ITPR1 autoimmunity: Frequency, neurologic phenotype, and cancer association

Autoantibodies specific for the neuronal (type 1) isoform of the ubiquitously expressed inositol trisphosphate receptor (ITPR) have been reported in 8 patients to date, 5 with cerebellar ataxia (1 with breast cancer) and 3 with peripheral neuropathy (1 with lung carcinoma and 1 with multiple myeloma).^1–4 We report in this study the frequency, neurologic presentations, and oncologic associations of 14 ITPR1-immunoglobulin G (IgG)-positive patients.

Methods.

The study was approved by the Mayo Clinic Institutional Review Board.

Patients.

In the course of clinical service evaluation for paraneoplastic neural autoantibodies in the Mayo Clinic Neuroimmunology Laboratory (1997–2017), 117 patients were classified as having an IgG that, by mouse tissue-based indirect immunofluorescence assay (IFA), bound to the cerebellum in a “Medusa head-like” cytoplasmic staining pattern (i.e., prominently immunoreactive Purkinje cell perikaryon and dendrites) and were stored.^1,3 Clinical information was obtained by physician telephonic interview and case record review.

Serologic testing.

ITPR1-specific IgG (see figure e-1 at Neurology.org/nn) was detected by transfected cell–based assay ([serum, 1:10; CSF, 1:2], Euroimmun, Germany) in 17 patients; clinical information was available in 14. Healthy control (100 adults and 45 children) and disease control subjects (30 neurologic [anti–neuronal nuclear antibody type 1 (ANNA-1/anti-Hu), Purkinje cell antibody type 1 (PCA-1/anti-Yo), PCA-2/anti-MAP1B, MS] and 60 nonneurologic [polyclonal gammopathies, Sjogren, and lupus] tested negative.

Results.

The median age at neurologic symptom onset was 64 years (range 7–83 years); 10 patients were women (71%). Data available for 14 seropositive cases (table) identified 4 major neurologic manifestations: (1) peripheral neuropathy (somatic, patients 1–4; autonomic, patient 5), (2) cerebellar ataxia (patients 6–10, see figure e-1), (3) encephalitis with seizures (patients 5, 11, and 12), and (4) myelopathy (patients 2, 12, and 13).

Table.

Neurologic, serologic, and oncologic characteristics of 14 ITPR1-IgG–positive patients

Patient 11 had a generalized tonic-clonic seizure, and focal status epilepticus followed. EEG showed unilateral periodic discharges. After an initially favorable response to IV methylprednisolone and antiseizure medications, the dose of prednisone was tapered off. Relapse 2 weeks later required intensive care unit care. Seizures thereafter were refractory to corticosteroid, IV immunoglobulin, and antiepileptic medications. Life support was withdrawn at the family's request. Seizure and encephalopathy in patient 12 were followed 1 week later by opsoclonus myoclonus syndrome. Quadriplegia developed 3 weeks later because of myelitis. This patient's CSF was additionally positive for NMDA-R-IgG and GFAP-IgG, which may explain encephalitis and myelitis, respectively.

CSF results available in 5 patients revealed elevated protein levels and pleocytosis in 4. ITPR1-IgG titer values did not correlate with the severity or the type of clinical or oncologic phenotype. Six cancers were found in 5 patients, 1 based on PET CT and the others proven histologically (3 breast, 1 renal, and 1 endometrial). Neurologic impairment did not improve significantly in any of the 10 patients who received immunotherapy.

Frequency.

Prospective detection of ITPR1-IgG in 8 patients (who were part of the 14 patients presented here) over a 12-month period represented 0.015% of 52,000 neurologic patients' specimens submitted for paraneoplastic autoantibody evaluation. By comparison, other recognized paraneoplastic antibodies' detection frequencies during that period were 0.20% for ANNA-1, 0.08% for PCA-1, 0.03% for ANNA-2 (anti-Ri), and 0.001% for PCA-Tr (delta/notch-like epidermal growth factor–related receptor).

Discussion.

The neurologic manifestations encountered in patients with ITPR1 autoimmunity are more diverse than previously described. Peripheral neuropathy was as common as cerebellar ataxia and was most strongly associated with malignancy (cancer was found in 3 of 5 patients with neuropathy). In addition to its high expression in the CNS (highest in Purkinje cells of the cerebellum), ITPR1 is also expressed in the peripheral nervous system, where it is implicated in synaptogenesis and axonal growth.⁵

The wide dissemination of cancer (including metastases to bone or liver) observed in ITPR1-IgG–positive patients contrasts with the limited stage cancers (usually restricted to regional lymph nodes) encountered in paraneoplastic neurologic autoimmunity related to small cell lung carcinomas.⁶ ITPR1 may be a biomarker of more aggressive tumor behavior, since ITPR1 is implicated in cell migration, and other ITPR isoforms in cancer dissemination. One study reported that resistance to conventional anticancer treatment in patients with renal cell carcinoma related to von Hippel-Lindau syndrome is associated with ITPR1 upregulation in the tumor, which protects against natural killer cell cytotoxicity through induction of autophagy.⁷ Other ITPR isoforms have been implicated in tumor growth promotion. Our data mandate a thorough search for cancer when ITPR1-IgG is detected, given the 36% frequency of malignancy we report. This estimate may be low because clinical information and follow-up for some patients were limited. The introduction of assays to detect ITPR1-IgG as part of service paraneoplastic neural antibody testing will allow for better definition of the full clinical and oncologic spectrum.