Chronic inflammatory demyelinating polyneuropathy associated with malignant melanoma
Citation Manager Formats
Make Comment
See Comments

NEUROLOGY 1996;46:822-824
Chronic inflammatory demyelinating polyneuropathy
(CIDP) is an acquired autoimmune disorder of unknown cause. CIDP most often occurs alone and not as a complication of other disorders but may accompany osteosclerotic myeloma and other plasma cell dyscrasias, human immunodeficiency virus (HIV) infection, and, rarely, systemic lupus erythematosus (SLE). [1] CIDP has not been associated with solid tumors and is not considered a direct or remote effect of cancer.
We report three patients with CIDP and malignant melanoma. The likelihood of a direct association is increased by the shared antigens between peripheral nerve and melanoma [2-4] and the appearance of demyelinating neuropathy after melanoma immunotherapy. [5,6]
Case reports.
Patient 1.
A 62-year-old man developed ascending leg numbness with paresthesias and progressive leg weakness. Four months after onset, he was unable to use stairs or get out of a chair without assistance because of weakness. At that time he had mild facial weakness, moderate limb weakness (proximal greater than distal), and areflexia. There was moderate loss of all sensory modalities in the distal limbs. He had patchy vitiligo, which had been present for 1 year.
The CSF protein was elevated to 104 mg/dl with no cells. Electrophysiologic studies were characteristic of an acquired demyelinating polyneuropathy Table 1. Other laboratory studies were normal, including a sedimentation rate, serum and urine immunoelectrophoresis with immunofixation, HIV serology, thyroid function studies, rheumatoid factor, and ANA. A skeletal survey was normal. Serum from this patient was assayed by ELISA for reactivity with myelin-associated glycoprotein (MAG) or GM1 ganglioside using previously described techniques. [7,8] Triplicate readings for his anti-MAG reactivity (0.157) and anti-GM1 ganglioside activity (0.124) did not differ from controls (0.137 and 0.140, respectively).
Table 1. Motor nerve conduction studies from three patients with melanoma and CIDP
He grew weaker and was unable to walk without assistance. He was treated with prednisone (60 mg daily) and plasmapheresis. An enlarged axillary lymph node (4 times 5 cm) was detected and excised. Pathologic studies of the node revealed malignant melanoma. Immunohistochemical stains, which were positive for S-100 protein (neuroectodermal origin) and HMB-45 (a melanocyte lineage, associated Ag present only in malignant melanoma cells) [9] but negative for monoclonal keratin (epithelial tumors) confirmed the nature of the tumor. No melanoma-like skin lesions were noted, and a search for other metastases was negative.
He became stronger over the next 3 months. He could walk independently and had only mild proximal leg weakness. His strength has remained stable on low doses of prednisone, and no tumor recurrence has developed over 2 years of follow-up.
Patient 2.
A 43-year-old man reported 6 months of progressive limb weakness, ascending leg numbness, and vitiligo. He had marked distal leg weakness and was unable to walk without assistance. There was distal loss of all sensory modalities in the legs. Reflexes were absent.
The CSF protein was elevated to 86 mg/dl with no cells. Electrophysiologic studies demonstrated an acquired demyelinating neuropathy (table). Other laboratory studies were unremarkable, including serum protein electrophoresis, sedimentation rate, rheumatoid factor, and ANA. A skeletal survey was normal.
An enlarged axillary lymph node was discovered at the initial evaluation. Pathologic examination demonstrated malignant melanoma. No melanotic skin lesions were found, and a search for metastases was negative. He was treated with prednisone (40 mg daily), and within 2 weeks he was much stronger. Within 3 months he could walk without help and had only mild distal leg weakness. Four years later he has remained strong on low-dose prednisone.
Patient 3.
A 49-year-old man noticed a multicolored nodule over his left breast. This regressed several months later. He then developed pain over his sternum, and a skin biopsy demonstrated malignant melanoma. He was treated with radiation therapy to the chest. One year later he developed numbness in his feet, and 6 months after that, distal leg weakness. Over the next 2 years, weakness and numbness remitted and recurred over several month-long cycles, and his disability worsened. He then had moderate to severe leg weakness but normal arm strength. Reflexes were absent, except for trace quadriceps jerks. He had decreased vibration and proprioception sensation, marked in the feet and mild in the hands, as well as diminished cold sensation below the knees.
The CSF protein was elevated to 60 mg/dl with no cells. Electrophysiologic studies demonstrated prolonged distal motor latencies and widespread prolonged or absent F-wave responses and were strongly suggestive of a demyelinating neuropathy (table). A sural nerve biopsy revealed a moderate loss of myelinated fibers, excessively thin myelin sheaths for the axon size, and small onion bulbs formations. Other laboratory studies were normal, including serum and urine protein electrophoresis and immunofixation, HIV serology, sedimentation rate, and rheumatoid factor. The patient declined treatment for the neuropathy.
Discussion.
These three patients developed a subacute progressive sensory-motor neuropathy in association with malignant melanoma. The clinical, electrophysiologic, and other laboratory findings indicated an acquired demyelinating polyneuropathy (CIDP). These features included facial diparesis and predominant proximal weakness (patient 1), areflexia out of proportion to weakness or sensory loss (patient 2), and a relapsing course (patient 3). CSF protein levels were elevated and electrophysiologic studies were characteristic of an acquired demyelinating polyneuropathy Table 1, as was the sural nerve biopsy. The two treated with corticosteroid therapy unequivocally improved, as is expected with CIDP.
We discovered biopsy-proven malignant melanoma in close proximity to the development of neuropathy in each case. In patients 1 and 2, the neuropathy was the first clinical manifestation, and the patients had lymph node involvement without detection of a primary skin site. Nodal involvement from an unknown primary skin site is an infrequent, but well-recognized manifestation of malignant melanoma thought to represent a metastatic growth from a regressed skin site. [9] The tumor histology in these isolated nodes can be confirmed with the detection of S-100 protein (neuroectodermal cells) and HMB-45 (malignant melanocytes) in the absence of monoclonal keratin (epithelial cell), as demonstrated in patient 1.
Two of our three patients had vitiligo. Vitiligo is a depigmentary disease of the skin that results from the destruction of melanocytes by antibodies against cell-surface antigens on melanocytes and may occur with melanoma. In fact, the presence of vitiligo may favorably influence the prognosis of this tumor. [10] The autoimmune mechanism in vitiligo that destroys normal melanocytes may slow the growth of melanoma. Consistent with this hypothesis, none of our patients had recurrent tumor on follow-up.
Although we did not detect anti-MAG or anti-GM1 antibodies in patient 1, the possibility of antibodies directed against a common antigen is intriguing. Melanocytes and Schwann cells are derived from the neural crest, share potential surface antigens including MAG [2] and gangliosides (GM2, GM3, GD3, GD2), [3] and thus may share common antigenic components with peripheral nerve. [4] With similar surface antigens, early trials of melanoma immunotherapy have produced demyelinating neuropathy. A trial of monoclonal anti-GD2 antibody in metastatic melanoma caused peripheral neurotoxicity in four patients, [5] one of whom had a sural nerve biopsy that demonstrated demyelination with inflammatory cells. In another report, two patients who received intradermal vaccinia melanoma cell lysates developed well-characterized demyelinating neuropathy. [6] Twelve percent of the patients treated in this way also developed vitiligo, providing further evidence for a shared immunopathogenic mechanism.
There is no proven association between CIDP and solid tumors, but we believe that the association reported here is not coincidental. CIDP and malignant melanoma are relatively rare. Two patients had an uncommon form of melanoma, with only nodal involvement and vitiligo, both of which imply an autoimmune response against melanoma. Schwann cells and melanoma cells share a number of surface antigens, so an association would not be unexpected. The immunopathologic mechanisms remain to be elucidated, but they are likely related to the process that causes melanoma associated with CIDP and vitiligo to have a relatively benign course.
Acknowledgment
We thank Dr. Arthur K. Asbury for his review of the manuscript.
Footnotes
-
.AB.-We report three patients who developed chronic inflammatory demyelinating polyneuropathy (CIDP) in association with malignant melanoma. In two cases, melanoma was discovered during the initial evaluation for neuropathy. Two patients also had vitiligo, an antibody-mediated disorder that may complicate melanoma. Melanoma cells and Schwann cells are both of neuroectodermal cell origin, with shared surface antigens. Shared immunoreactivity may account for the association between melanoma and CIDP, as with vitiligo.|
- Copyright 1996 by the Advanstar Communication Inc.
REFERENCES
- 1.↵
Albers JW, Kelly JJ. Acquired inflammatory demyelinating polyneuropathies: clinical and electrodiagnostic features. Muscle Nerve 1989; 12:435-451.
- 2.↵
Noronha AB, Harper JR, Ilyas AA, Reisfeld RA, Quarles RH. Myelin-associated glycoprotein shares an antigenic determinant with a glycoprotein of human melanoma cells. J Neurochem 1986; 47:1558-1565.
- 3.↵
Tsuchida T, Saxton RE, Morton DL, Irie RF. Gangliosides of human melanoma. Cancer 1989; 63:1166-1174.
- 4.↵
Fong JW, Ledeen RW, Kundu SK, Brostoff SW. Gangliosides of peripheral nerve myelin. J Neurochem 1976; 26:157-162.
- 5.↵
Saleh MN, Khazaeli MB, Wheeler RH, et al. Phase I trial of the murine monoclonal anti-GD2 antibody 14G2a in meastatic melanoma. Cancer Res 1992; 52:4342-4347.
- 6.↵
Fuller GN, Spies JM, Pollard JD, McLeod JG. Demyelinating neuropathy triggered by melanoma immunotherapy. Neurology 1994; 44:2404-2405.
- 7.↵
Shy ME, Evans VA, Lublin FD, et al. Antibodies to GM1 and GD1 in patients with motor neuron disease without plasma cell dyscrasia. Ann Neurol 1989; 25:511-513.
- 8.
Nobile-Orozio E, Vierotisz T, Messito MJ, Sherman WH, Latov N. Anti-MAG IgM antibodies in patients with neuropathy and IgM M-proteins: detection by ELISA. Neurology 1983; 33:939-942.
- 9.↵
Creagan ET. Regional and systemic strategies for metastatic malignant melanoma. Mayo Clin Proc 1989; 64:852-860.
- 10.↵
Bystryn JC, Rigel D, Friedman RJ, Kopf A. Prognostic significance of hypopigmentation in malignant melanoma. Arch Dermatol 1987; 123:1053-1055.
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Related Articles
- No related articles found.