Neurolymphomatosis (NL) is a rare clinical disease where neoplastic cells invade the cranial nerves and peripheral nerve roots, plexus, or other nerves in patients with hematologic malignancy. Most NL cases are caused by B-cell non-Hodgkin’s lymphoma (NHL). Diagnosis can be made by imaging with positron emission tomography (PET) and magnetic resonance imaging (MRI). We experienced two cases of NL involving the brachial plexus in patients with NHL. One patient, who had NHL with central nervous system (CNS) involvement, experienced complete remission after 8 cycles of R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) chemotherapy but relapsed into NL of the brachial plexus 5 months later. The other patient, who suffered from primary central nervous system lymphoma (PCNSL), had been undergoing chemoradiotherapy but progressed to NL of the brachial plexus.
Neurolymphomatosis (NL) is the term for nerve infiltration by neurotropic neoplastic cells in non-Hodgkin’s lymphoma (NHL) or acute leukemia [
A 60-year-old male patient came to the hospital complaining of right shoulder and upper arm pain with weakness that had appeared a month before. He had been diagnosed with stage IV diffuse large B-cell lymphoma (DLBCL) of the prostate 6 months previously. At the time when he was diagnosed with lymphoma, it had spread to the whole body, involving cranial nerves III, V, and VI. After 8 cycles of systemic chemotherapy with a R-CHOP regimen and intrathecal chemotherapy, he achieved complete remission and was free from disease until the right shoulder pain developed. In neurologic examination, measurement by manual muscle test found the muscular strength of the flexor of the right shoulder, adductor, abductor, and elbow extensor to be poor. However, the reflexes of the two arms were normal, and there was no hypoesthesia in a sensory test. No abnormal findings were revealed from a complete blood count, biochemical blood test, and blood coagulation test. However, polyneuropathy involving musculocutaneous nerve, axillary nerve, median nerve, and ulnar nerve in the branch level of the right brachial plexus was suspected from the nerve conduction velocity test. Accordingly, an MRI scan of the right shoulder was performed, but there was no evidence of neuropathy. The pain he experienced was considered to be postherpetic neuropathy because he had suffered from herpes zoster on C5 dermatome 5 months ago. Therefore conservative care and rehabilitation were undertaken. In spite of conservative management, his motor weakness in elbow flexor and wrist extensor got worse from grade 3 to grade 1. And hypoesthesia appeared in C4~T1 dermatome, especially marked in C5~C6 level. A cerebrospinal fluid (CSF) study and brain MRI were done to identify progression of disease. The CSF examination showed neither signs suggesting infection nor malignant tumor infiltration. No actual brain lesion was found, and the previous remission status seemed to have been maintained, according to the brain MRI. But PET-CT revealed strong FDG (fluorodeoxyglucose) uptake along the right brachial plexus (Figure
(a) PET-CT. Strong linear FDG uptake along the right brachial plexus was found, which had not been seen on PET-CT performed 2 months ago (white arrow). (b) Cervical spine MRI with contrast enhancement. The trunk of the right brachial plexus on the level of C5 and C6 were thickened with heterogeneous enhancement, suggesting lymphoma infiltration. (c) Follow-up PET-CT. Right brachial plexus invasion was improved after chemoradiotherapy, but new lesions were seen in right temporal base, right sciatic nerve, and right adrenal gland.
A 50-year-old female patient was diagnosed with primary CNS lymphoma (PCNSL) involving the left basal ganglion. PCNSL was pathologically confirmed as DLBCL. Her disease achieved partial remission after 10 weeks of scheduled chemotherapy with methotrexate/procarbazine/vincristine, followed by radiotherapy. While undergoing a 5-week course of whole brain radiotherapy, she began to complain of right arm weakness and pain, which gradually worsened. There was no trauma history that could be related to her symptoms. Neurologic examination of the right arm showed slightly compromised fine motor and sensory capabilities. Manual muscle testing (MMT) of right upper arm muscles revealed grade 4~5. Brain MRI was performed to identify seeding, infarction, or the regrowth of brain lesions. The MRI showed that the mass size of the previous lesion in the left basal ganglion was much reduced, and a CSF study did not identify CNS lymphoma progression or infection.
We decided to observe the patient because her arm pain was thought to have developed from vigorous rehabilitation, and it was controlled by analgesics. However, the weakness and hypoesthesia of her arm developed further.
Under the follow-up neurologic examination, decreased exercise ability was found to have progressed, especially in the C5 region from grade 4 to grade 2, and severe hypoesthesia was found in C6, 7, and 8 dermatomes. Right brachial plexopathy was found from a nerve conduction velocity test, and cervical spine MRI was undertaken. The latter revealed mass lesions at the extraforaminal portion at C5~C6 level, leading to a suggestion of lymphoma infiltration (Figure
(a) Cervical spine MRI (T2-weighted image). Cervical spine MRI demonstrated a 0.7 × 0.9 cm sized mass (white arrows) in the extraforaminal portion of C5 and C6, with T2 high signal intensity. Nerve roots of C5 and C6 were compressed by the mass, but there was no involvement of the spinal canal or epidural space and bone. These features were not found on the cervical MRI performed 5 months ago. (b) PET-CT. Strong linear FDG uptake along the right brachial plexus and strong focal FDG uptake in the extraforaminal portion of C5 and C6 was seen. (c) Follow-up PET-CT. There was no abnormal FDG uptake in the previous right brachial lesion after chemoradiotherapy.
NL is a rare disease, which has not been researched on a large scale, and its incidence has been poorly defined. According to a report by the International Primary CNS Lymphoma Collaborative Group (IPCG), NL developed in 50 patients over a 16-year period. Most cases of NL are related to non-Hodgkin’s lymphoma (90%), but some can be found in acute leukemia (10%) [
Despite its low incidence, consideration of neurolymphomatosis is important for diagnosis when a patient complains of related symptoms, especially a patient with a history of hematologic malignancy. Symptoms differ greatly according to the sites involved, and there are also some cases without specific symptoms. According to Baehring et al.’s report, NL presentation has 4 patterns: (1) painful involvement of nerves or roots, (2) cranial neuropathy, with or without pain, (3) painless involvement of peripheral nerves, or (4) painful or painless involvement of a single peripheral nerve [
In order to diagnose NL, radiological evaluation by, for example, MRI and PET-CT may be effective. The diagnostic sensitivity of MRI is reported to be around 40% [
The most important clues for diagnosis are a history of non-Hodgkin’s lymphoma and abnormal FDG uptake of the lesion. A nerve conduction test is useful for localization of the affected lesion [
There is currently no standard treatment for NL. There has been little extensive research on NL, so data comparing effects of regimens are unreliable. Generally, the treatment of NL is similar to that of PCNSL, involving systemic chemotherapy alone or a combination of systemic therapy and intrathecal chemotherapy or radiotherapy [
Rituximab is a monoclonal antibody to the protein CD20 and is used as the first-line regimen of NHL (R-CHOP). However, rituximab is a large molecule, so it cannot permeate the blood-brain barrier or the blood-nerve barrier [
Because lymphoma in our patients was localized in the right brachial plexus, limited-field radiation was an effective means of alleviating symptoms and improving the lesion. However, when the NL infiltrates into the CNS and multiple structures, extensive radiation is less effective and is poorly tolerated in most patients [
In a few cases, surgical intervention is needed, such as urgent decompressive laminectomy [
There is no standard evaluation of treatment response, but, based on symptom mitigation and improvement of radiographic signs, an improvement of 50~70% was recorded [
Despite the treatment, the prognosis for patients with NL is poor. The median survival of NL patients-is 10 months after diagnosis [
As already mentioned, the absence of a standard regime of management of NL can be one cause of its poor prognosis. Larger studies about NL must be performed, in order to yield the precise incidence, comparison between treatment results, and establishment of the standard management.
Therefore, when a patient who has a history of hematologic malignancy complains of neurologic symptoms, the rare condition of NL should be considered. For early diagnosis and treatment, MRI and PET-CT are very useful. Treatment can be systemic chemotherapy and/or intrathecal chemotherapy with or without radiotherapy. Regardless of treatment, the prognosis for patients with NL is poor.
The authors declare that there is no conflict of interests regarding the publication of this paper.