Perineural spread of cancer

Introduction

Perineural invasion (PNI) refers to a rare type of contiguous spread of neoplastic cells from their primary site along the potential space between or beneath the layers of perineurium. This entity is best described in cancers of the head and neck but is also well described in other solid tumors. Exact pathogenesis remains unclear, but proximity of the primary tumor to major nerves and plexus and the tumor cell’s ability to infiltrate and proliferate within the perineural space, as well as directed molecular interactions between the tumor and its microenvironment, are thought to play an important role. The clinical presentation varies depending on the nerves involved. Developing a broad differential to include direct compression of nerves by tumors, delayed effects of prior radiation therapy, infections, and paraneoplastic syndromes as well as work-up for other common causes of neuropathies is encouraged. Contrast-enhanced magnetic resonance imaging is the imaging modality of choice, but a biopsy is frequently required for confirmation of diagnosis. Accurate detection may help predict prognosis and guide therapy. We recommend a multidisciplinary approach in selecting the curative and palliative treatment options.

Classification of perineural invasion

PNI refers to a rare type of contiguous spread of neoplastic cells from their primary site along the potential space between or beneath the layers of perineurium ( Fig. 21.1 ). It is defined as “tumor in close proximity to nerve and involving at least 33% of its circumference or tumor cells within any of the three layers of the nerve sheath (epineurium, perineurium and endoneurium).” It was first described in 1952 by Dr. Frederick Mohs and Dr. Theodore Lathrop as “silent” extensions along the nerve sheath. Even though this phenomenon is frequently discussed in patients with primary head and neck malignancies, it is well defined in malignancies of the pancreas, prostate, and colorectum. Perineural invasion can be broadly divided into two main categories based on the patient’s clinical presentation:

•

Microscopic perineural invasion (MPNI) is the asymptomatic type (60–70%) where diagnosis is made on surveillance imaging, biopsy, or autopsy.
•

Clinical perineural invasion (CPNI) is the symptomatic type (30–40%). ^,

Another approach to their classification is to differentiate between the type and caliber of nerve that is involved:

•

Perineural invasion ( PNI) is used to describe involvement of small, unnamed nerves around a tumor.
•

Perineural spread (PNS) is the term used when there is evidence of invasion and spread along a larger, named nerve or cranial nerves.

Although both can result in symptoms, PNS is generally considered to be more aggressive and frequently presents with neuropathy of the affected nerve. ^,

Pathogenesis

The exact pathogenesis remains unclear, but the proximity of the primary tumor to major nerves and plexus and the tumor cell’s ability to infiltrate and proliferate within the perineural space, as well as directed molecular interactions between the tumor and its microenvironment, are thought to play an important role. Additionally, tumor-associated macrophages, Schwann cells, and other cells in the tumor microenvironment such as fibroblasts and extracellular matrix adhesion proteins may play an important role in perineural invasion and spread. A review of the literature suggests that cells capable of producing neurotropins, including nerve growth factor, brain-derived neurotropic factor, and glial cell line–derived neurotrophic factor, are linked to PNI in multiple type of cancers. Similarly, several studies have suggested that prostate tumor cells located adjacent to nerves have increased proliferation rates compared to those located further away, suggesting a dynamic interaction. ^, Evidence of PNI predisposes locoregional recurrence even after complete resection. The presence of PNI is independently associated with decreased survival. Due to its prognostic and predictive implications, the College of American Pathologists recommends including perineural space invasion in histopathologic reports for all cutaneous and head and neck malignancies. It is estimated that around 2–6% of patients with malignant head and neck tumors have evidence of PNS, whereas other tumors, such as prostate cancer, may demonstrate up to 75% PNI without significant PNS. ^, ^,

Clinical presentation

In head and neck cancers, PNS frequently affects cranial nerves V (CN V) and VII (CN VII). ^, ^, Patients with high-grade or poorly differentiated tumors, tumors involving the mid-face, male gender, and large primary tumors are at a higher risk of developing PNI. ^, Presenting symptoms are subtle, variable, and usually progress over the years. If the CN V is affected, symptoms include numbness (72%), pain (46%), or paresthesia. Patients may describe this sensation as ants crawling underneath the skin (formication) (10%). Similarly, patients may present with lower motor neuron type facial palsy if the CN VII is affected (33%). ^, ^, The pattern of nerve involvement is closely related to the primary location of the tumor. For example, tumors of the supraorbital region tend to invade CN V1 (ophthalmic nerve), whereas tumors of the medial cheek invade CN V2 (47%) (maxillary nerve). Similarly, tumors of the temporal region invade CN V2 and CN V3 (mandibular nerve), whereas tumors of the peri-auricular region invade CN VII (26%). ^, The subtleness of the symptoms and slow progression over months to years may cause an incorrect diagnosis, such as Bell’s palsy or trigeminal neuralgia. A high index of suspicion and good history will help broaden the differential diagnosis and aid in early detection.

Clinical cases

Case 21.1

Brachial Plexus Involvement in Breast Cancer: Perineural Spread Versus Radiation-Induced Plexopathy

Case. A 58-year-old woman presented with progressive radicular pain in her left arm. Pain was reported to start in the posterior neck, then radiate to the right axilla, posterior arm, and dorsal forearm into the thumb and index finger. She also reported weakness of hand grip and clawing of her left hand. Symptoms progressed very slowly after onset. On examination, atrophy but no fasciculation was noted, along with weakness in the flexors and extensors of the wrist and intrinsic muscles of the left hand. Hyporeflexia was noted in the left biceps, triceps, and brachioradialis. There was no evidence of Horner syndrome.

Her oncology history was significant for locally advanced right-sided breast cancer treated with surgical resection and local radiation therapy over 20 years prior to onset of her presenting symptoms. Due to her history of prior radiation, radiation-induced brachial plexopathy (RIBP) was considered in addition to disease progression with involvement of the brachial plexus. However, nerve conduction studies and electromyogram were unremarkable. Of note, evidence of myokymic discharges was not documented in this study. Contrast-enhanced MRI of the brachial plexus demonstrated abnormal enhancement and enlargement of the right brachial plexus along with involvement of the right pleura/lung ( Fig. 21.2 ). PET/CT did not demonstrate evidence of tumor recurrence. Follow-up imaging did not show significant changes in the pattern of enhancement seen in the brachial plexus, and a presumptive diagnosis of RIBP rather than perineural spread was made. Biopsy was deferred but was considered upon clinical or radiographic progression.

Teaching Points: Imaging Evaluation and Diagnostic Studies for Perineural Spread. This case highlights the important role of imaging in the evaluation of patients with suspected perineural spread. Imaging may include any combination of CT, MRI, and positron emission tomography (PET). For the patient in this case, serial neuroimaging was critical in differentiating between neoplastic and non-neoplastic process as MRI findings were not definitive, and the lack of tumor recurrence on FDG-PET favored a radiation-induced brachial plexopathy but was not definitive.

Imaging evaluation and diagnostic studies

Contrast-enhanced MRI is the gold standard for establishing an imaging diagnosis of PNI. Even with the best effort, imaging studies may be negative at the time of presentation. Sensitivity of imaging studies is higher in symptomatic patients (CPNI), but false-negative results have been reported even in this patient population. ^, ^, Overall, imaging studies can be negative in up to 45% of patients in PNI. ^, ^, Follow-up MRI may be useful in case of negative initial imaging.

Even though it is challenging, effort for early diagnosis is important, given its prognostic value. T1-weighted gadolinium-enhanced MRI is used to identify asymmetrical enlargement or abnormal enhancement around the cranial nerves. ^, ^, ^, Abnormal enhancement implies diffuse enhancement of the nerve with loss of the distinction between the nerve and the perineural vascular plexus. Fat-suppression sequences are helpful, since this decreases artifacts surrounding the skull base foramina. ^, ^, ^, Furthermore, sharing relevant findings such as trophy in facial muscles or muscles of mastication denervation may prompt the radiologist to look for subtle evidence on imaging. ^,

In some studies, the specificity of the MRI is reported to be increased to 95–100% by using targeted MRI sequences to the suspected nerves (aka MR neurography). ^, However, the long procedure time and movement artifacts can limit its utility. CT also plays an important role, especially for lesions that are close to bony structures and involve the skull base. In these cases, bone destruction or enlargement and erosion of foramina may be better appreciated on CT scans. ^, Although lacking the spatial resolution of CT or MRI, techniques such as FDG-PET/CT scans may further aid in diagnosis by demonstrating linear or curvilinear signal changes that represent their metabolic activity. A combination of CT and MRI with selected use of FDG-PET may increase the detection rate and improve the understanding of disease extent. ^,

Evaluation is more challenging in patients who have received prior radiotherapy where radiation-induced peripheral neuropathy (RIPN), such as radiation-induced plexitis, can present in a similar manner. In these cases, use of 3 Telsa MRI may increase sensitivity. RIPN is a rare but relatively well-described complication of radiation therapy. Radiation-related changes often occur early after radiation; however, as was the case for this patient, neuronal injury can occur even decades after initial exposure to radiation.

Metastasis to regional lymph nodes is uncommon, but some reports suggest that rates may be higher in patients with PNI. Evaluation of these lymph nodes with MRI or CT with contrast may assist in making a diagnosis. Ultrasound-guided fine-needle aspiration and/or sentinel lymph node biopsy can be obtained if lymph nodes are identified on imaging studies. ^, ^, Depending on the clinical presentation and clinical history, cerebrospinal fluid (CSF) studies can be considered, especially when leptomeningeal enhancement is noted by imaging. However, the yield is low and, in most cases, the only abnormality identified is high CSF protein level. ^,

Ultimately, biopsy of the involved nerve may be necessary to confirm the diagnosis. Even after an open biopsy, diagnosis of PNI may be challenging. Peritumoral fibrosis may mimic PNI by imaging as well as on histologic sections, especially on hematoxylin and eosin stains. ^, Addition of p75NGFR immunostaining with or without the S-100 immunostaining may be helpful in differentiating fibrosis versus invasion.

Clinical Pearls

1.

Contrast-enhanced MRI is the gold standard for establishing an imaging diagnosis of PNI, but false-negative rates are high and may fail to demonstrate active disease in at least 45% of patients.
2.

A combination of CT and MRI with selected use of FDG-PET may increase the detection rate and improve the understanding of disease extent.

Teaching Points: Differential Diagnosis of Perineural Spread. This case also highlights the role of imaging in guiding the clinical evaluation and informing the differential diagnosis. Imaging in this case revealed a neural lesion on contrast-enhanced MRI that did not show increased uptake on FDG-PET and showed stability on serial imaging that favored radiation-induced brachial plexitis as opposed to a direct neoplastic or perineural process.

Differential diagnosis of perineural spread

In cases where imaging studies are positive, the differential diagnosis is relatively narrowed. When imaging shows abnormal enhancement of the affected nerve with or without enlargement of the foramina or expansion of the nerve, the differential diagnosis includes neoplastic, inflammatory, and infectious causes ^, :

1.

Neoplastic differential diagnoses: primary neural tumors, PNI, traumatic injury, paraneoplastic syndromes, lymphoma
2.

Inflammatory differential diagnoses: granulomatous diseases (e.g., sarcoidosis)
3.

Infectious differential diagnoses: histiocytosis, mucormycosis, aspergillus, viral neuritis, IgG disease, syphilis, and Lyme disease

Biopsy or a close imaging follow-up may help in differentiating between these pathologies and PNI. CT-guided biopsy can be used for lesions involving V3 in the foramen ovale ( Fig. 21.3 ). Ultrasound-guided fine-needle aspiration is usually used for peripheral nerves suspected to be involved with tumor, especially the auriculotemporal and the greater auricular nerves.

When the imaging studies are negative, differential diagnosis varies depending on the nerve being affected. If CN V is affected, the differential diagnosis includes imaging-negative diagnoses like cluster headaches, dental pain, giant cell arteritis, glossopharyngeal neuralgia, vascular or direct compression of the trigeminal nerve, migraine, multiple sclerosis, otitis media, post-herpetic neuralgia, paroxysmal hemicranias, sinusitis, and temporomandibular joint syndrome. If CN VII is affected, the differential diagnosis will include Bell’s palsy, Lyme disease, stroke, multiple sclerosis, Guillain-Barre syndrome, sarcoidosis, direct compression from tumors (parotid, acoustic schwannoma, etc.), or Ramsay Hunt/Zoster syndrome. Although PNI is a feature of local invasiveness, sometimes it can be seen in certain benign breast lesions such as sclerosing adenosis, complex sclerosing lesion/radial scar, and ductal carcinoma in situ. ^, Some authors have suggested this is epithelial glands that are displaced into the perineural space which may mimic PNI. This may be a potential source of misdiagnosis and may not represent true invasion and thus may have minimal clinical impact on outcomes. ^, ^,

At times it may be challenging to differentiate between the possible etiologies of abnormal imaging given the complex clinical history of these patients. Abnormal enhancement of neural structures may be due to prior radiation to the field or perineural invasion ( Fig. 21.2 ).

Clinical Pearls

1.

The differential diagnosis of a new-onset brachial plexopathy in a cancer patient includes perineural spread and radiation-induced brachial plexopathy.
2.

For these patients, the presence of pain should favor perineural spread or locoregional cancer recurrence, whereas the absence of pain should favor radiation-induced brachial plexopathy.

Only gold members can continue reading. Log In or Register to continue