Salivary Gland Malignancies




Salivary gland malignant tumors represent a diverse group of neoplasms. Their low incidence makes research studies challenging, with most therapeutic recommendations based on case reviews, single-arm trials, or small randomized trials. The standard of care for localized disease is surgical resection. Radiotherapy is the preferred local therapy when surgery is not possible or if there is significant morbidity. When symptomatic metastatic disease develops, systemic therapy is considered. Recent trial accrual success with a cooperative group, treatments based on defined molecular targets, and the development of immunotherapies all hold promise in improving the care of patients with these tumors.


Key points








  • Salivary gland cancers are morphologically and biologically diverse.



  • Surgical resection with postoperative radiation is considered a treatment standard for localized disease.



  • Systemic therapy can be considered for patients with unresectable or metastatic disease with the goal of palliation of symptoms.






Introduction


Primary salivary gland malignancies represent less than 5% of all new head and neck cancers, with approximately 3000 new cases diagnosed in the United States annually. These diverse cancers arise from the malignant transformation of the various myoepithelial, ductal, and acinic components of 3 paired major (parotid, submandibular, and sublingual) and minor salivary glands distributed throughout the upper aerodigestive tract. The World Health Organization classifies 24 subtypes characterized by marked biological heterogeneity. For example, high-grade mucoepidermoid carcinomas, salivary duct carcinomas, malignant mixed tumors, and high-grade adenocarcinomas have the most aggressive clinical course, with frequent early spread to regional lymph nodes and distant sites. In contrast, adenoid cystic carcinomas generally display an indolent natural history with a propensity for local or distant recurrence even 10 to 15 years after initial treatment. Knowledge of these unique factors is essential in planning the nuances of therapy.




Introduction


Primary salivary gland malignancies represent less than 5% of all new head and neck cancers, with approximately 3000 new cases diagnosed in the United States annually. These diverse cancers arise from the malignant transformation of the various myoepithelial, ductal, and acinic components of 3 paired major (parotid, submandibular, and sublingual) and minor salivary glands distributed throughout the upper aerodigestive tract. The World Health Organization classifies 24 subtypes characterized by marked biological heterogeneity. For example, high-grade mucoepidermoid carcinomas, salivary duct carcinomas, malignant mixed tumors, and high-grade adenocarcinomas have the most aggressive clinical course, with frequent early spread to regional lymph nodes and distant sites. In contrast, adenoid cystic carcinomas generally display an indolent natural history with a propensity for local or distant recurrence even 10 to 15 years after initial treatment. Knowledge of these unique factors is essential in planning the nuances of therapy.




Initial evaluation


The clinical presentation of a salivary gland mass can reveal a great deal about its nature, with certain symptoms and physical findings associated with malignancy. Rapid growth and/or pain (either localized or referred to the temporomandibular joint or the ear), and paresthesias/hypesthesias caused by perineural spread are concerning signs for malignancy. Facial nerve dysfunction, firmness and fixation of a mass, the presence of trismus for tumors of the parapharyngeal space, and nodal involvement in the neck are findings that increase suspicion for a malignant process.


Computed tomography (CT) and MRI are the two most common imaging modalities used to evaluate salivary gland lesions, with the latter being the method of choice for patients with palpable masses and a strong suspicion for malignancy. Contrast enhancement per se cannot distinguish benign versus malignant processes but it can be critical in delineating the extent of the lesion. Irregular margins; bony invasion; presence of metastatic lymph nodes; and perineural spread along cranial nerve VII (stylomastoid foramen), cranial nerve V-3 (foramen ovale), or V-2 (foramen rotundum) can all be concerning signs of malignancy. Necrosis can also characterize malignancy, particularly in primary squamous cell carcinomas of the salivary glands (likely caused by squamous metaplasia in patients with chronic inflammation). In addition, PET is now being studied for its utility in evaluating salivary gland tumors. Keyes and colleagues found high sensitivity (100%) but low specificity (30%) in predicting malignancy in a cohort of 26 patients with parotid masses.


Histologic confirmation can be acquired before a definitive surgical procedure and this can be useful in planning the type and extent of definitive therapy. Fine-needle aspiration biopsy (FNAB) is the most widely used method for obtaining diagnostic tissue because of its convenience as an office procedure and its associated high sensitivity and specificity. In clinical scenarios in which surgical resection of a growing lesion is planned regardless of the histology, patients may opt to forgo FNAB. With respect to obtaining tissue for surgical preplanning, clinicians should consider the risk of performing unnecessary surgery before knowing the diagnosis based on permanent section evaluation. A frozen section intraoperatively can be useful for a diagnosis but its value compared with FNAB is controversial because its accuracy depends on the experience of the on-call pathologist. Some studies have shown that intraoperative diagnoses can change after permanent section examination. Others have shown the sensitivity and specificity of frozen sections to be 77% and 100%. The risk of either test is in performing unnecessary surgery, like a radical resection or a cervical lymphadenectomy, if a diagnosis of an aggressive malignancy is erroneously reported. Thus, the ultimate scope of treatment should be reserved until permanent section analysis is performed.




Surgical management


Surgical excision of a primary salivary gland malignancy can be curative for most cases when the tumor is small, low grade, and easily accessible. Parotid malignancies are most curable with surgery, followed by the submandibular and sublingual glands. Ultimately, prognosis depends on the gland of origin, histology, grade, and extent of disease (ie, American Joint Commission for Cancer stage). Bulkier tumors or those of aggressive histology/grade are best treated with surgery first, followed by adjuvant therapy.


Surgery of the Parotid Gland and Management of the Facial Nerve


Among parotid tumors, avoiding injury to the facial nerve is critical when tumors are not fixed to, or encasing, the nerve. Thus, surgery typically involves removing the superficial lobe of the gland above the plane of the facial nerve or the deep lobe just underneath it. A superficial parotidectomy can be considered the treatment of choice for most tumors in the superficial lobe of the gland. Enucleation of a tumor should be avoided because this has been associated with tumor spillage and higher recurrence rates. However, patients should be consented for total parotidectomy in cases in which there is deep lobe involvement, because preservation of the nerve often requires removing the superficial lobe to then dissect any deep lobe component from the nerve.


The decision on whether or not the facial nerve should be sacrificed is complex and should be considered on a case-by-case basis. When dealing with a malignant tumor, some general guidelines can be followed, such as preservation of the nerve when the tumor is not involving it. In contrast, sacrifice may be necessary if preservation would lead to grossly positive margins or tumor spillage. This decision can also be influenced by the options available postoperatively. For example, clinicians may err on the side of nerve sacrifice in cases of recurrent disease in which surgical salvage is considered after failed radiation therapy (RT). In cases in which the tumor is noted to abut the mastoid bone or the stylomastoid foramen, the patient should be consented for a mastoidectomy to identify the nerve trunk more proximally. This procedure would expose a portion of uninvolved nerve for grafting if sacrifice is required at the level of the stylomastoid foramen. Besides nerve grafting when the nerve is not working preoperatively or must be sacrificed, the surgeon should be prepared for facial reanimation procedures, such as a gold-weight eyelid implant and/or oral commissure facial sling suspension, to prevent corneal damage and preserve oral competence. This outcome can be accomplished either at the time of tumor resection or later, as an elective procedure. Larger tumors involving the ear canal or middle ear require a temporal bone resection. Patients should be counseled with respect to hearing loss if the ear canal is involved and requires obliteration.


Surgery of the Submandibular and Sublingual Glands


Surgery for tumors of the submandibular and sublingual glands also has its own set of special considerations. When submandibular malignancy is known a priori , clinicians must avoid shelling out the gland, and instead dissection should incorporate the adjacent lymph nodes of a level I cervical lymphadenectomy. Surgical management of a sublingual gland malignancy can be approached transorally. Obtaining negative margins for the sublingual gland can be more challenging because it is not encapsulated by the cervical fascia as is the submandibular gland, because of its location abutting the lingual cortex of the mandible, and its close relationship to the lingual nerve and Wharton duct of the submandibular gland. Often reposition of this duct is necessary to avoid chronic obstruction and inflammation of the submandibular gland.


Surgical considerations for minor salivary glands vary depending on their location. In general, the goal is to perform a wide local excision with clear margins confirmed intraoperatively with frozen sections. The procedure is determined by the location and extent of disease. Given the most common locations where tumors present, mainly the palate and base of tongue, surgical defects can have a significant impact on speech, swallowing, and velopharyngeal competence. Because of the low risk of occult metastatic spread, a prophylactic neck dissection is often not warranted. Smaller tumors of the parapharyngeal space may be approached transorally, whereas larger ones require a transcervical/transparotid approach with or without a mandibulotomy.


Management of the Neck


The overall incidence of cervical lymph metastases is low for patients with malignant salivary cancer, with about 16% of parotid carcinomas and 8% of submandibular/sublingual malignancies presenting with lymphadenopathy. Surgical management of cervical lymph nodes is usually performed in the presence of adenopathy or high-grade histology. Because of the low risk of occult disease, a prophylactic neck dissection is not often performed in other circumstances. Often the final diagnosis of the tumor depends on permanent pathologic analysis after tumor resection. Thus, treatment of the lymph nodes would be considered after the initial surgery. When the risk of occult disease is higher because of aggressive histology or high-grade features (such as advanced T and N descriptors, positive resection margins, perineural invasion) and the need for adjuvant treatment to the resected bed is indicated, management of the N0 neck can be accomplished effectively in this setting.




Radiation therapy


Unresectable locally advanced tumors with involvement of the skull base or encasing of the carotid vessels, and patients who decline surgery because of the requirement of sacrifice of an intact facial nerve or who are otherwise challenged with medical comorbidities, could be treated with primary RT. Mendenhall and colleagues reported on 160 patients who were treated with surgery and RT and compared outcomes with 64 patients treated with RT alone. The 10-year local control rate was 42% for RT alone versus 90% for surgery plus RT. There are obvious selection biases favoring combined modality treatment in these patients, because the patients receiving RT alone tend to have more medical comorbidities that often preclude effective control of the cancer by diminishing tolerance to treatment.


Small, well-localized, low-grade tumors excised with clear margins are best treated with surgery alone. High-grade, advanced stage (T3/4), and inadequately excised tumors treated with surgery alone have a poor prognosis compared with those treated with adjuvant RT, with an overall local regional control rate of 82% compared with 59% for the group treated with surgery alone in several nonrandomized studies. High tumor grade, advanced stage (T3/4), close or positive margins, and nodal involvement are the pathologic factors most commonly associated with a high risk of locoregional failure (30%–60%) after surgery. These patients are the ones most likely to benefit from the addition of postoperative radiotherapy.


A recurring theme with most high-grade salivary gland cancers is that, despite adequate long-term locoregional control rates of 80% to 90%, with surgery and RT, the overall rate of distant failures is in the order of 30% to 50%. Effective systemic therapy agents are likely to dramatically affect the survival outcome in these patients, and several novel agents are being studied.


Evaluation of response following primary RT to indolent varieties of salivary gland cancers, including adenoid cystic carcinomas, poses a problem. Radiological responses in these tumors are gradual, and often barely perceptible when annual interval images are compared. In addition, histologic appearances of posttreatment biopsy-detected residual tumor can be confusing, and do not necessarily represent active malignancy. Hence, patient management should take into consideration the presence of symptoms suggesting disease recurrence/progression in these situations.


Neutron Radiotherapy


Fast neutrons are a form of radiation with high linear energy transfer (LET) that directly damages DNA, independent of the presence of molecular oxygen. In comparison, low-LET radiation (X rays, electrons, protons) causes mostly indirect DNA damage, mediated by an oxygen-dependent pathway. In addition, high-LET RT has significantly higher relative biological effectiveness (RBE) for slowly cycling tumors. The RBE for neutron radiotherapy (NRT) versus fractionated RT was 8.0, compared with 3 to 3.5 for most late-responding normal tissues, with a therapeutic gain factor of 2 to 2.5. Damage by NRT is not readily repairable and there is less variation of sensitivity through the cell cycle. Therefore, NRT has a clear theoretic advantage compared with low-LET radiation in tumor models that have 1 or more of the mechanisms discussed earlier contributing to radioresistance to low-LET radiotherapy.


A prospective Radiation Therapy Oncology Group (RTOG)/Medical Research Council (MRC) randomized controlled study compared low-LET radiotherapy (X rays and electrons) with NRT for unresectable primary and recurrent salivary gland tumors. This study could only enroll 32 patients over a 6-year period, which indicates the rarity of the disease and the difficulty in conducting a prospective multicenter randomized study. An interim analysis at 2 years on 25 eligible patients showed that the NRT group achieved significantly better local control both at the primary site and in the lymph nodes (67% vs 17%; P <.005) and there was a trend toward improved survival (62% vs 25%; P = .1). Hence, the data monitoring committee deemed it unethical to offer any treatment other than NRT for salivary gland cancers. With longer follow-up the survival curves merged and patterns of failure analysis showed that delayed distant metastases accounted for most of the failures in the NRT arm, and local/regional failures predominated in the low-LET arm. The toxicity was worse with NRT in this study, with 9 patients having at least 1 severe or greater complication compared with 4 patients on the low-LET arm. There were no fatal events in either arm. In our experience at the University of Washington, using modern RT techniques, severe toxicity is rare. A retrospective study of 148 patients reported by Douglas and colleagues showed only 6% of patients developing grade 3 or 4 complications using the RTOG/European Organisation for Research and Treatment of Cancer grading system. To date, NRT is currently available only at our center, the University of Washington, creating an issue with access and logistics.


Using conventional RT, selected institutional outcomes seem favorable. For example, Pohar and colleagues reported a local control rate of 85% with primary RT, and Wang and Goodman reported a 5-year actuarial local control of 100% for 9 unresectable parotid gland tumors treated with RT. However, comparison between studies is hampered by methodological problems.


In the postoperative setting, multiple investigators showed good local control rates of approximately 80% to 85% for patients treated with conventional RT. There are also a few recently reported good outcomes with proton beam therapy ; however, longer follow-up will be important to interpret the results in the context of the natural history of salivary gland malignancies.




Systemic therapy


The use of systemic agents in salivary gland cancers is often considered among patients with recurrent/metastatic disease who are not deemed candidates for curative intent therapy. The different histologic subtypes and the heterogeneous natural history within these groups make the determination of the standard of care for systemic therapy challenging. Table 1 outlines the most common of these histologic subtypes. Several issues complicate the interpretation of the data:



  • 1.

    Metastatic salivary tumors are rare. Consequently, it is difficult to conduct randomized studies.


  • 2.

    Single or multi-institutional phase II trials almost always include all histologies, even though they may have different biological behaviors and responses to therapy. The small number of patients included in each histology may lead to misinterpretation of drug activity.


  • 3.

    Subsets of salivary gland cancers, such as adenoid cystic and acinic cell carcinomas, frequently have an indolent growth pattern. The development of asymptomatic, slowly growing pulmonary metastasis can be observed years after the initial diagnosis, which makes it difficult to interpret the time to progression and stable disease rates described in the studies, particularly considering the lack of randomization.


  • 4.

    As a consequence of the neurotropism of some of these histologies, particularly adenoid cystic carcinomas, patients may be very symptomatic (nerve palsies/paralysis). However, they may have disease that it is not easily measurable, particularly in the setting of prior surgeries and radiotherapy, and consequently may be excluded from clinical trials.


Mar 1, 2017 | Posted by in HEMATOLOGY | Comments Off on Salivary Gland Malignancies

Full access? Get Clinical Tree

Get Clinical Tree app for offline access