Autoimmune paraneoplastic syndromes associated to lung cancer: A systematic review of the literature Part 4: Neurological paraneoplastic syndromes, involving the peripheral nervous system and the neuromuscular junction and muscles

The development of new immune treatment in oncology and particularly for lung cancer may induce new complications, particularly activation or reactivation of auto-immune diseases. In this context, a systematic review on the auto-immune paraneoplastic syndromes that can complicate lung cancer appears useful. This article is the fourth of a series of five and deals mainly with neurological paraneoplastic syndromes involving the peripheral nervous system and the neuromuscular junction and muscles.

Immune-Mediated Neuropathies following Autologous Stem Cell Transplantation for Multiple Myeloma: Case Series and Review of the Literature

Neuropathy is a common finding in patients with multiple myeloma. Several different factors can cause neuropathy in these patients, such as the underlying disease itself, medications used for treatment, or immune-mediated processes. Immune-mediated neuropathies (IMN) consist of a heterogeneous spectrum of peripheral nerve disorders. Although IMN is associated with several hematological disorders, it remains a very rare complication of hematopoietic stem cell transplantation (HCT). We describe our experiences of 3 patients with multiple myeloma who experienced IMN following autologous HCT (auto-HCT). These 3 patients were felt to have IMN clearly attributable to auto-HCT because of a clear temporal association with auto-HCT and absence of any other obvious causative factor. The variety in their clinical presentations, diagnostic approach, and approaches to management are explained. The pathophysiology of how HCT may predispose to IMN remains poorly understood. Our report helps highlight several potential causes of this phenomenon, such as a paraneoplastic syndrome, immune reconstitution syndrome, or drug toxicity. We emphasize that a comprehensive approach is needed to address this rare entity, and that there should be a low threshold to initiate immune-specific therapy, such as plasmapheresis, if symptoms do not resolve spontaneously.

Paraneoplastic epilepsy

Epilepsy can be a manifestation of paraneoplastic syndromes which are the consequence of an immune reaction to neuronal elements driven by an underlying malignancy affecting other organs and tissues. The antibodies commonly found in paraneoplastic encephalitis can be divided into two main groups depending on the target antigen: 1) antibodies against neuronal cell surface antigens, such as against neurotransmitter (N-methyl-d-aspartate (NMDA), alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), gamma-aminobutyric acid (GABA)) receptors, ion channels (voltage-gated potassium channel (VGKC)), and channel-complex proteins (leucine rich, glioma inactivated-1 glycoprotein (LGI1) and contactin-associated protein-2 (CASPR2)) and 2) antibodies against intracellular neuronal antigens (Hu/antineuronal nuclear antibody-1 (ANNA-1), Ma2/Ta, glutamate decarboxylase 65 (GAD65), less frequently to CV2/collapsin response mediator protein 5 (CRMP5)). In this review, we provide a comprehensive survey of the current literature on paraneoplastic epilepsy indexed by the associated onconeuronal antibodies. While a range of seizure types can be seen with paraneoplastic syndromes, temporal lobe epilepsy is the most common because of the association with limbic encephalitis. Early treatment of the paraneoplastic syndrome with immune modulation/suppression may prevent the more serious potential consequences of paraneoplastic epilepsy.

[Paraneoplastic neurological syndromes: an update]

Paraneoplastic neurological syndromes are a group of neurological syndromes secondary to an underlying malignancy. Associated autoantibodies can be classified according to the cellular localization of the antigen target. Onconeuronal autoantibodies (targeting intracellular antigens) strongly associate with cancer and the response to immunotherapy is often disappointing. Identifying and treating the underlying malignancy is a high priority. However, immunomodulation can provide a favourable outcome for neurological symptoms associated with autoantibodies specific for cell membrane antigens. An early recognition of these disorders following the triad "clinical neurological syndrome--specific autoantibodies--tumour research" is important so that patients can benefit from appropriate targeted treatments.

Paraneoplastic disorders

PURPOSE OF REVIEW

Paraneoplastic disorders are autoimmune diseases associated with risks for specific cancers and marked by specific autoantibodies. They cause diverse clinical syndromes of the central and peripheral nervous systems.

RECENT FINDINGS

In the peripheral nervous system, autoimmunity to synaptic or axonal proteins has long been recognized to associate with specific cancers. In these disorders, typified by myasthenia gravis, the antibodies are directly toxic, and recovery with immunotherapy is the rule. In contrast, the classic paraneoplastic syndromes involve a higher risk of cancer, autoantibodies to intracellular proteins (eg, Hu proteins), T-cell-dependent disease mechanisms targeting the CNS or peripheral nervous system, and a poor response to treatment. Following the discovery of N-methyl-D-aspartate (NMDA) receptor antibodies, a new and expanding group of disorders involving autoantibodies to CNS synaptic and neuronal membrane proteins and a favorable response to immunotherapy emerged. A final group of disorders involves antibodies to intracellular synaptic proteins, such as glutamic acid decarboxylase 65 (GAD65), and it is unclear whether these diseases involve antibody or T-cell mechanisms.

SUMMARY

Neurologists should recognize the clinical syndromes associated with paraneoplastic disorders, utilize autoantibody and other testing to confirm the diagnosis, understand the pathologic basis of the diseases, and promptly give appropriate therapies.

Recent developments in paraneoplastic disorders of the nervous system

Paraneoplastic disorders of the nervous system (PNDs) are rare and unique disorders, where a specific pattern of neural damage occurs as a side effect of the interaction between the neoplasm and the host immune response. Clinical recognition of PNDs may be challenging but can lead to early detection of an occult neoplasm. Their study may lead to a better understanding of nervous system autoimmunity and even to devising novel immunotherapies against certain tumor types. Familiarity with the clinical syndromes, neuroradiological findings, autoantibodies, and tissue responses associated with PND may help arrive at a correct diagnosis in most cases.

[Paraneoplastic neurological syndromes and autoimmune encephalitis]

Paraneoplastic neurological syndromes (PNS) are defined as remote effects on the central and peripheral nervous system that are not caused directly by the tumor, its metastases and treatment, or metabolic disorders. The most probable cause is a falsely initiated immune reaction. Well-defined classical PNSs are associated with distinct tumors and occur with onconeural antibodies directed against intracellular neuronal antigens. However, response to therapy is limited. Recently, new antibodies directed against neuronal surface antigens were described in encephalitic syndromes of autoimmune origin. These probably antibody-mediated disorders are more frequent than classical PNS, occur with or without tumor association and often show a good response to immunosuppressive treatment.

[Paraneoplastic neurological syndrome and autoantibodies]

Paraneoplastic neurological syndrome (PNS) is a rare disorder caused by the remote effects of cancer is considered as caused by humoral or cell-mediated immunity. Several autoantibodies related to PNS have been discovered since the 1980s. These antibodies were classified into 2 categories based on the PNS diagnostic criteria recommended by PNS Euronetwork in 2004: well-characterized onconeural antibodies and partially characterized onconeural antibodies. Recently, techniques for the detection of these antibodies have been developed. Additional autoantibodies have been shown for neural surface antigens related to autoimmune-mediated encephalitis in patients with and without cancer. Because the PNS neurological symptoms often precede tumor diagnosis, the antibodies are useful diagnostic markers for PNS and occult tumors. Anti-tumor therapies are essential for PNS, and successful immunotherapies depend on the location of antigens, whether intracellular or on the surface of membranes. Generally, the latter cases are responsive to therapy. Only a limited numbers of cases with intracellular antigens have been improved by early and aggressive immunotherapies. Neurologists should be alert for PNS and check the presence of autoantibodies at the start of therapies when encountering rapidly progressive neurological symptoms of unknown origin. Here, we describe the relationship between onconeural antibodies, clinical features, tumor types, and response to immunotherapies.

[A man with sexual disinhibition caused by paraneoplastic limbic encephalitis]

A 63-year-old man with symptoms of depression and sexual disinhibition was admitted to a psychiatric clinic for the elderly. Because the man’s symptoms rapidly became more severe he was referred to the emergency room. There, his illness was diagnosed as paraneoplastic limbic encephalitis with positive anti-Hu antibodies; this is a paraneoplastic neurological syndrome presenting with short-term memory loss, epileptic seizures and psychiatric symptoms. For the prognosis of the illness it is essential that the syndrome is diagnosed as early as possible. Since patients sometimes present with mainly psychiatric symptoms it is important that psychiatrists are fully informed about the symptoms and are able to make an accurate diagnosis.

[Autoimmune encephalitis, clinical, radiological and immunological data]

INTRODUCTION

Encephalitis is an inflammatory or infectious disease with an acute or subacute presentation. Immunological abnormalities in serum can be found but may be underdiagnosed. In several cases, a paraneoplastic origin with anti-neuron antibodies is noted. In all cases, other auto-antibodies can be found with or without any neoplastic mechanism.

OBJECTIVES

The aim of our study was to describe a clinical, radiological and immunological cohort of patients with autoimmune encephalitis and suggest a diagnostic and therapeutic algorithm.

PATIENTS AND METHOD

We performed a retrospective study in an immunological unit of neurology. All patients with autoimmune encephalitis between March 2000 and October 2009 were included. The clinical, imaging and immunological evaluations were recorded for each patient.

RESULTS

Our cohort included 16 patients (eight men and eight women), mean age 45.3±10years. All patients had acute or subacute neuropsychological or neuropsychiatric impairment and all patients but one had temporal lobe dysfunction confirmed by cerebral MRI, PET or SPECT. Epilepsy was observed in 56% of cases, extra-temporal lobe impairment in 50%, including sleep disturbances. A cancer was found in only 25% (two small-cell lung cancers, one testis seminoma, one non-small-cell lung cancer with Merckel cells cancer). Anti-neuron antibodies were noted in 56% of cases (two with anti-voltage gate potassium channel complex antibodies (ab), two with anti-NMDA-R ab, two with anti-glutamate acid decarboxylase ab, one with anti-Ma2, two with anti-Hu ab and two remained uncharacterized). Systemic antibodies were found in 50% (one anti-gangliosides, one anti-SSA and one anti-DNA and four antinuclear ab uncharacterized, two anti-TPO and two anti-phospholipids). All patients received immunomodulatory treatments, including intravenous immunoglobulins (IgIV) and cancer was treated. Five patients achieved complete recovery, partial improvement was observed in 10 patients and two patients died.

DISCUSSION

Despite clinical homogeneity at presentation, clinical outcome seems to be different between patients with antibodies against neuronal surface antigens and those with antibodies against intracellular antigens, which are more likely refractory to immunotherapy and paraneoplastic. The frequency of extra-temporal lobe impairment suggests that the term of limbic encephalitis should be changed to autoimmune encephalitis.

Reversible paraneoplastic encephalitis in three patients with ovarian neoplasms

Anti-N-methyl-d-aspartate (NMDA) receptor encephalitis is a recently described potentially lethal but treatable disorder that often occurs as a paraneoplastic manifestation of ovarian teratomas. We report three women with this disorder who presented with subacute onset of delirium, seizures and autonomic instability. Anti-NMDA receptor antibodies were detectable in the serum or cerebrospinal fluid of each patient. Ovarian masses were detected in two patients, and subsequently excised. In the third patient, an empirical bilateral salpingo-oophorectomy was performed and revealed a microscopic neoplasm. All patients experienced slow reversal of the neurological symptoms following surgery and immunotherapy. Our experience suggests that prompt syndrome recognition followed by tumor removal and immunotherapy usually results in neurological recovery.

SOX1 antibodies in sera from patients with paraneoplastic neurological syndromes

OBJECTIVES

SOX1 antibodies have been described in patients with Lambert-Eaton myasthenic syndrome (LEMS) in association with voltage-gated calcium channel antibodies as serological markers of small cell lung cancer (SCLC). This study was aimed to screen for additional SOX1 autoimmunity in onconeural antibody-positive sera from patients with paraneoplastic neurological syndromes (PNS) other than LEMS and to identify the clinical-immunological profile and associated tumours of patients with coexisting SOX1 antibodies.

METHODS

We retrospectively analysed sera from 55 patients with different PNS positive for well-characterized antineuronal antibodies for the presence of SOX1 antibodies by recombinant ELISA and immunoblot.

RESULTS

Eight (14.5%) patients showed additional SOX1 antibodies in the ELISA and the recombinant immunoblot. Five patients had coexisting Hu antibodies, while the other three showed coexisting CV2/CRMP5, amphiphysin, and coexisting CV2/CRMP5 and Hu antibodies, respectively. PNS included (partially overlapping) subacute sensory neuropathy, subacute sensorimotor neuropathy, cerebellar degeneration, brainstem encephalitis, encephalomyelitis and limbic encephalitis. No tumour was detected in two patients, while the others had lung cancer (four SCLC and two non-SCLC). One patient showed SOX1-specific intrathecal antibody synthesis.

CONCLUSIONS

We describe SOX1 reactivity for the first time overlapping with CV2/CRMP5 and amphiphysin antibodies. SOX1 reactivity is predominantly associated with Hu antibodies and SCLC, but can occur also in other types of lung cancer. Neurological manifestations present in patients with coexisting SOX1 antibodies and well-characterized antineuronal antibodies do not differ from those previously described in patients positive for antineuronal antibodies but no SOX1-specific anti-glial antibodies.

Anti-Hu-associated paraneoplastic encephalomyelitis with esophageal small cell carcinoma

A 63-year-old woman had anti-Hu-associated paraneoplastic encephalomyelitis (anti-Hu syndrome) caused by esophageal small cell carcinoma (SCC). The patient developed bilateral limbic encephalitis, followed by myelitis, brain stem encephalitis, and autonomic failure. Extensive examination demonstrated SCC of the abdominal lymph nodes that was retrospectively diagnosed as metastasis of esophageal SCC on autopsy. The neuropathological findings were characterized by widespread neuronal loss and gliosis in the central nervous system, as well as patchy loss of myelin and axons in the spinal nerve roots with perivascular lymphocytic infiltration. This is the first detailed clinical and neuropathological report of anti-Hu syndrome caused by esophageal SCC.

Elderly-onset neuromyelitis optica which developed after the diagnosis of prostate adenocarcinoma and relapsed after a 23-valent pneumococcal polysaccharide vaccination

We report a case of elderly-onset neuromyelitis optica (NMO) positive for the anti-aquaporin-4 (AQP-4) antibody; symptoms developed after the diagnosis of prostate adenocarcinoma and relapsed after a 23-valent pneumococcal polysaccharide vaccination. We suggest that activation of CD4-positive T cells and secretion of interferon-gamma induced by adenocarcinoma and complement activation induced by vaccination are responsible for the onset and relapse of NMO, even if a patient is positive for the anti-AQP-4 antibody. This case supports the previous experimental finding that the anti-AQP-4 antibody does not cause NMO-like lesions when injected alone, but does so after the induction of T cell-mediated experimental autoimmune encephalomyelitis or when co-injected with human complement.

Movement disorders in paraneoplastic and autoimmune disease

PURPOSE OF REVIEW

The most relevant advances in immune-mediated movement disorders are described, with emphasis on the clinical--immunological associations, novel antigens, and treatment.

RECENT FINDINGS

Many movement disorders previously considered idiopathic or degenerative are now recognized as immune-mediated. Some disorders are paraneoplastic, such as anti-CRMP5-associated chorea, anti-Ma2 hypokinesis and rigidity, anti-Yo cerebellar ataxia and tremor, and anti-Hu ataxia and pesudoathetosis. Other disorders such as Sydenham's chorea, or chorea related to systemic lupus erythematosus and antiphospholipid syndrome occur in association with multiple antibodies, are not paraneoplastic, and are triggered by molecular mimicry or unknown mechanisms. Recent studies have revealed a new category of disorders that can be paraneoplastic or not, and associate with antibodies against cell-surface or synaptic proteins. They include anti-N-methyl-D-aspartate receptor (anti-NMDAR) encephalitis, which may cause dyskinesias, chorea, ballismus or dystonia (NMDAR antibodies), the spectrum of Stiff-person syndrome/muscle rigidity (glutamic acid decarboxylase, amphiphysin, GABA(A)-receptor-associated protein, or glycine receptor antibodies), neuromyotonia (Caspr2 antibodies), and opsoclonus--myoclonus--ataxia (unknown antigens).

SUMMARY

Neurologists should be aware that many movement disorders are immune-mediated. Recognition of these disorders is important because it may lead to the diagnosis of an occult cancer, and a substantial number of patients, mainly those with antibodies to cell-surface or synaptic proteins, respond to immunotherapy.

Onconeural antibodies in patients with neurological symptoms: detection and clinical significance

BACKGROUND

Onconeural antibodies are strongly associated with cancer and paraneoplastic neurological syndromes (PNS). Most of these antibodies are well-characterized (antibodies against Hu, Yo, Ri, CRMP5, amphiphysin, Ma2 and Tr) and are in common use for the diagnosis of definite PNS.

MATERIALS AND METHODS

Literature on detection and clinical significance of onconeural antibodies were identified by using relevant search terms in PubMed and reviewed.

CONCLUSIONS

The onconeural antibodies are directed against intracellular antigens and their pathogenic role is still largely unknown. They are highly specific markers of paraneoplastic aetiology in patients with neurological symptoms. Detection of an onconeural antibody in a patient with neurological symptoms should lead to prompt investigation for cancer. However, absence of detectable onconeural antibodies does not exclude the PNS diagnosis. In particular, failure to detect antibodies in patients without classical PNS symptoms may result in less vigorous cancer screening and diagnostic delay. Neuronal antibodies that are directed to synaptic proteins or proteins of the cell membrane are also associated with neurological symptoms, and probably have pathogenic effects. The association between these antibodies and cancer is less robust, and they are usually not included among the onconeural antibodies.

[Anti-NMDA receptor paraneoplastic encephalitis: complete recovery after ovarian teratoma removal]

INTRODUCTION

A paraneoplastic syndrome characterized by neuropsychiatric symptoms, involuntary movements and seizures has been recently associated with antibodies targeting NMDA (N-methyl-D-aspartate) receptor in patients with an ovarian teratoma. Severe neurological impairment is frequent and treatment in the intensive care unit is often required because of ventilatory failure and life-threatening autonomic instability. Tumor removal is curative in many cases and neurological improvement is demonstrated shortly after surgery.

CASE REPORT

Here we report on a patient with paraneoplastic encephalitis manifested by unconsciousness and coreo-athetosic movements related to NMDA receptor antibodies associated with an immature ovarian teratoma grade III. She made a complete recovery after oophorectomy, intravenous immunoglobulin and corticosteroids.

CONCLUSIONS

Treatment of paraneoplastic syndromes is based on specific therapy for underlying tumor associated to immunomodulators. As in this case, anti-NMDA encephalitis may significantly improve after tumor removal and intra-venous immunoglobuline.

Thymoma and paraneoplastic myasthenia gravis

Paraneoplastic autoimmune diseases associate occasionally with small cell lung cancers and gynecologic tumors. However, myasthenia gravis (MG) occurs in at least 30% of all patients with thymomas (usually present at MG diagnosis). These epithelial neoplasms almost always have numerous admixed maturing polyclonal T cells (thymocytes). This thymopoiesis-and export of mature CD4(+)T cells-particularly associates with MG, though there are rare/puzzling exceptions in apparently pure epithelial WHO type A thymomas. Other features potentially leading to inefficient self-tolerance induction include defective epithelial expression of the autoimmune regulator (AIRE) gene and/or of major histocompatibility complex class II molecules in thymomas, absence of myoid cells, failure to generate FOXP3(+) regulatory T cells, and genetic polymorphisms affecting T-cell signaling. However, the strong focus on MG/neuromuscular targets remains unexplained and suggests some biased autoantigen expression, T-cell selection, or autoimmunization within thymomas. There must be further clues in the intriguing serological and cellular parallels in some patients with late-onset MG but without thymomas-and in others with AIRE mutations-and in the contrasts with early-onset MG, as discussed here.

[Neurological syndromes, encephalitis]

The remote effects of malignant tumors in most cases of paraneoplastic neurological syndromes(PNS)are mediated by autoimmune processes against antigens shared by the tumor cells and the nervous tissue(onconeural antigens). Onconeural (or paraneoplastic)antibodies are broadly categorized into two groups according to the location of the corresponding onconeural antigens, inside or on the surface of neurons. Antibodies established as clinically relevant diagnostic markers for PNS are designated as well-characterized onconeural antibodies (or classical antibodies)that target intracellular antigens(Hu, Yo, Ri, CV2/CRMP5,Ma2, and amphiphysin). They also serve as useful markers in detecting primary tumors. Recent identification of new antibodies as markers of subtypes of limbic encephalitis has also expanded the concept of autoimmune limbic encephalitis. These autoantibodies are directed to neuronal cell-surface antigens including neurotransmitter receptors(NMDA, AMPA, and GABAB receptors)and ion channels(VGKC). They are less frequently associated with cancer, so that they cannot be used as specific markers for PNS. Autoimmune limbic encephalitis with anti-neuronal cell surface antobodies and paraneoplastic limbic encephalitis with classical antibodies overlap in some clinical features but are pathophysiologically distinct. Classical antibodies are not simple tumor markers. They seem to be closely related to the disease mechanisms because specific intrathecal synthesis has been shown in PNS patients. However, attempts to produce an animal model of PNS by passive transfer of these antibodies have been unsuccessful, and there is no direct evidence demonstrating the pathogenic role of classical antibodies. Instead, some circumstantial evidence, including pathological studies showing extensive infiltrates of T cells in the CNS of the patients, supports the hypothesis that cytotoxic-T cell mechanisms cause irreversible neuronal damage. On the other hand, humoral immune response is probably the principal mechanism in autoimmune encephalitis associated with antibodies against neuronal cell-surface antigens. Those antibodies are supposed to mediate neural dysfunction which may be reversed by immunosuppression therapy, while the exact mechanism remains to be elucidated. Further accumulation of the cases and longer observation would be necessary to delineate the clinical spectrum of each type of newly-identified autoimmune limbic encephalitis. Early diagnosis and optimal oncological treatment is a prerequisite for better prognosis of PNS patients. Detection of the primary tumor at very early stages including carcinoma in situ is a challenging issue. Optimization of immunosuppression/ immunomodulation therapy for each patient according to the underlying pathophysiological processes is another important clinical issue.

[Pathogenesis of paraneoplastic neurological syndromes]

Paraneoplastic neurological syndromes (PNS) are thought to be caused by autoimmune processes triggered by the cancer and directed against antigens common to both the cancer and the nervous system. There are several clinical phenotypes in combinations with the neurological syndromes, origin of cancer and the specific autoantibodies. PNS with antibodies against channel receptors on the cell surface tend to respond favorably to antibody-depletion therapies; this suggests that the antibodies detected in these PNS groups are closely related to their pathogenesis. PNS having the antibodies against intracellular proteins might be caused by cytotoxic T cell-mediated cell death. This is because the following findings; 1) the prominent mononuclear cells in CSF, 2) the infiltration of inflammatory cells, mainly CD8+ T lymphocytes, in the tumor and the nervous tissue. In addition, T cell receptor usage of infiltrated T lymphocytes in the affected CNS lesions has been shown to be oligoclonal. We observed that the disease model could not be produced using only anti-intracellular antibodies such as anti-Yo or anti-Hu antibodies, but that CTL activity could be induced in CD8+ T cells isolated from the peripheral mononuclear cells obtained from PNS patients with anti-Yo or anti-Hu antibodies. The anti-Yo- or anti-Hu-antibody-positive patients possess common human leukocyte antigen (HLA) class I motifs. This implies that in patients with anti-Yo or anti-Hu antibodies, the presentation of the certain antigen peptides on the cell surface could be used to stimulate CD8+ T lymphocytes that could attack their target tissues as effectors. Antigen-specific CTL-mediated cell death has been observed in cancer immunology and PNS appears to be a potential candidates for a future CTL-mediated neurological disease model.