Paraneoplastic antineuronal antibodies in patients with systemic autoimmune diseases

Initial screening for neuronal autoantibodies and their putative impact on survival in patients with small-cell lung cancer

INTRODUCTION

Small-cell lung cancer (SCLC) may be associated with neuronal autoantibodies and paraneoplastic neurological syndromes. It has been suggested that neuronal autoantibodies, especially antineuronal nuclear antibody type 1 (Hu) autoantibodies, are associated with longer survival of patients with SCLC. The objective of this study was to determine the frequency and distribution of neuronal autoantibodies at the time of diagnosis of SCLC patients and assess survival rates in relation to autoimmunity.

METHODS

In this retrospective study, serum from 40 patients with biopsy-proven SCLC at the time of diagnosis was studied prior to treatment. The sera originated from a cancer registry at the Oncology Department, Vejle Hospital from 2007 to 2010. The sera were analyzed blindly to clinical status for the presence of neuronal autoantibodies. Medical records were reviewed for neurological symptoms.

RESULTS

Neuronal autoantibodies were detected in 22/40 (55%) of the SCLC patients. A broad range of neurological symptoms was recorded in 28/40 (70%) patients, of which 14/28 (50%) were positive for neuronal autoantibodies. The most frequently detected autoantibodies were Hu (7/40, 17.5%) followed by GAD65 (6/22, 15.0%). Striational and P/Q- or N-type voltage-gated calcium channel antibodies were less common, with each found in five patients (12.5%). Eight patients (20%) had coexisting autoantibodies. Autoantibody-positivity was not associated with survival.

CONCLUSION

Neuronal autoantibodies were at time of diagnosis found in approximately half of the treatment-naïve SCLC patients. Neither autoantibody positivity at diagnosis nor neurological manifestations correlated with survival and their clinical importance requires further studies in larger, prospective cohorts.

Sensory neuronopathy and autoimmune diseases

Sensory neuronopathies (SNs) are a specific subgroup of peripheral nervous system diseases characterized by primary degeneration of dorsal root ganglia and their projections. Multifocal sensory symptoms often associated to ataxia are the classical features of SN. Several different etiologies have been described for SNs, but immune-mediated damage plays a key role in most cases. SN may herald the onset of some systemic autoimmune diseases, which further emphasizes how important the recognition of SN is in clinical practice. We have thus reviewed available clinical, neurophysiological, and therapeutic data on autoimmune disease-related SN, namely, in patients with Sjögren's syndrome, autoimmune hepatitis, and celiac disease.

Onconeural antibodies in sera from patients with various types of tumours

PURPOSE

We assessed the frequency and levels of onconeural antibodies in 974 patients with various types of tumours, but without apparent paraneoplastic neurological syndromes (PNS).

PATIENTS AND METHODS

We included patients with the following tumours: 200 small-cell lung cancer (SCLC) patients, 253 breast cancer patients, 182 ovarian cancer patients, 266 uterine cancer patients and 73 thymoma patients, as well as 52 patients with PNS and cancer and 300 healthy blood donors. Sera were screened for amphiphysin, CRMP5, Hu, Ma2, Ri and Yo antibodies using a multi-well immunoprecipitation technique.

RESULTS

The most frequently detected antibodies were Hu followed by CRMP5. Ma2, Yo, amphiphysin and Ri antibodies were less common, but each was found at similar frequencies. Onconeural antibodies were present at similar levels in sera from the PNS control group and from cancer patients. Hu antibodies were rare in cancers other than SCLC. CRMP5 was the only antibody found in patients with thymoma and this antibody was more common among patients with thymoma than in other tumour patients. With one exception, coexisting antibodies were only found in patients with SCLC. The presence of onconeural antibodies in SCLC patients was not associated with prolonged survival.

CONCLUSION

Onconeural antibodies are associated with various types of tumours suggesting that all antibodies should be included in the serological screening for possible PNS. The levels of onconeural antibody are not sufficiently sensitive to discriminate between cancer patients with PNS and those without.

Clinical relevance of non-neuronal auto-antibodies in patients with anti-Hu or anti-Yo paraneoplastic diseases

About half of the patients with paraneoplastic diseases develop an immune response against neuronal antigens expressed by both the tumor and the nervous system. In 31 patients with anti-Hu antibodies and 19 patients with anti-Yo antibodies, we searched for the presence of additional non-neuronal auto-antibodies and further studied whether the presence of such auto-antibodies was correlated with a specific oncological or neurological presentation. Positive antinuclear antibodies (ANA) were found at a titer of 1:40 or above in 48% patients with anti-Hu antibodies, and in 37% patients with anti-Yo antibodies. Anti-cytoplasmic antibodies were also detected in 42% patients with anti-Yo antibodies. No specific correlation between the presence of non-neuronal auto-antibodies and clinical characteristics of the patients could be identified. In particular, neither the type of underlying cancer, the overall survival, the tumor response to treatment, nor the neurological presentation were related to the serological status.

Paraneoplastic Neurological Antibodies: A Laboratory Experience

Antineuronal antibodies are associated with rare paraneoplastic neurological syndromes, and their identification alerts clinicians to examine for the presence of a tumor. Presented here is laboratory experience (prevalence, difficulties, and procedures) and several interesting but inconclusive results. A total of 1045 samples were screened over a 2-year period; 91 showed a degree of binding of antibodies to the cerebellum, and 22 of these 91 were confirmed, by Western blot, to have specific antineuronal antibodies. Thirteen of 22 were Hu-positive, and 6 of these also had antinuclear antibodies. Six were Yo-positive, 2 had anti-Ma antibodies, and 1 was Tr-positive. An additional 27 of 91 patients had cerebellar antibodies giving recognized staining patterns (Hu, Yo, and Ma). However, Western blot did not confirm these specificities, and hence they were reported as atypical. Six of 27 of these patients had neoplasms; 3 of the 6 gave nucleolar patterns (not Ma). Two appeared similar to Yo, and 1 similar to Hu. Antineuronal antibodies are rare, and in the absence of a specific etiology patients should be examined further for the possible presence of an underlying tumor. Methodical classification of the antibodies must be conducted to avoid incorrect reporting. Further criteria on the typing/reporting of atypical results may aid diagnosis of paraneoplastic neurological syndromes.

Modulation of autonomic balance by tumors and viruses

Autonomic balance, a function generally under host control, is subject to modulation by other signalers. In some cases, modulation of host autonomic function through behavioral and physical stressors exerted by another individual may have negative consequences for the stress recipient by inducing sympathetic bias. Modulation of autonomic function may sometimes benefit one party at the expense of another. Tumors and HIV are examples of illegitimate signalers who may induce host sympathetic bias to promote their own growth and evade host immune surveillance. Paraneoplastic and paraviral syndromes such as hypertrophic osteoarthopathy, QTc prolongation, insomnia, and cachexia could be viewed as epiphenomena related to the tumoral and viral manipulation of host autonomic balance. In a more general framework, other paraneoplastic and paraviral syndromes may represent epiphenomena related to modulation of endocrine, cytokine, and autonomic functions by tumors and viruses to promote their own survival. Spatial distribution of cancers and viruses within the host may reflect affinity for strategic locations that facilitate manipulation of a variety of host functions including autonomic, endocrine, and cytokine regulation. A more general for understanding spatial distribution of diseases based on gradients of autonomic balance in the body are explored. Darwinian perspectives are discussed.