Managing Paraneoplastic Neurological Disorders
http://www.100md.com
《肿瘤学家》
LEARNING OBJECTIVES
After completing this course, the reader will be able to:
Describe the autoimmune pathogenesis of paraneoplastic neurological syndromes.
Explain the clinical value of paraneoplastic antibody detection.
Describe the general treatment approach to paraneoplastic neurological syndromes.
ABSTRACT
Paraneoplastic neurological syndromes (PNS) are remote effects of cancer that are not caused by invasion of the tumor or its metastases. Immunologic factors appear important in the pathogenesis of PNS because antineuronal autoantibodies and T-cell responses against nervous system antigens have been defined for many of these disorders. The immunologic response is elicited by the ectopic expression of neuronal antigens by the tumor. Expression of these so-called "onconeural" antigens is limited to the tumor and the nervous system and sometimes also the testis. At the time of presentation of the neurological symptoms, most patients have not yet been diagnosed with cancer. Detection of paraneoplastic antibodies is extremely helpful in diagnosing an otherwise unexplained and often rapidly progressive neurological syndrome as paraneoplastic. In addition, the paraneoplastic antibodies may also direct the search for an underlying neoplasm. On the other hand, in patients known to have cancer, the presentation of a PNS may herald recurrence of the tumor or a second tumor. The number of paraneoplastic antibodies is still growing, and at least seven of these can now be considered well characterized. Based on the clinical syndrome, the type of antibody, and the presence or absence of cancer, patients are classified as having a "definite" or "possible" PNS. Despite the presumed autoimmune etiology of PNS, the results of various forms of immunotherapy have been disappointing, with some exceptions. Rapid detection and immediate treatment of the underlying tumor appears to offer the best chance of stabilizing the patient and preventing further neurological deterioration.
INTRODUCTION
Paraneoplastic neurological syndromes (PNS) are remote effects of cancer that are, by definition, caused neither by invasion of the tumor or its metastases nor by infection, ischemia, metabolic and nutritional deficits, surgery, or other forms of tumor treatment [1]. Immunologic factors are believed to be important in the pathogenesis of PNS because antibodies and T-cell responses against nervous system antigens have been defined for many of these disorders [1].
Presumably, the immunologic response is elicited by the ectopic expression of neuronal antigens by the tumor. Expression of these "onconeural" antigens is limited to the tumor and the nervous system, and sometimes also the testis. At the time of presentation of the neurological symptoms, most patients have not yet been diagnosed with cancer [2–5]. Detection of paraneoplastic antibodies can help diagnose the neurological syndrome as paraneoplastic and may direct the search for an underlying neoplasm. Often, the oncologist or hematologist will be involved in the tumor workup. On the other hand, in patients known to have cancer, the presentation of a PNS may herald recurrence of the tumor or a second tumor. In these patients, however, metastatic complications of the known cancer must be ruled out first. Despite the presumed autoimmune etiology of PNS, the results of various forms of immunotherapy have been disappointing, with some exceptions [2–5]. Rapid detection and immediate treatment of the underlying tumor appears to offer the best chance of stabilizing the patient and preventing further neurological deterioration [2–5].
PATHOGENESIS
Pathological examination of the nervous system generally shows loss of neurons in affected areas of the nervous system with inflammatory infiltration by CD4+ T-helper cells and B cells in the perivascular spaces and cytotoxic CD8+ T cells in the interstitial spaces [6–8]. Examination of the cerebrospinal fluid (CSF) frequently demonstrates pleocytosis, intrathecal synthesis of IgG, and oligoclonal bands, supporting an inflammatory or immune-mediated etiology.
The discovery of paraneoplastic antineuronal autoantibodies resulted in the general belief that these are immune-mediated disorders triggered by aberrant expression of onconeural antigens in the tumor. Support for this hypothesis comes from the fact that the target paraneoplastic antigens are expressed both in the tumor and in the affected parts of the nervous system. Furthermore, the tumors are usually small and heavily infiltrated with inflammatory cells, and spontaneous remissions at the time of neurological presentation have been described [9, 10]. These findings suggest that some PNS without an identifiable tumor may result from immune-mediated eradication of the tumor [9, 10]. In keeping with this hypothesis, one study found more limited disease distribution and better oncologic outcome in small cell lung cancer (SCLC) patients with paraneoplastic autoantibodies [11].
Although the paraneoplastic antibodies are synthesized intrathecally, a pathogenic role could only be proven for those paraneoplastic autoantibodies that are directed against easily accessible antigens located at the cell surface. Examples of such antigens are the acetylcholine receptor (anti-AChR muscle type in myasthenia gravis and neuronal ganglionic type in autonomic neuropathy), P/Q-type voltage-gated calcium channels (anti-VGCC in Lambert-Eaton myasthenic syndrome [LEMS]), voltage-gated potassium channels (anti-VGKC in neuromyotonia), and the metabotropic glutamate receptor mGluR1 (anti-mGluR1 in paraneoplastic cerebellar degeneration [PCD]). Most paraneoplastic antigens are located in the cytoplasm (e.g., the Yo antigen) or nucleus (e.g., the Hu and Ri antigens), and a pathogenic role for the respective antibodies has not be demonstrated [12]. In these disorders, indirect lines of evidence support the view that the cellular immune response against these antigens is responsible for the neurological damage [13–15]. The relative contribution of the cellular and humoral immunity to the clinical and pathological manifestations has not been resolved [13–15]. The paraneoplastic antibodies may, in these cases, be surrogate markers for T-lymphocyte activation [16].
A totally different mechanism seems at work in PCD in Hodgkin’s lymphoma because the target antigens of the associated anti-Tr and anti-mGluR1 autoantibodies are not expressed in Hodgkin’s tumor tissue [17]. Dysregulation of the immune system in Hodgkin’s lymphoma and an etiologic role for (viral?) infections have been postulated in this disorder.
INCIDENCE
The incidence of PNS varies with the neurological syndrome and with the tumor. Approximately 10% of patients with plasma cell disorders accompanied by malignant monoclonal gammopathies are affected by a paraneoplastic peripheral neuropathy. More than half of the patients with the rare osteosclerotic form of myeloma develop a severe predominantly motor paraneoplastic peripheral neuropathy. In other hematological malignancies, the incidence of PNS is very low, with the exception of Hodgkin’s disease. However, the incidence of PNS even in Hodgkin’s disease is well below 1%. In solid tumors, the more common neurological syndromes are myasthenia gravis, which occurs in 15% of patients with a thymoma, and LEMS, which affects 3% of patients with SCLC. For other solid tumors, the incidence of PNS is <1%.
DIAGNOSIS
Clinical syndromes are never pathognomonic for a paraneoplastic etiology, and a high index of clinical suspicion is important. Symptoms can be atypical, psychiatric, or even fluctuating, and PNS should often be in the differential diagnosis of otherwise unexplained neurological syndromes. Some neurological syndromes, such as limbic encephalitis and subacute cerebellar degeneration, are associated relatively often with cancer. These are called "classical" PNS and are in italics in Table 1 [18]. Other syndromes, such as sensorimotor polyneuropathy, are much more prevalent, and their association with cancer may be by chance. Detection of a "well-characterized" paraneoplastic antibody is extremely helpful because it proves the paraneoplastic etiology of the neurological syndrome. The paraneoplastic antibodies are generally divided into three categories (Table 2) [18]. The well-characterized antibodies are reactive with molecularly defined onconeural antigens. These antibodies are strongly associated with cancer and have been detected unambiguously by several laboratories in a reasonable number of patients with well-defined neurological syndromes [18]. The partially characterized antibodies are those with an unidentified target antigen and those that have either been described by a single group of investigators or have been reported in only a few patients. The third group consists of antibodies that are associated with specific disorders but do not differentiate between paraneoplastic and nonparaneoplastic cases.
Because different antibodies can be associated with the same clinical findings [4], and the same antibody can be associated with different clinical syndromes [2, 3], paraneoplastic antibodies should be searched for by screening rather than by focusing on a specific antibody. Recently, Pittock et al. [19] demonstrated, in a large prospective series, that approximately 30% of patients have more than one paraneoplastic antibody. The combination of paraneoplastic antibodies provides important additional information to narrow the search for an underlying malignancy [19].
In the absence of paraneoplastic antibodies, additional diagnostic tests may be helpful in some PNS, although these are never specific for a paraneoplastic etiology. Magnetic resonance imaging (MRI) can help diagnose limbic encephalitis and may demonstrate cerebellar atrophy several months after the onset of PCD. Examination of the CSF is generally not required for detection of paraneoplastic antibodies because these can almost always be detected in serum as well. CSF examination may, however, show signs of inflammation, such as elevated white cell counts, oligoclonal bands, and intrathecal synthesis of IgG, indicating an immune-mediated or inflammatory etiology. In patients known to have cancer, MRI and CSF cytology are important in ruling out leptomeningeal metastases. Some PNS of the peripheral nervous system, such as LEMS, myasthenia gravis, and neuromyotonia, are accompanied by characteristic electrophysiological changes. These findings, however, are also present in the absence of an underlying tumor. Determining the precise type of neurological syndrome may assist in the search for an underlying tumor, such as SCLC in LEMS and thymoma in myasthenia gravis.
Once a paraneoplastic diagnosis has been established or is suspected, rapid identification of the tumor becomes essential but may be difficult because most PNS develop in the early stages of cancer. The workup generally starts with a detailed history, including smoking habits, weight loss, night sweats, and fever. A thorough physical examination should include palpation for pathological lymph nodes, rectal and pelvic examination, and palpation of breasts and testis. Often, the tumor is detected by high-resolution computed tomography (CT) of the chest, abdomen, and pelvis. If the CT scan remains negative, whole-body fluorodeoxyglucose positron emission tomography (FDG-PET) or PET/CT is recommended to detect an occult tumor or its metastases [20–22]. In addition, the type of antibody and PNS may suggest a specific underlying tumor and indicate further diagnostic tests, such as mammography (may be replaced by MRI) or ultrasound of the testes or pelvis (Table 2). When all tests remain negative, repeat evaluation at 3- to 6-month intervals for 2–3 years is recommended.
Diagnosing a neurological syndrome as paraneoplastic requires the exclusion of other possible causes by a reasonably complete workup. Because of the difficulties in diagnosis, an international panel of neurologists has established diagnostic criteria that divide patients with a suspected PNS into "definite" and "probable" categories. These criteria are based on the presence or absence of cancer, the presence of well-characterized antibodies, and the type of clinical syndrome. Patients with a definite PNS include those with [18]:
A classical syndrome (i.e., encephalomyelitis, limbic encephalitis, subacute cerebellar degeneration, opsoclonus-myoclonus, subacute sensory neuronopathy, chronic gastrointestinal pseudo-obstruction, LEMS, or dermatomyositis) and cancer that develops within 5 years of the diagnosis of the neurological disorder, regardless of the presence of paraneoplastic antibodies.
A nonclassical syndrome that objectively improves or resolves after cancer treatment, provided that the syndrome is not susceptible to spontaneous remission.
A nonclassical syndrome with paraneoplastic antibodies (well characterized or not) and cancer that develops within 5 years of the diagnosis of the neurological disorder.
A neurological syndrome (classical or not) with well-characterized paraneoplastic antibodies (i.e., anti-Hu, anti-Yo, anti-Ri, antiamphiphysin, anti-CV2, or anti-Ma2).
Patients with a possible PNS include those with [18]:
A classical syndrome without paraneoplastic antibodies and no cancer but at high risk to have an underlying tumor (e.g., smoking habit).
A neurological syndrome (classical or not) without cancer but with partially characterized paraneoplastic antibodies.
A nonclassical neurological syndrome, no paraneoplastic antibodies, and cancer that presents within 2 years of the neurological syndrome.
TREATMENT AND PROGNOSIS
Despite the immunological etiology of most of the PNS, the results of immunotherapy have been disappointing [23]. Exceptions are the neurological syndromes associated with paraneoplastic antibodies that are directed against antigens that are located at the surface of the cell (i.e., antigens that are accessible to circulating antibodies). These include not only disorders of the peripheral nervous system (LEMS, myasthenia gravis, and neuromyotonia) but also anti-mGluR1-associated PCD and antiamphiphysin-associated stiff-person syndrome [24, 25]. Immunotherapy modalities that are recommended for these disorders include plasma exchange, immunoadsorption (extraction of patient IgG over a protein A column), steroids, and i.v. Ig.
For most PNS, when the antigen is cytoplasmic or nuclear, the nervous dysfunction is probably not caused by functional interference of antibodies with the target antigen. In disorders with intracellular target antigens and a strong cellular immune reaction, plasma exchange and immunoadsorption are not expected to give much benefit. In these cases, a trial of a treatment that modulates the activation and function of effector T cells makes more sense, but to date there is only limited evidence that steroids, cyclophosphamide, i.v. Ig, or other immunosuppressive therapies are effective [26].
Hence, the first goal of treatment for PNS is control of the tumor. In addition, antitumor therapy has been demonstrated to stop the paraneoplastic neurological deterioration and leave the patients, on average, in better condition [27]. In severely debilitated patients, for example, the elderly and bedridden, treatment of the underlying tumor is often withheld because of the very small chance of clinically relevant neurological improvement.
Table 3 provides a summary of treatment of PNS and the effect on neurological outcome.
CLINICAL SYNDROMES
The classical PNS are described below. Descriptions of the nonclassical syndromes, which usually are not paraneoplastic but may occur in association with cancer, can be found elsewhere [28].
Encephalomyelitis Paraneoplastic encephalomyelitis is characterized by involvement of several areas of the nervous system, including the temporal lobes and limbic system (limbic encephalitis), brainstem (brainstem encephalitis), cerebellum (subacute cerebellar degeneration), spinal cord (myelitis) dorsal root ganglia (subacute sensory neuronopathy), and autonomous nervous system (autonomic neuropathy) [29, 30]. Patients with predominant involvement of one area but clinical evidence of only mild involvement of other areas are usually classified according to the predominant clinical syndrome. Symptoms of limbic encephalitis, subacute cerebellar degeneration, subacute sensory neuronopathy, and autonomic neuropathy are described below. Symptoms of brainstem encephalitis can include diplopia, dysarthria, dysphagia, gaze abnormalities (nuclear, internuclear, or supranuclear), facial numbness, and subacute hearing loss.
Underlying Tumor
Although virtually all cancer types have been associated with paraneoplastic encephalomyelitis, the majority of patients have underlying SCLC [2, 3, 29–31]. Most patients are not known to have cancer when the neurological symptoms present, and the SCLC may be difficult to demonstrate because of its small size. When anti-Hu antibodies are detected or when the patient is at risk for lung cancer (smoking, age >50 years), a careful and repeated search for underlying SCLC is warranted. When the CT scan is negative, a total-body FDG-PET scan or FDG-PET/CT scan may detect the neoplasm [20,21]. When a tumor other than SCLC is detected in a patient with anti-Hu antibodies, it may unexpectedly express the Hu antigen [2] or may be an unrelated secondary neoplasm [31]. When tumor tissue is available for analysis and expresses the Hu antigen, a further workup for a second tumor (SCLC) can probably be safely deferred [2].
Diagnostic Evaluation
MRI or CT of the brain is normal or shows specific changes in most paraneoplastic encephalomyelitis patients with two exceptions [30]. In 65%–80% of patients with predominant limbic encephalitis, MRI and CT show temporal lobe abnormalities [32,33]. Patients with a predominant cerebellar syndrome will develop cerebellar atrophy in the chronic stage. CSF is abnormal in most patients, showing elevated protein, mild mononuclear pleocytosis, elevated IgG index, or oligoclonal bands [30].
Antineuronal Antibodies
Patients with paraneoplastic encephalomyelitis and SCLC often have anti-Hu antibodies (also called antineuronal nuclear autoantibodies [ANNA-1]) in their serum and CSF [2, 3, 30, 31]. Other antibodies associated with paraneoplastic encephalomyelitis include anti-CRMP5/CV2 [16], anti-amphiphysin [34], and the less well-characterized ANNA-3 [35] and Purkinje cell antibody, PCA-2 [36].
Treatment and Prognosis
Tumor treatment offers the best chance of stabilizing the patient’s neurological condition, while immunotherapy does not appear to modify the outcome of paraneoplastic encephalomyelitis [2, 3, 23]. Therefore, all efforts should be directed at early diagnosis of paraneoplastic encephalomyelitis and rapid identification and treatment of the tumor. Because of incidental reports of neurological improvement following various forms of immunosuppressive treatment, a trial of one or two immunosuppressive modalities may be warranted in a single patient. However, spontaneous neurological improvement has rarely been described [10]. The overall functional outcome is bad, and more than 50% of patients are confined to bed or chair in the chronic phase of the disease [2, 3, 23]. The median survival time of patients is approximately 1 year from diagnosis [2, 3]. Mortality is predicted by worse functional status at diagnosis, age >60 years, involvement of more areas of the nervous system, and absence of treatment [2].
Because of the limited efficacy of plasma exchange, i.v. Ig, and corticosteroids [2, 3, 23] and the presumed role of cellular immunity, more aggressive immunosuppression with cyclophosphamide, tacrolimus, or cyclosporine may be considered. To limit toxicity, these more aggressive immunosuppressive approaches should probably be reserved for patients who are not receiving chemotherapy.
Limbic Encephalitis
Paraneoplastic limbic encephalitis (PLE) is a rare disorder characterized by the subacute onset (in days to a few months) of short-term memory loss, seizures, confusion, and psychiatric symptoms suggesting involvement of the limbic system [29, 37]. Hypothalamic dysfunction may occur with somnolence, hyperthermia, and endocrine abnormalities. Selective impairment of recent memory is a hallmark of the disease but may not be evident in patients presenting with severe confusion or multiple seizures [33]. More than half of the patients presenting with limbic encephalitis may have an underlying neoplasm [33]. Clinically three groups of patients with PLE can be identified [33]. The first group consists of patients with anti-Hu antibodies and lung cancer (usually SCLC). The limbic encephalitis is part of paraneoplastic encephalomyelitis, and the patients have involvement of other areas outside the limbic system and brainstem. These patients are older (median age, 62 years), usually smoke, and are more often female [33, 38]. The second group consists of young males with testicular cancer and anti-Ma2 antibodies [39]. The median age is 34 years. Symptoms are usually confined to the limbic system, hypothalamus, and brainstem. The third group has no antineuronal antibodies (approximately 40% of patients with PLE) [32, 33]). In these patients, the symptoms are more often confined to the limbic system, the median age is around 57 years, and the associated tumor is often located in the lung [33, 38].
Underlying Tumor
The associated tumor is a lung tumor in 50%–60% of patients, usually SCLC (40%–55%), and the associated tumor is a testicular germ cell tumor in 20% of patients [32, 33, 38]. Other tumors include breast cancer, thymoma, Hodgkin’s disease, and immature teratomas [32, 33].
Diagnostic Evaluation
The diagnosis is often difficult because there are no specific clinical markers and symptoms usually precede the diagnosis of cancer [33]. MRI and CT scans are abnormal in 65%–80% of patients [32, 33]. Abnormalities consist of increased signal on T2-weighted and fluid-attenuated inversion-recovery images of one or both medial temporal lobes, hypothalamus, and brainstem. Early in the course of the disease, the MRI scan may be normal, and repeat imaging may be indicated. Co-registration of FDG-PET may further improve the sensitivity of imaging [40]. CSF examination is abnormal in 80% of patients, showing transient mild lymphocytic pleocytosis with elevated protein, IgG, or oligoclonal bands [32, 33]. Detection of paraneoplastic antibodies helps establish the diagnosis and direct a tumor search that should include the lung, breasts, and testicles in the absence of paraneoplastic antibodies.
Antineuronal Antibodies
Antineuronal antibodies are found in about 60% of patients with PLE. The most frequent related paraneoplastic antibodies are: anti-Hu, anti-Ma2 (with or without Ma1), anti-CV2/CRMP5, and antiamphiphysin [16, 33, 41]. The majority of patients with anti-Hu antibodies have symptoms that suggest dysfunction of areas of the nervous system outside the limbic system. The related tumor in these patients is usually SCLC. Patients with only anti-Ma2 antibodies (also called anti-Ta) are young males with testicular cancer. Patients with anti-Ma2 and anti-Ma1 antibodies are significantly older and are more often female [41]. Anti-Ma1 patients are more likely to develop cerebellar dysfunction and usually harbor tumors other than testicular cancer. Anti-CV2/CRMP5 antibodies are detected in patients with SCLC or thymoma [16]. Anti-VGKC antibodies can be associated with PLE and thymoma or with non–paraneoplastic limbic encephalitis [42, 43].
Treatment and Prognosis
Spontaneous complete recovery has been described, although very rarely [38, 44]. Immunotherapy is largely ineffective [33], but several cases benefiting from antitumor treatment have been reported [33, 38, 45]. Therefore, all efforts should be directed at identifying and treating the underlying tumor. If no tumor is found, the search should be repeated every 3–6 months for a total of 2–3years. Irrespective of treatment, partial neurological recovery was seen in 38% of patients with anti-Hu antibodies, 30% of patients with anti-Ta (anti-Ma2) antibodies, and 64% of patients without antibodies [33].
Subacute Cerebellar Degeneration
PCD is one of the most common and characteristic PNS [4, 29]. In a study of 137 consecutive patients with antibody-associated PNS, 50 (37%) presented with subacute cerebellar degeneration [4]. PCD usually starts acutely with nausea, vomiting, dizziness, and slight incoordination of walking, evolving rapidly over weeks to a few months with progressive ataxia of gait, limbs, and trunk, dysarthria, and often nystagmus associated with oscillopsia. The disease reaches its peak within months and then stabilizes. By this time, most patients are severely debilitated. They are generally unable to walk without support, they may be unable to sit unsupported, handwriting is often impossible, and feeding themselves has become difficult. The neurological signs are always bilateral but may be asymmetrical. Diplopia is common at presentation, although the investigator usually cannot detect abnormalities of ocular movement. The symptoms and signs are limited to the cerebellum and cerebellar pathways, but other mild neurological abnormalities may be found on careful examination. These include hearing loss, dysphagia, pyramidal and extrapyramidal tract signs, mental status change, and peripheral neuropathy [5, 46, 47].
Underlying Tumor
PCD can be associated with any cancer, but the most common tumors are lung cancer (usually SCLC), ovarian cancer, and lymphomas (particularly Hodgkin’s lymphoma). In 60%–70% of patients, neurological symptoms precede diagnosis of the cancer by a few months to 2–3 years and lead to its detection [4, 5, 17].
Diagnostic Evaluation
Subacute cerebellar degeneration is a rare disorder in cancer patients. On the other hand, 50% of patients presenting with acute or subacute nonfamilial ataxia are estimated to have an underlying malignancy [29]. MRI and CT scans are initially normal but often reveal cerebellar atrophy later in the course of the disease. CSF examination shows mild lymphocytic pleocytosis with elevated protein and IgG levels in the first weeks to months. Oligoclonal bands may be present. The diagnosis of PCD is established by demonstration of specific antineuronal antibodies. The type of antibody directs the search for an underlying neoplasm (Table 2).
Antineuronal Antibodies
PCD can be associated with various antineuronal autoantibodies. The clinical and tumor specificities of each of the antibodies are summarized in Table 2.
Anti-Yo (also called PCA-1), anti-Tr (PCA-Tr), and anti-mGluR1 antibodies are associated with relatively "pure" cerebellar syndromes. Anti-Yo antibodies are associated with breast cancer and tumors of the ovaries, endometrium, and fallopian tubes [4, 5, 48]. These antibodies are directed against the cerebellar degeneration–related (CDR) proteins that are expressed by Purkinje cells and the associated tumors [48, 49]. CDR-2–specific cytotoxic T cells have been identified in the serum from patients with PCD, suggesting a pathogenic role for the cellular immune response in this PNS [14]. Anti-Tr (PCA-Tr) antibodies are directed against an unidentified cytoplasmic Purkinje cell antigen and appear specific for Hodgkin’s disease [17]. Anti-mGluR1 antibodies have been found in two patients with PCD and Hodgkin’s disease. Passive transfer of patient anti-mGluR1 IgG into CSF of mice induced severe, transient ataxia [25].
Approximately50%ofpatientswithcerebellardegeneration and underlying SCLC have high titers of anti-Hu antibodies [50]. The remaining patients are likely to have anti-P/Q-type VGCC antibodies. These antibodies were present in all patients who also had LEMS and in some patients with cerebellar degeneration without LEMS. In patients with antiamphiphysin or anti-CV2/CRMP5 antibodies, the cerebellar degeneration is often part of the paraneoplastic encephalomyelitis syndrome, and more widespread neurological symptoms and signs are usually found.
The more recently discovered PCA-2 antibody and the ANNA-3 antibody are associated with lung cancer and a variety of neurological syndromes including cerebellar degeneration [36]. Anti-Zic4 antibodies are strongly associated with SCLC, and most patients have paraneoplastic encephalomyelitis, often presenting with cerebellar dysfunction [51]. These patients often have concurrent anti-Hu or anti-CV2/CRMP5 antibodies. Patients with isolated antiZic4 antibodies are more likely to develop cerebellar symptoms.
Treatment and Prognosis
The outcome of PCD is generally poor, and the best chance to at least stabilize the syndrome is to treat the underlying tumor [4]. Incidental improvement has been reported either spontaneously or in association with plasma exchange, steroids, i.v. Ig, or rituximab [52]. In patients with anti-Yo–associated cerebellar degeneration, the prognosis is better for patients with breast cancer than for those with gynecologic cancer [5]. The prognosis is better in patients with PCD associated with Hodgkin’s disease and anti-Tr (PCA-Tr) or anti-mGluR1 antibodies. With successful treatment of the tumor and/or immunotherapy, symptoms may disappear and the antibodies vanish [17, 25].
Opsoclonus-Myoclonus
Opsoclonus is a disorder of ocular motility that consists of involuntary, arrhythmic, high-amplitude conjugate saccades in all directions. Opsoclonus may occur intermittently or, if more severe, constantly, and it does not remit in the darkness or when the eyes are closed. Opsoclonus is often associated with diffuse or focal myoclonus, the "dancing eyes and dancing feet syndrome," and other cerebellar and brainstem signs [28, 53, 54]. An excessive startle response reminiscent of hyperekplexia may also occur in opsoclonus-myoclonus patients [55]. In contrast to most PNS, the course of opsoclonus-myoclonus may be remitting and relapsing [54].
Underlying Tumor
Approximately 20% of adult patients with opsoclonus-myoclonus have a previously undiscovered malignancy [53]. The most commonly associated neoplasms are SCLC and breast and gynecologic cancers [55, 56]. Many other tumors, including thyroid and bladder cancer, have also been reported [57].
Almost 50% of children with opsoclonus-myoclonus have an underlying neuroblastoma. Conversely, approximately 2%–3% of children with neuroblastoma have paraneoplastic opsoclonus-myoclonus [58, 59]. Tumors in children with paraneoplastic opsoclonus-myoclonus apparently have a better prognosis than tumors in patients without this PNS.
Diagnostic Evaluation
MRI scans are usually normal but may show hyperintensities in the brainstem on T2-weighted images [60]. Examination of the CSF may show mild pleocytosis and protein elevation. In some patients, paraneoplastic opsoclonus-myoclonus resembles PCD. The prominent opsoclonus and truncal, rather than appendicular, ataxia distinguish this syndrome from anti-Yo– and anti-Hu–associated PCD [28]. Adult patients with paraneoplastic opsoclonus-myoclonus are older (median age, 66 years) than patients with the idiopathic syndrome (median age, 40 years). In adult patients, the tumor search should be directed at the most common underlying tumors, that is, high-resolution CT of the chest and abdomen, and gynecological examination and mammography (or MRI of the breasts) [56]. When this is negative, FDG-PET should be considered [22, 61].
In children, nonparaneoplastic opsoclonus–myoclonus occurs as a self-limited illness and is probably the result of a viral infection of the brainstem. The search for an occult neuroblastoma should include imaging of the chest and abdomen (CT or MRI scan), urine catecholamine measurements, and metaiodobenzylguanidine scan [62]. When negative, the evaluation should be repeated after several months [63].
Antineuronal Antibodies
Specific antibodies are found in only a minority of patients with paraneoplastic opsoclonus-myoclonus [56]. In women, anti-Ri antibodies (or ANNA-2) are mostly associated with breast and gynecologic tumors. Anti-Ri has occasionally been found in bladder cancer and SCLC and may then occur in male patients [28, 57]. Anti-Ri antibodies are directed against the Nova proteins [64, 65]. Paraneoplastic opsoclonus-myoclonus can also be associated with anti-Hu antibodies, usually as part of a more widespread paraneoplastic encephalomyelitis. Bataller et al. [66] screened a brainstem cDNA library with sera from 21 patients with (paraneoplastic) opsoclonus-myoclonus. Twenty-five proteins were identified, recognized by one or two sera each, demonstrating that immunity to neuronal autoantigens in opsoclonus-myoclonus is both frequent and heterogeneous.
In children presenting with opsoclonus-myoclonus, the detection of anti-Hu antibodies is diagnostic of an underlying neuroblastoma [67]. The frequency of anti-Hu antibodies in neuroblastoma with paraneoplastic opsoclonus-myoclonus is approximately 10% [67–69]. This finding differs little from the 4%–15% of anti-Hu positive sera in children with neuroblastoma who do not have opsoclonus-myoclonus [67, 68].
Treatment and Prognosis
In contrast to most of the other PNS, paraneoplastic opsoclonus-myoclonus may remit either spontaneously, following treatment of the tumor, or in association with clonazepam or thiamine treatment. Most patients with idiopathic opsoclonus-myoclonus make a good recovery that seems to be accelerated by steroids or i.v. Ig. Paraneoplastic opsoclonus-myoclonus usually has a more severe clinical course, and treatment with steroids or i.v. Ig appears ineffective. In a series of 14 patients with paraneoplastic opsoclonus-myoclonus, eight patients whose tumors were treated showed complete or partial neurological recovery. In contrast, five of the six patients whose tumors were not treated died of the neurological syndrome despite steroids, i.v. Ig, or plasma exchange [56]. However, improvement following the administration of steroids, cyclophosphamide, azathioprine, i.v. Ig, plasma exchange, or plasma filtration with a protein A column has been described in single cases [55, 70–72].
In children, paraneoplastic opsoclonus-myoclonus may improve following treatment with adrenocorticotropic hormone, prednisone, azathioprine, or i.v. Ig, but residual central nervous system signs are frequent [59, 63, 73]. Treatment of the tumor with chemotherapy is the most important predictor of good neurological recovery [74].
Subacute Sensory Neuronopathy
Subacute sensory neuronopathy is an uncommon disorder that is probably paraneoplastic in about 20% of patients [75, 76]. The symptoms begin with pain and paraesthesia. Clumsiness and unsteady gait then develop and usually become predominant. The distribution of symptoms is often asymmetrical or multifocal. The upper limbs are often affected first and are almost invariably involved with evolution. Sensory loss may also affect the face, chest, or abdomen. On examination, all sensory modalities are affected, but the most striking abnormality is loss of deep sensation causing sensory ataxia with pseudoathetosis of the hands. Tendon reflexes are depressed or absent. In most patients, the disease progresses rapidly over weeks to months, leaving the patient severely disabled. In a few patients, the neuronopathy remains stable for months with mild neurological deficits [77]. Subacute sensory neuronopathy occurs in approximately 75% of patients with paraneoplastic encephalomyelitis, is predominant in 50%, and is clinically pure in 25% [2, 3]. Autonomic neuropathy, including gastrointestinal pseudo-obstruction, is common.
Underlying Tumor
Subacute sensory neuronopathy is associated with lung cancer, usually SCLC, in 70%–80% of patients [2, 3, 31]. Other associated tumors include breast cancer, ovarian cancer, sarcoma, and Hodgkin’s lymphoma [75, 76]. Subacute sensory neuronopathy usually predates the diagnosis of cancer, with a median delay of 3.5–4.5 months [2, 3].
Diagnostic Evaluation
Electrophysiologically, the hallmark of subacute sensory neuronopathy is the absence of, or marked reduction in, sensory nerve action potentials. Motor conduction velocities may be mildly reduced. Early in the course of the disease, CSF examination shows mild pleocytosis, with an elevated IgG level and oligoclonal bands [3, 75, 76]. Sural nerve biopsy is rarely required for the diagnosis but may differentiate this disorder from vasculitic neuropathy.
Antineuronal Antibodies
Anti-Hu is the most frequent paraneoplastic antibody in subacute sensory neuronopathy [2, 3, 30, 31]. In this setting, anti-Hu antibody detection has a specificity of 99% and sensitivity of 82% [78]. The absence of anti-Hu antibodies does not rule out an underlying cancer. Anti-CRMP5/CV2 antibodies also occur with paraneoplastic peripheral neuropathies [79]. These patients usually have a sensory or sensorimotor neuropathy, with less frequent involvement of the arms but often associated with cerebellar ataxia [16, 79, 80]. Anti-CRMP5/CV2 antibodies are usually associated with SCLC, neuroendocrine tumors, and thymoma. Antiamphiphysin antibodies are associated with multifocal paraneoplastic encephalomyelitis, and symptoms often include sensory or sensorimotor neuropathy [34, 81, 82]. Associated tumors (mostly limited) are mainly SCLC, breast cancer, and melanoma.
Treatment and Prognosis
Immunotherapy consisting of plasma exchange, steroids, and i.v. Ig is ineffective in most cases [27, 83]. There may be some exceptions to this rule [23, 84]. In one study, 2 of 10 patients stabilized in relatively good clinical condition following intensive treatment with a combination of steroids, cyclophosphamide, and i.v. Ig [23]. Early detection and treatment of the underlying neoplasm, usually SCLC, appears to offer the best chance of stabilizing the neurological symptoms [3, 27]. In patients with an identifiable tumor, antitumor treatment is recommended. In the absence of a tumor, antitumor treatment may be considered in patients with anti-Hu antibodies, age >50 years, and with a history of smoking. In patients not receiving antitumor therapy, a short course of immunotherapy can be considered.
Symptomatic treatment is directed at neuropathic pain and dysautonomic symptoms such as orthostatic hypotension.
LEMS
Patients with LEMS present with proximal weakness of the lower extremities and fatigability. Bulbar symptoms may occur more frequently than previously reported [85] but are generally milder than with myasthenia gravis. Respiratory weakness can occur. Deep tendon reflexes, especially those in the legs, are diminished or absent but may reappear after exercise. Autonomic features, especially dryness of the mouth, impotence, and mild/moderate ptosis, ultimately develop in 95% of patients [85–87]. In some patients, LEMS may develop in association with other PNS, including PCD and encephalomyelitis [50].
Underlying Tumor
Approximately 70% of patients have cancer, almost always SCLC [86, 88]. Other tumors include small cell carcinomas of the prostate and cervix, lymphomas, and adenocarcinomas. The prevalence of LEMS in patients with SCLC is estimated to be around 3% [87, 89]. Clinically and serologically, the 30% without identifiable tumors are indistinguishable from the paraneoplastic LEMS patients, although LEMS may have a more progressive course in patients with SCLC [85]. In patients presenting with LEMS, a smoking history and absence of the HLA-B8 genotype strongly predict underlying SCLC [90]. Patients with SCLC and LEMS survive significantly longer than SCLC patients who do not have this PNS [91].
Diagnostic Evaluation
The typical pattern of electromyographic abnormalities is the hallmark of LEMS. This includes a low compound muscle action potential at rest with a decreased response at low rates of repetitive stimulation (3 Hz) and an incremental response at high rates of repetitive stimulation (50 Hz) or 15–30 seconds of maximal voluntary contraction [92].
Antineuronal Antibodies
Most patients with LEMS have antibodies against P/Q-type calcium channels that are located presynaptically in the neuromuscular junction [92]. About 20% have anti-MysB antibodies reactive with the ß subunit of neuronal calcium channels [93].
Treatment and Prognosis
Treatment of LEMS must be tailored to the individual based on severity of the symptoms, underlying disease, life expectancy, and previous response to treatment. In patients with paraneoplastic LEMS, treatment of the tumor frequently leads to neurological improvement [94]. Symptomatic treatment is with drugs that facilitate the release of acetylcholine from motor nerve terminals, such as 3,4-diaminopyridine (DAP) [95]. In a placebo-controlled randomized trial, DAP (5–20 mg three to four times daily) was effective for long-term treatment, alone or in combination with other treatments [96]. The maximum recommended daily dose of DAP is 80 mg; at higher doses, seizures occur [96]. Cholinesterase inhibitors (pyridostigmine, 30–60 mg, every 6 hours) may improve dryness of the mouth but rarely relieve weakness. If these treatments are not effective enough, it must be decided if immunosuppressive therapy with steroids, azathioprine, or cyclosporine is appropriate. Removal of the pathogenic anti-P/Q-type calcium channel antibodies by plasma exchange [97] and i.v. Ig can give quick but transient relief [86, 98]. LEMS responds less favorably to immunotherapy than myasthenia gravis.
Dermatomyositis
In dermatomyositis, the characteristic heliotrope rash (purplish discoloration of the eyelids) often precedes the appearance of proximal muscle weakness. Other manifestations include arthralgia, myocarditis and congestive heart failure, and interstitial lung disease. Clinical, electromyographical, and pathological findings of dermatomyositis are similar in patients with and without cancer.
Underlying Tumor
The standardized incidence ratio for a malignant disease in dermatomyositis is 6.2 (95% confidence interval, 3.9–10.0) [99]. Dermatomyositis is associated with cancer of the ovary, lung, pancreas, stomach, colorectum, and breast, and with non-Hodgkin’s lymphoma [100].
Diagnostic Evaluation
Most patients have elevated serum creatine kinase levels and electromyographic evidence of myopathy. Muscle imaging (CT or MRI) may help in confirming the diagnosis and determining the type of inflammatory myopathy and in selecting an appropriate biopsy site. Muscle or skin biopsy is the definitive diagnostic procedure and shows inflammatory infiltrates [101].
Antineuronal Antibodies
Antibodies to the Mi-2 protein complex are specific for dermatomyositis and are present in high titers in about 35% of cases [102].
Treatment and Prognosis
Treatment of paraneoplastic dermatomyositis is generally the same as for patients without a tumor. Nearly all patients respond to corticosteroids [103]. Refractory patients and patients requiring a lower dose of steroids can be treated with azathioprine. Methotrexate and cyclophosphamide may also be considered [103].
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.
REFERENCES
Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med 2003;349:1543–1554.
Graus F, Keime-Guibert F, Rene R et al. Anti-Hu-associated paraneoplastic encephalomyelitis: analysis of 200 patients. Brain 2001;124: 1138–1148.
Sillevis Smitt P, Grefkens J, De Leeuw B et al. Survival and outcome in 73 anti-Hu positive patients with paraneoplastic encephalomyelitis/sensory neuronopathy. J Neurol 2002;249:745–753.
Shams’ili S, Grefkens J, de Leeuw B et al. Paraneoplastic cerebellar degeneration associated with antineuronal antibodies: analysis of 50 patients. Brain 2003;126:1409–1418.
Rojas I, Graus F, Keime-Guibert F et al. Long-term clinical outcome of paraneoplastic cerebellar degeneration and anti-Yo antibodies. Neurology 2000;55:713–715.
Denny-Brown D. Primary sensory neuropathy with muscular changes associated with carcinoma. J Neurol Neurosurg Psychiatry 1948;11: 73–87.
Jean WC, Dalmau J, Ho A et al. Analysis of the IgG subclass distribution and inflammatory infiltrates in patients with anti-Hu-associated paraneoplastic encephalomyelitis. Neurology 1994;44:140–147.
Bernal F, Graus F, Pifarre A et al. Immunohistochemical analysis of anti-Hu-associated paraneoplastic encephalomyelitis. Acta Neuropathol (Berl) 2002;103:509–15.
Darnell RB, DeAngelis LM. Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet 1993;341: 21–22.
Byrne T, Mason WP, Posner JB et al. Spontaneous neurological improvement in anti-Hu associated encephalomyelitis. J Neurol Neurosurg Psychiatry 1997;62:276–278.
Graus F, Dalmou J, Rene R et al. Anti-Hu antibodies in patients with small-cell lung cancer: association with complete response to therapy and improved survival. J Clin Oncol 1997;15:2866–2872.
Sillevis Smitt PA, Manley GT, Posner JB. Immunization with the paraneoplastic encephalomyelitis antigen HuD does not cause neurologic disease in mice. Neurology 1995;45:1873–1878.
Benyahia B, Liblau R, Merle-Beral H et al. Cell-mediated autoimmunity in paraneoplastic neurological syndromes with anti-Hu antibodies. Ann Neurol 1999;45:162–167.
Albert ML, Austin LM, Darnell RB. Detection and treatment of activated T cells in the cerebrospinal fluid of patients with paraneoplastic cerebellar degeneration. Ann Neurol 2000;47:9–17.
Tanaka M, Tanaka K, Tokiguchi S et al. Cytotoxic T cells against a peptide of Yo protein in patients with paraneoplastic cerebellar degeneration and anti-Yo antibody. J Neurol Sci 1999;168:28–31.
Yu Z, Kryzer TJ, Griesmann GE et al. CRMP-5 neuronal autoantibody: marker of lung cancer and thymoma-related autoimmunity. Ann Neurol 2001;49:146–154.
Bernal F, Shams’ili S, Rojas I et al. Anti-Tr antibodies as markers of paraneoplastic cerebellar degeneration and Hodgkin’s disease. Neurology 2003;60:230–234.
Graus F, Delattre JY, Antoine JC et al. Recommended diagnostic criteria for paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry 2004;75:1135–1140.
Pittock SJ, Kryzer TJ, Lennon VA. Paraneoplastic antibodies coexist and predict cancer, not neurological syndrome. Ann Neurol 2004;56:715–719.
Antoine JC, Cinotti L, Tilikete C et al. [18F]fluorodeoxyglucose positron emission tomography in the diagnosis of cancer in patients with paraneoplastic neurological syndrome and anti-Hu antibodies. Ann Neurol 2000;48:105–108.
Rees JH, Hain SF, Johnson MR et al. The role of [18F]fluoro-2-deoxyglucose-PET scanning in the diagnosis of paraneoplastic neurological disorders. Brain 2001;124:2223–2231.
Younes-Mhenni S, Janier MF, Cinotti L et al. FDG-PET improves tumour detection in patients with paraneoplastic neurological syndromes. Brain 2004;127:2331–2338.
Keime-Guibert F, Graus F, Fleury A et al. Treatment of paraneoplastic neurological syndromes with antineuronal antibodies (Anti-Hu, anti-Yo) with a combination of immunoglobulins, cyclophosphamide, and methyl-prednisolone. J Neurol Neurosurg Psychiatry 2000;68:479–482.
Sommer C, Weishaupt A, Brinkhoff J et al. Paraneoplastic stiff-person syndrome: passive transfer to rats by means of IgG antibodies to amphiphysin. Lancet 2005;365:1406–1411.
Sillevis Smitt P, Kinoshita A, DeLeeuw B et al. Paraneoplastic cerebellar ataxia due to autoantibodies against a glutamate receptor. N Engl J Med 2000;342:21–27.
Vernino S, O’Neill BP, Marks RS et al. Immunomodulatory treatment trial for paraneoplastic neurological disorders. Neuro-oncol 2004;6:55–62.
Keime-Guibert F, Graus F, Broet P et al. Clinical outcome of patients with anti-Hu-associated encephalomyelitis after treatment of the tumor. Neurology 1999;53:1719–1723.
Posner JB. Paraneoplastic syndromes. In: Posner JB, ed., Neurologic Complications of Cancer. Philadelphia: FA Davis; 1995:353–385.
Henson RA, Urich H. Cancer and the Nervous System: The Neurologic Complications of Systemic Malignant Disease. Oxford: Blackwell Scientific, 1982.
Dalmau J, Graus F, Rosenblum MK et al. Anti-Hu–associated paraneoplastic encephalomyelitis/sensory neuronopathy. A clinical study of 71 patients. Medicine (Baltimore) 1992;71:59–72.
Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and oncologic profiles of patients seropositive for type 1 antineuronal nuclear autoantibodies. Neurology 1998;50:652–657.
Lawn ND, Westmoreland BF, Kiely MJ et al. Clinical, magnetic resonance imaging, and electroencephalographic findings in paraneoplastic limbic encephalitis. Mayo Clin Proc 2003;78:1363–1368.
Gultekin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain 2000;123:1481–1494.
Dropcho EJ. Antiamphiphysin antibodies with small-cell lung carcinoma and paraneoplastic encephalomyelitis. Ann Neurol 1996;39:659–667.
Chan KH, Vernino S, Lennon VA. ANNA-3 anti-neuronal nuclear antibody: marker of lung cancer-related autoimmunity. Ann Neurol 2001;50:301–311.
Vernino S, Lennon VA. New Purkinje cell antibody (PCA-2): marker of lung cancer-related neurological autoimmunity. Ann Neurol 2000;47:297–305.
Corsellis JA, Goldberg GJ, Norton AR. "Limbic encephalitis" and its association with carcinoma. Brain 1968;91:481–496.
Alamowitch S, Graus F, Uchuya M et al. Limbic encephalitis and small cell lung cancer. Clinical and immunological features. Brain 1997;120:923–928.
Voltz R, Gultekin SH, Rosenfeld MR et al. A serologic marker of paraneoplastic limbic and brain-stem encephalitis in patients with testicular cancer. N Engl J Med 1999;340:1788–1795.
Kassubek J, Juengling FD, Nitzsche EU et al. Limbic encephalitis investigated by 18FDG-PET and 3D MRI. J Neuroimaging 2001;11:55–59.
Dalmau J, Graus F, Villarejo A et al. Clinical analysis of anti-Ma2-associated encephalitis. Brain 2004;127:1831–1844.
Vincent A, Buckley C, Schott JM et al. Potassium channel antibody-associated encephalopathy: a potentially immunotherapy-responsive form of limbic encephalitis. Brain 2004;127:701–712.
Pozo-Rosich P, Clover L, Saiz A et al. Voltage-gated potassium channel antibodies in limbic encephalitis. Ann Neurol 2003;54:530–533.
Taylor RB, Mason W, Kong K et al. Reversible paraneoplastic encephalomyelitis associated with a benign ovarian teratoma. Can J Neurol Sci 1999;26:317–320.
Rosenfeld MR, Eichen JG, Wade DF et al. Molecular and clinical diversity in paraneoplastic immunity to Ma proteins. Ann Neurol 2001;50: 339–348.
Hammack J, Kotanides H, Rosenblum MK et al. Paraneoplastic cerebellar degeneration. II. Clinical and immunologic findings in 21 patients with Hodgkin’s disease. Neurology 1992;42:1938–1943.
Posner JB. Paraneoplastic cerebellar degeneration. Can J Neurol Sci 1993;20:S117–S122.
Furneaux HM, Rosenblum MK, Dalmau J et al. Selective expression of Purkinje-cell antigens in tumor tissue from patients with paraneoplastic cerebellar degeneration. N Engl J Med 1990;322:1844–1851.
Fathallah-Shaykh H, Wolf S, Wong E et al. Cloning of a leucine-zipper protein recognized by the sera of patients with antibody-associated paraneoplastic cerebellar degeneration. Proc Natl Acad Sci U S A 1991;88:3451–3454.
Mason WP, Graus F, Lang B et al. Small-cell lung cancer, paraneoplastic cerebellar degeneration and the Lambert-Eaton myasthenic syndrome. Brain 1997;120:1279–1300.
Bataller L, Wade DF, Graus F et al. Antibodies to Zic4 in paraneoplastic neurologic disorders and small-cell lung cancer. Neurology 2004;62: 778–782.
Shams’ili S, de Beukelaar J, Gratama JW et al. An uncontrolled trial of rituximab for antibody associated paraneoplastic neurological syndromes. J Neurol 2006;253:16–20.
Digre KB. Opsoclonus in adults. Report of three cases and review of the literature. Arch Neurol 1986;43:1165–1175.
Anderson NE, Budde-Steffen C, Rosenblum MK et al. Opsoclonus, myoclonus, ataxia, and encephalopathy in adults with cancer: a distinct paraneoplastic syndrome. Medicine (Baltimore) 1988;67:100–109.
Wirtz PW, Sillevis Smitt PA, Hoff JI et al. Anti-Ri antibody positive opsoclonus-myoclonus in a male patient with breast carcinoma. J Neurol 2002;249:1710–1712.
Bataller L, Graus F, Saiz A et al. Clinical outcome in adult onset idiopathic or paraneoplastic opsoclonus-myoclonus. Brain 2001;124:437–443.
Prestigiacomo CJ, Balmaceda C, Dalmau J. Anti-Ri-associated paraneoplastic opsoclonus-ataxia syndrome in a man with transitional cell carcinoma. Cancer 2001;91:1423–1428.
Altman AJ, Baehner RL. Favorable prognosis for survival in children with coincident opso-myoclonus and neuroblastoma. Cancer 1976;37:846–852.
Rudnick E, Khakoo Y, Antunes NL et al. Opsoclonus-myoclonus-ataxia syndrome in neuroblastoma: clinical outcome and antineuronal antibodies-a report from the Children’s Cancer Group Study. Med Pediatr Oncol 2001;36:612–622.
Hormigo A, Dalmau J, Rosenblum MK et al. Immunological and pathological study of anti-Ri-associated encephalopathy. Ann Neurol 1994;36:896–902.
Linke R, Schroeder M, Helmberger T et al. Antibody-positive paraneoplastic neurologic syndromes: value of CT and PET for tumor diagnosis. Neurology 2004;63:282–286.
Swart JF, de Kraker J, van der Lely N. Metaiodobenzylguanidine total-body scintigraphy required for revealing occult neuroblastoma in opsoclonus-myoclonus syndrome. Eur J Pediatr 2002;161:255–258.
Hayward K, Jeremy RJ, Jenkins S et al. Long-term neurobehavioral outcomes in children with neuroblastoma and opsoclonus-myoclonus-ataxia syndrome: relationship to MRI findings and anti-neuronal antibodies. J Pediatr 2001;139:552–559.
Yang YY, Yin GL, Darnell RB. The neuronal RNA-binding protein Nova-2 is implicated as the autoantigen targeted in POMA patients with dementia. Proc Natl Acad Sci U S A 1998;95:13254–13259.
Buckanovich RJ, Yang YY, Darnell RB. The onconeural antigen Nova-1 is a neuron-specific RNA-binding protein, the activity of which is inhibited by paraneoplastic antibodies. J Neurosci 1996;16:1114–1122.
Bataller L, Rosenfeld MR, Graus F et al. Autoantigen diversity in the opsoclonus-myoclonus syndrome. Ann Neurol 2003;53:347–353.
Dalmau J, Graus F, Cheung NK et al. Major histocompatibility proteins, anti-Hu antibodies and paraneoplastic encephalomyelitis in neuroblastoma and small cell lung cancer. Cancer 1995;75:99–109.
Antunes NL, Khakoo Y, Matthay KK et al. Antineuronal antibodies in patients with neuroblastoma and paraneoplastic opsoclonus-myoclonus. J Pediatr Hematol Oncol 2000;22:315–320.
Pranzatelli MR, Tate ED, Wheeler A et al. Screening for autoantibodies in children with opsoclonus-myoclonus-ataxia. Pediatr Neurol 2002;27:384–387.
Jongen JL, Moll WJ, Sillevis Smitt PA et al. Anti-Ri positive opsoclonus-myoclonus-ataxia in ovarian duct cancer. J Neurol 1998;245:691–692.
Dropcho EJ, Kline LB, Riser J. Antineuronal (anti-Ri) antibodies in a patient with steroid-responsive opsoclonus-myoclonus. Neurology 1993;43:207–211.
Nitschke M, Hochberg F, Dropcho E. Improvement of paraneoplastic opsoclonus-myoclonus after protein A column therapy. N Engl J Med 1995;332:192.
Mitchell WG, Davalos-Gonzalez Y, Brumm VL et al. Opsoclonus-ataxia caused by childhood neuroblastoma: developmental and neurologic sequelae. Pediatrics 2002;109:86–98.
Russo C, Cohn SL, Petruzzi MJ et al. Long-term neurologic outcome in children with opsoclonus-myoclonus associated with neuroblastoma: a report from the Pediatric Oncology Group. Med Pediatr Oncol 1997;28:284–288.
Chalk CH, Windebank AJ, Kimmel DW et al. The distinctive clinical features of paraneoplastic sensory neuronopathy. Can J Neurol Sci 1992;19:346–351.
Horwich MS, Cho L, Porro RS et al. Subacute sensory neuropathy: a remote effect of carcinoma. Ann Neurol 1977;2:7–19.
Graus F, Bonaventura I, Uchuya M et al. Indolent anti-Hu-associated paraneoplastic sensory neuropathy. Neurology 1994;44:2258–2261.
Molinuevo JL, Graus F, Serrano C et al. Utility of anti-Hu antibodies in the diagnosis of paraneoplastic sensory neuropathy. Ann Neurol 1998;44:976–980.
Antoine JC, Honnorat J, Camdessanche JP et al. Paraneoplastic anti-CV2 antibodies react with peripheral nerve and are associated with a mixed axonal and demyelinating peripheral neuropathy. Ann Neurol 2001;49:214–221.
Honnorat J, Antoine JC, Derrington E et al. Antibodies to a subpopulation of glial cells and a 66 kDa developmental protein in patients with paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry 1996;61:270–278.
Pittock SJ, Lucchinetti CF, Parisi JE et al. Amphiphysin autoimmunity: Paraneoplastic accompaniments. Ann Neurol 2005;58:96–107.
Saiz A, Dalmau J, Butler MH et al. Anti-amphiphysin I antibodies in patients with paraneoplastic neurological disorders associated with small cell lung carcinoma. J Neurol Neurosurg Psychiatry 1999;66: 214–217.
Uchuya M, Graus F, Vega F et al. Intravenous immunoglobulin treatment in paraneoplastic neurological syndromes with antineuronal autoantibodies. J Neurol Neurosurg Psychiatry 1996;60:388–392.
OhSJ, Dropcho EJ, Claussen GC. Anti-Hu-associated paraneoplastic sensory neuropathy responding to early aggressive immunotherapy: report of two cases and review of literature. Muscle Nerve 1997;20:1576–1582.
Wirtz PW, Wintzen AR, Verschuuren JJ. Lambert-Eaton myasthenic syndrome has a more progressive course in patients with lung cancer. Muscle Nerve 2005;32:226–229.
O’Neill JH, Murray NM, Newsom-Davis J. The Lambert-Eaton myasthenic syndrome. A review of 50 cases. Brain 1988;111:577–596.
Elrington GM, Murray NM, Spiro SG et al. Neurological paraneoplastic syndromes in patients with small cell lung cancer. A prospective survey of 150 patients. J Neurol Neurosurg Psychiatry 1991;54:764–767.
Wirtz PW, Nijnuis MG, Sotodeh M et al. The epidemiology of myasthenia gravis, Lambert-Eaton myasthenic syndrome and their associated tumours in the northern part of the province of South Holland. J Neurol 2003;250:698–701.
Hawley RJ, Cohen MH, Saini N, Armbrustmacher VW. The carcinomatous neuromyopathy of oat cell lung cancer. Ann Neurol 1980;7:65–72.
Wirtz PW, Willcox N, van der Slik AR et al. HLA and smoking in prediction and prognosis of small cell lung cancer in autoimmune Lambert-Eaton myasthenic syndrome. J Neuroimmunol 2005;159:230–237.
Wirtz PW, Lang B, Graus F et al. P/Q-type calcium channel antibodies, Lambert-Eaton myasthenic syndrome and survival in small cell lung cancer. J Neuroimmunol 2005;164:161–165.
Motomura M, Johnston I, Lang B et al. An improved diagnostic assay for Lambert-Eaton myasthenic syndrome. J Neurol Neurosurg Psychiatry 1995;58:85–87.
Rosenfeld MR, Wong E, Dalmau J et al. Cloning and characterization of a Lambert-Eaton myasthenic syndrome antigen. Ann Neurol 1993;33: 113–120.
Chalk CH, Murray NM, Newsom-Davis J et al. Response of the Lambert-Eaton myasthenic syndrome to treatment of associated small-cell lung carcinoma. Neurology 1990;40:1552–1556.
McEvoy KM, Windebank AJ, Daube JR et al. 3,4-Diaminopyridine in the treatment of Lambert-Eaton myasthenic syndrome. N Engl J Med 1989;321:1567–1571.
Sanders DB, Massey JM, Sanders LL et al. A randomized trial of 3,4-diaminopyridine in Lambert-Eaton myasthenic syndrome. Neurology 2000;54:603–607.
Newsom-Davis J, Murray NM. Plasma exchange and immunosuppressive drug treatment in the Lambert-Eaton myasthenic syndrome. Neurology 1984;34:480–485.
Bird SJ. Clinical and electrophysiologic improvement in Lambert-Eaton syndrome with intravenous immunoglobulin therapy. Neurology 1992;42:1422–1423.
Buchbinder R, Forbes A, Hall S et al. Incidence of malignant disease in biopsy-proven inflammatory myopathy. A population-based cohort study. Ann Intern Med 2001;134:1087–1095.
Hill CL, Zhang Y, Sigurgeirsson B et al. Frequency of specific cancer types in dermatomyositis and polymyositis: a population-based study. Lancet 2001;357:96–100.
Mastaglia FL, Garlepp MJ, Phillips BA et al. Inflammatory myopathies: clinical, diagnostic and therapeutic aspects. Muscle Nerve 2003;27: 407–425.
Targoff IN. Update on myositis-specific and myositis-associated autoantibodies. Curr Opin Rheumatol 2000;12:475–481.
Griggs RC, Mendell JR, Miller RG. Evaluation and Treatment of Myopathies. Philadelphia: F.A. Davis Company; 1995:1–464.(Janet W. de Beukelaar, Pe)
After completing this course, the reader will be able to:
Describe the autoimmune pathogenesis of paraneoplastic neurological syndromes.
Explain the clinical value of paraneoplastic antibody detection.
Describe the general treatment approach to paraneoplastic neurological syndromes.
ABSTRACT
Paraneoplastic neurological syndromes (PNS) are remote effects of cancer that are not caused by invasion of the tumor or its metastases. Immunologic factors appear important in the pathogenesis of PNS because antineuronal autoantibodies and T-cell responses against nervous system antigens have been defined for many of these disorders. The immunologic response is elicited by the ectopic expression of neuronal antigens by the tumor. Expression of these so-called "onconeural" antigens is limited to the tumor and the nervous system and sometimes also the testis. At the time of presentation of the neurological symptoms, most patients have not yet been diagnosed with cancer. Detection of paraneoplastic antibodies is extremely helpful in diagnosing an otherwise unexplained and often rapidly progressive neurological syndrome as paraneoplastic. In addition, the paraneoplastic antibodies may also direct the search for an underlying neoplasm. On the other hand, in patients known to have cancer, the presentation of a PNS may herald recurrence of the tumor or a second tumor. The number of paraneoplastic antibodies is still growing, and at least seven of these can now be considered well characterized. Based on the clinical syndrome, the type of antibody, and the presence or absence of cancer, patients are classified as having a "definite" or "possible" PNS. Despite the presumed autoimmune etiology of PNS, the results of various forms of immunotherapy have been disappointing, with some exceptions. Rapid detection and immediate treatment of the underlying tumor appears to offer the best chance of stabilizing the patient and preventing further neurological deterioration.
INTRODUCTION
Paraneoplastic neurological syndromes (PNS) are remote effects of cancer that are, by definition, caused neither by invasion of the tumor or its metastases nor by infection, ischemia, metabolic and nutritional deficits, surgery, or other forms of tumor treatment [1]. Immunologic factors are believed to be important in the pathogenesis of PNS because antibodies and T-cell responses against nervous system antigens have been defined for many of these disorders [1].
Presumably, the immunologic response is elicited by the ectopic expression of neuronal antigens by the tumor. Expression of these "onconeural" antigens is limited to the tumor and the nervous system, and sometimes also the testis. At the time of presentation of the neurological symptoms, most patients have not yet been diagnosed with cancer [2–5]. Detection of paraneoplastic antibodies can help diagnose the neurological syndrome as paraneoplastic and may direct the search for an underlying neoplasm. Often, the oncologist or hematologist will be involved in the tumor workup. On the other hand, in patients known to have cancer, the presentation of a PNS may herald recurrence of the tumor or a second tumor. In these patients, however, metastatic complications of the known cancer must be ruled out first. Despite the presumed autoimmune etiology of PNS, the results of various forms of immunotherapy have been disappointing, with some exceptions [2–5]. Rapid detection and immediate treatment of the underlying tumor appears to offer the best chance of stabilizing the patient and preventing further neurological deterioration [2–5].
PATHOGENESIS
Pathological examination of the nervous system generally shows loss of neurons in affected areas of the nervous system with inflammatory infiltration by CD4+ T-helper cells and B cells in the perivascular spaces and cytotoxic CD8+ T cells in the interstitial spaces [6–8]. Examination of the cerebrospinal fluid (CSF) frequently demonstrates pleocytosis, intrathecal synthesis of IgG, and oligoclonal bands, supporting an inflammatory or immune-mediated etiology.
The discovery of paraneoplastic antineuronal autoantibodies resulted in the general belief that these are immune-mediated disorders triggered by aberrant expression of onconeural antigens in the tumor. Support for this hypothesis comes from the fact that the target paraneoplastic antigens are expressed both in the tumor and in the affected parts of the nervous system. Furthermore, the tumors are usually small and heavily infiltrated with inflammatory cells, and spontaneous remissions at the time of neurological presentation have been described [9, 10]. These findings suggest that some PNS without an identifiable tumor may result from immune-mediated eradication of the tumor [9, 10]. In keeping with this hypothesis, one study found more limited disease distribution and better oncologic outcome in small cell lung cancer (SCLC) patients with paraneoplastic autoantibodies [11].
Although the paraneoplastic antibodies are synthesized intrathecally, a pathogenic role could only be proven for those paraneoplastic autoantibodies that are directed against easily accessible antigens located at the cell surface. Examples of such antigens are the acetylcholine receptor (anti-AChR muscle type in myasthenia gravis and neuronal ganglionic type in autonomic neuropathy), P/Q-type voltage-gated calcium channels (anti-VGCC in Lambert-Eaton myasthenic syndrome [LEMS]), voltage-gated potassium channels (anti-VGKC in neuromyotonia), and the metabotropic glutamate receptor mGluR1 (anti-mGluR1 in paraneoplastic cerebellar degeneration [PCD]). Most paraneoplastic antigens are located in the cytoplasm (e.g., the Yo antigen) or nucleus (e.g., the Hu and Ri antigens), and a pathogenic role for the respective antibodies has not be demonstrated [12]. In these disorders, indirect lines of evidence support the view that the cellular immune response against these antigens is responsible for the neurological damage [13–15]. The relative contribution of the cellular and humoral immunity to the clinical and pathological manifestations has not been resolved [13–15]. The paraneoplastic antibodies may, in these cases, be surrogate markers for T-lymphocyte activation [16].
A totally different mechanism seems at work in PCD in Hodgkin’s lymphoma because the target antigens of the associated anti-Tr and anti-mGluR1 autoantibodies are not expressed in Hodgkin’s tumor tissue [17]. Dysregulation of the immune system in Hodgkin’s lymphoma and an etiologic role for (viral?) infections have been postulated in this disorder.
INCIDENCE
The incidence of PNS varies with the neurological syndrome and with the tumor. Approximately 10% of patients with plasma cell disorders accompanied by malignant monoclonal gammopathies are affected by a paraneoplastic peripheral neuropathy. More than half of the patients with the rare osteosclerotic form of myeloma develop a severe predominantly motor paraneoplastic peripheral neuropathy. In other hematological malignancies, the incidence of PNS is very low, with the exception of Hodgkin’s disease. However, the incidence of PNS even in Hodgkin’s disease is well below 1%. In solid tumors, the more common neurological syndromes are myasthenia gravis, which occurs in 15% of patients with a thymoma, and LEMS, which affects 3% of patients with SCLC. For other solid tumors, the incidence of PNS is <1%.
DIAGNOSIS
Clinical syndromes are never pathognomonic for a paraneoplastic etiology, and a high index of clinical suspicion is important. Symptoms can be atypical, psychiatric, or even fluctuating, and PNS should often be in the differential diagnosis of otherwise unexplained neurological syndromes. Some neurological syndromes, such as limbic encephalitis and subacute cerebellar degeneration, are associated relatively often with cancer. These are called "classical" PNS and are in italics in Table 1 [18]. Other syndromes, such as sensorimotor polyneuropathy, are much more prevalent, and their association with cancer may be by chance. Detection of a "well-characterized" paraneoplastic antibody is extremely helpful because it proves the paraneoplastic etiology of the neurological syndrome. The paraneoplastic antibodies are generally divided into three categories (Table 2) [18]. The well-characterized antibodies are reactive with molecularly defined onconeural antigens. These antibodies are strongly associated with cancer and have been detected unambiguously by several laboratories in a reasonable number of patients with well-defined neurological syndromes [18]. The partially characterized antibodies are those with an unidentified target antigen and those that have either been described by a single group of investigators or have been reported in only a few patients. The third group consists of antibodies that are associated with specific disorders but do not differentiate between paraneoplastic and nonparaneoplastic cases.
Because different antibodies can be associated with the same clinical findings [4], and the same antibody can be associated with different clinical syndromes [2, 3], paraneoplastic antibodies should be searched for by screening rather than by focusing on a specific antibody. Recently, Pittock et al. [19] demonstrated, in a large prospective series, that approximately 30% of patients have more than one paraneoplastic antibody. The combination of paraneoplastic antibodies provides important additional information to narrow the search for an underlying malignancy [19].
In the absence of paraneoplastic antibodies, additional diagnostic tests may be helpful in some PNS, although these are never specific for a paraneoplastic etiology. Magnetic resonance imaging (MRI) can help diagnose limbic encephalitis and may demonstrate cerebellar atrophy several months after the onset of PCD. Examination of the CSF is generally not required for detection of paraneoplastic antibodies because these can almost always be detected in serum as well. CSF examination may, however, show signs of inflammation, such as elevated white cell counts, oligoclonal bands, and intrathecal synthesis of IgG, indicating an immune-mediated or inflammatory etiology. In patients known to have cancer, MRI and CSF cytology are important in ruling out leptomeningeal metastases. Some PNS of the peripheral nervous system, such as LEMS, myasthenia gravis, and neuromyotonia, are accompanied by characteristic electrophysiological changes. These findings, however, are also present in the absence of an underlying tumor. Determining the precise type of neurological syndrome may assist in the search for an underlying tumor, such as SCLC in LEMS and thymoma in myasthenia gravis.
Once a paraneoplastic diagnosis has been established or is suspected, rapid identification of the tumor becomes essential but may be difficult because most PNS develop in the early stages of cancer. The workup generally starts with a detailed history, including smoking habits, weight loss, night sweats, and fever. A thorough physical examination should include palpation for pathological lymph nodes, rectal and pelvic examination, and palpation of breasts and testis. Often, the tumor is detected by high-resolution computed tomography (CT) of the chest, abdomen, and pelvis. If the CT scan remains negative, whole-body fluorodeoxyglucose positron emission tomography (FDG-PET) or PET/CT is recommended to detect an occult tumor or its metastases [20–22]. In addition, the type of antibody and PNS may suggest a specific underlying tumor and indicate further diagnostic tests, such as mammography (may be replaced by MRI) or ultrasound of the testes or pelvis (Table 2). When all tests remain negative, repeat evaluation at 3- to 6-month intervals for 2–3 years is recommended.
Diagnosing a neurological syndrome as paraneoplastic requires the exclusion of other possible causes by a reasonably complete workup. Because of the difficulties in diagnosis, an international panel of neurologists has established diagnostic criteria that divide patients with a suspected PNS into "definite" and "probable" categories. These criteria are based on the presence or absence of cancer, the presence of well-characterized antibodies, and the type of clinical syndrome. Patients with a definite PNS include those with [18]:
A classical syndrome (i.e., encephalomyelitis, limbic encephalitis, subacute cerebellar degeneration, opsoclonus-myoclonus, subacute sensory neuronopathy, chronic gastrointestinal pseudo-obstruction, LEMS, or dermatomyositis) and cancer that develops within 5 years of the diagnosis of the neurological disorder, regardless of the presence of paraneoplastic antibodies.
A nonclassical syndrome that objectively improves or resolves after cancer treatment, provided that the syndrome is not susceptible to spontaneous remission.
A nonclassical syndrome with paraneoplastic antibodies (well characterized or not) and cancer that develops within 5 years of the diagnosis of the neurological disorder.
A neurological syndrome (classical or not) with well-characterized paraneoplastic antibodies (i.e., anti-Hu, anti-Yo, anti-Ri, antiamphiphysin, anti-CV2, or anti-Ma2).
Patients with a possible PNS include those with [18]:
A classical syndrome without paraneoplastic antibodies and no cancer but at high risk to have an underlying tumor (e.g., smoking habit).
A neurological syndrome (classical or not) without cancer but with partially characterized paraneoplastic antibodies.
A nonclassical neurological syndrome, no paraneoplastic antibodies, and cancer that presents within 2 years of the neurological syndrome.
TREATMENT AND PROGNOSIS
Despite the immunological etiology of most of the PNS, the results of immunotherapy have been disappointing [23]. Exceptions are the neurological syndromes associated with paraneoplastic antibodies that are directed against antigens that are located at the surface of the cell (i.e., antigens that are accessible to circulating antibodies). These include not only disorders of the peripheral nervous system (LEMS, myasthenia gravis, and neuromyotonia) but also anti-mGluR1-associated PCD and antiamphiphysin-associated stiff-person syndrome [24, 25]. Immunotherapy modalities that are recommended for these disorders include plasma exchange, immunoadsorption (extraction of patient IgG over a protein A column), steroids, and i.v. Ig.
For most PNS, when the antigen is cytoplasmic or nuclear, the nervous dysfunction is probably not caused by functional interference of antibodies with the target antigen. In disorders with intracellular target antigens and a strong cellular immune reaction, plasma exchange and immunoadsorption are not expected to give much benefit. In these cases, a trial of a treatment that modulates the activation and function of effector T cells makes more sense, but to date there is only limited evidence that steroids, cyclophosphamide, i.v. Ig, or other immunosuppressive therapies are effective [26].
Hence, the first goal of treatment for PNS is control of the tumor. In addition, antitumor therapy has been demonstrated to stop the paraneoplastic neurological deterioration and leave the patients, on average, in better condition [27]. In severely debilitated patients, for example, the elderly and bedridden, treatment of the underlying tumor is often withheld because of the very small chance of clinically relevant neurological improvement.
Table 3 provides a summary of treatment of PNS and the effect on neurological outcome.
CLINICAL SYNDROMES
The classical PNS are described below. Descriptions of the nonclassical syndromes, which usually are not paraneoplastic but may occur in association with cancer, can be found elsewhere [28].
Encephalomyelitis Paraneoplastic encephalomyelitis is characterized by involvement of several areas of the nervous system, including the temporal lobes and limbic system (limbic encephalitis), brainstem (brainstem encephalitis), cerebellum (subacute cerebellar degeneration), spinal cord (myelitis) dorsal root ganglia (subacute sensory neuronopathy), and autonomous nervous system (autonomic neuropathy) [29, 30]. Patients with predominant involvement of one area but clinical evidence of only mild involvement of other areas are usually classified according to the predominant clinical syndrome. Symptoms of limbic encephalitis, subacute cerebellar degeneration, subacute sensory neuronopathy, and autonomic neuropathy are described below. Symptoms of brainstem encephalitis can include diplopia, dysarthria, dysphagia, gaze abnormalities (nuclear, internuclear, or supranuclear), facial numbness, and subacute hearing loss.
Underlying Tumor
Although virtually all cancer types have been associated with paraneoplastic encephalomyelitis, the majority of patients have underlying SCLC [2, 3, 29–31]. Most patients are not known to have cancer when the neurological symptoms present, and the SCLC may be difficult to demonstrate because of its small size. When anti-Hu antibodies are detected or when the patient is at risk for lung cancer (smoking, age >50 years), a careful and repeated search for underlying SCLC is warranted. When the CT scan is negative, a total-body FDG-PET scan or FDG-PET/CT scan may detect the neoplasm [20,21]. When a tumor other than SCLC is detected in a patient with anti-Hu antibodies, it may unexpectedly express the Hu antigen [2] or may be an unrelated secondary neoplasm [31]. When tumor tissue is available for analysis and expresses the Hu antigen, a further workup for a second tumor (SCLC) can probably be safely deferred [2].
Diagnostic Evaluation
MRI or CT of the brain is normal or shows specific changes in most paraneoplastic encephalomyelitis patients with two exceptions [30]. In 65%–80% of patients with predominant limbic encephalitis, MRI and CT show temporal lobe abnormalities [32,33]. Patients with a predominant cerebellar syndrome will develop cerebellar atrophy in the chronic stage. CSF is abnormal in most patients, showing elevated protein, mild mononuclear pleocytosis, elevated IgG index, or oligoclonal bands [30].
Antineuronal Antibodies
Patients with paraneoplastic encephalomyelitis and SCLC often have anti-Hu antibodies (also called antineuronal nuclear autoantibodies [ANNA-1]) in their serum and CSF [2, 3, 30, 31]. Other antibodies associated with paraneoplastic encephalomyelitis include anti-CRMP5/CV2 [16], anti-amphiphysin [34], and the less well-characterized ANNA-3 [35] and Purkinje cell antibody, PCA-2 [36].
Treatment and Prognosis
Tumor treatment offers the best chance of stabilizing the patient’s neurological condition, while immunotherapy does not appear to modify the outcome of paraneoplastic encephalomyelitis [2, 3, 23]. Therefore, all efforts should be directed at early diagnosis of paraneoplastic encephalomyelitis and rapid identification and treatment of the tumor. Because of incidental reports of neurological improvement following various forms of immunosuppressive treatment, a trial of one or two immunosuppressive modalities may be warranted in a single patient. However, spontaneous neurological improvement has rarely been described [10]. The overall functional outcome is bad, and more than 50% of patients are confined to bed or chair in the chronic phase of the disease [2, 3, 23]. The median survival time of patients is approximately 1 year from diagnosis [2, 3]. Mortality is predicted by worse functional status at diagnosis, age >60 years, involvement of more areas of the nervous system, and absence of treatment [2].
Because of the limited efficacy of plasma exchange, i.v. Ig, and corticosteroids [2, 3, 23] and the presumed role of cellular immunity, more aggressive immunosuppression with cyclophosphamide, tacrolimus, or cyclosporine may be considered. To limit toxicity, these more aggressive immunosuppressive approaches should probably be reserved for patients who are not receiving chemotherapy.
Limbic Encephalitis
Paraneoplastic limbic encephalitis (PLE) is a rare disorder characterized by the subacute onset (in days to a few months) of short-term memory loss, seizures, confusion, and psychiatric symptoms suggesting involvement of the limbic system [29, 37]. Hypothalamic dysfunction may occur with somnolence, hyperthermia, and endocrine abnormalities. Selective impairment of recent memory is a hallmark of the disease but may not be evident in patients presenting with severe confusion or multiple seizures [33]. More than half of the patients presenting with limbic encephalitis may have an underlying neoplasm [33]. Clinically three groups of patients with PLE can be identified [33]. The first group consists of patients with anti-Hu antibodies and lung cancer (usually SCLC). The limbic encephalitis is part of paraneoplastic encephalomyelitis, and the patients have involvement of other areas outside the limbic system and brainstem. These patients are older (median age, 62 years), usually smoke, and are more often female [33, 38]. The second group consists of young males with testicular cancer and anti-Ma2 antibodies [39]. The median age is 34 years. Symptoms are usually confined to the limbic system, hypothalamus, and brainstem. The third group has no antineuronal antibodies (approximately 40% of patients with PLE) [32, 33]). In these patients, the symptoms are more often confined to the limbic system, the median age is around 57 years, and the associated tumor is often located in the lung [33, 38].
Underlying Tumor
The associated tumor is a lung tumor in 50%–60% of patients, usually SCLC (40%–55%), and the associated tumor is a testicular germ cell tumor in 20% of patients [32, 33, 38]. Other tumors include breast cancer, thymoma, Hodgkin’s disease, and immature teratomas [32, 33].
Diagnostic Evaluation
The diagnosis is often difficult because there are no specific clinical markers and symptoms usually precede the diagnosis of cancer [33]. MRI and CT scans are abnormal in 65%–80% of patients [32, 33]. Abnormalities consist of increased signal on T2-weighted and fluid-attenuated inversion-recovery images of one or both medial temporal lobes, hypothalamus, and brainstem. Early in the course of the disease, the MRI scan may be normal, and repeat imaging may be indicated. Co-registration of FDG-PET may further improve the sensitivity of imaging [40]. CSF examination is abnormal in 80% of patients, showing transient mild lymphocytic pleocytosis with elevated protein, IgG, or oligoclonal bands [32, 33]. Detection of paraneoplastic antibodies helps establish the diagnosis and direct a tumor search that should include the lung, breasts, and testicles in the absence of paraneoplastic antibodies.
Antineuronal Antibodies
Antineuronal antibodies are found in about 60% of patients with PLE. The most frequent related paraneoplastic antibodies are: anti-Hu, anti-Ma2 (with or without Ma1), anti-CV2/CRMP5, and antiamphiphysin [16, 33, 41]. The majority of patients with anti-Hu antibodies have symptoms that suggest dysfunction of areas of the nervous system outside the limbic system. The related tumor in these patients is usually SCLC. Patients with only anti-Ma2 antibodies (also called anti-Ta) are young males with testicular cancer. Patients with anti-Ma2 and anti-Ma1 antibodies are significantly older and are more often female [41]. Anti-Ma1 patients are more likely to develop cerebellar dysfunction and usually harbor tumors other than testicular cancer. Anti-CV2/CRMP5 antibodies are detected in patients with SCLC or thymoma [16]. Anti-VGKC antibodies can be associated with PLE and thymoma or with non–paraneoplastic limbic encephalitis [42, 43].
Treatment and Prognosis
Spontaneous complete recovery has been described, although very rarely [38, 44]. Immunotherapy is largely ineffective [33], but several cases benefiting from antitumor treatment have been reported [33, 38, 45]. Therefore, all efforts should be directed at identifying and treating the underlying tumor. If no tumor is found, the search should be repeated every 3–6 months for a total of 2–3years. Irrespective of treatment, partial neurological recovery was seen in 38% of patients with anti-Hu antibodies, 30% of patients with anti-Ta (anti-Ma2) antibodies, and 64% of patients without antibodies [33].
Subacute Cerebellar Degeneration
PCD is one of the most common and characteristic PNS [4, 29]. In a study of 137 consecutive patients with antibody-associated PNS, 50 (37%) presented with subacute cerebellar degeneration [4]. PCD usually starts acutely with nausea, vomiting, dizziness, and slight incoordination of walking, evolving rapidly over weeks to a few months with progressive ataxia of gait, limbs, and trunk, dysarthria, and often nystagmus associated with oscillopsia. The disease reaches its peak within months and then stabilizes. By this time, most patients are severely debilitated. They are generally unable to walk without support, they may be unable to sit unsupported, handwriting is often impossible, and feeding themselves has become difficult. The neurological signs are always bilateral but may be asymmetrical. Diplopia is common at presentation, although the investigator usually cannot detect abnormalities of ocular movement. The symptoms and signs are limited to the cerebellum and cerebellar pathways, but other mild neurological abnormalities may be found on careful examination. These include hearing loss, dysphagia, pyramidal and extrapyramidal tract signs, mental status change, and peripheral neuropathy [5, 46, 47].
Underlying Tumor
PCD can be associated with any cancer, but the most common tumors are lung cancer (usually SCLC), ovarian cancer, and lymphomas (particularly Hodgkin’s lymphoma). In 60%–70% of patients, neurological symptoms precede diagnosis of the cancer by a few months to 2–3 years and lead to its detection [4, 5, 17].
Diagnostic Evaluation
Subacute cerebellar degeneration is a rare disorder in cancer patients. On the other hand, 50% of patients presenting with acute or subacute nonfamilial ataxia are estimated to have an underlying malignancy [29]. MRI and CT scans are initially normal but often reveal cerebellar atrophy later in the course of the disease. CSF examination shows mild lymphocytic pleocytosis with elevated protein and IgG levels in the first weeks to months. Oligoclonal bands may be present. The diagnosis of PCD is established by demonstration of specific antineuronal antibodies. The type of antibody directs the search for an underlying neoplasm (Table 2).
Antineuronal Antibodies
PCD can be associated with various antineuronal autoantibodies. The clinical and tumor specificities of each of the antibodies are summarized in Table 2.
Anti-Yo (also called PCA-1), anti-Tr (PCA-Tr), and anti-mGluR1 antibodies are associated with relatively "pure" cerebellar syndromes. Anti-Yo antibodies are associated with breast cancer and tumors of the ovaries, endometrium, and fallopian tubes [4, 5, 48]. These antibodies are directed against the cerebellar degeneration–related (CDR) proteins that are expressed by Purkinje cells and the associated tumors [48, 49]. CDR-2–specific cytotoxic T cells have been identified in the serum from patients with PCD, suggesting a pathogenic role for the cellular immune response in this PNS [14]. Anti-Tr (PCA-Tr) antibodies are directed against an unidentified cytoplasmic Purkinje cell antigen and appear specific for Hodgkin’s disease [17]. Anti-mGluR1 antibodies have been found in two patients with PCD and Hodgkin’s disease. Passive transfer of patient anti-mGluR1 IgG into CSF of mice induced severe, transient ataxia [25].
Approximately50%ofpatientswithcerebellardegeneration and underlying SCLC have high titers of anti-Hu antibodies [50]. The remaining patients are likely to have anti-P/Q-type VGCC antibodies. These antibodies were present in all patients who also had LEMS and in some patients with cerebellar degeneration without LEMS. In patients with antiamphiphysin or anti-CV2/CRMP5 antibodies, the cerebellar degeneration is often part of the paraneoplastic encephalomyelitis syndrome, and more widespread neurological symptoms and signs are usually found.
The more recently discovered PCA-2 antibody and the ANNA-3 antibody are associated with lung cancer and a variety of neurological syndromes including cerebellar degeneration [36]. Anti-Zic4 antibodies are strongly associated with SCLC, and most patients have paraneoplastic encephalomyelitis, often presenting with cerebellar dysfunction [51]. These patients often have concurrent anti-Hu or anti-CV2/CRMP5 antibodies. Patients with isolated antiZic4 antibodies are more likely to develop cerebellar symptoms.
Treatment and Prognosis
The outcome of PCD is generally poor, and the best chance to at least stabilize the syndrome is to treat the underlying tumor [4]. Incidental improvement has been reported either spontaneously or in association with plasma exchange, steroids, i.v. Ig, or rituximab [52]. In patients with anti-Yo–associated cerebellar degeneration, the prognosis is better for patients with breast cancer than for those with gynecologic cancer [5]. The prognosis is better in patients with PCD associated with Hodgkin’s disease and anti-Tr (PCA-Tr) or anti-mGluR1 antibodies. With successful treatment of the tumor and/or immunotherapy, symptoms may disappear and the antibodies vanish [17, 25].
Opsoclonus-Myoclonus
Opsoclonus is a disorder of ocular motility that consists of involuntary, arrhythmic, high-amplitude conjugate saccades in all directions. Opsoclonus may occur intermittently or, if more severe, constantly, and it does not remit in the darkness or when the eyes are closed. Opsoclonus is often associated with diffuse or focal myoclonus, the "dancing eyes and dancing feet syndrome," and other cerebellar and brainstem signs [28, 53, 54]. An excessive startle response reminiscent of hyperekplexia may also occur in opsoclonus-myoclonus patients [55]. In contrast to most PNS, the course of opsoclonus-myoclonus may be remitting and relapsing [54].
Underlying Tumor
Approximately 20% of adult patients with opsoclonus-myoclonus have a previously undiscovered malignancy [53]. The most commonly associated neoplasms are SCLC and breast and gynecologic cancers [55, 56]. Many other tumors, including thyroid and bladder cancer, have also been reported [57].
Almost 50% of children with opsoclonus-myoclonus have an underlying neuroblastoma. Conversely, approximately 2%–3% of children with neuroblastoma have paraneoplastic opsoclonus-myoclonus [58, 59]. Tumors in children with paraneoplastic opsoclonus-myoclonus apparently have a better prognosis than tumors in patients without this PNS.
Diagnostic Evaluation
MRI scans are usually normal but may show hyperintensities in the brainstem on T2-weighted images [60]. Examination of the CSF may show mild pleocytosis and protein elevation. In some patients, paraneoplastic opsoclonus-myoclonus resembles PCD. The prominent opsoclonus and truncal, rather than appendicular, ataxia distinguish this syndrome from anti-Yo– and anti-Hu–associated PCD [28]. Adult patients with paraneoplastic opsoclonus-myoclonus are older (median age, 66 years) than patients with the idiopathic syndrome (median age, 40 years). In adult patients, the tumor search should be directed at the most common underlying tumors, that is, high-resolution CT of the chest and abdomen, and gynecological examination and mammography (or MRI of the breasts) [56]. When this is negative, FDG-PET should be considered [22, 61].
In children, nonparaneoplastic opsoclonus–myoclonus occurs as a self-limited illness and is probably the result of a viral infection of the brainstem. The search for an occult neuroblastoma should include imaging of the chest and abdomen (CT or MRI scan), urine catecholamine measurements, and metaiodobenzylguanidine scan [62]. When negative, the evaluation should be repeated after several months [63].
Antineuronal Antibodies
Specific antibodies are found in only a minority of patients with paraneoplastic opsoclonus-myoclonus [56]. In women, anti-Ri antibodies (or ANNA-2) are mostly associated with breast and gynecologic tumors. Anti-Ri has occasionally been found in bladder cancer and SCLC and may then occur in male patients [28, 57]. Anti-Ri antibodies are directed against the Nova proteins [64, 65]. Paraneoplastic opsoclonus-myoclonus can also be associated with anti-Hu antibodies, usually as part of a more widespread paraneoplastic encephalomyelitis. Bataller et al. [66] screened a brainstem cDNA library with sera from 21 patients with (paraneoplastic) opsoclonus-myoclonus. Twenty-five proteins were identified, recognized by one or two sera each, demonstrating that immunity to neuronal autoantigens in opsoclonus-myoclonus is both frequent and heterogeneous.
In children presenting with opsoclonus-myoclonus, the detection of anti-Hu antibodies is diagnostic of an underlying neuroblastoma [67]. The frequency of anti-Hu antibodies in neuroblastoma with paraneoplastic opsoclonus-myoclonus is approximately 10% [67–69]. This finding differs little from the 4%–15% of anti-Hu positive sera in children with neuroblastoma who do not have opsoclonus-myoclonus [67, 68].
Treatment and Prognosis
In contrast to most of the other PNS, paraneoplastic opsoclonus-myoclonus may remit either spontaneously, following treatment of the tumor, or in association with clonazepam or thiamine treatment. Most patients with idiopathic opsoclonus-myoclonus make a good recovery that seems to be accelerated by steroids or i.v. Ig. Paraneoplastic opsoclonus-myoclonus usually has a more severe clinical course, and treatment with steroids or i.v. Ig appears ineffective. In a series of 14 patients with paraneoplastic opsoclonus-myoclonus, eight patients whose tumors were treated showed complete or partial neurological recovery. In contrast, five of the six patients whose tumors were not treated died of the neurological syndrome despite steroids, i.v. Ig, or plasma exchange [56]. However, improvement following the administration of steroids, cyclophosphamide, azathioprine, i.v. Ig, plasma exchange, or plasma filtration with a protein A column has been described in single cases [55, 70–72].
In children, paraneoplastic opsoclonus-myoclonus may improve following treatment with adrenocorticotropic hormone, prednisone, azathioprine, or i.v. Ig, but residual central nervous system signs are frequent [59, 63, 73]. Treatment of the tumor with chemotherapy is the most important predictor of good neurological recovery [74].
Subacute Sensory Neuronopathy
Subacute sensory neuronopathy is an uncommon disorder that is probably paraneoplastic in about 20% of patients [75, 76]. The symptoms begin with pain and paraesthesia. Clumsiness and unsteady gait then develop and usually become predominant. The distribution of symptoms is often asymmetrical or multifocal. The upper limbs are often affected first and are almost invariably involved with evolution. Sensory loss may also affect the face, chest, or abdomen. On examination, all sensory modalities are affected, but the most striking abnormality is loss of deep sensation causing sensory ataxia with pseudoathetosis of the hands. Tendon reflexes are depressed or absent. In most patients, the disease progresses rapidly over weeks to months, leaving the patient severely disabled. In a few patients, the neuronopathy remains stable for months with mild neurological deficits [77]. Subacute sensory neuronopathy occurs in approximately 75% of patients with paraneoplastic encephalomyelitis, is predominant in 50%, and is clinically pure in 25% [2, 3]. Autonomic neuropathy, including gastrointestinal pseudo-obstruction, is common.
Underlying Tumor
Subacute sensory neuronopathy is associated with lung cancer, usually SCLC, in 70%–80% of patients [2, 3, 31]. Other associated tumors include breast cancer, ovarian cancer, sarcoma, and Hodgkin’s lymphoma [75, 76]. Subacute sensory neuronopathy usually predates the diagnosis of cancer, with a median delay of 3.5–4.5 months [2, 3].
Diagnostic Evaluation
Electrophysiologically, the hallmark of subacute sensory neuronopathy is the absence of, or marked reduction in, sensory nerve action potentials. Motor conduction velocities may be mildly reduced. Early in the course of the disease, CSF examination shows mild pleocytosis, with an elevated IgG level and oligoclonal bands [3, 75, 76]. Sural nerve biopsy is rarely required for the diagnosis but may differentiate this disorder from vasculitic neuropathy.
Antineuronal Antibodies
Anti-Hu is the most frequent paraneoplastic antibody in subacute sensory neuronopathy [2, 3, 30, 31]. In this setting, anti-Hu antibody detection has a specificity of 99% and sensitivity of 82% [78]. The absence of anti-Hu antibodies does not rule out an underlying cancer. Anti-CRMP5/CV2 antibodies also occur with paraneoplastic peripheral neuropathies [79]. These patients usually have a sensory or sensorimotor neuropathy, with less frequent involvement of the arms but often associated with cerebellar ataxia [16, 79, 80]. Anti-CRMP5/CV2 antibodies are usually associated with SCLC, neuroendocrine tumors, and thymoma. Antiamphiphysin antibodies are associated with multifocal paraneoplastic encephalomyelitis, and symptoms often include sensory or sensorimotor neuropathy [34, 81, 82]. Associated tumors (mostly limited) are mainly SCLC, breast cancer, and melanoma.
Treatment and Prognosis
Immunotherapy consisting of plasma exchange, steroids, and i.v. Ig is ineffective in most cases [27, 83]. There may be some exceptions to this rule [23, 84]. In one study, 2 of 10 patients stabilized in relatively good clinical condition following intensive treatment with a combination of steroids, cyclophosphamide, and i.v. Ig [23]. Early detection and treatment of the underlying neoplasm, usually SCLC, appears to offer the best chance of stabilizing the neurological symptoms [3, 27]. In patients with an identifiable tumor, antitumor treatment is recommended. In the absence of a tumor, antitumor treatment may be considered in patients with anti-Hu antibodies, age >50 years, and with a history of smoking. In patients not receiving antitumor therapy, a short course of immunotherapy can be considered.
Symptomatic treatment is directed at neuropathic pain and dysautonomic symptoms such as orthostatic hypotension.
LEMS
Patients with LEMS present with proximal weakness of the lower extremities and fatigability. Bulbar symptoms may occur more frequently than previously reported [85] but are generally milder than with myasthenia gravis. Respiratory weakness can occur. Deep tendon reflexes, especially those in the legs, are diminished or absent but may reappear after exercise. Autonomic features, especially dryness of the mouth, impotence, and mild/moderate ptosis, ultimately develop in 95% of patients [85–87]. In some patients, LEMS may develop in association with other PNS, including PCD and encephalomyelitis [50].
Underlying Tumor
Approximately 70% of patients have cancer, almost always SCLC [86, 88]. Other tumors include small cell carcinomas of the prostate and cervix, lymphomas, and adenocarcinomas. The prevalence of LEMS in patients with SCLC is estimated to be around 3% [87, 89]. Clinically and serologically, the 30% without identifiable tumors are indistinguishable from the paraneoplastic LEMS patients, although LEMS may have a more progressive course in patients with SCLC [85]. In patients presenting with LEMS, a smoking history and absence of the HLA-B8 genotype strongly predict underlying SCLC [90]. Patients with SCLC and LEMS survive significantly longer than SCLC patients who do not have this PNS [91].
Diagnostic Evaluation
The typical pattern of electromyographic abnormalities is the hallmark of LEMS. This includes a low compound muscle action potential at rest with a decreased response at low rates of repetitive stimulation (3 Hz) and an incremental response at high rates of repetitive stimulation (50 Hz) or 15–30 seconds of maximal voluntary contraction [92].
Antineuronal Antibodies
Most patients with LEMS have antibodies against P/Q-type calcium channels that are located presynaptically in the neuromuscular junction [92]. About 20% have anti-MysB antibodies reactive with the ß subunit of neuronal calcium channels [93].
Treatment and Prognosis
Treatment of LEMS must be tailored to the individual based on severity of the symptoms, underlying disease, life expectancy, and previous response to treatment. In patients with paraneoplastic LEMS, treatment of the tumor frequently leads to neurological improvement [94]. Symptomatic treatment is with drugs that facilitate the release of acetylcholine from motor nerve terminals, such as 3,4-diaminopyridine (DAP) [95]. In a placebo-controlled randomized trial, DAP (5–20 mg three to four times daily) was effective for long-term treatment, alone or in combination with other treatments [96]. The maximum recommended daily dose of DAP is 80 mg; at higher doses, seizures occur [96]. Cholinesterase inhibitors (pyridostigmine, 30–60 mg, every 6 hours) may improve dryness of the mouth but rarely relieve weakness. If these treatments are not effective enough, it must be decided if immunosuppressive therapy with steroids, azathioprine, or cyclosporine is appropriate. Removal of the pathogenic anti-P/Q-type calcium channel antibodies by plasma exchange [97] and i.v. Ig can give quick but transient relief [86, 98]. LEMS responds less favorably to immunotherapy than myasthenia gravis.
Dermatomyositis
In dermatomyositis, the characteristic heliotrope rash (purplish discoloration of the eyelids) often precedes the appearance of proximal muscle weakness. Other manifestations include arthralgia, myocarditis and congestive heart failure, and interstitial lung disease. Clinical, electromyographical, and pathological findings of dermatomyositis are similar in patients with and without cancer.
Underlying Tumor
The standardized incidence ratio for a malignant disease in dermatomyositis is 6.2 (95% confidence interval, 3.9–10.0) [99]. Dermatomyositis is associated with cancer of the ovary, lung, pancreas, stomach, colorectum, and breast, and with non-Hodgkin’s lymphoma [100].
Diagnostic Evaluation
Most patients have elevated serum creatine kinase levels and electromyographic evidence of myopathy. Muscle imaging (CT or MRI) may help in confirming the diagnosis and determining the type of inflammatory myopathy and in selecting an appropriate biopsy site. Muscle or skin biopsy is the definitive diagnostic procedure and shows inflammatory infiltrates [101].
Antineuronal Antibodies
Antibodies to the Mi-2 protein complex are specific for dermatomyositis and are present in high titers in about 35% of cases [102].
Treatment and Prognosis
Treatment of paraneoplastic dermatomyositis is generally the same as for patients without a tumor. Nearly all patients respond to corticosteroids [103]. Refractory patients and patients requiring a lower dose of steroids can be treated with azathioprine. Methotrexate and cyclophosphamide may also be considered [103].
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.
REFERENCES
Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med 2003;349:1543–1554.
Graus F, Keime-Guibert F, Rene R et al. Anti-Hu-associated paraneoplastic encephalomyelitis: analysis of 200 patients. Brain 2001;124: 1138–1148.
Sillevis Smitt P, Grefkens J, De Leeuw B et al. Survival and outcome in 73 anti-Hu positive patients with paraneoplastic encephalomyelitis/sensory neuronopathy. J Neurol 2002;249:745–753.
Shams’ili S, Grefkens J, de Leeuw B et al. Paraneoplastic cerebellar degeneration associated with antineuronal antibodies: analysis of 50 patients. Brain 2003;126:1409–1418.
Rojas I, Graus F, Keime-Guibert F et al. Long-term clinical outcome of paraneoplastic cerebellar degeneration and anti-Yo antibodies. Neurology 2000;55:713–715.
Denny-Brown D. Primary sensory neuropathy with muscular changes associated with carcinoma. J Neurol Neurosurg Psychiatry 1948;11: 73–87.
Jean WC, Dalmau J, Ho A et al. Analysis of the IgG subclass distribution and inflammatory infiltrates in patients with anti-Hu-associated paraneoplastic encephalomyelitis. Neurology 1994;44:140–147.
Bernal F, Graus F, Pifarre A et al. Immunohistochemical analysis of anti-Hu-associated paraneoplastic encephalomyelitis. Acta Neuropathol (Berl) 2002;103:509–15.
Darnell RB, DeAngelis LM. Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet 1993;341: 21–22.
Byrne T, Mason WP, Posner JB et al. Spontaneous neurological improvement in anti-Hu associated encephalomyelitis. J Neurol Neurosurg Psychiatry 1997;62:276–278.
Graus F, Dalmou J, Rene R et al. Anti-Hu antibodies in patients with small-cell lung cancer: association with complete response to therapy and improved survival. J Clin Oncol 1997;15:2866–2872.
Sillevis Smitt PA, Manley GT, Posner JB. Immunization with the paraneoplastic encephalomyelitis antigen HuD does not cause neurologic disease in mice. Neurology 1995;45:1873–1878.
Benyahia B, Liblau R, Merle-Beral H et al. Cell-mediated autoimmunity in paraneoplastic neurological syndromes with anti-Hu antibodies. Ann Neurol 1999;45:162–167.
Albert ML, Austin LM, Darnell RB. Detection and treatment of activated T cells in the cerebrospinal fluid of patients with paraneoplastic cerebellar degeneration. Ann Neurol 2000;47:9–17.
Tanaka M, Tanaka K, Tokiguchi S et al. Cytotoxic T cells against a peptide of Yo protein in patients with paraneoplastic cerebellar degeneration and anti-Yo antibody. J Neurol Sci 1999;168:28–31.
Yu Z, Kryzer TJ, Griesmann GE et al. CRMP-5 neuronal autoantibody: marker of lung cancer and thymoma-related autoimmunity. Ann Neurol 2001;49:146–154.
Bernal F, Shams’ili S, Rojas I et al. Anti-Tr antibodies as markers of paraneoplastic cerebellar degeneration and Hodgkin’s disease. Neurology 2003;60:230–234.
Graus F, Delattre JY, Antoine JC et al. Recommended diagnostic criteria for paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry 2004;75:1135–1140.
Pittock SJ, Kryzer TJ, Lennon VA. Paraneoplastic antibodies coexist and predict cancer, not neurological syndrome. Ann Neurol 2004;56:715–719.
Antoine JC, Cinotti L, Tilikete C et al. [18F]fluorodeoxyglucose positron emission tomography in the diagnosis of cancer in patients with paraneoplastic neurological syndrome and anti-Hu antibodies. Ann Neurol 2000;48:105–108.
Rees JH, Hain SF, Johnson MR et al. The role of [18F]fluoro-2-deoxyglucose-PET scanning in the diagnosis of paraneoplastic neurological disorders. Brain 2001;124:2223–2231.
Younes-Mhenni S, Janier MF, Cinotti L et al. FDG-PET improves tumour detection in patients with paraneoplastic neurological syndromes. Brain 2004;127:2331–2338.
Keime-Guibert F, Graus F, Fleury A et al. Treatment of paraneoplastic neurological syndromes with antineuronal antibodies (Anti-Hu, anti-Yo) with a combination of immunoglobulins, cyclophosphamide, and methyl-prednisolone. J Neurol Neurosurg Psychiatry 2000;68:479–482.
Sommer C, Weishaupt A, Brinkhoff J et al. Paraneoplastic stiff-person syndrome: passive transfer to rats by means of IgG antibodies to amphiphysin. Lancet 2005;365:1406–1411.
Sillevis Smitt P, Kinoshita A, DeLeeuw B et al. Paraneoplastic cerebellar ataxia due to autoantibodies against a glutamate receptor. N Engl J Med 2000;342:21–27.
Vernino S, O’Neill BP, Marks RS et al. Immunomodulatory treatment trial for paraneoplastic neurological disorders. Neuro-oncol 2004;6:55–62.
Keime-Guibert F, Graus F, Broet P et al. Clinical outcome of patients with anti-Hu-associated encephalomyelitis after treatment of the tumor. Neurology 1999;53:1719–1723.
Posner JB. Paraneoplastic syndromes. In: Posner JB, ed., Neurologic Complications of Cancer. Philadelphia: FA Davis; 1995:353–385.
Henson RA, Urich H. Cancer and the Nervous System: The Neurologic Complications of Systemic Malignant Disease. Oxford: Blackwell Scientific, 1982.
Dalmau J, Graus F, Rosenblum MK et al. Anti-Hu–associated paraneoplastic encephalomyelitis/sensory neuronopathy. A clinical study of 71 patients. Medicine (Baltimore) 1992;71:59–72.
Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and oncologic profiles of patients seropositive for type 1 antineuronal nuclear autoantibodies. Neurology 1998;50:652–657.
Lawn ND, Westmoreland BF, Kiely MJ et al. Clinical, magnetic resonance imaging, and electroencephalographic findings in paraneoplastic limbic encephalitis. Mayo Clin Proc 2003;78:1363–1368.
Gultekin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain 2000;123:1481–1494.
Dropcho EJ. Antiamphiphysin antibodies with small-cell lung carcinoma and paraneoplastic encephalomyelitis. Ann Neurol 1996;39:659–667.
Chan KH, Vernino S, Lennon VA. ANNA-3 anti-neuronal nuclear antibody: marker of lung cancer-related autoimmunity. Ann Neurol 2001;50:301–311.
Vernino S, Lennon VA. New Purkinje cell antibody (PCA-2): marker of lung cancer-related neurological autoimmunity. Ann Neurol 2000;47:297–305.
Corsellis JA, Goldberg GJ, Norton AR. "Limbic encephalitis" and its association with carcinoma. Brain 1968;91:481–496.
Alamowitch S, Graus F, Uchuya M et al. Limbic encephalitis and small cell lung cancer. Clinical and immunological features. Brain 1997;120:923–928.
Voltz R, Gultekin SH, Rosenfeld MR et al. A serologic marker of paraneoplastic limbic and brain-stem encephalitis in patients with testicular cancer. N Engl J Med 1999;340:1788–1795.
Kassubek J, Juengling FD, Nitzsche EU et al. Limbic encephalitis investigated by 18FDG-PET and 3D MRI. J Neuroimaging 2001;11:55–59.
Dalmau J, Graus F, Villarejo A et al. Clinical analysis of anti-Ma2-associated encephalitis. Brain 2004;127:1831–1844.
Vincent A, Buckley C, Schott JM et al. Potassium channel antibody-associated encephalopathy: a potentially immunotherapy-responsive form of limbic encephalitis. Brain 2004;127:701–712.
Pozo-Rosich P, Clover L, Saiz A et al. Voltage-gated potassium channel antibodies in limbic encephalitis. Ann Neurol 2003;54:530–533.
Taylor RB, Mason W, Kong K et al. Reversible paraneoplastic encephalomyelitis associated with a benign ovarian teratoma. Can J Neurol Sci 1999;26:317–320.
Rosenfeld MR, Eichen JG, Wade DF et al. Molecular and clinical diversity in paraneoplastic immunity to Ma proteins. Ann Neurol 2001;50: 339–348.
Hammack J, Kotanides H, Rosenblum MK et al. Paraneoplastic cerebellar degeneration. II. Clinical and immunologic findings in 21 patients with Hodgkin’s disease. Neurology 1992;42:1938–1943.
Posner JB. Paraneoplastic cerebellar degeneration. Can J Neurol Sci 1993;20:S117–S122.
Furneaux HM, Rosenblum MK, Dalmau J et al. Selective expression of Purkinje-cell antigens in tumor tissue from patients with paraneoplastic cerebellar degeneration. N Engl J Med 1990;322:1844–1851.
Fathallah-Shaykh H, Wolf S, Wong E et al. Cloning of a leucine-zipper protein recognized by the sera of patients with antibody-associated paraneoplastic cerebellar degeneration. Proc Natl Acad Sci U S A 1991;88:3451–3454.
Mason WP, Graus F, Lang B et al. Small-cell lung cancer, paraneoplastic cerebellar degeneration and the Lambert-Eaton myasthenic syndrome. Brain 1997;120:1279–1300.
Bataller L, Wade DF, Graus F et al. Antibodies to Zic4 in paraneoplastic neurologic disorders and small-cell lung cancer. Neurology 2004;62: 778–782.
Shams’ili S, de Beukelaar J, Gratama JW et al. An uncontrolled trial of rituximab for antibody associated paraneoplastic neurological syndromes. J Neurol 2006;253:16–20.
Digre KB. Opsoclonus in adults. Report of three cases and review of the literature. Arch Neurol 1986;43:1165–1175.
Anderson NE, Budde-Steffen C, Rosenblum MK et al. Opsoclonus, myoclonus, ataxia, and encephalopathy in adults with cancer: a distinct paraneoplastic syndrome. Medicine (Baltimore) 1988;67:100–109.
Wirtz PW, Sillevis Smitt PA, Hoff JI et al. Anti-Ri antibody positive opsoclonus-myoclonus in a male patient with breast carcinoma. J Neurol 2002;249:1710–1712.
Bataller L, Graus F, Saiz A et al. Clinical outcome in adult onset idiopathic or paraneoplastic opsoclonus-myoclonus. Brain 2001;124:437–443.
Prestigiacomo CJ, Balmaceda C, Dalmau J. Anti-Ri-associated paraneoplastic opsoclonus-ataxia syndrome in a man with transitional cell carcinoma. Cancer 2001;91:1423–1428.
Altman AJ, Baehner RL. Favorable prognosis for survival in children with coincident opso-myoclonus and neuroblastoma. Cancer 1976;37:846–852.
Rudnick E, Khakoo Y, Antunes NL et al. Opsoclonus-myoclonus-ataxia syndrome in neuroblastoma: clinical outcome and antineuronal antibodies-a report from the Children’s Cancer Group Study. Med Pediatr Oncol 2001;36:612–622.
Hormigo A, Dalmau J, Rosenblum MK et al. Immunological and pathological study of anti-Ri-associated encephalopathy. Ann Neurol 1994;36:896–902.
Linke R, Schroeder M, Helmberger T et al. Antibody-positive paraneoplastic neurologic syndromes: value of CT and PET for tumor diagnosis. Neurology 2004;63:282–286.
Swart JF, de Kraker J, van der Lely N. Metaiodobenzylguanidine total-body scintigraphy required for revealing occult neuroblastoma in opsoclonus-myoclonus syndrome. Eur J Pediatr 2002;161:255–258.
Hayward K, Jeremy RJ, Jenkins S et al. Long-term neurobehavioral outcomes in children with neuroblastoma and opsoclonus-myoclonus-ataxia syndrome: relationship to MRI findings and anti-neuronal antibodies. J Pediatr 2001;139:552–559.
Yang YY, Yin GL, Darnell RB. The neuronal RNA-binding protein Nova-2 is implicated as the autoantigen targeted in POMA patients with dementia. Proc Natl Acad Sci U S A 1998;95:13254–13259.
Buckanovich RJ, Yang YY, Darnell RB. The onconeural antigen Nova-1 is a neuron-specific RNA-binding protein, the activity of which is inhibited by paraneoplastic antibodies. J Neurosci 1996;16:1114–1122.
Bataller L, Rosenfeld MR, Graus F et al. Autoantigen diversity in the opsoclonus-myoclonus syndrome. Ann Neurol 2003;53:347–353.
Dalmau J, Graus F, Cheung NK et al. Major histocompatibility proteins, anti-Hu antibodies and paraneoplastic encephalomyelitis in neuroblastoma and small cell lung cancer. Cancer 1995;75:99–109.
Antunes NL, Khakoo Y, Matthay KK et al. Antineuronal antibodies in patients with neuroblastoma and paraneoplastic opsoclonus-myoclonus. J Pediatr Hematol Oncol 2000;22:315–320.
Pranzatelli MR, Tate ED, Wheeler A et al. Screening for autoantibodies in children with opsoclonus-myoclonus-ataxia. Pediatr Neurol 2002;27:384–387.
Jongen JL, Moll WJ, Sillevis Smitt PA et al. Anti-Ri positive opsoclonus-myoclonus-ataxia in ovarian duct cancer. J Neurol 1998;245:691–692.
Dropcho EJ, Kline LB, Riser J. Antineuronal (anti-Ri) antibodies in a patient with steroid-responsive opsoclonus-myoclonus. Neurology 1993;43:207–211.
Nitschke M, Hochberg F, Dropcho E. Improvement of paraneoplastic opsoclonus-myoclonus after protein A column therapy. N Engl J Med 1995;332:192.
Mitchell WG, Davalos-Gonzalez Y, Brumm VL et al. Opsoclonus-ataxia caused by childhood neuroblastoma: developmental and neurologic sequelae. Pediatrics 2002;109:86–98.
Russo C, Cohn SL, Petruzzi MJ et al. Long-term neurologic outcome in children with opsoclonus-myoclonus associated with neuroblastoma: a report from the Pediatric Oncology Group. Med Pediatr Oncol 1997;28:284–288.
Chalk CH, Windebank AJ, Kimmel DW et al. The distinctive clinical features of paraneoplastic sensory neuronopathy. Can J Neurol Sci 1992;19:346–351.
Horwich MS, Cho L, Porro RS et al. Subacute sensory neuropathy: a remote effect of carcinoma. Ann Neurol 1977;2:7–19.
Graus F, Bonaventura I, Uchuya M et al. Indolent anti-Hu-associated paraneoplastic sensory neuropathy. Neurology 1994;44:2258–2261.
Molinuevo JL, Graus F, Serrano C et al. Utility of anti-Hu antibodies in the diagnosis of paraneoplastic sensory neuropathy. Ann Neurol 1998;44:976–980.
Antoine JC, Honnorat J, Camdessanche JP et al. Paraneoplastic anti-CV2 antibodies react with peripheral nerve and are associated with a mixed axonal and demyelinating peripheral neuropathy. Ann Neurol 2001;49:214–221.
Honnorat J, Antoine JC, Derrington E et al. Antibodies to a subpopulation of glial cells and a 66 kDa developmental protein in patients with paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry 1996;61:270–278.
Pittock SJ, Lucchinetti CF, Parisi JE et al. Amphiphysin autoimmunity: Paraneoplastic accompaniments. Ann Neurol 2005;58:96–107.
Saiz A, Dalmau J, Butler MH et al. Anti-amphiphysin I antibodies in patients with paraneoplastic neurological disorders associated with small cell lung carcinoma. J Neurol Neurosurg Psychiatry 1999;66: 214–217.
Uchuya M, Graus F, Vega F et al. Intravenous immunoglobulin treatment in paraneoplastic neurological syndromes with antineuronal autoantibodies. J Neurol Neurosurg Psychiatry 1996;60:388–392.
OhSJ, Dropcho EJ, Claussen GC. Anti-Hu-associated paraneoplastic sensory neuropathy responding to early aggressive immunotherapy: report of two cases and review of literature. Muscle Nerve 1997;20:1576–1582.
Wirtz PW, Wintzen AR, Verschuuren JJ. Lambert-Eaton myasthenic syndrome has a more progressive course in patients with lung cancer. Muscle Nerve 2005;32:226–229.
O’Neill JH, Murray NM, Newsom-Davis J. The Lambert-Eaton myasthenic syndrome. A review of 50 cases. Brain 1988;111:577–596.
Elrington GM, Murray NM, Spiro SG et al. Neurological paraneoplastic syndromes in patients with small cell lung cancer. A prospective survey of 150 patients. J Neurol Neurosurg Psychiatry 1991;54:764–767.
Wirtz PW, Nijnuis MG, Sotodeh M et al. The epidemiology of myasthenia gravis, Lambert-Eaton myasthenic syndrome and their associated tumours in the northern part of the province of South Holland. J Neurol 2003;250:698–701.
Hawley RJ, Cohen MH, Saini N, Armbrustmacher VW. The carcinomatous neuromyopathy of oat cell lung cancer. Ann Neurol 1980;7:65–72.
Wirtz PW, Willcox N, van der Slik AR et al. HLA and smoking in prediction and prognosis of small cell lung cancer in autoimmune Lambert-Eaton myasthenic syndrome. J Neuroimmunol 2005;159:230–237.
Wirtz PW, Lang B, Graus F et al. P/Q-type calcium channel antibodies, Lambert-Eaton myasthenic syndrome and survival in small cell lung cancer. J Neuroimmunol 2005;164:161–165.
Motomura M, Johnston I, Lang B et al. An improved diagnostic assay for Lambert-Eaton myasthenic syndrome. J Neurol Neurosurg Psychiatry 1995;58:85–87.
Rosenfeld MR, Wong E, Dalmau J et al. Cloning and characterization of a Lambert-Eaton myasthenic syndrome antigen. Ann Neurol 1993;33: 113–120.
Chalk CH, Murray NM, Newsom-Davis J et al. Response of the Lambert-Eaton myasthenic syndrome to treatment of associated small-cell lung carcinoma. Neurology 1990;40:1552–1556.
McEvoy KM, Windebank AJ, Daube JR et al. 3,4-Diaminopyridine in the treatment of Lambert-Eaton myasthenic syndrome. N Engl J Med 1989;321:1567–1571.
Sanders DB, Massey JM, Sanders LL et al. A randomized trial of 3,4-diaminopyridine in Lambert-Eaton myasthenic syndrome. Neurology 2000;54:603–607.
Newsom-Davis J, Murray NM. Plasma exchange and immunosuppressive drug treatment in the Lambert-Eaton myasthenic syndrome. Neurology 1984;34:480–485.
Bird SJ. Clinical and electrophysiologic improvement in Lambert-Eaton syndrome with intravenous immunoglobulin therapy. Neurology 1992;42:1422–1423.
Buchbinder R, Forbes A, Hall S et al. Incidence of malignant disease in biopsy-proven inflammatory myopathy. A population-based cohort study. Ann Intern Med 2001;134:1087–1095.
Hill CL, Zhang Y, Sigurgeirsson B et al. Frequency of specific cancer types in dermatomyositis and polymyositis: a population-based study. Lancet 2001;357:96–100.
Mastaglia FL, Garlepp MJ, Phillips BA et al. Inflammatory myopathies: clinical, diagnostic and therapeutic aspects. Muscle Nerve 2003;27: 407–425.
Targoff IN. Update on myositis-specific and myositis-associated autoantibodies. Curr Opin Rheumatol 2000;12:475–481.
Griggs RC, Mendell JR, Miller RG. Evaluation and Treatment of Myopathies. Philadelphia: F.A. Davis Company; 1995:1–464.(Janet W. de Beukelaar, Pe)