Antineuronal antibodies in small cell lung carcinoma--a significance for paraneoplastic syndromes?

Immune diagnosis of paraneoplastic neurological disease

A continuous stream of new information on clinical, pathological and immunological aspects of paraneoplastic neurological syndromes has been published in recent years. In this survey, we will discuss current opinions on the value of anti-neuronal antibody detection for establishing a diagnosis of one of the paraneoplastic syndromes of the central nervous system.

Paraneoplastic encephalo-myelo-ganglionitis: cellular binding sites of the antineuronal antibody

The cellular binding sites of an antineuronal antibody were characterized in an autopsy case of the paraneoplastic encephalo-myelo-ganglionitis. A 61 year-old woman developed a subacute sensorimotor polyneuropathy and, later, multiple involvement of cranial nerves, disturbance of consciousness, and generalized seizure. An autopsy revealed a small cell lung carcinoma and neuropathological changes that included disseminated encephalitis, spinal anterior horn lesions, severe loss of dorsal root ganglion neurons, and secondary degeneration and loss of the nerve fibers in the spinal posterior column and peripheral nerves. The serum IgG from the patient contained antineuronal antibody(s) including an antibody to 35- to 37-kDa neuronal antigens called anti-Hu as demonstrated in Western blot. In immunohistochemical studies, the serum IgG immunostained neurons of the brains, spinal cords, and dorsal root ganglia of humans or rats. Confocal laser-scanning microscopy revealed binding of the patient's IgG in the neuronal nuclei and cytoplasm, but not in the nucleoli. In immunoelectron microscopic studies, immunolabelling with the IgG was found diffusely in the karyoplasm, excluding nucleoli, and in the cytoplasmic matrix between the cisternae of the reticulums, Golgi apparatus, and mitochondria. Encephalo-myeloganglionitis is a clinicopathological entity frequently associated with the presence of neoplasm and antineuronal antibody, however, the role of the antibody in the pathogenesis remains to be elucidated.

Limbic encephalitis occurring in association with Alzheimer's disease

Paraneoplastic limbic encephalitis is a rare cause of subacute dementia. A patient with limbic encephalitis and small-cell lung carcinoma is reported in whom the onset of subacute cognitive impairment was obscured by concurrent Alzheimer's disease. MRI revealed increased T2 signal in medial temporal lobes which corresponded to inflammatory pathology demonstrated at necropsy. High titres of antineuronal antibody (type II, anti-Hu) were present at death. Direct immunofluorescent staining of necropsy tissue revealed IgG bound to most remaining neurons in the temporal lobe. Antineuronal antibody screening and MRI are important in detecting limbic encephalitis.

Sensory neuronopathy and small cell lung cancer: antineuronal antibody reacting with neuroblastoma cells

The anti-Hu antibody is associated with a paraneoplastic subacute sensory neuronopathy (SSN) described in cases of small cell lung cancer (SCLC). The Hu antigen is a pan-neuronal nuclear antigen with a molecular weight of 35-40 kDa. In this study we demonstrated the presence of the paraneoplastic Hu antigen in different neuroblastoma cell lines. We showed that by indirect immunocytochemistry the serum of patients with SSN and SCLC reacts with the nuclei of neuroblastoma cell lines SKN-SH and LAN-1. Western blot analysis of nuclear extracts from neuroblastoma cell lines SKN-SH, IMR-32 and LAN-1 confirmed the presence of the Hu antigen in these neuroblastoma cell lines. By comparing the immunocytochemical method and the Western blot analysis we were able to determine that the Western blot analysis was a more sensitive test. Screening of the sera of a large population (a total of 122 patients with SCLC, 17 with paraneoplastic disorders as well as 121 controls with other neurological disorders) was performed and showed all 5 of the patients with SSN and SCLC to be positive for the anti-Hu antibody, whereas only 11 of the 122 SCLC patients and none of the controls were positive, thereby suggesting that this test has a very high degree of sensitivity.

Circulating antineuronal antibodies reach neurons in vivo: an autopsy study

In serum and cerebrospinal fluid of 6 out of 15 patients with small cell lung cancer circulating antineuronal antibodies could be detected by indirect immunofluorescence. None of the patients showed signs of a neurological paraneoplastic syndrome, usually associated with the presence of these antibodies. On sections of frontal brain, cerebellum, dorsal root ganglia and peripheral nerve, obtained at autopsy, a direct immunofluorescence test was performed. Only in antibody-positive individuals were membrane-bound immunoglobulin deposits detected on neurons of dorsal root ganglia and Purkinje cells. The present study showed that circulating antineuronal antibodies reach the central and peripheral nervous system in vivo. These findings support the theory that these antibodies might be directly involved in the pathogenesis of neurological paraneoplastic syndromes.

Cerebrospinal fluid pheresis in Guillain Barré syndrome

While plasmapheresis is established in the treatment of acute polyneuroradiculitis, disabling pareses may last long, persisting neurological deficits remain frequent, and costs and side-effects are considerable. Repeated filtration of cerebrospinal fluid may remove pathogenetically relevant cells and polypeptides. Observations in 12 severe Guillain-Barré patients treated with CSF pheresis indicate that it is a safe and effective procedure. We hypothesize mechanisms of action of and potential indications for CSF pheresis as a more general concept. In inflammatory demyelinating polyneuropathy, CSF filtration could be combined with 'dynamic' cerebrospinal fluid pheresis, intravenous immunoglobulin therapy, cryoprecipitation, and/or immuno-adsorption to increase its effectiveness.

Diagnostic value of anti-neuronal antibodies for paraneoplastic disorders of the nervous system

The diagnostic value of the presence of anti-neuronal antibodies in serum was examined in 21 patients suspected of paraneoplastic disorders of the nervous system (NS) (group 1) and was compared to three control groups; group 2: 25 patients with a neurological disease, without cancer and no sign of paraneoplastic disorder; group 3: 27 patients with neurological disease and cancer and no signs of a paraneoplastic disorder; group 4: 94 patients with cancer and without neurological disease. In group 1, anti-neuronal nuclear antibodies were detected in eight patients (38%), in titres from 1:1000 to 1:32,000. A small cell lung cancer was present in six patients, ovarian cancer in one patient and in one patient no tumour could be detected. The neurological symptoms preceded a diagnosis of cancer in five out of eight patients. Anti-neuronal antibodies were found in the serum of two out of 94 patients (2%) from control group 3 but not in serum from any of the other control groups. These data indicate a moderate sensitivity of 38%, but a high specificity of 98.6% (95% confidence interval 95.5-99.8%) for the presence of anti-neuronal nuclear antibodies if a paraneoplastic NS disorder is suspected.

Neuropathologic findings after bone marrow transplantation: an autopsy study

Clinicopathologic records and neuropathologic tissues of 109 patients who underwent necropsy after treatment with bone marrow transplantation (BMT) were examined. Underlying disorders included leukemia (70), aplastic anemia (25), solid tumors (7), lymphoma (5), Hodgkin's disease (1) and Wiskott-Aldrich syndrome (1). There were 34 females and 75 males, ranging in age from 2 to 56 years. Survival after transplantation averaged 3.6 months. The most common findings were cerebrovascular lesions (29), including hematomas, hemorrhagic necrosis, and infarcts. Central nervous system infections comprised the next most common finding, including 10 fungal and four bacterial infections. A recurrence of underlying malignancy for which transplant had been performed occurred in five patients. Leukoencephalopathy of varying severity was found in eight patients, half of whom had received intrathecal chemotherapy and/or cranial radiation. Patients with systemic graft-versus-host disease had a variety of nonspecific neuropathologic findings in the nervous system; however, nearly half (44%) showed no detectable changes. Other nonspecific alterations included hypoxic/ischemic changes, vascular siderocalcinosis, and neuroaxonal spheroids (associated with hemorrhage or necrosis). These findings provide a guide as to likely causes of a neurologic syndrome in a patient who has undergone BMT, and can be compared with neuropathologic findings in other forms of immunosuppression.

Detection of the anti-Hu antibody in the serum of patients with small cell lung cancer--a quantitative western blot analysis

We looked for the presence of the anti-Hu antibody in the sera from 50 normal subjects; 44 patients with small cell lung cancer, not associated with paraneoplastic disease; and 25 patients with small cell lung cancer associated with paraneoplastic sensory neuropathy, encephalomyelitis, or both. Using the avidin-biotin immunoperoxidase method and a highly sensitive quantitative Western blot analysis, the anti-Hu antibody was not detected in the 50 normal human sera. Seven of the 44 patients with small cell lung cancer but no paraneoplastic syndrome had detectable levels (average titer, 76 U/ml) of anti-Hu antibody on Western blot. These levels are significantly lower than the average titer of the 25 patients who had small cell lung cancer and paraneoplastic sensory neuropathy or encephalomyelitis (average titer, 4,592 U/ml). In the group with nonparaneoplastic small cell lung cancer (low anti-Hu titer) there was a predominance of women (5 women: 2 men), and all patients had "limited" disease when diagnosed. In the antibody-negative group the sex ratio was 16 women to 21 men and 51% of the patients had "extensive" disease. None of the 7 patients with a low-titer anti-Hu antibody developed a paraneoplastic syndrome by the time of writing. The anti-Hu antibody appears, when present, to be a good marker for small cell lung cancer and, when present at high titer, for small cell lung cancer associated with a paraneoplastic syndrome.

Cerebellar cortical degeneration in association with small-cell carcinoma of the oesophagus

We report the association of paraneoplastic cerebellar cortical degeneration with small-cell oesophageal carcinoma in a 60-year-old woman and describe the histopathological findings. We believe this to be the first report of such an occurrence. The cytoplasmic antigen PGP 9.5, which is strongly expressed in cerebellar Purkinje cells, was identified immunohistologically in the primary tumour and immunoglobulin demonstrated on Purkinje cells. These findings support an immunological pathogenesis for this condition and suggest that it is mediated by antibodies directed against tumour antigen which cross-react with Purkinje cells.

Encephalomyelitis and demyelinating diseases in patients with extracerebral malignant tumors

We examined central nervous system (CNS) lesions in 456 patients with primary extracerebral malignant tumors. Inflammatory reactions caused by viral (progressive multifocal leukoencephalopathy (PML), herpes zoster varicella), fungal, or bacterial infections could be demonstrated in 20 patients. In a further 19 patients, the brain tissue showed lymphocyte infiltrates of unknown etiology and, in four of these, autopsy revealed probable paraneoplastic, non-bacterial, endocarditis as a possible explanation for the local inflammatory reaction. The frequency of thrombophlebitis, non-arteriosclerotic thrombosis and arteritis was significantly higher than in a control group of 2052 tumor-free patients. Focal spongiform-axonopathic lesions (24 cases) as well as diffuse leukoencephalopathy (11 cases) were interpreted as probably being at least in part paraneoplastic because the same alterations could also be observed in patients who had never undergone cytostatic or radiation therapy. The possible pathogenetic conditions are discussed and a classification of these tumor-accompanying, but not always tumor-dependent, lesions suggested.

Circulating antineuronal antibodies in small cell lung cancer

Antibodies directed against neuronal cells were found in the serum of 10 out of 27 patients with small cell lung cancer (SCLC) by means of an indirect immunofluorescent test. Positive staining was found either in nuclei of neurons with sparing of nucleoli or in the cytoplasm only. Antibodies belonged to IgG or IgM. A comparison with a control group of 60 patients yielded a high specificity of the circulating antibodies for SCLC. Contrary to recent reports, we were not able to confirm an association between the existence of circulating antibodies against neurons and the incidence of neurologic paraneoplastic disorders.

Reactivity of circulating antineuronal antibodies (CANA) on peripheral nervous system structures

The appearance of circulating antineuronal antibodies (CANA) in patients with malignant tumors has been correlated with the occurrence of paraneoplastic neurological syndromes. However, the effect of CANA on the peripheral nervous system is poorly understood. The reactivity of sera from CANA-positive and -negative patients were investigated on cryostat sections of peripheral nerves and skeletal muscle, and on nerve tease-fiber preparations. Only CANA-positive sera showed staining of Schwann cell nuclei on cryostat sections, whereas nerve tease-fiber preparations and sections of skeletal muscle remained negative. Positive direct immunofluorescence of small cell lung cancer (SCLS) cells was confined to CANA-positive patients only. These findings suggest the existence of a common antigen between SCLC and Schwann cells.