Paraneoplastic Neurological Syndromes (PNS) are rare remote effects of cancer, not directly attributed to mass lesions, metastases, infections, ischemia, coagulopathy, metabolic disruptions or tumour treatment. Currently, PNS treatment is mostly limited to tumour treatment. Because of an initial inflammatory stage early in the evolution of the PNS several immunotherapy modalities have been tried. Intravenous human immunoglobulins could be expected to provide a stabilization or even improvement of PNS, if administered early enough to prevent permanent neuronal damage.

Paraneoplastic neurological syndromes (PNS) are remote effects of malignant neoplasia on neural tissue, not directly caused by mass lesions, metastases, infections, ischemia, coagulopathy, metabolic disruptions or treatment. The symptoms and signs of PNS are diverse, usually acute or subacute. Onconeural antibodies are detected in their serum, and sometimes in their Cerebrospinal Fluid (CSF); onconeural antibodies are highly specific for identifying patients with neurologic symptoms as having a paraneoplastic syndrome. The most frequent ones are anti-Hu, anti-Yo and anti-CV2/CRMP5. Although a known cancer patient may present with a PNS, neurologic symptoms precede other manifestations of a tumor in about 65% of patients.

Currently, autoimmunity is postulated to underlie the pathophysiology of PNS. Tumor cells can sometimes express antigens normally found only in the nervous system, an immunologically privileged site. Onconeural antigens on tumors can be identified as foreign by the immune system and an immune attack can be elicited against them. In the subset of PNS patients, the immune system could also recognize and attack onconeural antigens in normal nervous tissue. This autoimmune hypothesis for the pathogenesis of PNS is supported by a series of arguments: Onconeural antibodies are found in serum and CSF; Cerebrospinal fluid studies show inflammation in 93% of patients; In the acute phase of paraneoplastic cerebellar degeneration hypermetabolism in 18F-Fluoro-2-Desoxy-Glucose Positron Emission Tomography (FDG-PET) scan and increased perfusion on Single Photon Emission Computed Tomography (SPECT) have been described and attributed to inflammatory changes; Pathological examination of the nervous system from patients with anti-Hu associated paraneoplastic encephalomyelitis demonstrates loss of neurons in affected areas with extensive T-cells infiltration and immunohistochemical studies on damaged neural tissue after autopsy reveal inflammatory infiltrations.

Effective treatment of PNS requires an early clinical suspicion followed by rapid diagnosis, through detection of onconeural antibodies, and identification of the underlying tumor. Unfortunately appropriate treatment of the cancer is most often unsuccessful to improve neurological symptoms and these patients are generally left bedridden until death. It's seems important to associate an immune treatment but has never been adequately tested. Experience with immunotherapy modalities, such as corticosteroids, plasma exchange, immunosuppressants or human intravenous immunoglobulin (IVIg), relies on case reports, retrospective studies, and a couple of prospective studies in patients generally treated at a late stage. To date, it has not been possible to set up a prospective therapeutic trial evaluating the role of early immune treatment, administered less than 6 months after the onset of symptoms, in these patients. Considerable evidence for an initial inflammatory stage (early in the evolution of the PNS), and isolated case reports of early immunological intervention suggest that this approach may be valid to treat these disorders. Because IVIg are often use in neuro-immunological disorders (ex. first-line therapy in Guillain-Barre syndrome, chronic inflammatory demyelinating polyradiculoneuropathy, multifocal motor neuropathy and dermatomyositis) with little and well known side effects and no interaction with cancer treatment we decide to use IVIg for our trial.

The Aim of the study is to improve neurological impairment and function in patients with early diagnosis of various PNS associated with well-characterized onconeural antibodies anti-Hu, anti-Yo, anti-CV2/CRMP5. The primary endpoint of the study is the percentage of patients with neurological improvement after 3 months of immunotherapy with IVIg. "Success" is defined by ≥ 1 point lower score in the modified Rankin Scale (mRS) after treatment compared to baseline. This is a prospective, open-label trial explores the efficacy of drugs approved for treatment of immune-mediated neurological disorders given early in the evolution of PNS associated with well-characterized onconeural antibodies. A total number of 17 patients will be treated. This is a multicenter study carried under the auspices of the French National Reference Center on PNS.

Inclusion criteria :

• Age ≥ 18 years
• Clinical diagnosis of PNS according to published criteria
• Positive well-characterized onconeural antibodies (Hu, Yo, CV2/CRMP5) in serum or CSF samples
• Rankin score between 2 and 4
• Less than 6 months since onset of symptoms
• Less than 3 weeks in a Rankin score between 2 and 3
• Patients who have given written informed consent

Exclusion criteria :

• Patients not be able to receive IVIg
• Patients who receive or will receive concomitant immunotherapy different from that in the protocol
• Patients with known selective deficiency of IgA
• Women of childbearing potential who are pregnant or lactating, seeking pregnancy or failing to take adequate contraceptive precautions
• Patients with psychiatric or systemic diseases that prevent the proposed treatment
• Patients who will not be able to attend the required follow-up visits
• Renal, hepatic or cardiac insufficiency, coagulopathy