellular differentiation of the tumor is one of several important factors affecting patient survival. The other indicators include the histologic diagnosis, site, size, degree of fixation, or local extension and facial nerve involvement. Also, regional lymph node and distant metastases are of major importance relating to survival. However, the histologic grade of the tumor is applicable only for adenocarcinoma not otherwise specified and mucoepidermoid carcinomas, or when either of these is the malignant component of carcinoma in pleomorphic adenoma. For all of the other subtypes, it is the actual histologic type that defines the grade. For instance, salivary duct carcinoma is high-grade whereas basal cell adenocarcinoma is low-grade by definition.

Clinical Presentation and Evaluation

Parotid

Primary Site

A thorough history and carefully detailed physical examination are always crucial first steps in the evaluation of patients with a parotid mass. The differential diagnoses of a parotid mass include a malignant tumor as well as several types of benign causes (Table 32-3). Parotid malignancy is usually an asymptomatic mass. However, as the lesion progresses and enlarges, episodic pain occurs in 10% to 20% of cases; subsequently, significant pain can result and become constant. Patients can present with complaints of an inability to move one side of the face (cranial nerve VII), a shoulder (cranial nerve IX), or one side of the tongue (cranial nerve XII) as the adjacent cranial nerves become involved by the malignant tumor. Clinical presentations can vary according to the histopathologic findings (Table 32-4).

A careful and methodic examination of the patient can provide valuable information regarding the stage of the tumor. The mandibular opening should be measured to rule out trismus resulting from tumor involvement of the pterygoid plates. Bimanual palpation of the tumor will provide data regarding its dimensions, texture, and mobility. The overlying soft tissues, adjacent mandible, and ipsilateral external auditory canal are sites where tumors can infiltrate and should be carefully evaluated.

The functional status of the adjacent cranial nerves must be documented. This always includes evaluation of the facial nerve (cranial nerve VII), which is involved in 2% to 14% of cases. However, if the deep lobe of the parotid is involved, cranial nerves IX, X, XI, and XII may be affected.

Workup with computed tomography (CT) or magnetic resonance imaging (MRI) can be helpful in selected cases. Slices should cover from above the ears to below the clavicles, thus imaging the base of skull, parotid, retropharyngeal nodes, cervical nodes, and supraclavicular nodes. There are four indications for a pretreatment CT or MRI in parotid gland tumors4: (1) deep lobe parotid tumors, (2) neurologically symptomatic tumors, (3) recurrent tumors, and (4) large tumors.

CT and MRI cannot differentiate a benign from a malignant tumor and thus are not routinely used for small, mobile, circumscribed lesions. An MRI scan with gadolinium best delineates deep-tissue infiltration or perineural extent of the tumor with respect to the facial nerve (cranial VII).

Both benign and malignant salivary tumors may appear well circumscribed on CT or MRI. This may cause one to underestimate the actual extent of infiltration of the cancer.

In the initial evaluation of salivary gland tumors, 2-fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) scans must be judiciously employed. The salivary gland takes up FDG from the blood and subsequently excretes it in the saliva. There is normal variability in the degree of such uptake, however. Parotid and submandibular glands generally have a mild to moderate degree of physiologic FDG uptake. Salivary gland tumors, whether benign or malignant, may or may not show a significant uptake of FDG. Tumors with low metabolic activity may be associated with minimal or no uptake; this increases the risk for a false-negative result. Benign (e.g., Warthin tumor, pleomorphic adenoma) and malignant (e.g., primary parotid lymphoma) parotid tumors with uptake are indistinguishable and may give rise to a high false-positive rate. Also, a malignancy may cause bilateral FDG uptake in parotid or submandibular glands, which can mimic a normal physiologic pattern. There are some salivary gland malignancies that have little or no FDG uptake. The use of a PET scan in the preoperative setting has not been useful. However, once a histologic diagnosis has been made establishing a malignancy of a salivary gland, a PET scan may of be helpful for evaluation of distant metastasis. In general, one can consider using a PET scan in high-grade salivary gland malignancies.

The initial diagnostic procedure in a patient suspected of having a parotid gland malignancy can be a fine-needle aspiration (FNA) biopsy. This is a safe technique that is not associated with tumor spread along the fine-gauge needle tract, which can occur with large-bore needles.

The harvested cells allow for cytopathologic diagnosis based on the appearance of individual cells and cell clusters. It is important to note, however, that such a limited sampling of cells may not provide an accurate analysis of the full morphologic spectrum and this can be associated with major difficulties in cytologic diagnosis. Different salivary tumors that are both clinically and biologically different may appear quite similar on a cellular evaluation. Also, some malignant salivary tumors microscopically do not look frankly malignant.

In general, 75% to 80% of parotid masses are benign and 20% to 25% are malignant. An FNA biopsy is an accurate diagnostic technique with overall sensitivities of greater than 90% and specificities of greater than 95%.510 FNA biopsies are as accurate as frozen section analysis in parotid masses. Note that a malignant salivary gland tumor is more likely to be inaccurately diagnosed as a benign lesion than an adenoma is to be erroneously called a malignant lesion. Salivary gland lesions that are particularly associated with difficulties in achieving an accurate diagnosis include acinic cell carcinomas (false-negative confusion for a benign process), monomorphic adenomas (false-positive confusion with an adenoid cystic carcinoma), and lymphoid lesions (both false-negative and false-positive diagnoses).

If a malignancy is thus diagnosed, one can assess if it is a primary parotid lesion or if it is a metastatic lesion to the parotid gland from another primary site. In the former situation, appropriate radiographic evaluation and surgical planning can occur and important discussions can be held with the patient. For metastatic lesions to the parotid gland from another primary site, appropriate workup can be initiated to determine the anatomic origin. The most frequent primary malignant sites include skin areas with lymphatic drainage to the parotid gland lymph nodes (80%), the thyroid gland, kidney, and breast.

It should be noted that the routine use of an FNA biopsy in this fashion is controversial because some surgeons believe that it would not change their standard approach, a parotidectomy, and thus would add little to the management of such tumors. Some institutions reserve the FNA biopsy for inoperable or recurrent lesions. However, incisional or excisional biopsies are never performed to establish a tissue diagnosis because such procedures can increase the likelihood of recurrence, the risk of injury to the facial nerve, and subsequent surgical morbidity (because a wide removal of the biopsy site must then be added to the cancer surgery).

Lymph Nodes

Although the overall risk (18%)11 of lymphatic spread is less common than for mucosal squamous cell carcinomas of the head and neck, it is still essential to carefully evaluate the regional lymph nodes. The influence of tumor histologic features on the frequency of clinically involved nodes at presentation in 474 previously untreated major salivary gland cancers was reviewed by Armstrong from MSKCC.12 These data are presented in Table 32-5. The risk of nodal involvement increases with grade and size. However, adenoid cystic cancer, despite often being aggressive histologically and large in size, had only a 2% rate of clinically evident metastases.

Table 32-5 Clinically Involved Nodes at Presentation in Major Salivary Gland Cancers: The Influence of Tumor Histology

Histologic Subtypes Number (%)
Anaplastic 6/7 (86)
Epidermoid 6/28 (21)
Adenocarcinoma 11/49 (22)
Mucoepidermoid 30/209 (14)
Malignant mixed 11/69 (16)
Acinic 2/55 (2)
Adenoid cystic 1/55 (2)
Oncocytoma 0/2 (0)
Total 67/474 (14)

Data from Armstrong et al.12

In the series reported by Armstrong and associates,12 overall clinically occult, pathologically positive nodes occurred in 12% (47/407). By univariate analysis, several factors appeared to predict the risk of occult metastases, but multivariate analysis revealed that only size and grade were significant risk factors. Tumors of 4 cm or more had a 20% (32/164) risk of occult metastases, compared with a 4% (9/220) risk for smaller tumors (P < 0.00001). High-grade tumors (regardless of histologic type) had a 49% (29/59) risk of occult metastases, compared with a 7% (15/221) risk for intermediate or low-grade tumors (P < 0.00001). The effects of tumor histologic anatomy and grade on occult nodal metastases are detailed in Tables 32-6 and 32-7.

Table 32-6 Effect of Tumor Histology on Occult Lymph Node Involvement in Major Salivary Gland Cancers

Histologic Subtypes Number (%)
Epidermoid 9/22 (41)
Adenocarcinoma 7/38 (18)
Mucoepidermoid 26/179 (14)
Acinic 2/53 (4)
Adenoid cystic 2/54 (4)
Anaplastic 1/1  
Malignant mixed 0/58  
Oncocytoma 0/2  
Total 47/407 (12)

Data from Armstrong et al.12

Table 32-7 Effect of Tumor Grade on Occult Lymph Node Involvement in Major Salivary Gland Cancer

Tumor Grade Number (%)
Low 2/125 (2)
Intermediate 13/96 (14)
High 29/59 (49)

Data from Armstrong et al.12

It is rare for a low-grade tumor to metastasize to the regional nodes. All high-grade lesions carry a significant risk for nodal metastasis irrespective of the histologic type. High–T stage lesions are associated with an increased incidence of nodal metastasis. Submandibular or sublingual primary salivary malignancies have an increased risk for nodal spread compared with parotid salivary malignancies. The histologic type can influence the risk for nodal metastasis as well. Mucoepidermoid carcinoma, squamous cell carcinoma, and undifferentiated carcinoma are associated with a high incidence of nodal metastasis; however, adenoid cystic carcinoma, malignant mixed carcinoma, and acinic cell carcinoma have a low incidence (see Tables 32-5 and 32-6).

The physical examination includes palpation of the preauricular-parotid region (preauricular, intraparotid, and periparotid nodes); submandibular nodes (level IB); upper cervical nodes, especially the superior deep jugular nodes and jugulodigastric nodes (level II); and mid cervical nodes (level III). The ipsilateral tonsillar region should be inspected for fullness, which may indicate gross retropharyngeal nodal involvement, particularly when the deep lobe of the parotid is infiltrated by tumor. Any adenopathy is documented by location, size, number, texture, tenderness, and mobility. The initial radiographic evaluation of the parotid gland may also provide important clinical information regarding the nodal status.

Submandibular Gland

Primary Site

A submandibular salivary gland tumor usually presents as an asymptomatic mass but can be associated with episodic pain in 6% to 7% of cases. However, as the lesion progresses and enlarges, significant pain can develop and become constant. Patients can infrequently present with complaints of decreased sensation in the ipsilateral lower teeth, gums, and lower lip (cranial nerve V3); an inability to move the ipsilateral oral tongue (cranial nerve XII); or facial weakness associated with the marginal mandibular or cervical branch of the facial nerve (cranial VII) as the adjacent cranial nerves become involved with the malignant tumor (14%).

A careful and methodic examination of the patient can provide valuable information regarding the stage of the tumor. Bimanual palpation of the lateral floor of mouth area will provide data regarding tumor dimension, texture, and mobility. The overlying soft tissues and adjacent mandible are sites where tumors can infiltrate and should be carefully evaluated. The functional status of the adjacent cranial nerves must be documented. This includes the marginal mandibular or cervical branch of the facial nerve (cranial nerve VII) or both, the lingual nerve (cranial nerve V3), or the ipsilateral hypoglossal nerve (cranial nerve XII).

Submandibular gland tumors are seldom imaged because they usually present early and are easily palpated. Note that large primary parotid gland lesions can push down into the submandibular space, thus giving the erroneous clinical impression that there is enlargement of the submandibular gland. If this is suspected in a particular case, one must radiographically evaluate the parotid gland with a CT or MRI. An obstructing calculus within the Wharton duct can cause retrograde pressure and thus result in enlargement of the submandibular gland. Enlargement of upper cervical lymph nodes can displace the submandibular gland, thus forming a “pseudomass.”

Based on the particular characteristics of a submandibular tumor, an MRI scan may be indicated to assist in evaluating for deep-tissue infiltration or perineural extent of the tumor with respect to the lingual nerve (cranial nerve V3), the hypoglossal nerve (cranial XII), and the marginal mandibular and cervical branches of the facial nerve (cranial nerve VII).

CTs or MRIs cannot generally differentiate between a benign or malignant submandibular mass and thus are not routinely used for small, mobile, circumscribed tumors. However, these studies can determine whether the mass is within the submandibular gland or extrinsic to it (e.g., dermoid, neural lesion, ranula, thyroglossal duct cyst, lymph node). Both benign and malignant salivary tumors may appear well circumscribed on a CT or MRI. This may cause one to underestimate the actual extent of infiltration of the cancer.

Concerns regarding the use of an FDG PET scan are similar to those discussed previously for parotid salivary gland malignancies.

The initial biopsy of a suspicious submandibular salivary gland mass can be an FNA biopsy. Such a procedure is important because only a minority of submandibular masses are primary tumors originating in that gland. Note that, in general, a submandibular gland mass has a 50% chance of being malignant. An FNA biopsy is useful only if it reveals a malignancy. It is important to recognize that the mass may be an enlarged submandibular lymph node (level IB), and a careful evaluation of the head and neck for a primary mucosal lesion (e.g., oral cavity, maxillary antrum) must be performed. All suspicious lesions are approached surgically with a submandibular triangle dissection despite a negative FNA biopsy. Incisional or excisional biopsies are never performed to establish a tissue diagnosis because such procedures increase the probability of recurrence and the risk of surgical morbidity because a wide removal of the biopsy site must be added to the final definitive surgical procedure. The only exception to this rule is when there is a small tumor of the submandibular gland that is surrounded by normal parenchyma—then a simple excision of the submandibular gland may be performed.

Sublingual Salivary Gland

Primary Site

A sublingual salivary gland tumor usually presents as an asymptomatic swelling in the anterior floor of mouth. It may be difficult differentiating a primary lesion of the sublingual gland from a primary lesion of the floor of mouth. Up to 15% of patients with such a malignancy can have associated pain. Patients may complain of ipsilateral loss of tongue sensation (cranial nerve V3).

A careful and methodic examination of the patient can provide valuable information regarding the stage of the tumor. Bimanual palpation of the floor of the mouth will provide data regarding tumor dimension, texture, and mobility. Because adjacent soft tissues are sites where tumor can infiltrate, they must be carefully evaluated. The functional status of the ipsilateral lingual nerve (cranial nerve V3) must be documented.

Considerations regarding the use of CT, MRI, and FDG PET imaging are similar to those noted previously for submandibular salivary gland tumors.

The initial biopsy procedure of a suspicious sublingual region mass can be an FNA biopsy. Note that, in general, most sublingual salivary gland masses are malignant. Such a procedure is useful only if the biopsy reveals a malignancy. It is important to recognize that the mass may represent a primary minor salivary gland tumor in the anterior floor of mouth. These lesions, despite a negative FNA biopsy, should be resected with a formal cancer procedure. Incisional or excisional biopsies are never performed to establish a tissue diagnosis because such procedures increase the recurrence risk and the risk of surgical morbidity because a wide removal of the biopsy site must be added to the final definitive cancer surgery.

Staging

The American Joint Committee on Cancer (AJCC) has established the staging criteria for major salivary gland tumors. This system is based on an extensive retrospective review of the world’s literature.

The AJCC clinical staging is based on three major categories of evaluation: (1) history—pain, trismus, unilateral facial weakness; (2) clinical examination findings—size of mass, fixation, extension to adjacent tissues, involvement of adjacent cranial nerves; and (3) radiologic findings.

Several factors influence patient survival and are reflected in the AJCC staging criteria: histologic type, cellular differentiation of the tumor (grade), site, size, degree of fixation or local extension, facial nerve involvement, status of regional lymph nodes, status of distant metastasis.

Periodically, the staging criteria are reviewed and modified to better establish a system that reflects contemporary technology and therapeutic interventions. The current AJCC Cancer Staging Manual (Table 328) represents the seventh edition, published in 2010.14 The following examines the changes from the sixth edition (2002)15 as they relate to major salivary gland tumors.

Treatment

Surgery

Surgical resection of the involved major salivary gland is the primary treatment of choice. However, patients must be carefully selected to ensure resectability (see “Staging”).

Parotid Gland

In patients with small lesions located in the tail of the parotid gland, a limited local excision including a margin of normal appearing parotid tissue without dissection of the facial nerve (cranial nerve VII) can be considered. Otherwise, a superficial parotidectomy is the surgical treatment of choice.

For lesions confined to the superficial lobe of the parotid gland, which accounts for approximately 90% of cases, the entire superficial lobe is resected in an en bloc fashion (superficial parotidectomy, subtotal parotidectomy, lateral lobe parotidectomy). Generally, an excisional biopsy alone is not done because it would violate the surgical field and is associated with a high recurrence rate. If the tumor is adjacent to the deep lobe, it may be necessary to resect both the superficial lobe and the deep lobe together (total parotidectomy). When the tumor extends to involve the adjacent soft tissues (skin, muscles) or bone, a large en bloc resection of these structures may be required (radical parotidectomy). The Stensen duct is divided and ligated during a typical superficial parotidectomy. Tumor tracking along this duct is rarely seen. The adjacent facial nerve (cranial nerve VII) must be carefully evaluated, and every effort should be made to preserve it. Loss of the facial nerve results in significant disability for the patient. Its innervation of the muscles of facial expression affects mastication, speaking, emotional expression, and socialization. However, those cases in which the facial nerve is enveloped by or adherent to tumor require sacrifice of the nerve. All other situations undergo a dissection and preservation of the facial nerve. Under no circumstance should a piecemeal excision of the parotid tumor be performed in an effort to preserve the facial nerve. In certain instances, it may be possible to preserve part of the uninvolved facial nerve and sacrifice only the branches involved by tumor.

In selected cases in which the facial nerve has been sacrificed, it is desirable to perform an immediate cable grafting as long as there are disease-free proximal and distal branches of the facial nerve. Donor nerves include the sensory nerve branches of the cervical plexus or the sural nerves. In cases in which the proximal stump of the facial nerve is not available, the hypoglossal nerve (cranial nerve XII) can be anastomosed to the remaining distal portion of the facial nerve. The functional outcome of such nerve grafts are often quite suboptimal, although it usually allows an improvement in muscle tone. It should be noted that the use of postoperative radiation therapy in these patients is not contraindicated. Radiation can begin 3 to 4 weeks postoperatively with a completely healed incision without any adverse effect on the nerve graft.17

With the sacrifice of the facial nerve, closure of the ipsilateral upper eyelid is compromised. This can cause corneal exposure with resultant ulceration, pain, and ultimately blindness. Therefore, it is important to address this functional defect with the placement of a gold weight in the upper eyelid or a tarsorrhaphy to protect the eye.

Following a parotidectomy, it is not uncommon for patients to develop ipsilateral facial sweating and flushing along the distribution of the auriculotemporal nerve during chewing, deglutition, or even when thinking of food. This is known as Frey syndrome (gustatory sweating) and is due to misdirected regeneration of parasympathetic and sympathetic nerve fibers to the cholinergic receptors of the dermal sweat glands; it is generally not a major problem.

For the 10% to 12% of cases involving the deep lobe of the parotid gland, a total parotidectomy is performed. This involves the removal of the entire parotid gland with a cuff of a normal tissue. However, it is difficult to distinguish between the borders of the gland and the adjacent fat, so in reality the absolute removal of all glandular tissues is nearly impossible. The surgical principles presented earlier for lesions of the superficial lobes still apply to deep lobe lesions.

The overall incidence of occult regional lymph node metastasis is low for parotid gland tumors. Therefore, in general, an elective neck dissection is not performed routinely. However, if the primary lesion is sizable with a high-grade mucoepidermoid carcinoma or squamous cell carcinoma histopathologic finding in which there is an increased risk of nodal spread, selective neck dissection is performed.

For patients who present with clinically palpable adenopathy, a therapeutic neck dissection is the procedure of choice. However, if there is limited adenopathy located only in the first echelon nodes, a selective neck dissection can be considered.

Radiation Therapy

External-Beam Radiation Therapy

There is no evidence to indicate that unresectable malignant salivary gland tumors do not respond to conventionally fractionated photon radiation therapy. The literature contains reports of partially resected and unresected salivary gland cancers that were successfully treated in this fashion (Table 32-9) and the response rates are similar to those obtained from treating equivalent-sized squamous cell carcinomas of the head and neck.

The use of altered fractionation has been reported with encouraging results. Wang and Goodman18 from Massachusetts General Hospital treated 14 patients with unresectable malignant parotid carcinomas using accelerated hyperfractionation (1.60 Gy twice a day) photon therapy and various boost techniques to a total dosage of 65 to 70 cGy. The 5-year actuarial local control rate was 82% and the disease-specific survival rate was 55%.

The therapeutic approach of accelerated fractionation external-beam photon irradiation with a delayed concomitant boost with concurrent chemotherapy is used at MSKCC. Future studies using this combined modality attack are required. Our experience has shown that a pretreatment evaluation for a percutaneous endoscopic gastrostomy tube for sustenance and close aggressive monitoring of these patients are required because of the prominent acute side effects associated with the resulting mucositis and esophagitis.

Brachytherapy

For technically implantable lesions, brachytherapy alone or in conjunction with external-beam therapy can be used for unresectable malignant parotid tumors. Particularly if the patient was previously treated with radiation therapy and has an unresectable recurrence, brachytherapy alone can be considered as the sole therapeutic modality.

Armstrong et al.19 reported on 20 patients with recurrent or advanced disease treated with brachytherapy alone using 192Ir or 125I. Previously, radiation therapy had been administered to 15 of these patients. The implant was to gross disease in 15 of the 20 patients. The actuarial local control rate at 5 years was 60%. Two patients had soft tissue necrosis as a complication and were treated with conservative management. However, two patients with extensive skull base tumors treated with partial surgery and 125I implants for gross residual disease had cerebral abscesses, of which one was fatal.

In unresectable cases with no prior history of radiation, combined external-beam and brachytherapy techniques have been employed. King and Fletcher20 reported on 16 such patients treated in this fashion with 10 (63%) achieving local control.

Fast neutrons are a densely ionizing, high linear energy transfer type of particulate radiation that have a limited role in clinical radiation oncology (Fig. 32-15). They are contrasted with photons in the following fashion: (1) the biologic effectiveness of fast neutrons is much less affected by a hypoxic environment; (2) the lethal effects of fast neutrons are less dependent on the cell cycle phase compared with photons; (3) the repair of sublethal damage in malignant cells matters less; (4) fast neutrons are biologically more effective (relative biologic effectiveness [RBE] > 2.6). Also, fast neutrons lack skin sparing and thus can cause a more prominent acute dermal reaction than photons.

Batterman and associates21 measured the RBE of fast neutrons to cobalt-60 for human tumors metastatic to lung. RBE was inferred by observation of growth delay. A metastatic adenoid cystic cancer had an RBE of 8 compared with 2.5 to 4 for most other tumors. This observation on limited clinical material prompted an enthusiastic evaluation of the role of neutrons for the treatment of localized salivary cancers.

Fast neutrons have a high RBE compared with photons. This does not mean, however, that they have a superior therapeutic ratio. Estimates of the exact RBE for salivary cancers are variable and it may simply be that more “radiation effect” was given with the fast neutron doses than with the photon doses. This may have had two effects—one was the increased local control, and the other was the doubling of severe late effects. In the paper from Laramore and associates,22 severe late effects were detailed comparing the two treatments. Severe late effects can be assessed by measuring the occurrence of each category of toxicity in each patient. For photons, there were 10 events in 12 categories of toxicity in 12 patients. This represents a 7% occurrence rate (10/144 = 7%). For fast neutrons, there were 26 events in 12 categories of toxicity in 13 patients, representing a 17% occurrence rate (26/156 = 17%). The outstanding question about fast neutron therapy is whether equivalent local control could be achieved with photons if the dose of photons were escalated to a point at which the toxicity was equivalent to that of fast neutron therapy.

The Radiation Therapy Oncology Group (RTOG; United States) and the Medical Research Council (MRC; England)22 conducted a major randomized phase III multi-institutional clinical trial of the treatment of 32 patients with unresectable salivary gland tumors with either fast neutrons, photons or electrons, or both. The 10-year likelihood of local control was 56% with fast neutrons and 17% with photons and electrons. The fast neutron group had a higher incidence of distant metastasis. The overall survival was not improved with fast neutrons.

Caterall et al.23 reported on the MRC cyclotron experience of 65 patients with locally advanced or recurrent malignant salivary gland tumors, 89% of which were stage IV, who were treated with fast neutron radiotherapy. They achieved a 72% local control rate. The 5-year survival rate was 50%. The facial nerve was not damaged by fast neutron therapy.

Prott et al.24 reviewed their nonrandomized experience in treating patients with unresectable adenoid cystic carcinoma of the salivary glands using fast neutron therapy. Seventy-two patients were followed for a median of 50 months; 39.1% had a complete remission and 48.6% had a partial remission. The recurrence-free survival was 86% at 1 year, 71% at 2 years and 45% at 5 years. It was concluded that fast neutron therapy for unresectable adenoid cystic carcinoma of the salivary gland was an effective treatment.

Huber et al.25 evaluated and compared their patients with unresectable or partially resectable adenoid cystic carcinoma of major and minor salivary glands that were treated with one of the following approaches: (1) fast neutron therapy, (2) photon therapy, or (3) photon and neutron mixed-beam therapy. The 75 patients in this nonrandomized study were followed for a median of 51 months. The 5-year actuarial local control was as follows: (1) fast neutron therapy: 75%; (2) photon therapy: 32%; (3) photon and neutron mixed-beam therapy: 32%. Postoperative radiation and small tumor size were associated with high local control rates on multivariate analysis. There was no statistically significant difference in overall survival. However, those patients who were treated for primary disease and had negative nodes were associated with improved survival rate on multivariate analysis. Overall acute toxicity was comparable; but there was a higher percentage of grade 3 or 4 late toxicity with fast neutrons (19%) compared with photons (4%) or photon and neutron mixed-beam (10%) therapy. It was concluded that fast neutron therapy could be recommended for unresectable or partially resected adenoid cystic carcinomas of the salivary gland.

Douglas et al.26 reviewed their nonrandomized experience in treating patients with malignant salivary gland tumors using fast neutron therapy. Of 279 evaluable patients, 263 had gross residual disease and 16 patients had no gross disease. There were 141 patients with major salivary gland tumors and 138 patients with minor salivary gland tumors. These patients were followed for a median of 36 months. The 6-year actuarial local-regional control rate was 59%. Patients who had no gross disease enjoyed a 100% 6-year actuarial local-regional control rate. Those patients with a tumor size less than or equal to 4 cm, with no base-of-skull invasion and prior surgical resection but no prior radiation therapy had a statistically significant improvement in local-regional control on a multivariate analysis. The 6-year actuarial cause-specific survival rate was 67%. Those patients with stage I–II disease, minor salivary gland sites, no base-of-skull invasion, and primary disease had a statistically significant improvement in cause-specific survival on multivariate analysis. The 6-year actuarial freedom from metastatic disease was 64%. Those patients without lymph node involvement and no base-of-skull invasion predominated this group. The long-term rate of grade 3 or 4 toxicity was 10%. It was concluded that fast neutron therapy for malignant salivary gland tumors with gross disease resulted in good local-regional control. However, for patients with no gross disease, this treatment resulted in excellent long-term control.

Indications for Postoperative Radiation Therapy

The indications for postoperative radiation therapy for major salivary gland malignancies include (1) resectable primary T3-T4 and recurrent tumors, (2) unresectable or gross residual disease, (3) close or microscopically positive surgical margins, (4) deep lobe parotid tumors, (5) perineural (including facial nerve) and vascular invasion, (6) positive or close surgical margins, (7) high-grade histologic features, (8) adenoid cystic carcinoma, (9) locoregional lymph node metastasis, and (10) concern of the surgeon over the margins or conduct of the procedure (e.g., piecemeal resection, tumor spillage) irrespective of the pathology report (Table 32-12).

Table 32-12 Indications for Postoperative Radiation Therapy for Major Salivary Gland Malignancies

The use of postoperative radiation therapy has an established and important role in the treatment of major salivary gland tumors. Its employment substantially improves locoregional control in selected cases. However, there are no data from randomized trials in the worlds’ literature.

Spiro and colleagues27 reviewed the experience at MSKCC of 288 patients with parotid malignancies who were primarily treated with surgery (only 12 patients underwent postoperative radiation therapy). Recurrence rates by stage were as follows: (1) stage I: 7%; (2) stage II: 21%; and (3) stage III: 58%.

In comparison, Harrison and colleagues28 reported on the 5-year actuarial local control rates by T classification of MSKCC patients with major salivary gland malignancies who were treated with surgery and postoperative radiation therapy: (1) T1: 100%; (2) T2: 83%; (3) T3: 80%; (4) T4: 43%. Patients without neck node involvement had an 83% local control rate versus 58% for those with nodal metastasis.

At the University of California, San Francisco, Fu and colleagues29 reported on 35 patients with minor and major salivary gland carcinomas who had known microscopic disease at or close to the surgical margins following curative surgery. There was a 54% (7/13) recurrence rate in patients who underwent surgery alone compared with a 14% (3/22) rate for those treated with surgery and postoperative radiation therapy.

Various other retrospective reports are reported in the literature detailing the value of adjuvant postoperative radiation therapy (Table 32-13).

The biggest criticism of these studies is that they do not accurately compare the distribution of the many prognostic factors between the surgery-alone patients and the combined-treatment group. These factors include three primary sites; a large number of histologic features; and the effect of grade, tumor-node-metastasis (TNM) staging, extent of surgery, margin status, and deep lobe involvement. To control for this large number of variables, Armstrong and associates30 performed a matched-pair analysis comparing 46 patients treated with surgery and postoperative radiation with 46 prognostically matched patients treated with surgery alone. The large population of patients treated with surgery alone at MSKCC allowed selection (without reference to outcome) of very well-matched pairs. This analysis demonstrated that outcome was excellent for stage I and II cancers treated with surgery alone. For patients with T3 and T4 tumors or nodal disease, postoperative radiation therapy significantly improved locoregional control and survival (Table 32-14). There was a trend toward improved outcome when high-grade tumors were treated with postoperative radiation. Despite the enhancement of local control with radiation, local failures occurred in 51% of stage III and IV tumors, and 37% of high-grade tumors. Harrison and associates28 reported that local failure was particularly high for T4 tumors, despite the use of surgery and postoperative radiation. This provides a rationale for adjuvant brachytherapy in selected cases. Although the data are limited for adenoid cystic cancer (Table 32-15), most authorities recommend postoperative radiation for all but the smallest tumors.

Table 32-14 Postoperative Radiation Therapy: A Matched-Pair Analysis

Reference Surgery, % Surgery + Radiation, %
Stage I/II    
5-yr survival 96 82
5-yr local control 91 79
Stage III/IV    
5-yr survival 10 51
5-yr local control 17 51

Data from Armstrong et al.12

Table 32-15 Effect of Postoperative Radiation Therapy on Local Control of Adenoid Cystic Cancer

Reference Surgery, % (n) Surgery + Radiation, % (n)
Fu et al26 30 (10) 90 (10)
Matsuba et al* 24 (19) 60 (36)
Miglianico et al 44 (38) 78 (43)

* From Matsuba H, Spector G, Thawley SE, et al: Adenoid cystic salivary gland carcinoma: a histopathologic review of treatment failure patterns, Cancer 57:519, 1986.

From Miglianico L, Eschwege F, Marandas P, et al: Cervico-facial adenoid cystic carcinoma: study of 102 cases. Influence of radiation therapy, Int J Radiat Oncol Biol Phys 13:673, 1987.

Radiation Oncology Treatment Planning

Simulation

Patients undergo simulation in a supine position. The head is placed in a holder that allows hyperextension of the neck. All incisional scars and masses are wired with material appropriate for visualization on regular fluoroscopic simulation films or on the CT scans, depending on which unit is used. A bite block is employed for patients with tumors of the submandibular or sublingual glands; however, it is usually unnecessary for parotid gland tumors. A customized thermoplastic face mask (Fig. 32-16 and Fig. 32-17) or extended head and shoulder immobilizaton unit is created for immobilization and consistency of the head and neck position throughout treatment. A shoulder pull board is employed to bring the shoulder maximally in a caudad direction, particularly when the neck nodes are treated.

At this time, the patient undergoes a fluoroscopically assisted simulation or a CT simulation, depending on whether they will be treated with a two-dimensional or three-dimensional conformal or intensity-modulated radiation therapy (IMRT) treatment plan.31

For the patients who undergo a two-dimensional or fluoroscopically assisted simulation, the borders are based on the anatomic structures within the treatment volume that are peculiar to the salivary gland subsite to be treated. Once the simulation films are evaluated and approved, blocks are drawn on them to shield areas not requiring treatment. Cerrobend blocks are then manufactured.

For the patients who undergo a CT simulation (Fig. 32-18), the area within the superior and inferior borders of the field, which includes the treatment volume, are scanned using 3-mm slices. No intravenous contrast is needed. The radiation oncologist then selects the isocenter location on the CT scan, which is then marked on the patient with a permanent tattoo.

Computer Treatment Planning

The computer data from the CT simulation is downloaded from the CT simulator and transferred to treatment planning computers. The clinical target volume, which represents the visualized tumor (gross target volume) and regions at risk for microscopic disease, such as adjacent tissues or lymph nodes as well as the important adjacent normal structures including the spinal cord, brainstem, optic nerves and chiasm, orbits, cochlea, and the contralateral parotid gland are contoured. When contouring the postoperative surgical bed, the remaining intact contralateral salivary gland is used as an anatomic reference point. A generous margin around the anatomic region of at least 2 cm should be considered to cover the entire operative bed with sufficient borders. For patients who have adenoid cystic carcinoma of the salivary gland, the usual approach is to irradiate the pathways of the adjacent cranial nerve up to the base of the skull. The detailed and complex pathways of the cranial nerve must be carefully and accurately contoured. LeBlanc32 has correlated gross anatomic information with associated CT axial images delineating these pathways. The regional lymph node areas require contouring in specific instances, as noted earlier. Several excellent references are available that explore the anatomic region of the respective surgical lymph node group levels (I to VI) with respect to their location on axial CT slices. Nowak et al.33 correlated the borders of the surgical levels in the neck (I to VI) with structures seen on a CT scan. They defined the six cervical lymph node regions and noted their respective reproducible landmarks on the CT scan. Wijers et al.34 presented a more simplified approach for delineating cervical nodal target volume based on CT scans. Chao et al.35 published guidelines for target volume determination of head and neck lymph nodes. This was based on their analysis of lymph node failure in IMRT-treated patients. The Danish Head and Neck Cancer Group, the European Organisation for Research and Treatment of Cancer, the Groupe d’Oncologie Radiothérapie Tête Et Cou, and the RTOG consensus guidelines give a very nicely detailed color atlas of CT-based delineation of lymph node levels in the N0 neck (www.rtog.org.).

This anatomic data is then used for computer-based complex treatment planning employing the three-dimensional conformal radiation therapy (3D-CRT) or IMRT technique, whichever is most appropriate. A thorough discussion between the radiation oncologist and dosimetry staff is held to outline the disease status, treatment goals, areas of concern, and dosage planned. The resulting isodose curves, dose-volume histograms for the tumor area as well as the adjacent critical structures, and the digitally reconstructed radiographs are carefully evaluated (Fig. 32-19, Fig. 32-20, and Fig. 32-21).

Radiation Therapy Techniques

Parotid Gland Tumors

Primary Site

During the past several decades, multiple techniques have been used for the radiation treatment of parotid gland tumors. These have ranged from very basic to extremely complex plans and are related to the evolution of the therapy units, treatment planning sophistication, and our understanding of normal tissue tolerance.

Radiation therapy approaches include (1) electrons—lateral en face electron beam (Fig. 32-22), (2) combination of electrons and photons (50% to 80% weighting toward electrons)—lateral en face electron beam and photons using a direct lateral field or a wedge-pair technique, and (3) photons—wedge pair using 3D-CRT or IMRT approach (see Figs. 32-19 through 32-21).

image

FIGURE 32-22 • Postoperative left lateral electron beam portal for a parotid gland carcinoma. Note that the dashed line is 1 cm larger than photon portals due to the constriction of the electron isodose lines at depth.

(From Levitt ST, Khan FM, Potish RA, et al (eds): Levitt and Tapley’s technological basis of radiation therapy: practical clinical applications, ed 3, Philadelphia, 1999, Lea & Febiger, p 294.)

The most frequently used techniques have been with a combination of electrons and photons as a so-called mixed beam (Fig. 32-23) or photons alone using a wedge pair (Fig 32-24). Electrons are delivered with a direct lateral en face approach covering the tumor (parotid gland) or tumor bed with a 2- to 3-cm margin (Fig. 32-25).

image

FIGURE 32-23 • Transverse image of a parotid gland mixed beam (photon and electron) therapy isodose distribution.

(From Perez CA, Brady LW, eds: Principles and practice of radiation oncology, ed 2, Philadelphia, 1998, Lippincott-Raven, p 970.)

image

FIGURE 32-24 • Transverse image of a parotid gland wedge pair photon beam therapy isodose distribution.

(From Wang CC: Radiation therapy for head and neck neoplasms, ed 3, New York, 1997, Wiley-Liss, p 317.)

image

FIGURE 32-25 • Transverse image of the electron beam isodose distribution for the parotid gland.

(From Thawley SE, Panje WR, Batsakis JG, et al: Comprehensive management of head and neck tumors, vol 2, ed 2, Philadelphia, 1999, Saunders, p 1190.)

The general portal margins that would encompass the planning target volume are (1) superior—top of the zygomatic bone, (2) inferior—hyoid bone–thyroid notch interspace, (3) anterior—2 cm anterior to the upper second molar, (4) posterior—posterior to the mastoid tip, (5) lateral—2 cm flash of the cheek, and (6) medial—2 cm medial from the ipsilateral oropharyngeal area. However, if the accessory parotid gland is involved with tumor, an additional 2-cm margin must be added anteriorly because this is the location of this parotid gland anatomic variation.

The electron portal margins are 1 cm larger than those for photons because of the constriction of the electron isodose curves at depth. The energy of the electron that is chosen depends on the anatomic distance from the skin of the ipsilateral cheek to the oral mucosa and generally ranges between 12 and 16 meV. In some institutions, the ipsilateral external auditory canal, tympanic membrane, and middle ear are protected from radiation with the use of a Cerrobend ear plug.

The photon technique using a wedge pair involves an anterior and a posterior oblique ipsilateral attack with wedges (45 and 30 degrees) on a cobalt 60 1.25 MV gamma ray unit or a linear accelerator with 4- to 6-MV x-rays. The basic clinical margin setup is similar to that noted previously for electrons but without the added extra 1-cm margin. To avoid an exit dose through the contralateral eye, some practitioners use a slight inferior angulation of the beam. A bolus is placed over the scar if clinically indicated.

When a combination of electrons and photons are used, either modality can start first. There is a weighting between 50% and 80% with electrons, but this can also be with the photon component depending on the discretion of the radiation oncologist. By mixing the two different beams, one can decrease the irradiation of the contralateral parotid gland, acute radiation skin reaction, and mucositis.

With the development of CT simulation and sophisticated computer-based treatment planning, the state of the art and the preferred radiation techniques now mandate the use of 3D-CRT or IMRT depending on the needs and complexity of the plan. For the majority of cases, 3D-CRT using either a two- or three-field approach including wedges is appropriate. However, when the histopathologic anatomy is adenoid cystic carcinoma, the increased risk of perineural invasion and travel along the pathways of the adjacent cranial nerves (beware of perineural skip lesions) require the treatment volume to include the neural pathways to the base of the skull. In such cases, IMRT treatment plans give the best approach.

Sparing the contralateral parotid gland is a very important consideration during the complex treatment planning process for 3D-CRT and IMRT. The dose contraints to the contralateral parotid gland are that (1) the mean dose to the gland should be limited to less than or equal to 26 Gy and (2) the dose to at least 50% of the gland should be limited to less than 30 Gy.

Lymph Nodes

When there is an indication to treat the regional lymph nodes as presented earlier in the chapter, either electrons (9 meV) or photons (6 MV to 25 MV) are employed. Only ipsilateral regional lymph nodes are treated because it is quite rare for there to be involvement of the contralateral cervical nodes, even with multiple ipsilateral cervical nodal involvement. However, if significant bulky nodes are encountered, some practitioners consider the contralateral cervical nodes to be at sufficient risk for metastatic disease and therefore treat the bilateral neck nodes. Otherwise, the usual approach is to treat the entire ipsilateral cervical and supraclavicular nodes.

With electrons, a direct lateral or en face field is matched to the primary parotid field. Special care must be taken to get the ipsilateral shoulder as far down as possible. This is to allow the electron cone to be positioned to include the nodes as far down the chain as possible. When multiple cervical lymph nodes are involved, photons should be used to treat the ipsilateral lower cervical nodes as well as the adjacent supraclavicular nodes.

When photons are used, either anteroposterior–posteroanterior (AP/PA) or direct AP fields are used and positioned off the spinal cord. If there is concern regarding the nodal regions in the mid to posterior neck, then the AP/PA approach is employed. These fields are matched with the primary parotid fields, and asymmetric jaws or half-beam blocks at the superior margin are used to decrease the overlap of the adjoining field.

Submandibular Gland

Sublingual Gland

Side Effects

Critical Normal Tissues: Radiation Injury

Pathophysiology of Radiation Injury to Salivary Tissue

The normal salivary glands produce approximately 1 to 1.5 L of saliva daily. The major salivary glands produce 80% of this volume. The parotid glands consist predominantly of serous acini, the submandibular glands contain both mucous and serous acini, and the minor salivary glands are predominantly mucus secretors. During radiation therapy, acute, transient, often painless sialadenitis may occur in the first 24 hours. Significant swelling of the gland within the port is the usual feature. Management is conservative using mild analgesia or anti-inflammatory medication. Serous salivary tissue seems to be more sensitive than mucous tissue, and the saliva becomes reduced in quantity and more ropy, mucous-like, and tenacious. This alteration in character can cause significant symptoms. The magnitude of clinical effect depends largely on the volume of salivary tissue being irradiated, because most clinically useful dose fractionation schedules cause xerosis of any irradiated salivary tissue. Put another way, the only method of sparing salivary function is to avoid irradiating as much salivary tissue as possible.

Permanent xerostomia causes discomfort, alters taste acuity, promotes poor oral hygiene, and accelerates dental decay. The remaining salivary gland tissues are largely obliterated after radiation damage and replaced by collagen. The saliva that is produced is reduced in quantity and has an altered electrolyte content and reduced pH level. The reduced pH favors the growth of bacteria that cause caries and decay. Fluoride supplementation and regular dental hygiene and professional assessment are vital. A homemade mouthwash containing baking soda and salt promotes hygiene and can restore pH toward alkaline values.

Tolerance Doses and Fractionation

Loss of salivary function is usually complete and permanent after doses of 35 Gy in conventional fractionation. Marks and associates36 noted that only one-fifth of patients who received between 40 and 60 Gy had any measurable salivary flow after salivary stimulation. Young patients have better chances of recovery of some salivary function, and the higher the rate of pretreatment salivary flow, the greater the chance of recovery.37

Acute and Late Side Effects

Special Cases

Adenoid Cystic Carcinoma (Cylindroma)

Adenoid cystic carcinoma is a relatively uncommon pathologic subtype comprising approximately 2% to 5% of salivary gland malignancies. It affects men and women equally and is the most common malignant tumor of the minor salivary glands and submandibular.

The initial treatment of choice for adenoid cystic carcinoma is surgical resection. However, it should be noted that this type of malignancy has the propensity not only for perineural spread, but also for insidious local extension, both of which can make it difficult to surgically encompass the potentially involved region adequately. Thus, postoperative radiation therapy is recommended. The routine use of combined surgery and postoperative radiation therapy has been associated with improved local control for both major and minor salivary gland adenoid cystic carcinomas.

There is a 50% incidence of microscopic perineural invasion, particularly of the adjacent cranial nerves. Perineural spread can be associated with skip areas of involvement and therefore one cannot be comfortable with negative neural margins on frozen section. Thus, postoperative radiation therapy of the entire pathway of the adjacent cranial nerve to the base of the skull is employed. The resultant base-of-skull failure from perineural invasion and extension of adenoid cystic carcinoma is associated with significant morbidity and is very difficult to treat effectively. Garden et al.45 found that the base-of-skull failure was rare whether or not elective radiation was given to the skull base. The policy at MSKCC, however, is to irradiate the adjacent cranial nerve pathway to the base of the skull in all patients who have adenoid cystic carcinoma of the salivary gland (see Fig. 32-20).

The overall incidence of regional lymph node metastasis at presentation is 15%. Because the risk for occult nodal metastasis is rare, elective surgical dissection or irradiation is not employed.

Spiro and Huvos reviewed the MSKCC experience of 184 previously untreated patients with adenoid cystic carcinoma of the minor and major salivary glands.46 Of these, 63% were stage I or II, 43% stage III, and 21% stage IV. Sixty-eight percent were grade 1 (cribriform pattern only), 26% were grade 2 (mixed cribriform pattern and solid features), and only 5% were grade 3 (solid only). All patients were treated with surgery but relatively few received postoperative radiation therapy. The cumulative 10-year survival rate was as follows: (1) stage I/II, 75%; (2) stage III, 43%; and (3) stage IV, 15%. The cause-specific survival at 10 years was 94% for stage I. They arrived at two important conclusions: (1) The clinical stage was the only factor that had significant effect on survival, and (2) the tumor grade alone was not predictive of survival, regional nodal metastasis, or distant metastasis. Surgery alone has been associated with a local control rate of 25% to 40%; however, when postoperative radiation therapy is employed, local control rates increase 75% to 80%.

Of recurrences from salivary gland malignancies, 70% occur generally within the first 2 years of treatment. However, with adenoid cystic carcinomas, most of these local failures occur much later. These patients must be followed closely for many years post-treatment.

Adenoid cystic carcinomas can have a natural history of several years even when presenting with metastatic disease. Consideration of controlling the primary tumor is prudent (see Fig. 32-13).

Recurrent Major Salivary Gland Cancers

Surgery is the treatment of choice for locally recurrent major salivary gland cancers. To maximize the chances of resectability and control, recurrent lesions must be diagnosed as early as possible. Thus, close follow-up is mandatory for the first 2 to 3 years after the initial treatment.

Unresectable, previously unirradiated, locally recurrent disease can be considered for concurrent chemotherapy and accelerated fractionation external-beam radiation therapy if neutron therapy is not a viable option. Brachytherapy can be considered in selected cases.

The role of postoperative radiotherapy in patients with local recurrence who did not receive radiation as part of their initial treatment is not clear. In patients with resected large recurrent tumors and high risk features, postoperative external-beam radiation should be considered. Brachytherapy can also be used in selected cases.

Armstrong and associates47 reviewed the experience at MSKCC of recurrent major salivary gland malignancies in 78 patients who underwent resection of locally confined recurrent major salivary gland cancers. Thirty-eight patients had completely resected tumors, with low- or intermediate-grade histologic anatomy, without lymph node spread, and the cause-specific survival rate was 83% at 5 years, 70% at 10 years, and 48% at 15 years. Local control for these 38 patients was superior for those with tumors of 3 cm or less compared with those with larger tumors: 80% versus 62% at 5 years, and 73% versus 25% at 10 years (P < 0.05).

The remaining 40 patients had high-risk features (high-grade histologic anatomy, lymph node metastases, and close or positive margins of resection). Half of the high-risk patients received postoperative radiation therapy and half did not because radiation had been given before the development of local recurrence. Consequently, although local control was enhanced with radiation, this analysis could not determine the role of postoperative radiation. However, the data indicate that surgery alone yielded good local control in patients with small tumors (<3 cm) and no high-risk features, suggesting that postoperative radiation may be unnecessary for these patients if they can be closely followed.

However, another study by Spiro and Spiro at MSKCC reviewed the results of attempted salvage of patients developing locally confined recurrences following initial therapy of their primary tumor.48 Between 1966 and 1982, 155 patients presented with locally confined new primary carcinomas and were treated surgically with or without radiation therapy.48 Of the latter group, local failure occurred in 15% (23/155) and only 7 of the 23 were suitable for salvage surgical therapy. All but one of the seven died of their disease, with an actuarial survival rate of only 21% 5 years after recurrence.48a In an attempt to explain this discrepancy, it has been postulated that the extent or completeness of therapy administered at the time of the initial diagnosis, before the development of a recurrence, may be an important predictor of the success of management of locally recurrent salivary cancer. The mechanisms of local recurrence following minimal or inadequate initial therapy may differ from the mechanisms of recurrence following aggressive initial therapy. The former may be related to technical factors, whereas the latter may be a manifestation of biologic aggressiveness.

Few other data address the role of postoperative conventional external-beam radiotherapy in the management of recurrent tumors of the major salivary glands. Guillamondejui and associates49 from the MD Anderson Hospital treated 12 patients with surgery or radiotherapy or both for locoregional recurrences. Locoregional control was achieved in 42% (5/12). In contrast, King and Fletcher,50 from the same institution, reported a crude locoregional control rate of 81% (25/31), with minimum follow-up of 2 years, among patients presenting with postoperative local recurrences treated with radiotherapy. Tu and associates,51 from the Chinese Academy of Medical Sciences, reported a significant survival advantage with the use of postoperative radiotherapy for recurrent parotid cancers. At 5 years, 89% of the patients (17/19) were alive in the combined-therapy arm compared with 59% (10/17) in the surgery-alone arm.52 This paper did not compare prognostic factors between the two groups, nor did it detail the patterns of treatment failure. In addition, as the numbers of patients were small, it is difficult to make firm conclusions as to the efficacy of postoperative radiation therapy.

Armstrong and associates reported on 20 patients with recurrent19 or advanced12 disease who were treated with 192Ir and 125I brachytherapy as described earlier.19 Previous radiation therapy had been administered to 15 patients. The implant was to gross disease in 15 of the 20 patients. Actuarial local control was 60% at 5 years. Two patients had soft tissue necrosis as a complication, which was resolved with conservative management. There were two cerebral abscesses, one of which was fatal. Both patients had extensive skull-base tumors treated with partial resection and 125I brachytherapy for gross residual disease.

Brachytherapy can be used in conjunction with resection if the resection margins are close or positive, or in the presence of a T4 tumor. It may also be used as sole treatment of accessible localized recurrent disease previously treated with radiation therapy. Permanent 125I seeds can be implanted directly into the tumor or embedded in a synthetic absorbable suture and sewn into tumor beds. The typical matched peripheral dose for 125I implants is 160 Gy. Afterloaded 192Ir can be used as a temporary implant in single or multiple planes for either microscopic residual disease or unresected disease. Typical median peripheral doses are 45 Gy for microscopic disease and 60 Gy for gross disease. A patient with a resectable submandibular cancer with lung metastases who was treated with 125I is illustrated in Fig. 32-26.

Pleomorphic Adenoma (Benign Mixed Tumor)

Pleomorphic adenomas are benign tumors of salivary glands, but they can also occur in lacrimal glands. The pleomorphic adenoma is the most common benign tumor of the parotid glands and accounts for 75% of all parotid epithelial tumors. These adenomas occur rather frequently and generally are found in a relatively young population.

The initial treatment of choice is excision of the tumor with a margin of normal tissue with preservation of the facial nerve. Simple enucleation of the tumor is contraindicated. The likelihood of local recurrence for most patients who undergo uncomplicated surgery is only 1%. The few cases that have local postoperative recurrences can generally be treated again with surgery. However, multiple recurrent tumors carry up to a 25% risk for further recurrences. With each recurrence, the probability of malignant transformation (e.g., carcinoma ex pleomorphic adenoma) or facial nerve damage increases.

Pleomorphic adenomas are radioresponsive tumors. In selected cases that are at high risk for local recurrence, the use of postoperative radiation therapy is associated with a cumulative risk of recurrence of 8% at 20 years.53

There is an established role for the use of radiation therapy in this benign lesion.5458 The indications for postoperative radiation therapy include (1) more than three histologically benign local recurrences, with each recurrence associated with an increasing degree of infiltration; (2) a large lesion (>3 to 5 cm) for which surgery is associated with inadequate margins; (3) microscopically positive surgical margins; (4) macroscopic residual disease; and (5) malignant transformation54 (Table 32-16). Previously, postoperative radiation therapy was indicated if there was tumor spill during surgery, but now this is not felt to be the case.58a

Table 32-16 Indications for Postoperative Radiation Therapy for Pleomorphic Adenomas

The use of postoperative radiation therapy must be based on selective and judicious indications, as noted earlier, as well as astute clinical judgment. Despite its effectiveness in controlling pleomorphic adenomas, its liberal administration is contraindicated. Because these lesions are benign and are usually found in young patients, the exposure of such patients to the small but definite risk of developing a radiation-induced malignancy as a late complication is unjustified.

The radiation oncology technical approaches are as presented previously for malignant tumors of the salivary gland. Because this is a benign lesion, there is no attempt to include the regional lymph nodes. However, when malignant transformation is noted in a recurrent lesion, the technique chosen is identical to that for a similar malignant lesion of the salivary gland.

The total cumulative dosage of radiation therapy that is administered for pleomorphic adenomas is 50 to 60 Gy depending on the particular surgical and pathologic characteristics of the case.

Minor Salivary Gland Tumors

Clinical Presentation and Evaluation

Regional Lymph Nodes

There is an overall 15% risk for cervical lymph node metastasis. This risk, however, is related to the site of origin and the grade of tumor.11 The physical examination should include palpation of regional nodes. There may be an increased risk for bilateral or contralateral nodal metastasis, particularly when the primary site is in the oral cavity or oropharynx because of the rich lymphatic pathways in these regions.

Treatment

Surgery

Surgical resection of malignant minor salivary gland tumors is the primary treatment of choice. However, patients must be carefully selected to ensure resectability. A review of the MSKCC experience by Spiro noted that there were fewer early-stage lesions for the minor salivary gland lesions compared with the major salivary gland malignancies.11 In fact, almost all of the minor salivary gland malignancies were locally advanced.

The procedure used depends on the anatomic site of the lesion as well as the tumor size, extent, and grade. For low-grade lesions, a more conservative surgical resection is employed with the goal of achieving negative margins and minimizing extensive surgery with its associated cosmetic and functional impairments. For high-grade lesions, surgery is more aggressive and radical with consideration given to including an ipsilateral neck dissection. The surgical approach is similar to that used for squamous cell carcinomas of similar stage in that site. This may mean that important functional structures such as the larynx or orbit need to be sacrificed. It is important to note that more radical and extensive surgery does not necessarily translate to further improvement in control of locally advanced tumors. Such procedures, however, are associated with increased morbidity with an adverse effect on the patient’s quality of life. Thus, in such cases a less radical surgical approach and the use of planned postoperative radiation therapy can maintain and even improve the chances for control.

The overall incidence of occult regional lymph node metastasis is low for minor salivary gland tumors, especially for adenoid cystic carcinomas. Therefore, in general, an elective neck dissection is not performed routinely. However, if the primary lesion is sizeable with a high-grade mucoepidermoid carcinoma or squamous cell carcinoma histopathologic anatomy, there is an increased risk of nodal spread, and a selective neck dissection is performed in clinically N0 patients.

For patients who present with clinically palpable adenopathy, a therapeutic neck dissection is performed. However, if there is limited adenopathy in the first echelon nodes, a selective neck dissection can be considered.

Radiation Oncology

Indications for Postoperative Radiation Therapy

Few data are available to support the use of postoperative radiation therapy for minor salivary gland malignancies compared with the data available for such treatment in major salivary gland malignancies. Postoperative external-beam radiation therapy has been reported for minor salivary gland malignancies with a local control rate of 75%,6164 compared with 50% reported for surgery alone.60,61,64 Several series using postoperative radiation therapy for minor salivary gland malignancies are detailed in Table 32-17.

It is hard to make conclusions based on such limited data; however, it appears that postoperative radiation may improve outcome in selected patients. There is no good hypothesis to explain why postoperative radiation is effective for selected major salivary gland cancers and ineffective for selected minor salivary gland cancers. Therefore, it is reasonable to recommend postoperative radiation therapy for the following indications: (1) resectable primary T3-T4 and recurrent tumors, (2) unresectable or partially resected gross residual disease, (3) close or microscopically positive surgical margins, (4) perineural or vascular invasion, (5) positive or close surgical margins, (6) high-grade histologic findings, (7) adenoid cystic carcinoma, (8) locoregional lymph node metastasis, (9) concern of the surgeon over the margins of resection or conduct of the procedure (e.g., piecemeal resection, tumor spillage) irrespective of the pathology report (Table 32-18).

Table 32-18 Indications for Postoperative Radiation Therapy for Minor Salivary Gland Malignancies

The basic pretreatment evaluation and preparation of a patient for the radiation therapy of minor salivary gland malignancies is identical to that presented previously for the major salivary gland tumors.

Radiation Therapy Technique

References

1 Frommer J. The human accessory parotid gland: its incidence, nature and significance. Oral Surg. 1977;43:671.

2 Johnson FE, Spiro RH. Tumors arising in accessory parotid tissue. Am J Surg. 1979;138:576.

3 Spiro JD, Spiro RH. Salivary tumors. In: Shah J, editor. Cancer of the head and neck. Hamilton, Ontario/London: BC Decker; 2001:240.

4 Eisele DW, Kleinberg LR, O’Malley BB. Management of malignant salivary gland tumors. In: Harrison LB, Sessions RB, Hong WK, editors. Head and neck cancer. A multidisciplinary approach. Philadelphia: Lippincott-Raven; 1999:721.

5 O’Dwyer P, Farrar WB, James A. Needle aspiration biopsy of major salivary gland tumors: its value. Cancer. 1986;57:554.

6 Frable MAS, Frable WJ. Fine needle aspiration biopsy of salivary glands. Laryngoscope. 1991;101:245.

7 Roland NJ, Caslin AW, Smith PA, et al. Fine needle aspiration cytology of salivary gland lesions reported immediately in a head and neck clinic. J Laryngol Otol. 1993;107:1025.

8 Zurrida S, Alasio L, Tradati N, et al. Fine needle aspiration of parotid masses. Cancer. 1993;72:2306.

9 Layfield JL, Tan P, Glasgow BJ. Fine needle aspiration of salivary gland lesions. Arch Pathol Lab Med. 1987;111:346.

10 Jayaram G, Verma AK, Sood N, et al. Fine needle aspiration cytology of salivary gland lesions. J Oral Pathol Med. 1994;23:256.

11 Spiro RH. Salivary neoplasms: overview of a 35-year experience with 2807 patients. Head Neck Surg. 1986;8:177.

12 Armstrong JG, Harrison LG, Spiro RH, et al. The indications for elective treatment of the neck in cancer of the major salivary glands. Cancer. 1992;69:615.

13 Wang CC. Tumors of the salivary glands. In: Wang CC, editor. Radiation therapy for head and neck neoplasms. ed 3. New York: Wiley-Liss; 1997:311.

14 American Joint Committee on Cancer, Major salivary glands (parotid, submandibular, and sublingual). ed 7. Edge SB, Byrd DR, Compton CC, et al, editors. AJCC cancer staging manual. 2010; Springer, New York:79.

15 American Joint Committee on Cancer. Major salivary glands (parotid, submandibular, and sublingual. In: Greene Fl, Page DL, Fleming ID, et al, editors. AJCC cancer staging manual. ed 6. New York: Springer; 2002:69.

16 Personal communication with Jatin P. Shah, Chief, Head and Neck Service, Memorial Sloan-Kettering Cancer Center.

17 McGuitt WF. Effect of radiation therapy on facial nerve cable autografts. Trans Am Acad Ophthalmol Otolaryngol. 1976;82:487.

18 Wang CC, Goodman M. Photon irradiation of unresectable carcinomas of salivary glands. Int J Radiat Oncol Biol Phys. 1991;21:569.

19 Armstrong JG, Harrison LB, Spiro RH, et al. Brachytherapy of malignant tumors of salivary gland origin. Endocurither Hypertherm Oncol. 1996;6:19.

20 King JJ, Fletcher GH. Malignant tumors of the major salivary glands. Radiology. 1971;123:49.

21 Batterman J, Breuer K, Hart G, et al. Observations on pulmonary metastases in patients after single doses and multiple fractions of fast neutrons and cobalt-60 gamma rays. Eur J Cancer. 1981;17:539.

22 Laramore G, Krall J, Griffin T, et al. Neutron versus photon irradiation for unresectable salivary gland tumors: final report of an RTOG-MRC randomized trial. Int J Radiat Oncol Biol Phys. 1993;27:235.

23 Catterall M, Errington RD. The implications of improved treatment of malignant salivary gland tumors by fast neutron radiotherapy. Int J Radiat Oncol Biol Phys. 1987;13:1314.

24 Prott FJ, Micke O, Haverkamp U, et al. Results of fast neutron therapy of adenoid cystic carcinomas of the salivary glands. Anticancer Res. 2000;20(5C):3743.

25 Huber PE, Debus J, Latz D, et al. Radiotherapy for advanced adenoid cystic carcinoma: neutrons, photons or mixed beam? Radiother Oncol. 2001;59(2):161.

26 Douglas JG, Koh WJ, Austin-Seymour M, et al. Treatment of salivary gland neoplasms with fast neutron radiotherapy. Arch Otolaryngol Head Neck Surg. 2003;129(9):944.

27 Spiro RH, Huvos AG, Strong EW. Cancer of the parotid gland: a clinicopathologic study of 288 primary cases. Am J Surg. 1975;130:457.

28 Harrison LB, Armstrong JG, Spiro RH. Postoperative radiation therapy for major salivary gland malignancies. J Surg Oncol. 1990;45:52.

29 Fu KK, Leibel SA, Levine ML, et al. Carcinoma of the major and minor salivary glands. Cancer. 1977;40:2882.

30 Armstrong JG, Harrison LB, Spiro RH, et al. Malignant tumors of major salivary gland origin: a matched pair analysis of the role of combined surgery and postoperative radiation therapy. Arch Otolaryngol Head Neck Surg. 1990;116:290.

31 Chong LM, Hunt M. Intensity modulated radiation therapy for head and neck cancer. In: Fuks Z, Leibel SA, Ling CC, editors. A practical guide to intensity-modulated radiation therapy. Madison, Wisc: Medical Physics Publishing, 2003.

32 Leblanc A. The cranial nerves, ed 2. Berlin: Springer-Verlag; 1995.

33 Nowak PJ, Wijers OB, Lagerwaard FJ, et al. A three-dimensional CT-based target definition for elective radiation of the neck. Int J Radiat Oncol Biol Phys. 1999;45:33.

34 Wijers OB, Levendag PC, Tan T, et al. A simplified CT-based definition of the lymph node levels in the node negative neck. Radiother Oncol. 1999;52:35.

35 Chao KSC, Wippold FJ, et al. Determination and delineation of nodal target volumes for head and neck cancer based on patterns of failure in patients receiving definitive and postoperative IMRT. Int J Radiat Oncol Biol Phys. 2002;53:1174.

36 Marks J, Davis CC, Gottsman VL, et al. The effects of radiation on parotid salivary function. Int J Radiat Oncol Biol Phys. 1981;7:1013.

37 Mira J, Wescott W, Starcke E, et al. Some factors influencing salivary function when treating with radiotherapy. Int J Radiat Oncol Biol Phys. 1981;7:535.

38 Lawson W, Som ML. Second primary cancer after irradiation of laryngeal cancer. Ann Otol Rhinol Laryngol. 1975;84:771.

39 Belsky JL, Takechi N, Yamamoto T, et al. Salivary gland neoplasms following atomic radiation: additional cases and reanalysis of combined data in a fixed population, 1957–1970. Cancer. 1975;35:555.

40 Dreyfus AI, Clark JR, Fallon BG, et al. Cyclophosphamide, doxorubicin, cisplatin combination chemotherapy for advanced carcinomas of salivary gland origin. Cancer. 1987;60:2869.

41 Kaplan MJ, Johns ME, Cantrell RW. Chemotherapy for salivary gland cancer. Otolaryngol Head Neck Surg. 1966;95:165.

42 Dimery IW, Legha SS, Shirinian M, et al. Fluorouracil, doxorubicin, cyclophosphamide, and cisplatin combination chemotherapy in advanced or recurrent salivary gland carcinoma. J Clin Oncol. 1990;8:1056.

43 Licitra L, Marchini S, Spinazze S, et al. Cisplatin in advanced salivary gland carcinoma. Cancer. 1991;68:1874.

44 Creaghan ET, Woods JE, Rubin J, et al. Cisplatin-based chemotherapy for neoplasms arising from salivary glands and contiguous structures in the head and neck. Cancer. 1988;62:2313.

45 Garden AS, Weber RS, Morrison WH, et al. The influence of positive margins and nerve invasion in adenoid cystic carcinoma of the head and neck treated with surgery and radiation. Int J Radiat Oncol Biol Phys. 1995;32:619.

46 Spiro RH, Huvos AG. Stage means more than grade in adenoid cystic carcinoma. Am J Surg. 1992;164:230.

47 Armstrong JG, Harrison LB, Spiro RH, et al. Observation on the natural history and treatment of recurrent major salivary gland cancer. J Surg Oncol. 1990;44:138.

48 Spiro R, Spiro J. Cancer of the salivary glands. In: Meyers E, Sven J, editors. Cancer of the head and neck. ed 2. London: Churchill Livingstone; 1984:645.

48a Armstrong JG: Unpublished data.

49 Guillamondejui OM, Byers RM, Luna MH, et al. Aggressive surgery in treatment for parotid cancer. The role of adjunctive postoperative radiotherapy. Am J Roentgenol. 1975;123:49.

50 King JJ, Fletcher GH. Malignant tumors of the major salivary glands. Radiology. 1971;100:381.

51 Tu G-Y, Hu Y, Jiang P, et al. The superiority of combined therapy (surgery and postoperative irradiation) in parotid cancer. Arch Otolaryngol. 1982;108:710.

52 Woods J, Weiland L, Chong G, et al. Pathology and surgery of primary tumors of the parotid. Surg Clin North Am. 1977;57:565.

53 Dawson AK, Orr JA. Long-term results of local excision and radiotherapy in pleomorphic adenoma of the parotid gland. Int J Radiat Ncol Biol Phys. 1985;11:4.

54 Chen AM, Garcia J, Bucci MK, et al. Recurrent pleomorphic adenoma of the parotid gland: long-term outcome of patients treated with radiation therapy. Int J Radiat Oncol Biol Phys. 2006;66(4):1031.

55 Leverstein H, Tiwari RM, Snow GB. The surgical management of recurrent or residual pleomorphic adenomas of the parotid gland. Analysis and results in 40 patients. Eur Arch Otorhinolaryngol.. 1997;254(7):313.

56 Barton J, Slevin NJ, Gleave EN. Radiotherapy for pleomorphic adenoma of the parotid gland. Int J Radiat Oncol Biol Phys. 1992;22(5):925.

57 Touquet R, Mackenzie IJ, Carruth JA. Management of the parotid pleomorphic adenoma, the problem of exposing tumour tissue at operation. The logical pursuit of treatment policies. Br J Oral Maxillofac Surg. 1990;28(6):404.

58 Carew JF, Spiro RH, Singh B, et al. Treatment of recurrent pleomorphic adenomas of the parotid gland. Otolaryngol Head Neck Surg. 1999;121(5):539.

58a Buchman C, Stringer SP, Mendenhall WM, et al. Pleomorphic adenoma: effect of tumor spill and inadequate resection on tumor recurrence. Laryngoscope. 1994;104:1231-1234.

59 DeVries EJ, Johnson JT, Meyers EN, et al. Base of tongue salivary gland tumors. Head Neck Surg. 1987;9:329.

60 Spiro RH, Thaler HT, Hicks WF. The importance of clinical staging of minor salivary gland carcinoma. Am J Surg. 1991;162:330.

61 Eapen LJ, Gerig LH, Catton GE, et al. Impact of local radiation in the management of salivary gland carcinomas. Head Neck Surg. 1988;10:239.

62 Garden AS, Weber RS, Ang KK, et al. Postoperative radiation therapy for malignant tumors of minor salivary glands. Outcomes and patterns of failure. Cancer. 1994;73:2563.

63 Parson JT, Mendenhall WM, Strenger SP, et al. Management of minor salivary gland carcinomas. Int J Radiat Oncol Biol Phys. 1996;35:443.

64 Sadeghi A, Tran LM, Mark R, et al. Minor salivary gland tumors of the head and neck: treatment strategies and prognosis. Am J Clin Oncol. 1993;16:3.