Radial nerve myofibroma: a rare benign tumor with perineural infiltration. Illustrative case

Abstract

BACKGROUND

Myofibromas are benign mesenchymal tumors, classically presenting in infants and young children in the head and neck region. Perineural involvement, especially in peripheral nerves within the upper extremity, is extremely rare in myofibromas.

OBSERVATIONS

The authors present the case of a 16-year-old male with a 4-month history of an enlarging forearm mass and rapidly progressive dense motor weakness in wrist, finger, and thumb extension. Preoperative imaging and fine needle biopsy confirmed the diagnosis of a benign isolated myofibroma. Given the dense paralysis, operative management was indicated, and intraoperative exploration showed extensive involvement of tumor within the radial nerve. The infiltrated nerve segment was excised along with the tumor, and the resulting 5-cm nerve gap was reconstructed using autologous cabled grafts.

LESSONS

Perineural pseudoinvasion can be an extremely rare and atypical feature of nonmalignancies, resulting in dense motor weakness. Extensive nerve involvement may still necessitate nerve resection and reconstruction, despite the benign etiology of the lesion.

Myofibromas are rare benign mesenchymal tumors, classically presenting in infants and young children as either solitary tumors or multicentric myofibromatosis.^1,2 Solitary lesions are seen most commonly in males, involving the soft tissues of the head and neck region, whereas the multicentric form more often affects females and can also involve bone and viscera.² Solitary tumors can demonstrate some regression through central necrosis. Generalized myofibromatosis, however, is associated with poorer outcomes and can be fatal, especially in cases with visceral involvement.^2,3

The clinical presentation of myofibromas can be nonspecific. Patients most often present with a painless subcutaneous mass, with accompanying overlying skin changes or secondary ulceration.² Motor weakness and sensory disturbances are less commonly seen. Oculomotor, facial, and median nerve pareses have been reported in lesions involving the orbit, parotid gland, and palm, respectively.^2,4–7 However, most reports do not specify if this is a result of nerve compression from mass effect or actual perineural tumor infiltration. Perineural invasion is a pathological characteristic of numerous malignancies that portend a worse prognosis and higher morbidity and mortality; however, it is an unusual feature of benign tumors.^3,8,9 Here, we present a rare case of a solitary myofibroma in the upper extremity with extensive perineural involvement of the radial nerve that necessitated total resection of the nerve and reconstruction.

Illustrative Case

History and Examination

A 16-year-old right-hand–dominant man presented with a 4-month history of progressive weakness in left wrist, finger, and thumb extension with an associated enlarging mass in the proximal forearm (Fig. 1). There was no preceding trauma and no past history of neurofibromatosis, schwannomatosis, or other neurocutaneous syndromes. Physical examination revealed normal triceps and brachioradialis function but weak wrist dorsiflexion with radial deviation and absent finger and thumb extension, with accompanying muscle atrophy. He had decreased sensation in the dorsal forearm, thumb, and index finger. Motor and sensory examination of the median and ulnar nerves was normal. The forearm mass was firm, nonmobile, nonpulsatile, minimally tender to palpation, and had a negative Tinel’s sign.

FIG. 1.

At clinical presentation, a mass in the left dorsal radial forearm was noted (A, B) with rapidly progressive motor weakness in the wrist, finger, and thumb extension.

Noncontrast magnetic resonance imaging (MRI) demonstrated a 3 × 1 cm irregular, heterogeneous lesion arising in the lateral portion of the brachialis and the brachioradialis muscles, encasing and infiltrating the radial nerve (Fig. 2A–D). The radial nerve itself was enlarged and hyperintense proximal and distal to the mass. The imaging characteristics did not meet the classic criteria for a benign neurogenic tumor. The denervation changes observed in the extensor muscles in the proximal forearm (Fig. 2C) were also atypical for a benign neurogenic tumor, so this lesion was radiologically indeterminate.

FIG. 2.

A: Sagittal T2-weighted MRI with fat suppression demonstrating an irregular mass (arrowhead) associated with the enlarged and hyperintense radial nerve (arrow). B: Coronal T2-weighted image redemonstrates the elongated mass (arrowhead) and radial nerve (arrow). C: Axial T2-weighted image shows denervation of the extensor musculature (asterisks) in the proximal forearm. D: Axial T1-weighted image shows an irregular mass (arrow) that is hyperintense compared to the surrounding skeletal muscle. E: PET-CT showed mild to moderate FDG uptake with a maximum SUV of 2.9 within the proximal forearm corresponding to the location of the myofibroma.

Positron emission tomography-computed tomography (PET-CT) showed a 2.2 × 1.1 cm lesion with mild to moderate fluorodeoxyglucose (FDG) uptake and a maximum standardized uptake value (SUV) of 2.9 (Fig. 2E). Nerve conduction studies demonstrated low amplitudes in the extensor digitorum communis and absent radial sensory nerve action potentials. Needle electromyography showed dense fibrillations and no activation in any radial innervated muscles distal to the brachioradialis, confirming a severe left radial mononeuropathy. Due to the short time course of symptom onset and rapid tumor growth in an otherwise healthy young patient, there was high suspicion for a malignant neurogenic tumor or soft tissue sarcoma. Therefore, an ultrasound-guided fine needle biopsy specimen was obtained and subsequently confirmed the diagnosis of a benign myofibroma. Given the dense paralysis of the posterior interosseus nerve muscles, surgical intervention was recommended.

Surgical Findings

The patient underwent excision of the radial nerve myofibroma through an anterior approach and curvilinear incision across the antecubital fossa. Proximally, the radial nerve was identified between the brachioradialis and brachialis. Distally, the superficial and deep branches of the radial nerve were dissected (Fig. 3A). The mass involved not only the brachioradialis and brachialis muscles, but also the radial nerve itself. An attempt was made to excise the tumor while preserving the deep branch of the radial nerve; however, the lesion was densely adherent and deeply infiltrative within the nerve, without a plane of separation. An extensor carpi radialis brevis motor branch was directly coursing through and encased in tumor (Fig. 3B). Intraoperative nerve stimulation of the posterior interosseous nerve and extensor carpi radialis motor branch did not produce any motor responses. Therefore, a decision was made to resect the nerve along with the tumor (Fig. 4A). After negative margins were confirmed with frozen pathology, there was a 5-cm gap in the radial nerve (Fig. 3C). A dorso-radial incision was made in the ipsilateral distal forearm, and 18 cm of the superficial branch of the radial nerve was harvested for nerve grafting. Three 6-cm cabled nerve grafts were used to bridge the defect (Fig. 3D). The patient was immobilized with the elbow flexed and the wrist maintained in neutral for 3 weeks.

FIG. 3.

Resection and reconstruction. A: Intraoperative exploration demonstrated extensive perineural involvement of the myofibroma within the superficial and deep branches of the radial nerve without a plane of separation. B: Extensor carpi radialis brevis motor branch coursing directly into the tumor, with absent responses to intraoperative nerve stimulation. C: Resection of the radial nerve with the tumor, resulting in a 5-cm nerve gap. D: Nerve reconstruction using the superficial branch of the radial nerve as three cabled autologous nerve grafts to bridge the defect. Coaptation performed with 9–0 nylon suture and fibrin glue. Distal = left; proximal = right.

FIG. 4.

Myofibroma pathology and histology. Excised surgical specimen (A) demonstrating extensive involvement of the myofibroma within the superficial (white arrow) and deep branches (asterisk) of the radial nerve and extensor carpi radialis brevis motor (ECRB) branch. H&E-stained section (B) showing myofibroma with characteristic morphology completely encasing a central nerve fascicle (original magnification ×40). Immunohistochemistry for S100 (C) and SOX10 (D) demonstrate axons within nerve fascicles (brown) and highlight extensive infiltration by myofibroma tumor cells (negative staining, original magnification ×200).

Histopathological Analysis

The hematoxylin and eosin (H&E)-stained sections confirmed the diagnosis of a myofibroma, with a characteristic biphasic appearance of hyalinized, vaguely myoid nodules juxtaposed with the cellular myofibroblastic center. The associated nerve fascicle was completely encased within the myofibroma (Fig. 4B), without a clear tissue plane. Immunohistochemical stains for S100 and SOX10 highlighted Schwann cells associated with axons within the nerve fascicles and demonstrated extensive perineural invasion of the myofibroma (Fig. 4C, D). The magnetic resonance images, surgical findings, and histopathological analysis were all consistent with an outside-in mechanism for tumor spread.

Outcome and Follow-Up

The patient had an uneventful postoperative course. There were no postoperative complications, and he had the expected donor sensory deficit in the superficial branch distribution of the radial nerve. His ulnar and median nerve function remained intact and strong, preserving the opportunity to augment function with tendon transfers in the future.

Discussion

Observations

In patients presenting with rapidly progressive motor weakness and an accompanying mass, there must be a high index of suspicion for either malignant peripheral nerve tumor or secondary perineural invasion from another malignancy. In this case, although the MRI differential diagnosis included a schwannoma, this was not in keeping with the clinically rapid growth and deterioration of motor function. Therefore, the utility of obtaining a definitive tissue diagnosis outweighed the risks of performing biopsy of a potentially benign nerve tumor. Preoperative biopsies of peripheral nerve tumors are associated with a high rate of new or worsening neurological symptoms and can induce scarring and distortion of anatomical planes that make definitive resection more difficult.¹⁰ The ultrasound-guided needle biopsy in this case established the nonneurogenic origin of this tumor and confirmed its benign pathology as a myofibroma.

Myofibromas are pericytic (perivascular) myoid neoplasms characterized by a distinctive biphasic growth pattern consisting of a center of immature-appearing, plump, spindled cells amid hemangiopericytoma-like branching vasculature, surrounded by a periphery of nodules and fascicles of variably hyalinized, myoid, or chondroid cells.^4,11 By immunohistochemistry, myofibromas express smooth muscle actin and are typically negative for desmin, CD34, and S100. Rarely, myofibromas may present with atypical features such as predominantly fascicular architecture with inconspicuous myoid nodules, high cellularity, infiltrative growth pattern, and/or perineural invasion.⁴ The presence of these features can lead to a misdiagnosis of low-grade myofibroblastic sarcoma; however, the identification of foci with more typical morphology by critical histological review, immunohistochemical evaluation, and genetic testing can aid in differentiating these from other spindle cell sarcomas.⁴ In the case presented here, S100 was negative in the tumor cells and, in conjunction with SOX10, served to highlight infiltrative growth of the myofibroma into associated nerve fascicles (Fig. 4).

The term perineural invasion was first broadly defined by Batsakis⁹ as tumor cells invading in, around, or through nerve fibers and was later refined by Dunn et al.⁸ to emphasize the malignant potential and worse prognosis associated with this finding. However, perineural involvement can be seen with many benign tumors, including myofibromas, leading Moghimi et al.³ to suggest the term perineural pseudoinvasion or involvement instead in nonmalignancies. The mechanism of perineural invasion in malignant and benign lesions is not well understood.¹² The main difference between perineural pseudoinvasion and true invasion is the underlying histopathology of the tumor itself and whether it is malignant or benign.

The majority of reports of perineural involvement or nerve compression from myofibromas involve lesions within the oral and maxillofacial region.^2,4,13 Linos et al.⁴ have reported on a series of 24 cases with atypical features, 6 of which demonstrated perineural involvement/nerve compression; however, the specific location of only one case involving the facial nerve was discussed. Similarly, in the series by Foss and Ellis,¹³ of 79 oral myofibromas, 25 had nerve involvement. In many of these case series, it is unclear whether nerve involvement refers to extrinsic compression from mass effect or true perineural involvement. Hoang et al.⁵ have presented a skull base myofibroma causing oculomotor nerve palsy. This patient underwent a craniotomy in which the tumor was “found intraoperatively to invade and distort the [oculomotor] nerve.”⁵ However, there was no discussion of the resulting deficits, reconstruction, or patient follow-up.⁵

There are even fewer described cases of myofibromas affecting peripheral nerves in the extremities. Two cases describe myofibromas in the palm causing median nerve compression from mass effect.^6,7 Another recent case report highlights a myofibroma in the medial upper arm causing extrinsic compression on the ulnar nerve resulting in clawing and a positive Wartenberg’s and Froment’s sign but without nerve involvement or infiltration.¹⁴ There is only one case describing perineural involvement in a 33-year-old woman with a 6-month history of spontaneous radial nerve palsy.¹⁵ She underwent neurolysis, excision of what was suspected to be a neuroma-in-continuity from inflammatory or compressive neuropathy and primary nerve repair; however, postoperative histology confirmed the diagnosis of a myofibroma/myopericytoma.¹⁵ In this case, there was an hourglass deformity seen during intraoperative exploration and no identified mass on preoperative MRI. Our case would be the second in the literature with true perineural infiltration of a major peripheral nerve within the extremity, and both cases required resection and reconstruction of the involved nerve. In our case, there was a clear exophytic mass, confirmed on histological analysis to arise from outside the nerve before invading it, which differs from that reported by Hinchcliff et al.,¹⁵ in which there was no defined mass on preoperative imaging. This perineural invasion is a highly unusual feature of benign tumors. However, as illustrated in our case, nerve resection and reconstruction may be necessary, depending on the extent of nerve involvement.

Lessons

Patients with rapidly progressive motor weakness resulting from a mass necessitate a high index of suspicion for malignancy. In these cases, the ability to obtain a definitive diagnosis outweighs the potential risks of biopsy to obtain a definitive diagnosis outweighs the potential risks of the biopsy. Although rare, perineural pseudoinvasion can be a feature of nonmalignancies, including benign isolated myofibromas. Symptoms may be a result of extrinsic nerve compression and mass effect, resulting from a well-encapsulated tumor that is easily excised with minimal deficits. However, extensive nerve infiltration can also occur, resulting in the need for nerve resection and reconstruction.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Spinner, Amrami. Acquisition of data: Spinner, Wu, Amrami. Analysis and interpretation of data: Spinner, Wu, Cook. Drafting the article: all authors. Critically revising the article: Spinner, Wu, Amrami. Reviewed submitted version of manuscript: Spinner, Wu, Amrami. Approved the final version of the manuscript on behalf of all authors: Spinner. Study supervision: Spinner.