Perineural Spread in Head and Neck Malignancies: Clinical Significance and Evaluation with¹⁸F-FDG PET/CT

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights. HEAD AND NECK NEOPLASMS 1717 Perineural Spread in Head and Neck Malignancies: Clinical Significance and Evaluation with 18F-FDG PET/CT1 INVITED COMMENTARY See฀discussion฀on฀ this฀article฀by฀Escott฀ (pp฀1736–1738). Fabio฀M.฀Paes,฀MD฀•฀Adam฀D.฀Singer,฀MD฀•฀Adam฀N.฀Checkver,฀MD Ricardo฀A.฀Palmquist,฀MD฀•฀Gabriela฀De฀La฀Vega,฀MD฀•฀Charif฀Sidani,฀MD Certain tumors of the head and neck use peripheral nerves as a direct conduit for tumor growth away from the primary site by a process known as perineural spread. Perineural spread is associated with decreased survival and a higher risk of local recurrence and metastasis. Radiologists play an important role in the assessment and management of head and neck cancer, and positron emission tomography/computed tomography (PET/CT) with 2-[fluorine 18]fluoro-2-deoxy-D-glucose (FDG) is part of the work-up and follow-up of many affected patients. Awareness of abnormal FDG uptake patterns within the head and neck is fundamental for diagnosing perineural spread. The cranial nerves most commonly affected by perineural spread are the trigeminal and facial nerves. Risk of perineural spread increases with a midface location of the tumor, male gender, increasing tumor size, recurrence after treatment, and poor histologic differentiation. Focal or linear increased FDG uptake along the V2 division of the trigeminal nerve or along the medial surface of the mandible, or asymmetric activity in the masticator space, foramen ovale, or Meckel cave should raise suspicion for perineural spread. If FDG PET/ CT findings suggest perineural spread, the radiologist should look at available results of other imaging studies, especially magnetic resonance imaging, to confirm the diagnosis. Knowledge of common FDG PET/CT patterns of neoplastic involvement along the cranial nerves and potential diagnostic pitfalls is of the utmost importance for adequate staging and treatment planning. © RSNA, 2013฀•฀radiographics.rsna.org Abbreviations: FDG = 2-[fluorine 18] fluoro-2-deoxy-D-glucose, MIP = maximum intensity projection, PET/CT = positron emission tomography/ computed tomography, SCCA = squamous cell carcinoma RadioGraphics 2013;฀33:1717–1736฀•฀Published online฀10.1148/rg.336135501฀•฀Content Codes: 1 From the Department of Radiology, Miller School of Medicine, University of Miami, Jackson Memorial Hospital, West Wing-279, 1611 NW 12th Ave, Miami, FL 33136. Recipient of a Certificate of Merit award for an education exhibit at the 2012 RSNA Annual Meeting. Received January 31, 2013; revision requested February 26 and received March 15; accepted April 9. All authors have no financial relationships to disclose. Address correspondence to F.M.P. (e-mail: fpaes@med.miami.edu). © RSNA, 2013 1718 October Special Issue 2013 Introduction Head and neck cancer is a commonly used, broad label for a heterogeneous group of malignancies that affect the oral cavity, pharynx, larynx, nasal cavity, paranasal sinuses, and salivary glands. Given the great diversity of structures and cell types, head and neck cancer encompasses a wide variety of histologic subtypes. Overall, squamous cell carcinoma (SCCA) is the most common type of head and neck cancer (1,2). Tumors of the head and neck may spread by a variety of mechanisms, including direct extension, hematogenous dissemination, or lymphatic dissemination. In addition, certain tumors are able to use peripheral nerves as a direct conduit for tumor growth away from the primary site by a process known as perineural spread. The presence of perineural spread is of great clinical importance, because, even when asymptomatic, it implies a poor prognosis. Furthermore, it is often not recognized at the time of surgery and may occur in the absence of hematogenous or lymphatic metastasis (3,4). It is well established that imaging plays a key role in the assessment and management of head and neck malignancies. Among the available modalities, positron emission tomography/computed tomography (PET/CT) with 2-[fluorine 18]fluoro2-deoxy-D-glucose (FDG) has been proved fundamental in the detection of unknown primary tumors, staging, planning for radiation therapy, and follow-up of most head and neck cancers (5,6). Even though FDG PET/CT is now part of the routine initial work-up and follow-up of patients with head and neck cancer, only a few cases have been reported that demonstrate the diagnosis of perineural spread by using FDG PET/CT (7–10). This oversight in the literature could be related to the interpreting physician’s lack of suspicion or unfamiliarity with the imaging findings and also to the probable low sensitivity of FDG PET/CT for the detection of perineural spread. To our knowledge, no data exist on sensitivity or specificity of this imaging modality for the diagnosis of perineural spread in head and neck malignancies. In this article, we aim to increase awareness of perineural spread in head and neck cancer and describe its imaging features on FDG PET/ radiographics.rsna.org CT images. We summarize the utility of FDG PET/CT in the evaluation of head and neck cancer and discuss the pathogenesis of perineural spread, as well as its clinical and prognostic implications. Different FDG PET/CT imaging patterns suggestive of perineural spread are discussed and illustrated, with an emphasis on the neuroanatomy of the commonly involved cranial nerves. Correlation with other imaging modalities is provided when available. Finally, possible diagnostic pitfalls for interpretation of FDG PET/CT images in the setting of perineural spread are presented. Usefulness of FDG PET/CT for Evaluating Head and Neck Cancer Clinical applications of PET/CT have broadened substantially over recent years because of the increasing availability of the modality and greater experience with its use. FDG PET/CT is an effective imaging modality for the assessment and management of head and neck malignancies. It is superior to PET or CT alone for detection of primary neoplastic lesions in the head and neck and superior to magnetic resonance (MR) imaging or CT alone in the assessment of recurrent cancers after surgery, chemotherapy, or radiation therapy (11–14). The fusion of PET and CT decreases the probability of misinterpreting the PET data by increasing anatomic accuracy and by depicting physiologic function in vivo. It plays a vital role in the staging of newly diagnosed head and neck cancer, specifically in the detection of cervical nodal involvement (sensitivity, 87%–90%; specificity, 80%–93%; and negative predictive value, 89%–95%) (15). PET/CT has greater sensitivity and specificity than either CT or MR imaging for the assessment of lymph node metastasis (15–17). In addition, it helps in the exclusion of distant metastasis and synchronous primary malignancy (18,19). Staging head and neck cancer cases with FDG PET/CT alters the TNM classification in 34% of cases, resulting in a change in radiation therapy technique in 29% (19). Furthermore, the incorporation of FDG PET/ CT into nodal assessment has been proved to significantly reduce the number of unnecessary neck dissections and to generate considerable cost savings (20,21). RG฀ •฀ Volume฀33฀ Number฀6฀ FDG PET/CT is useful for evaluating the response of tumors to treatment, as reduction in uptake correlates with a decline in the number of viable hypermetabolic tumor cells (22). The sensitivity and specificity of FDG PET/CT for detection of residual disease after chemotherapy have been reported as high as 90% and 83%, respectively (23). A meta-analysis evaluation of posttreatment FDG PET/CT studies of the primary tumor site in more than 2300 patients showed pooled sensitivity, specificity, positive predictive value, and negative predictive value of 79.9%, 87.5%, 58.6%, and 95.1%, respectively (24). Another important application of FDG PET/ CT is in the detection of recurrent neoplasm after treatment, as this setting often represents a diagnostic dilemma in the interpretation of images from anatomic-based modalities (14,19,25). Neck dissection, surgical reconstruction, and radiation therapy can make detection of recurrent disease difficult because of granulation tissue formation, necrosis, anatomic changes, loss of normal fat planes, edema, and fibrosis (26,27). On contrast material–enhanced CT and MR images obtained after treatment, disease progression or recurrence are suggested when a lesion increases in size, enhances, or changes in morphology. In reality, however, granulation tissue may also enhance, indolent tumors may be stable in size for a long time, and apparent lesion growth may occur because of normal response to treatment. Relying on an “anatomic” approach or modality to monitor head and neck cancers after therapy may result in treatment delay while waiting for visible tumor progression (28). Because of its “metabolic” approach, FDG PET/CT enables recurrent disease in the treated neck to be detected earlier and with greater sensitivity than when CT or MR imaging are used to monitor cancer treatment. The sensitivity and specificity of FDG PET/CT for detecting recurrent head and neck tumors varies from 95% and 60%, respectively, to as high as 96% and 72% (27,29). FDG PET/CT has also been used for directing biopsy of areas that appear most suspicious for viable tumor (16,30,31). The overall diagnostic performance of FDG PET/CT for assessment of treatment response and surveillance is good, but its positive predictive value is suboptimal. Its negative predictive Paes฀et฀al฀ 1719 value remains high, and negative results from posttreatment FDG PET/CT studies are highly suggestive of remission (24). Unfortunately, aside from a few case reports, no data exist about the usefulness and diagnostic performance of FDG PET/CT for the detection of perineural spread of head and neck cancer. Perineural Spread in Head and Neck Cancer It is important to distinguish perineural spread from perineural invasion because these terms are often used interchangeably in the literature. Perineural฀invasion is a histologic diagnosis that is beyond the resolution of macroscopic imaging modalities, whereas perineural฀spread is dissemination of tumor cells along a nerve that can be detected with imaging techniques (32). The exact pathologic definition of perineural spread continues to be ambiguous (4,33). One challenge posed to the radiologist is whether tumor directly adjacent to a nerve constitutes perineural spread. Clearly, imaging findings along a nerve distant from the primary tumor can be suspected of representing perineural spread, but when a nerve is located within or right next to a primary tumor, the presence of perineural spread and the resultant clinical implications are less clear. Pathogenesis of Perineural Spread Despite the first description of perineural spread of head and neck cancer in 1835, the pathogenesis of perineural spread remains incompletely understood (33). Earlier proposed mechanisms for perineural dissemination included direct invasion and spread through the loose collagenous network of the nerve sheath or lymphatic channels associated with the nerves. However, these theories have fallen out of favor because studies have shown that there is no lymphatic network within the nerve and that the collagenous perineurium actually creates a tight boundary between the nerve and its surrounding environment (4,33). Although some hypothesize that perineural spread is a reflection of increased aggressiveness of tumors, the failure of rapidly growing neoplasms to disseminate along the peripheral nerves, even in advanced stage tumors, 1720 October Special Issue 2013 radiographics.rsna.org Table 1 Perineural Spread in Head and Neck Cancers: Histologic Type and Clinical Presentation* Histologic Type SCCA Adenoid cystic carcinoma Mucoepidermoid carcinoma Melanoma—desmoplastic variant Lymphoma Comments Most prevalent histologic subtype, therefore accounting for largest number of cases of perineural spread Highest association with perineural spread (approximately 60%); accounts for 1%–3% of head and neck malignancies Rare tumor with high association with perineural spread Rare, slow-growing tumor with high rate of intracranial extension Usually advanced cases of non-Hodgkin lymphoma Sources.—References 38–41. *Perineural spread is asymptomatic in up to 40% of patients with radiographic evidence of disease. Symptoms are specific to perineural spread in a particular cranial nerve distribution: Cranial nerve V, facial pain, burning, formication or numbness; cranial nerve VII, facial weakness and droop; mandibular nerve (V3), difficulty chewing; cranial nerve XII, tongue atrophy; cranial nerve X, vocal cord paralysis. makes this theory improbable. Currently, it is hypothesized that perineural spread occurs by means of complex interactions between specific tumor cell types, supporting stroma, and nerves. Genetic and posttranslational adaptations are necessary to provide tumor cells with the ability to break down the supportive collagenous networks and to adhere to and spread along the nerve sheath complex. These abilities are accomplished in part by increasing the activity of a family of proteinases known as the matrix metalloproteinases and by expressing cell surface receptors or their ligands, such as the chemokine CX3CL1 or its receptor CX3CR1 (4,34). Other secreted molecules such as glial cell line– derived neurotrophic factor (GDNF), neurotrophins, and neuron growth factor (NGF) also facilitate tumor growth and perineural spread by acting as chemotactic and chemokinetic agents (35). There has been extensive investigation about the tumor expression of the neural cell adhesion molecule (NCAM), also known as CD56, which is strongly associated with perineural spread. In one study, 89% of all adenoid cystic carcinomas, and 93% of those associated with perineural spread, expressed NCAM (36). NCAM expression has also been described in cases of head and neck SCCA that show perineural spread (37). Histologic Subtypes Because biopsy has been used as the standard of reference for documenting tumor spread along nerves, most of the incidence and prevalence data available refer to perineural invasion rather than perineural spread. The frequency of perineural invasion in head and neck cancer varies with tumor histologic type and location (Table 1). Most cases of perineural invasion occur in SCCA because this malignancy has the highest incidence among all head and neck cancers. The most common sites of SCCA are the tonsils, oral cavity, tongue, floor of the mouth, hypopharynx, and larynx (42,43). Higher rates of perineural invasion (2%–30%) are seen in SCCA cases derived from the oral cavity and larynx, compared with those derived from the skin (2,33). In a series of SCCA of the larynx and hypopharynx, perineural invasion was present in 34% of patients (44). Adenoid cystic carcinoma constitutes 7.5%– 10% of salivary gland cancers but only 1%–3% of head and neck malignancies overall (36,45). These tumors are most often found in the minor salivary glands, and, although relatively rare among the head and neck malignancies, adenoid cystic carcinomas have the highest relative incidence of perineural invasion, with rates as high as 50.7%–56.4% (36,46). The risk of perineural invasion is higher among adenoid cystic carci- RG฀ •฀ Volume฀33฀ Number฀6฀ nomas that arise from the major salivary glands compared with those from the minor salivary glands (36). Mucoepidermoid carcinoma is a rare malignant neoplasm that often originates from the major salivary glands. It represents up to 50% of parotid gland cancers and 30%–40% of salivary gland malignancies overall (47). The overall 5-year survival rate for patients with mucoepidermoid carcinoma is approximately 79%, but those with perineural invasion have a significantly worse prognosis, independent of histologic grade, tumor size, or positive surgical margins (47). Although not typically included in the head and neck cancer category, cutaneous malignancies, particularly SCCA and desmoplastic melanomas, may exhibit perineural invasion, especially in the presence of recurrent disease (48–52). Among nonmelanotic skin cancers, the reported incidence of perineural invasion is low, encompassing 2% among cases of basal cell carcinomas and 3% among cases of cutaneous SCCA (2,32). Among the cutaneous melanoma subtypes encountered, the desmoplastic neurotrophic tumors represent about 1% of cases. The growth pattern of this tumor may be slower than that of the conventional subtype, but the higher prevalence of perineural invasion increases its risk of intracranial dissemination (7,38). In addition to the cancers just mentioned, other malignancies, regardless of their histologic type or primary location, may also be associated with perineural invasion if they involve anatomic sites such as the masticator space, pterygopalatine fossa, the Meckel cave, or cavernous sinus (32). Clinical Features of Perineural Spread Any cranial nerve and its branches can be involved by perineural spread, but the most commonly affected are the trigeminal nerve (cranial nerve V) and the facial nerve (cranial nerve VII), most likely because of their extensive innervation of the head and neck structures (39,53). The risk of perineural spread increases with a midface location of the tumor, male gender, increasing tumor size, recurrence after treatment, and poor histologic differentiation (2). Clinically, it is difficult to diagnose perineural spread because as many as 40% of patients are asymptomatic (4,37). When present, the signs Paes฀et฀al฀ 1721 and symptoms most commonly associated with perineural spread are pain, paresthesia including formication (ie, the sensation of ants or worms under the skin), numbness, and motor weakness (Table 1). The most frequently encountered denervation symptoms are either facial paralysis or masticator muscle weakness when cranial nerve VII and the V3 branch of cranial nerve V, respectively, are involved (32,53). Multiple neuropathies of the cranial nerves are also an ominous finding, suggesting either perineural spread proximal to the cavernous sinus, tumor spread from one cranial nerve to another, or leptomeningeal disease (49). Importance฀of฀Detecting฀Perineural฀Spread฀for฀Treatment฀Planning Therapeutic strategies depend on histologic subtypes and staging at the time of treatment planning. Management of most SCCAs includes single-modality treatment with curative intent (surgery or radiation therapy) for early-stage disease and a multimodality approach for advancedstage disease, often including neoadjuvant and adjuvant therapies. Independent of the nodal status, when perineural spread is present in SCCA, treatment can be changed to include neck dissection, adjuvant therapy, or a larger radiation field (54,55). The primary treatment of choice for salivary gland neoplasms is surgery (45,56). In cases of parotid tumors, facial nerve–sparing surgery may be possible, and it is therefore essential to determine whether perineural spread is present along the course of cranial nerve VII (45). Nevertheless, tumors associated with a higher incidence of perineural spread such as adenoid cystic carcinomas are often treated with surgery followed by radiation therapy because of their higher rate of local recurrence and spread to the skull base (1,45,56). The same principle applies to treatment of skin malignancies. The Mohs procedure is a microscopically controlled surgery that allows for removal of a skin cancer lesion with a very narrow surgical margin and that yields a high cure rate. During the procedure, after each surgical resection, the pathologist examines frozen sections of the skin tissue for cancer cells, informing the surgeon if the circumferential and deep margins are clear of neoplasm. In patients who undergo Mohs 1722 October Special Issue 2013 radiographics.rsna.org Table 2 Head฀and฀Neck฀Tumor฀Sites฀and฀Common฀Cranial฀Nerves฀Involved Common Cranial Nerve Involved Primary Tumor Sites Palate, maxilla, nasopharynx, midface skin, upper lip Lower face, mandible, masticator space, retromolar trigone, parapharyngeal space Temporal bone or scalp, parotid gland, external auditory canal Tongue base Nasopharynx, lacrimal gland Locations of Abnormal FDG Uptake Maxillary nerve Pterygopalatine fossa, foramen rotundum, cavernous sinus Foramen ovale, cavernous sinus, lateral mandible Mandibular nerve Facial nerve, auriculotemporal nerve (mandibular nerve branch) Hypoglossal nerve Maxillary nerve, Vidian nerve, greater superficial petrosal nerve Parotid space, infratemporal fossa, stylomastoid foramen, temporal bone Hypoglossal canal, paraspinal region Vidian canal (difficult to see), pterygopalatine fossa Sources.—References 32, 37, 39, 62, and 63. surgery, even when clear margins are achieved, if perineural invasion is detected histologically, use of adjuvant radiation therapy is indicated (57). Regardless of histologic type and location of the primary head and neck cancer, in advanced cases of perineural spread, postoperative radiation therapy can be used to improve local control (45,55). Prognostic Value of Detecting Perineural Spread There has been considerable historical debate over the prognostic value of detecting perineural spread in cases of head and neck cancer. Although some investigators have failed to demonstrate worsened prognosis with statistical significance, the vast majority have concluded that the presence of perineural invasion is associated with higher risk of local recurrence, higher risk of metastasis, and decreased survival rates (1,3,4,33,35,57–60). The prognostic value of detecting perineural spread varies among the histologic type and location of the head and neck malignancies. In mucosal-derived SCCA, the 3-year survival rate is approximately 23% in those with perineural spread, compared with 49% in those without (4). In cases of SCCA of the oropharynx with invasion into the soft palate, perineural spread is associated with decreased survival rates and shorter disease-free intervals, compared with those associated with invasion of the soft palate alone (61). The 5-year survival rate associated with adenoid cystic carcinoma is only 37% in the presence of perineural spread, compared with 94% if no perineural spread is present (46). Among skin cancers, there is a worse prognosis associated with clinically evident perineural spread, compared with occult cases. Patients with skin cancer and clinically apparent perineural spread had a lower rate of local control at 5 years follow-up (54% of cases), compared with those with occult perineural spread (80%) (2). The most dreaded complication of perineural spread is leptomeningeal carcinomatosis, in which the continuity between the perineurium and leptomeninges and the communication of the perineural space and subarachnoid space permit the dissemination of neoplastic cells from the involved nerves into the meninges (57). When leptomeningeal carcinomatosis is present, the prognosis is poor (57). PET/CT฀Imaging฀Findings Accurate interpretation of FDG PET/CT studies in a patient with known or suspected head and neck malignancy requires a general foundation of knowledge about cranial nerve anatomy. Particular attention should be given to the courses of these nerves, specifically divisions of the trigeminal and facial nerves, as these are most often RG฀ •฀ Volume฀33฀ Number฀6฀ Paes฀et฀al฀ 1723 Figure 1. Drawing illustrates the anatomic distribution of the trigeminal nerve (cranial nerve V); ophthalmic, maxillary, and mandibular nerves (V1, V2, and V3 branches); and geniculate ganglion. CN = cranial nerve, GSPN = greater superficial petrosal nerve, N = nerve. involved in perineural spread (Table 2) (37). In the following subsections, pertinent anatomy and associated patterns of FDG uptake are discussed according to the associated cranial nerve. Any linear or curvilinear focus of abnormal FDG uptake within the head and neck region, especially in the distributions of cranial nerves V and VII, should raise high suspicion for perineural involvement. All three planes and maximum intensity projection (MIP) images must be evaluated. Retrospective evaluation of previous imaging studies, such as contrast-enhanced MR imaging examinations, should be performed for correlation, which may lend to increased sensitivity in a “second look” cranial nerve evaluation. Trigeminal฀Nerve฀(Cranial฀Nerve฀V) The trigeminal nerve is responsible for sensation of the face and motor functions related to mastication. It is the largest of the cranial nerves and has three major divisions: the ophthalmic nerve (V1 branch), the maxillary nerve (V2 branch), and the mandibular nerve (V3 branch) (Fig 1). The ophthalmic and maxillary nerves transmit sensory and autonomic fibers. The mandibular nerve has both sensory and motor functions. Of the trigeminal divisions, the ophthalmic nerve is not often involved by perineural spread and thus will not be discussed in detail. The maxillary division of cranial nerve V (the V2 branch) supplies sensory innervation to the midface, maxillary teeth, and palatine mucosa. It courses from the cavernous sinus through the foramen rotundum to the pterygopalatine fossa and continues through the inferior orbital fissure to the infraorbital groove along the orbital floor into the infraorbital foramen. Along its course, the maxillary nerve gives branches to the nasal, zygomatic, and alveolar nerves (62). SCCA of the skin and desmoplastic melanomas are often seen on the malar surface adjacent to the nose or upper lip, which explains the higher risk of tumor cells spreading through the maxillary nerve (7). Primary tumors of the palate and nasopharynx may gain access to the maxillary nerve and the pterygopalatine fossa through the palatine nerves (32,40). 1724 October Special Issue 2013 radiographics.rsna.org Figure 2. Maxillary nerve (V2) involvement (infraorbital branch) in a 55-year-old man with SCCA of the left cheek who presented with left facial numbness. (a, b) Sequential axial fused PET/CT images show abnormal FDG activity (arrowhead) along the inferior orbital nerve, extending from the left cheek lesion, a finding compatible with perineural spread. (c) Axial contrast-enhanced angled T1-weighted fat-saturated MR image demonstrates enlargement and enhancement along V2 (arrowheads). (d–f) Sagittal fused PET/CT (d), PET MIP (e), and contrastenhanced T1-weighted fat-saturated MR (f) images demonstrate the neoplastic involvement along V2 (arrowheads), extending from the left cheek to the pterygopalatine fossa (arrow). The abnormal FDG activity in perineural spread may be linear or focal along the V2 division. The FDG uptake may be present asymmetrically at the foramen rotundum as a route of central extension (Fig 2). Nonfused, bone-window CT im- ages should be evaluated for asymmetric widening of the foramina (39). The mandibular division of the trigeminal nerve gives sensory innervation to the lower face, mandibular teeth, floor of the mouth, anterior two-thirds of the tongue, and buccal mucosa. The mandibular nerve also provides motor func- RG฀ •฀ Volume฀33฀ Number฀6฀ Paes฀et฀al฀ 1725 Figure 3. Inferior alveolar nerve involvement in a 49-year-old man with mucoepidermoid carcinoma that involves the right mandible; the patient had previously undergone partial right hemi-mandibulectomy, chemotherapy, and radiation therapy. (a) Axial fused PET/CT image demonstrates a focus of uptake (arrow) adjacent to the remaining right mandibular ramus, a finding suggestive of perineural spread along the course of the inferior alveolar nerve. (b) Coronal MIP volumerendered FDG PET image shows a focus of intense FDG uptake (arrow) that corresponds to the recurrent neoplasm, which extends superiorly along the nerve course (arrowheads). (c) Axial contrastenhanced T1-weighted fat-saturated image reveals abnormal enhancement (arrow) in the same location. Perineural spread was proved at biopsy. tion for mastication. The nerve courses from the Meckel cave through the foramen ovale to the masticator space, where it divides into an anterior trunk (terminating in the lingual nerve) and posterior trunk (becoming the often-involved inferior alveolar nerve, which terminates at the mental nerve) (32,62). Tumors along the V3 distribution have the potential of retrograde perineural spread into the masticator space and through the foramen ovale into the Meckel cave. Examples include skin cancers of the lower lip that spread through the mental nerve to the inferior alveolar nerve, and parotid, infratemporal fossa, or lateral facial malignancies that spread through the auriculotemporal branch to the main trunk (39,62,64). Focal or linear increased FDG uptake along the medial surface of the mandible (Figs 3, 4) or asymmetric 1726฀ October฀Special฀Issue฀2013 radiographics.rsna.org Figure 4. Mandibular nerve involvement (inferior alveolar and mental nerve branches) in a 76-year-old woman with advanced SCCA of the right lower lip. (a) Axial fused PET/CT image shows an intense focus of abnormal FDG uptake in the region of the foramen ovale (arrow), a finding highly suggestive of perineural spread along V3. (b) Axial CT image, obtained at the same level but with bone windows, demonstrates enlargement and irregularity of the right foramen ovale (arrow), compared with the contralateral side (arrowhead). (c) Sagittal fused PET/CT image demonstrates intense abnormal linear activity, a finding compatible with perineural spread along the mandibular nerve course (arrowheads), particularly the inferior alveolar branch (arrow), to the level of the right mandibular body. (d) Axial fused PET/CT image demonstrates a hypermetabolic focus in the soft tissues anterior to the right mandibular body (arrow). (e) Axial CT image shows corresponding asymmetric soft-tissue thickening (arrow) and erosion of cortical bone (arrowhead) in the region of the mental foramen, findings compatible with perineural spread involving the mental nerve. activity in the masticator space, foramen ovale, or Meckel cave (Figs 4, 5) should raise suspicion for perineural spread. CT images should be used to evaluate for foraminal widening (ovale, mandibular, and mental foramina) (Figs 4, 6). Cranial nerves III, IV, and VI arise from cerebral peduncles in the posterior aspect of the brainstem and ventral pons, respectively. They RG฀ •฀ Volume฀33฀ Number฀6฀ Paes฀et฀al฀ 1727 Figure฀5.฀ Trigeminal nerve branch and ganglion involvement in a 76year-old woman with advanced SCCA of the right lower lip (same patient as in Fig 4, now referred for restaging). (a–c) Axial PET (a), CT (b), and fused PET/CT (c) images show a hyperattenuating mass with intense curvilinear FDG activity that involves the right Meckel cave/cavernous sinus (arrow), petrous apex, and cerebellopontine angle (arrowhead), compatible with perineural spread. (d) Axial contrast-enhanced MR image shows corresponding enhancing soft tissue that involves the intracranial portions of the trigeminal nerve and ganglion (arrow), compatible with perineural spread. (e–g) Coronal PET (e), CT (f), and fused PET/ CT (g) images show intense abnormal FDG activity along the course of V3 (arrowheads in e and g) associated with enlargement and erosion of the foramen ovale (black arrow in f). Note the normal size and lack of FDG uptake in the left foramen ovale (white arrow in f and g). 1728 October Special Issue 2013 radiographics.rsna.org Figure฀6.฀ Inferior alveolar nerve focal involvement in a 54-year-old man with SCCA of the buccal mucosa. (a–c) Axial PET (a) and fused axial (b) and coronal (c) PET/CT images show a single focus of intense increased FDG uptake in the region of the right mandibular foramen (arrowhead), a finding highly suggestive of perineural involvement of the right inferior alveolar nerve. (d, e) Soft-tissue (d) and bone-window (e) CT images, obtained in a “second look” evaluation after PET/CT, demonstrate subtle effacement of the fat planes (arrow in d) and mild widening of the mandibular foramen (arrow in e). The contralateral side was normal (arrowhead). share a similar course with the ophthalmic nerve along the cavernous sinus into the superior orbital fissure and into the orbit. In rare cases, increased FDG activity along this pathway may be related to perineural spread involving one, a few, or all of these nerves. Facial฀Nerve฀(Cranial฀Nerve฀VII) The facial nerve controls the muscles of facial expression and taste sensations from the anterior two-thirds of the tongue and oral cavity. It emerges from the brainstem between the pons and the medulla, continues as canalicular and labyrinthine segments, geniculate ganglion, and tympanic and mastoid segments before it exits the skull through the stylomastoid foramen. The nerve then courses through the parotid gland, dividing the superficial and deep lobes (65). The geniculate ganglion also gives rise to the greater superficial petrosal nerve, which continues anteriorly in the Vidian canal as the Vidian nerve (39). The facial nerve is most often involved by parotid malignancies or tumors from the adjacent skin that invade the parotid gland (32,62). In these cases, perineural spread should be suspected anytime there is abnormal FDG activity that extends superiorly and centrally toward the stylomastoid foramen or within the temporal bone itself (37). RG฀ •฀ Volume฀33฀ Number฀6฀ Paes฀et฀al฀ 1729 Figure 7. Hypoglossal nerve involvement in a 54-year-old man with tongue weakness 1 year after completion of definitive chemotherapy and radiation therapy for SCCA of the right tongue base. (a) Sagittal MIP PET image shows a large focus of intense FDG accumulation in the right tongue base (arrow), compatible with recurrent SCCA, and associated linear activity along the course of the hypoglossal nerve (arrowheads). (b–d) Axial fused PET/CT (b), coronal PET (c), and coronal fused PET/ CT (d) images show a linear focus of FDG uptake that extends along the right paraspinal region (arrowheads in c and d) to the skull base and hypoglossal canal (arrow in b). Biopsy demonstrated perineural spread. Hypoglossal Nerve฀(Cranial฀Nerve฀XII) The hypoglossal nerve courses from its nucleus at the level of the medulla, exits the skull through the hypoglossal canal, continues below along the digastric muscles, and loops over the hyoid bone to innervate the tongue. Although less common, tongue base cancers may involve the hypoglossal nerve; in these cases, FDG ac- tivity tracks superiorly and centrally through the hypoglossal canal (63) (Fig 7). It should be emphasized that neoplasms may spread from one cranial nerve to other cranial nerves because of their proximity and their communication between the peripheral branches (Fig 8), particularly between cranial nerves V 1730 October Special Issue 2013 radiographics.rsna.org Figure 8. Trigeminal nerve involvement in a 75-year-old man 12 years after surgical resection of a melanoma of the right paranasal skin. (a) Axial fused PET/CT image shows intense FDG accumulation in the region of the right inferior orbital fissure, infraorbital groove (arrow), and right foramen ovale (arrowhead), compatible with perineural spread along V2 and V3. (b) Axial fused PET/CT image shows involvement of V3 at the mandibular foramen (arrowhead). (c) Axial CT image depicts enlargement and erosion of the mandibular foramen (arrow). (d) Sagittal PET/CT image shows FDG activity along V2 (arrow) and V3 (arrowheads). (e, f) Coronal fused PET/CT (e) and contrast-enhanced T1-weighted fat-saturated MR (f) images show intense FDG accumulation and abnormal enhancement along V3 (arrowheads). We believe the patient initially developed V2 perineural spread from the original site and, from peripheral branch interconnection with V3 within the pterygopalatine fossa, subsequently developed V3 perineural spread. The patient also had involvement of the Meckel cave and cavernous sinus (disseminated through another possible pathway and not shown here). and VII. Facial nerve fibers, via the greater superficial petrosal nerve, communicate with small V2 branches in the pterygopalatine ganglion after coursing through the Vidian canal (Fig 1). Other regions of cranial nerve interconnection include the ophthalmic and maxillary nerves (V1 and V2) in the cavernous sinus; the ophthalmic, maxillary, and mandibular nerves (V1, V2, and V3) in the Meckel cave; and the mandibular nerve (V3) and cranial nerve VII through the auriculotemporal nerve (32,37,62). RG฀ •฀ Volume฀33฀ Number฀6฀ Paes฀et฀al฀ 1731 Table 3 Imaging฀Evaluation฀of฀Perineural฀Spread฀in฀Head฀and฀Neck฀Malignancies Imaging Modality Primary Use in Head and Neck Cancer Imaging Findings Suggestive of Perineural Spread FDG PET/CT Nodal and metastatic assessment; evaluation of unknown primary site, recurrence, and response to treatment Contrast-enhanced MR imaging Staging and follow-up; evaluation of local extent of primary tumor and perineural spread Linear or focal uptake along the cranial nerves or neural foramina; asymmetric uptake in the pterygopalatine fossa, Meckel cave, or cavernous sinus; bone erosion or widening of the neural foramina Thickening and/or enhancement along the cranial nerves; enhancement of foramina, pterygopalatine fossa, Meckel cave, or cavernous sinus; effacement of perineural fat planes; muscle atrophy Contrastenhanced CT Staging and follow-up; bone involvement or destruction Thickening and/or enhancement along the cranial nerves; loss of adjacent fat planes; bone erosion or widening of the neural foramina Although it is possible to diagnose perineural spread on FDG PET/CT images, as exemplified in this article, we emphasize that there are no available data about the sensitivity, specificity, positive predictive value, and negative predictive value of FDG PET/CT in the detection of perineural spread. Correlation of PET/CT with฀Other฀Imaging฀Modalities If FDG PET/CT findings suggest perineural spread, it is pertinent to look at the available results of other imaging studies to confirm the diagnosis (Table 3). Because MR imaging is the standard of reference for imaging of suspected perineural spread, correlation is advised also to evaluate the complete extent of perineural spread. When correlating FDG PET/CT findings with those of MR imaging, it is important to search for direct and indirect signs of perineural spread. Direct signs include prominence; irregularity; and asymmetric, often avid, enhancement of the nerves. Indirect signs include findings of muscle denervation. In the acute and subacute phases of denervation, denervated muscles show high T2 signal intensity; in the chronic phase, they demonstrate fatty replacement. In addition, particular at- Comments Highest negative predictive value for metastasis and tumor recurrence; limited spatial resolution; attention to physiologic uptake and mimickers Standard of reference for evaluation of perineural spread; posttreatment changes may lead to falsepositive findings; cannot be used in patients with pacemakers, certain medical devices, claustrophobia, renal failure, etc Initial diagnostic tool due to speed and availability; limited soft-tissue contrast resolution; best for bone evaluation; beam hardening artifact at the skull base limits evaluation tention should be given to symmetry of fat planes that surround the nerves, to ensure they are not obliterated by enhancing soft tissue (37,62). Suspicious findings seen during the CT portion of the PET/CT study can be confirmed in a separate dedicated CT examination that focuses on the region of interest. Findings such as foraminal destruction, erosion, or widening are better depicted with thinner bone-window CT sections (64). Potential Pitfalls FDG PET/CT is effective in staging and restaging of malignancies of the head and neck, but imaging pitfalls exist. Physiologic uptake, asymmetric tracer distribution in different physiologic states, and altered anatomic landmarks after treatment often make image interpretation challenging. Variability in the physiologic uptake in normal structures such as the nasal turbinates, pterygoid muscles, extraocular muscles, and lymphoid tissue of the Waldeyer ring can confound interpretation and result in false-positive findings (30). Uptake in muscles, which may asymmetrically contract during the examination, may also lead to false-positive results, and this scenario can 1732 October Special Issue 2013 radiographics.rsna.org Figure฀9.฀ Pitfall: paraganglioma in a 62-year-old woman with a history of treated nasopharyngeal cancer who presented with a soft-tissue mass in the left skull base during imaging follow-up. Axial fused PET/CT (a), coronal PET (b), and coronal fused PET/ CT (c) images show intense abnormal FDG uptake in the region of the jugular foramen (arrow). Although the patient’s history suggested recurrence of nasopharyngeal cancer with perineural spread, surgical biopsy revealed a glomus jugulare tumor. be particularly problematic, since one looks for possible asymmetric activity along cranial nerves as a sign of perineural spread. Often, a review of the CT portion of the PET/CT study allows correlation of findings with normal anatomic structures, thus reducing the rate of false-positive results (27). The nearness of perineural spread to anatomic structures that have inherently high uptake may be problematic, as coregistration algorithms for fusion of PET and CT images are not perfect and can be compromised by motion. Coregistration artifacts may be particularly evident in evaluations of the base of skull, where the cranial nerves exiting through neural foramina can be obscured by normal brain FDG activity (66). Also, insufficient spatial resolution for lesions smaller than 1 cm limits the sensitivity of PET/CT for detection of perineural spread, unless there is significant increased FDG uptake that contrasts with low background activity. This is the case for small nerve branches, which can be involved by neoplastic cells and not visualized on FDG PET/CT images because of volume averaging with normal adjacent tissues (66,67). Posttreatment changes are a challenge for interpretation because inflammation results in increased FDG uptake, mainly when PET/CT RG฀ •฀ Volume฀33฀ Number฀6฀ is performed within the 1st month after surgery, chemotherapy, or radiation therapy (29,68). The current consensus is that FDG PET/CT should be performed at least 6–8 weeks after surgery and 8–10 weeks after chemotherapy or radiation therapy, because the delayed follow-up studies provide more accurate information about residual or recurrent tumor (overall sensitivity, 90%; specificity, 93.3%) (6,14,19,66,69). Not every case of abnormal FDG uptake along a nerve distribution is caused by malignant neoplasm. There are reports of perineural spread associated with mucormycosis, aspergillosis, and sinonasal sarcoidosis (37). Entities such as paragangliomas, neurofibromas, or schwannomas may give rise to a similar linear appearance of increased FDG activity, although cross-sectional images may show the different spaces involved (Fig 9). In some cases, meningiomas may protrude through the skull base foramina, mimicking perineural spread. There are also numerous benign dermatologic conditions associated with perineural spread (70). Correlation with patient history, observations from physical examination, and, often, analysis of a histologic specimen play an important role in these cases. Conclusions Head and neck cancer is a dreadful group of diseases associated with high morbidity and mortality when patients present with advanced-stage disease. Certain tumors possess the ability to spread along the cranial nerves through a series of complex interactions between tumor cells, the nerve sheath complex, and stromal cells. Regardless of the histologic subtype, the presence of perineural spread is associated with poor prognosis because of the increased risk of local recurrence, increased risk of metastasis, and decreased rates of survival. With knowledge of the cranial nerve anatomy, potential pathways of perineural tumor spread can be predicted on the basis of the primary tumor location. Even though MR imaging is the standard of reference for the imaging evaluation of perineural spread, the radiologist or nuclear medicine physician should not miss the opportunity of reporting the presence of perineural spread on FDG PET/CT images obtained during the staging or follow-up of head and neck cancers. Awareness of the described patterns of abnormal FDG uptake within the head and neck region is fundamental for diagnosing perineural Paes฀et฀al฀ 1733 involvement. Axial, coronal, sagittal, and MIP PET images should be evaluated thoroughly, and, whenever available, correlation of FDG PET/ CT images with those from other studies such as contrast-enhanced MR imaging should be used to avoid misinterpretation. Finally, it is important to report the suspicion of perineural spread when interpreting FDG PET/CT images because, if confirmed, the presence of perineural involvement may change case management. Acknowledgment.—The authors thank Christopher Granville, MD, Miller School of Medicine, University of Miami, Miami, Florida, for the outstanding illustration of the trigeminal nerve included in the article. References 1. Johnston M, Yu E, Kim J. Perineural invasion and spread in head and neck cancer. Expert Rev Anticancer Ther 2012;12(3):359–371. 2. Balamucki CJ, Mancuso AA, Amdur RJ, et al. Skin carcinoma of the head and neck with perineural invasion. Am J Otolaryngol 2012;33(4):447–454. 3. Fowler BZ, Crocker IR, Johnstone PA. Perineural spread of cutaneous malignancy to the brain: a review of the literature and five patients treated with stereotactic radiotherapy. Cancer 2005;103(10): 2143–2153. 4. Liebig C, Ayala G, Wilks JA, Berger DH, Albo D. Perineural invasion in cancer: a review of the literature. Cancer 2009;115(15):3379–3391. 5. Chatterjee S, Frew J, Mott J, et al. Variation in radiotherapy target volume definition, dose to organs at risk and clinical target volumes using anatomic (computed tomography) versus combined anatomic and molecular imaging (positron emission tomography/computed tomography): intensity-modulated radiotherapy delivered using a tomotherapy Hi Art machine—final results of the VortigERN study. Clin Oncol (R Coll Radiol) 2012;24(10):e173–e179. 6. Zollinger LV, Wiggins RH. A head and neck radiologist’s perspective on best practices for the usage of PET/CT scans for the diagnosis and treatment of head and neck cancers. Arch Otolaryngol Head Neck Surg 2012;138(8):754–758. 7. Moulin-Romsee G, Benamor M, Neuenschwander S. Desmoplastic neurotropic melanoma in a patient with trigeminal neuralgia: FDG PET/CT and MRI. Clin Nucl Med 2008;33(5):353–355. 8. Conrad GR, Sinha P, Holzhauer M. Perineural spread of skin carcinoma to the base of the skull: detection with FDG PET and CT fusion. Clin Nucl Med 2004;29(11):717–719. 9. Liu SH, Kao PF, Chiu CH, Liu DW, Hsieh HJ, Chen PR. Perineural recurrence of buccal cancer demonstrated by F-18 FDG PET/CT. Clin Nucl Med 2010;35(3):189–191. 1734 October Special Issue 2013 10. Nguyen BD, Roarke MC. Salivary duct carcinoma with perineural spread to facial canal: F-18 FDG PET/CT detection. Clin Nucl Med 2008;33(12): 925–928. 11. Branstetter BF 4th, Blodgett TM, Zimmer LA, et al. Head and neck malignancy: is PET/CT more accurate than PET or CT alone? Radiology 2005;235 (2):580–586. 12. Anzai Y, Carroll WR, Quint DJ, et al. Recurrence of head and neck cancer after surgery or irradiation: prospective comparison of 2-deoxy-2-[F-18]fluoroD-glucose PET and MR imaging diagnoses. Radiology 1996;200(1):135–141. 13. Kitagawa Y, Nishizawa S, Sano K, et al. Prospective comparison of 18F-FDG PET with conventional imaging modalities (MRI, CT, and 67Ga scintigraphy) in assessment of combined intraarterial chemotherapy and radiotherapy for head and neck carcinoma. J Nucl Med 2003;44(2):198–206. 14. Kubota K, Yokoyama J, Yamaguchi K, et al. FDGPET delayed imaging for the detection of head and neck cancer recurrence after radio-chemotherapy: comparison with MRI/CT. Eur J Nucl Med Mol Imaging 2004;31(4):590–595. 15. Sadick M, Schoenberg SO, Hoermann K, Sadick H. Current oncologic concepts and emerging techniques for imaging of head and neck squamous cell cancer. GMS Curr Top Otorhinolaryngol Head Neck Surg 2012;11:Doc08. 16. Laubenbacher C, Saumweber D, Wagner-Manslau C, et al. Comparison of fluorine-18-fluorodeoxyglucose PET, MRI and endoscopy for staging head and neck squamous-cell carcinomas. J Nucl Med 1995; 36(10):1747–1757. 17. Subramaniam RM, Truong M, Peller P, Sakai O, Mercier G. Fluorodeoxyglucose-positron-emission tomography imaging of head and neck squamous cell cancer. AJNR Am J Neuroradiol 2010;31(4): 598–604. 18. Strobel K, Haerle SK, Stoeckli SJ, et al. Head and neck squamous cell carcinoma (HNSCC): detection of synchronous primaries with (18)F-FDGPET/CT. Eur J Nucl Med Mol Imaging 2009;36(6): 919–927. 19. Mak D, Corry J, Lau E, Rischin D, Hicks RJ. Role of FDG-PET/CT in staging and follow-up of head and neck squamous cell carcinoma. Q J Nucl Med Mol Imaging 2011;55(5):487–499. 20. Pryor DI, Porceddu SV, Scuffham PA, Whitty JA, Thomas PA, Burmeister BH. Economic analysis of FDG-PET-guided management of the neck after primary chemoradiotherapy for node-positive head and neck squamous cell carcinoma. Head Neck 2012; Sep 18. [Epub ahead of print] 21. Sher DJ, Tishler RB, Annino D, Punglia RS. Costeffectiveness of CT and PET-CT for determining the need for adjuvant neck dissection in locally advanced head and neck cancer. Ann Oncol 2010;21 (5):1072–1077. radiographics.rsna.org 22. Abe Y, Matsuzawa T, Fujiwara T, et al. Assessment of radiotherapeutic effects on experimental tumors using 18F-2-fluoro-2-deoxy-D-glucose. Eur J Nucl Med 1986;12(7):325–328. 23. Lowe VJ, Dunphy FR, Varvares M, et al. Evaluation of chemotherapy response in patients with advanced head and neck cancer using [F-18]fluorodeoxyglucose positron emission tomography. Head Neck 1997;19(8):666–674. 24. Gupta T, Master Z, Kannan S, et al. Diagnostic performance of post-treatment FDG PET or FDG PET/CT imaging in head and neck cancer: a systematic review and meta-analysis. Eur J Nucl Med Mol Imaging 2011;38(11):2083–2095. 25. Ong SC, Schöder H, Lee NY, et al. Clinical utility of 18F-FDG PET/CT in assessing the neck after concurrent chemoradiotherapy for locoregional advanced head and neck cancer. J Nucl Med 2008;49 (4):532–540. 26. Rege SD, Chaiken L, Hoh CK, et al. Change induced by radiation therapy in FDG uptake in normal and malignant structures of the head and neck: quantitation with PET. Radiology 1993;189(3): 807–812. 27. Blodgett TM, Fukui MB, Snyderman CH, et al. Combined PET-CT in the head and neck. I. Physiologic, altered physiologic, and artifactual FDG uptake. RadioGraphics 2005;25(4):897–912. 28. Peterson AM, Meltzer CC, Evanson EJ, Flickinger JC, Kondziolka D. MR imaging response of brain metastases after gamma knife stereotactic radiosurgery. Radiology 1999;211(3):807–814. 29. Wong RJ, Lin DT, Schöder H, et al. Diagnostic and prognostic value of [(18)F]fluorodeoxyglucose positron emission tomography for recurrent head and neck squamous cell carcinoma. J Clin Oncol 2002; 20(20):4199–4208. 30. Jabour BA, Choi Y, Hoh CK, et al. Extracranial head and neck: PET imaging with 2-[F-18]fluoro2-deoxy-D-glucose and MR imaging correlation. Radiology 1993;186(1):27–35. 31. Wahl RL, Quint LE, Cieslak RD, Aisen AM, Koeppe RA, Meyer CR. “Anatometabolic” tumor imaging: fusion of FDG PET with CT or MRI to localize foci of increased activity. J Nucl Med 1993;34(7): 1190–1197. 32. Ginsberg LE. Imaging of perineural tumor spread in head and neck cancer. Semin Ultrasound CT MR 1999;20(3):175–186. 33. Binmadi NO, Basile JR. Perineural invasion in oral squamous cell carcinoma: a discussion of significance and review of the literature. Oral Oncol 2011; 47(11):1005–1010. 34. Marchesi F, Piemonti L, Mantovani A, Allavena P. Molecular mechanisms of perineural invasion: a forgotten pathway of dissemination and metastasis. Cytokine Growth Factor Rev 2010;21(1):77–82. 35. Iwamoto S, Odland PB, Piepkorn M, Bothwell M. Evidence that the p75 neurotrophin receptor mediates perineural spread of desmoplastic melanoma. J Am Acad Dermatol 1996;35(5 Pt 1):725–731. RG฀ •฀ Volume฀33฀ Number฀6฀ 36. Barrett AW, Speight PM. Perineural invasion in adenoid cystic carcinoma of the salivary glands: a valid prognostic indicator? Oral Oncol 2009;45(11): 936–940. 37. Gandhi D, Gujar S, Mukherji SK. Magnetic resonance imaging of perineural spread of head and neck malignancies. Top Magn Reson Imaging 2004; 15(2):79–85. 38. Chang PC, Fischbein NJ, McCalmont TH, et al. Perineural spread of malignant melanoma of the head and neck: clinical and imaging features. AJNR Am J Neuroradiol 2004;25(1):5–11. 39. Caldemeyer KS, Mathews VP, Righi PD, Smith RR. Imaging features and clinical significance of perineural spread or extension of head and neck tumors. RadioGraphics 1998;18(1):97–110. 40. Parker GD, Harnsberger HR. Clinical-radiologic issues in perineural tumor spread of malignant diseases of the extracranial head and neck. RadioGraphics 1991;11(3):383–399. 41. Warden KF, Parmar H, Trobe JD. Perineural spread of cancer along the three trigeminal divisions. J Neuroophthalmol 2009;29(4):300–307. 42. Hayat MJ, Howlader N, Reichman ME, Edwards BK. Cancer statistics, trends, and multiple primary cancer analyses from the Surveillance, Epidemiology, and End Results (SEER) Program. Oncologist 2007;12(1):20–37. 43. Carvalho AL, Nishimoto IN, Califano JA, Kowalski LP. Trends in incidence and prognosis for head and neck cancer in the United States: a site-specific analysis of the SEER database. Int J Cancer 2005; 114(5):806–816. 44. Chirilă M, Bolboacă SD, Cosgarea M, Tomescu E, Mureşan M. Perineural invasion of the major and minor nerves in laryngeal and hypopharyngeal cancer. Otolaryngol Head Neck Surg 2009;140(1): 65–69. 45. Bell RB, Dierks EJ, Homer L, Potter BE. Management and outcome of patients with malignant salivary gland tumors. J Oral Maxillofac Surg 2005;63 (7):917–928. 46. Majoie CB, Hulsmans FJ, Verbeeten B Jr, et al. Perineural tumor extension along the trigeminal nerve: magnetic resonance imaging findings. Eur J Radiol 1997;24(3):191–205. 47. McHugh CH, Roberts DB, El-Naggar AK, et al. Prognostic factors in mucoepidermoid carcinoma of the salivary glands. Cancer 2012;118(16): 3928–3936. 48. Goepfert H, Dichtel WJ, Medina JE, Lindberg RD, Luna MD. Perineural invasion in squamous cell skin carcinoma of the head and neck. Am J Surg 1984; 148(4):542–547. 49. Woodruff WW Jr, Yeates AE, McLendon RE. Perineural tumor extension to the cavernous sinus from superficial facial carcinoma: CT manifestations. Radiology 1986;161(2):395–399. 50. Majoie CB, Hulsmans FJ, Castelijns JA, Walter A, Bras J, Peeters FL. Perineural tumor extension of facial malignant melanoma: CT and MRI. J Comput Assist Tomogr 1993;17(6):973–975. Paes฀et฀al฀ 1735 51. Batsakis JG, Raymond AK. Desmoplastic melanoma. Ann Otol Rhinol Laryngol 1994;103(1): 77–79. 52. Catalano PJ, Sen C, Biller HF. Cranial neuropathy secondary to perineural spread of cutaneous malignancies. Am J Otol 1995;16(6):772–777. 53. Mendenhall WM, Ferlito A, Takes RP, et al. Cutaneous head and neck basal and squamous cell carcinomas with perineural invasion. Oral Oncol 2012;48 (10):918–922. 54. Tai SK, Li WY, Yang MH, et al. Treatment for T1-2 oral squamous cell carcinoma with or without perineural invasion: neck dissection and postoperative adjuvant therapy. Ann Surg Oncol 2012;19(6): 1995–2002. 55. Expert Panel on Radiation Oncology-Head and Neck, Salama JK, Saba N, et al. ACR appropriateness criteria® adjuvant therapy for resected squamous cell carcinoma of the head and neck. Oral Oncol 2011;47(7):554–559. 56. Shen C, Xu T, Huang C, Hu C, He S. Treatment outcomes and prognostic features in adenoid cystic carcinoma originated from the head and neck. Oral Oncol 2012;48(5):445–449. 57. Dunn M, Morgan MB, Beer TW. Perineural invasion: identification, significance, and a standardized definition. Dermatol Surg 2009;35(2):214–221. 58. Farasat S, Yu SS, Neel VA, et al. A new American Joint Committee on Cancer staging system for cutaneous squamous cell carcinoma: creation and rationale for inclusion of tumor (T) characteristics. J Am Acad Dermatol 2011;64(6):1051–1059. 59. Lee KJ, Abemayor E, Sayre J, Bhuta S, Kirsch C. Determination of perineural invasion preoperatively on radiographic images. Otolaryngol Head Neck Surg 2008;139(2):275–280. 60. Panizza B, Solares CA, Redmond M, Parmar P, O’Rourke P. Surgical resection for clinical perineural invasion from cutaneous squamous cell carcinoma of the head and neck. Head Neck 2012;34(11): 1622–1627. 61. Meng FY, Ko JY, Lou PJ, et al. The determining risk factors for treatment outcomes in patients with squamous cell carcinoma of the hard palate. Ann Surg Oncol 2012;19(6):2003–2010. 62. Nemec SF, Herneth AM, Czerny C. Perineural tumor spread in malignant head and neck tumors. Top Magn Reson Imaging 2007;18(6):467–471. 63. Chong VF, Fan YF. Hypoglossal nerve palsy in nasopharyngeal carcinoma. Eur Radiol 1998;8(6): 939–945. 64. Ong CK, Chong VF. Imaging of perineural spread in head and neck tumours. Cancer Imaging 2010;10 (Spec No A):S92–S98. 65. Sheth S, Branstetter BF 4th, Escott EJ. Appearance of normal cranial nerves on steady-state free precession MR images. RadioGraphics 2009;29(4): 1045–1055. 1736฀ October฀Special฀Issue฀2013 66. Fukui MB, Blodgett TM, Snyderman CH, et al. Combined PET-CT in the head and neck. II. Diagnostic uses and pitfalls of oncologic imaging. RadioGraphics 2005;25(4):913–930. 67. Avril NE, Weber WA. Monitoring response to treatment in patients utilizing PET. Radiol Clin North Am 2005;43(1):189–204. radiographics.rsna.org 68. Rogers JW, Greven KM, McGuirt WF, et al. Can post-RT neck dissection be omitted for patients with head-and-neck cancer who have a negative PET scan after definitive radiation therapy? Int J Radiat Oncol Biol Phys 2004;58(3):694–697. 69. Minn H, Paul R, Ahonen A. Evaluation of treatment response to radiotherapy in head and neck cancer with fluorine-18 fluorodeoxyglucose. J Nucl Med 1988;29(9):1521–1525. 70. Ronaghy A, Yaar R, Goldberg LJ, Mahalingam M, Bhawan J. Perineural involvement: what does it mean? Am J Dermatopathol 2010;32(5):469–476. Teaching Points October Special Issue 2013 Perineural Spread in Head and Neck Malignancies: Clinical Significance and Evaluation with 18F-FDG PET/CT Fabio฀M.฀Paes,฀MD฀•฀Adam฀D.฀Singer,฀MD฀•฀Adam฀N.฀Checkver,฀MD฀•฀Ricardo฀A.฀Palmquist,฀MD฀•฀Gabriela฀ De฀La฀Vega,฀MD฀•฀Charif฀Sidani,฀MD RadioGraphics 2013;฀33:1717–1736฀•฀Published online฀10.1148/rg.336135501฀•฀Content Codes: Page 1718 The presence of perineural spread is of great clinical importance, because, even when asymptomatic, it implies a poor prognosis. Furthermore, it is often not recognized at the time of surgery and may occur in the absence of hematogenous or lymphatic metastasis. Page 1720 The frequency of perineural invasion in head and neck cancer varies with tumor histologic type and location. Most cases of perineural invasion occur in SCCA because this malignancy has the highest incidence among all head and neck cancers. The most common sites of SCCA are the tonsils, oral cavity, tongue, floor of the mouth, hypopharynx, and larynx. Higher rates of perineural invasion (2%–30%) are seen in SCCA cases derived from the oral cavity and larynx, compared with those derived from the skin. In a series of SCCA of the larynx and hypopharynx, perineural invasion was present in 34% of patients. Adenoid cystic carcinoma constitutes 7.5%–10% of salivary gland cancers but only 1%–3% of head and neck malignancies overall. These tumors are most often found in the minor salivary glands, and, although relatively rare among the head and neck malignancies, adenoid cystic carcinomas have the highest relative incidence of perineural invasion, with rates as high as 50.7%–56.4%. Page 1721 Any cranial nerve and its branches can be involved by perineural spread, but the most commonly affected are the trigeminal nerve (cranial nerve V) and the facial nerve (cranial nerve VII), most likely because of their extensive innervation of the head and neck structures. The risk of perineural spread increases with a midface location of the tumor, male gender, increasing tumor size, recurrence after treatment, and poor histologic differentiation. Page 1723 Any linear or curvilinear focus of abnormal FDG uptake within the head and neck region, especially in the distributions of cranial nerves V and VII, should raise high suspicion for perineural involvement. All three planes and maximum intensity projection (MIP) images must be evaluated. Retrospective evaluation of previous imaging studies, such as contrast-enhanced MR imaging examinations, should be performed for correlation, which may lend to increased sensitivity in a “second look” cranial nerve evaluation. Pages฀1729–1730 It should be emphasized that neoplasms may spread from one cranial nerve to other cranial nerves because of their proximity and their communication between the peripheral branches, particularly between cranial nerves V and VII. Facial nerve fibers, via the greater superficial petrosal nerve, communicate with small V2 branches in the pterygopalatine ganglion after coursing through the Vidian canal. Other regions of cranial nerve interconnection include the ophthalmic and maxillary nerves (V1 and V2) in the cavernous sinus; the ophthalmic, maxillary, and mandibular nerves (V1, V2, and V3) in the Meckel cave; and the mandibular nerve (V3) and cranial nerve VII through the auriculotemporal nerve.