Article Text

PDF

Paraneoplastic cerebellar degeneration in olfactory neuroepithelioma
  1. K Maeda1,
  2. T Sasaki1,
  3. Y Murata1,
  4. M Kanasaki1,
  5. T Terashima1,
  6. H Kawai1,
  7. H Yasuda1,
  8. H Okabe2,
  9. K Tanaka3
  1. 1Division of Neurology, Department of Medicine, Shiga University of Medical Science, Otsu, Shiga, Japan 520–2192
  2. 2Department of Clinical Laboratory Medicine, Shiga University of Medical Science, Otsu, Shiga, Japan 520–2192
  3. 3Department of Neurology, Brain Research Institute, Niigata University, Niigata, Japan
  1. Correspondence to:
 Kengo Maeda
 Division of Neurology, Department of Medicine, Shiga University of Medical Science, Otsu, Shiga, Japan 520–2192; kengo{at}belle.shiga-med.ac.jp

Statistics from Altmetric.com

Anti-Hu antibody was first discovered in patients with paraneoplastic encephalomyelitis associated with small cell lung cancer (SCLC). This antibody recognises proteins comprised in the Hu family expressed by neuronal cells as well as SCLC. After the first report, anti-Hu antibody was found in other neoplasms including prostate and breast cancer, adrenal carcinoma, chondromyxosarcoma, neuroblastoma, and neuroendocrine neoplasms at other sites.1 Olfactory neuroepithelioma (9523/3)2 is thought to differ from classic neuroblastoma (9500/3) in its expression pattern of tyrosine hydroxylase, MYCN amplification, and fusion of the Ewing sarcoma gene and the Friend leukaemia virus integration 1 gene or the ETS related gene.3

Anti-Hu antibody in association with olfactory neuroepithelioma has not been reported previously. We report a patient with cerebellar ataxia that paralleled the recurrence of the tumour. Serum and cerebrospinal fluid (CSF) from the patient contained anti-Hu antibody, and the olfactory neuroepithelioma resected from the patient expressed Hu antigen.

CASE REPORT

Seven years before admission, a 65 year old man presented with olfactory neuroepithelioma that had invaded the orbit and frontal lobe. The tumour was dissected surgically, and dura mater graft was not used in the surgery. The patient underwent irradiation (total dose of 50 Gy). The tumour recurred at the parotid gland in January 2001, and there was gait instability. The patient consulted a neurologist, but there was no specific finding. The recurrent tumour was surgically dissected; however, the instability progressed rapidly, and at the patient’s admission in November 2001, he needed support when walking. There was neither alcoholism nor family history of cerebellar ataxia. His parents were not consanguineous.

General physical examination was negative. There was no lymphoadenopathy. He was alert and mentally normal. Olfactory sensation had been decreased since the first surgery, there was a downbeat nystagmus, and muscle strength was maximum. Both superficial and deep sensation were normal. Deep tendon reflex was symmetrical and normal, Romberg test was negative, and no pathological reflex was found. Nose–finger–nose test was normal, but heel–shin test was poor. Dysmetria was marked in both legs. His gait was wide based and ataxic, and tandem gait was impossible. There was no dysarthria.

Haematological studies, blood chemical analyses, and serological studies were normal. Tumour markers including α-fetoprotein, prostate specific antigen, pro-gastrin releasing peptide, neurone specific enolase, sialyl Lewis (a) (CA19–9), and sialyl Lewis (x) (SLX) were within normal limits. Levels of vitamin B1 and B12 were normal. Protein level in cerebrospinal fluid (CSF) was increased to 105 mg/dl with normal cellularity. Myelin basic protein and oligoclonal IgG band was negative. IgG index was 0.6. No malignant cells were found in the CSF. Nerve conduction study was normal. Short sensory evoked potentials of upper and lower limbs were normal. Electroencephalogram showed beta rhythm at the bilateral frontal region, with otherwise normal findings.

Computed tomography (CT) showed no lung tumour. Magnetic resonance imaging (MRI) showed bilateral leukoaraiosis at bilateral frontal lobes that had been present since after the first surgery. The cerebellum was slightly atrophic.

Titres of anti-Hu antibody in the serum and CSF were 1:1920 and 1:64, respectively (indirect immunofluorescence and Western blotting for recombinant HuD). Serum:CSF antibody titre ratio was 30. The ratio for (CSF/serum antibody titre)/(CSF/serum albumin) was 1.8. These values indicated that intrathecal synthesis of anti-Hu antibody had stopped at this time point. Other anti-neuronal antibodies including anti-Yo, Ri, CV2, Tr, Ma, amphiphysin, and glutamic acid decarboxylase were all negative. Systemic examination including 67Ga-citrate scintigraphy did not disclose malignant tumours. Immunohistochemistry with anti-HuD antibody (Santa Cruz, sc-5977, ×100) revealed that a part of the tumour expressed Hu protein (fig 1).

Figure 1

 Immunohistochemistry using anti-HuD antibody. A part of the patient’s tumour expressed HuD antigen (×400).

Over the course of 4 years after discharge, the cerebellar ataxia did not worsen further in the absence of immunological treatment. Follow up thoracic CT and tumour marker study did not disclose other malignant tumours. There was no evidence of the recurrence of olfactory neuroepithelioma.

CONCLUSION

This patient presented cerebellar ataxia of the trunk and lower limbs that progressed rapidly within approximately 6 months after the second surgery and stabilised thereafter. This clinical course is not inconsistent with the natural course of paraneoplastic cerebellar degeneration. Although isolated cerebellar ataxia in anti-Hu antibody positive patients is rare (4/200 patients),1 a high titre of serum anti-Hu antibody (1:1920) corroborated the diagnosis of paraneoplastic syndrome.1 The expression of the HuD protein by the olfactory neuroepithelioma confirmed the diagnosis.

Olfactory neuroepithelioma is a neuroectodermal neoplasm that arises from the olfactory epithelium. It is distinguished from classic neuroblastoma as described by Sorensen et al.3 Unlike neuroblastoma, olfactory neuroepithelioma shows differentiation to the neural processes and glandular structure and is rarely associated with catecholamine secretion. In addition, olfactory neuroepithelioma expresses epithelial markers such as cytokeratin and a 34 kDa epithelial membrane glycoprotein recognised by monoclonal antibody named Ber-EP4. The tumour in this case expressed both Ber-EP4 and cytokeratin (see Okabe et al4, case no. 6). Moreover, it also expressed luteinising hormone releasing hormone. The expression pattern of Ber-EP4 and cytokeratin was heterogeneous in this tumour.4 These findings suggest that the tumour in this case had arisen from the olfactory placode and was distinct from classic neuroblastoma arising from the neural crest.5 This neuroepithelial tumour has not been reported to be associated with paraneoplastic syndrome. Our data clearly demonstrate the expression of Hu antigen by the olfactory neuroepithelioma cells and the presence of Hu antibody in his serum and CSF. It is interesting that neurological manifestations developed in parallel with the recurrence of the tumour. The recurrence might have enhanced immune response. Despite resection of the recurrent tumour, the cerebellar ataxia worsened for several months after surgery. However, it did not progress thereafter. In patients with neurological symptoms and Hu antibody, olfactory neuroepithelioma should be considered when a neoplasm is not found at the common sites such as the lung or breast.

REFERENCES

View Abstract

Footnotes

  • Competing interests: none

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.