Article Text

Short report
Characterisation of TRIM46 autoantibody-associated paraneoplastic neurological syndrome
  1. Cristina Valencia-Sanchez1,
  2. Andrew M Knight2,
  3. M Bakri Hammami2,
  4. Yong Guo1,
  5. John R Mills2,
  6. Thomas J Kryzer2,
  7. Amanda L Piquet3,
  8. Anik Amin4,
  9. Morgan Heinzelmann4,
  10. Claudia F Lucchinetti1,
  11. Vanda A Lennon1,2,5,
  12. Andrew McKeon1,2,
  13. Sean J Pittock1,2,
  14. Divyanshu Dubey1,2
  1. 1 Department of Neurology, Mayo Clinic, Rochester, Minnesota, USA
  2. 2 Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
  3. 3 Department of Neurology, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA
  4. 4 Department of Neurology, University of Texas Southwestern Medical Center, Dallas, Texas, USA
  5. 5 Department of Immunology, Mayo Clinic, Rochester, Minnesota, USA
  1. Correspondence to Dr Divyanshu Dubey, Department of Neurology, Mayo Clinic, Rochester, MN 55905, USA; Dubey.Divyanshu{at}


Objectives To report the expanded neurological presentations and oncological associations of tripartite motif-containing protein 46 (TRIM46)-IgG seropositive patients.

Methods Archived sera/cerebrospinal fluid (CSF) were evaluated by tissue-based immunofluorescence assay to identify patients with identical axon initial segment (AIS)-specific staining pattern. Phage immunoprecipitation sequencing (PhIP-Seq) was used to identify the putative autoantigen.

Results IgG in serum (17) and/or CSF (16) from 25 patients yielded unique AIS-specific staining on murine central nervous system (CNS) tissue. An autoantibody specific for TRIM46 was identified by PhIP-Seq, and autoantigen specificity was confirmed by transfected COS7 cell-based assay. Clinical information was available for 22 TRIM46-IgG seropositive patients. Fifteen were female (68%). Median age was 67 years (range 25–87). Fifteen (68%) patients presented with subacute cerebellar syndrome (six isolated; nine with CNS accompaniments: encephalopathy (three), brainstem signs (two), myelopathy (two), parkinsonism (one)). Other phenotypes included limbic encephalitis (three), encephalopathy with/without seizures (two), myelopathy (two). Eighteen (82%) had cancer: neuroendocrine carcinomas (9; pancreatic (3), small-cell lung (4), oesophagus (1), endometrium (1)), adenocarcinomas (6; lung (2), ovarian (2), endometrial (1), breast (1)), sarcoma (2) and gastrointestinal tumour (1). Neurological symptoms in three followed immune checkpoint inhibitor (ICI) administration.

Conclusions This study supports TRIM46-IgG being a biomarker of paraneoplastic CNS disorders and expands the neurological phenotypes, oncological and ICI-related adverse event associations.

  • neuroimmunology
  • paraneoplastic syndrome

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Tripartite motif-containing protein 46 (TRIM46) IgG staining pattern was initially described by Shams’ili and colleagues in a patient with paraneoplastic encephalomyelitis and small cell lung cancer (SCLC).1 The autoantigen was later identified using immunoprecipitation mass spectrometry.2 Two additional cases were subsequently reported, one of whom had SCLC.3 Herein, we describe the neurological presentations and oncological associations of 22 additional cases.


Retrospective review of the patients’ charts and human specimen acquisition was approved by the Mayo Clinic Institutional Review Board.

Between 2007 and 2020, 264 archived specimens (158 sera, 106 cerebrospinal fluid (CSF)) with a filamentous IgG staining pattern on cryosectioned mouse tissues assessed by indirect immunofluorescence assay (IFA) were retested to identify identical axon initial segment (AIS) immunoreactivity. Phage immunoprecipitation sequencing (PhIP-Seq)4 5 was used to search for a putative autoantigen among patient samples demonstrating autoantibody binding to AIS (CDI labs, Baltimore, Maryland, USA). TRIM46 was confirmed as the autoantigen by TRIM46 cell-based assay (CBA) and colocalisation on mouse brain sections (online supplemental methods). Serum samples from patients with SCLC (n=69) and Merkel cell carcinoma (n=10) but no neurological disorder, Sjogren’s syndrome (n=15), systemic lupus erythematosus (n=35) and healthy control subjects (n=127), and CSF samples from multiple sclerosis (MS, n=42), acute disseminated encephalomyelitis (n=12) and normal pressure hydrocephalus (NPH, n=13) patients were tested by tissue IFA. In addition, 24 SCLC sera, 27 healthy sera, 5 NPH CSFs and 8 MS CSFs were tested on TRIM46 CBA.

Supplemental material

Formalin-fixed paraffin-embedded 5 µm thick sections of the breast adenocarcinoma (patient 3) tissue was evaluated for TRIM46 immunoreactivity (online supplemental methods).


Confirmation of TRIM46 as the autoantigen

Thirty-three specimens corresponding to 25 patients (9 serum only, 8 CSF only, 8 both) demonstrated an identical staining pattern of the AIS in cerebellar Purkinje cell layer, cerebral cortex and hippocampus by indirect IFA on murine brain (figure 1A). Non-neural tissues (murine kidney and gastric mucosa) were non-reactive.

Figure 1

Distinct immunofluorescence pattern of patient IgG binding to mouse tissues, confirmation of tipartite motif-containing protein 46 (TRIM46) as the target antigen. (A) Patient IgGs yielded an indirect immunofluorescence staining pattern characterised by staining of the axonal initial segment in cerebellar Purkinje neurons at ×20 magnification (A1), cerebral cortex at ×20 (A2) and hippocampus at ×10 (A3). (B) Dual immunostaining yielded by a commercial mouse IgG specific for TRIM46 (B1) and by a patient’s IgG (B2), binding to the axonal initial segment of Purkinje neurons; merged image (B3). (C) GFP-tagged TRIM46 protein expressed in transfected COS7 cells (C1), and patient IgG binding (C2); merged image (C3).

Evaluation of sera from three patients by PhIP-Seq identified TRIM46 peptides (online supplemental table 1).

Supplemental material

Murine tissue IFA demonstrated colocalisation between commercial mouse TRIM46-IgG and patient-IgG on murine brain tissue (figure 1B). All 33 specimens tested positive on TRIM46 CBA (figure 1C). All the control specimens tested by tissue IFA and/or CBA were negative.

After identification of the target antigen, a literature search revealed three previously reported cases.1 3

Histopathological assessment of breast adenocarcinoma tissue revealed cytoplasmic TRIM46 immunoreactivity (figure 2).

Figure 2

Expression of tripartite motif-containing protein 46 (TRIM46) in breast carcinoma tissue. H&E stain shows excised breast carcinoma tissue at different magnifications (A–D, G). The poorly differentiated tumour tissue (B, C) shows no TRIM46 expression on immunohistochemistry (E, F). However, the peripheral tumour tissue with better differentiation (D, G) shows TRIM46 immunoreactivity (H, I). Another example of peripheral relative normal tissue shows TRIM46 in the epithelium (J). Normal breast control tissue is negative for TRIM46 (K, L). Scale bars: 100 µm (A), 50 µm (B, D, E, H, K), 20 µm (C, F, G, I, J, L).

Clinical and diagnostic findings

Clinical information was available for 22 TRIM46-IgG seropositive patients. Fifteen were female (68%). Median age was 67 years (range 25–87, table 1 and online supplemental table 2).

Table 1

Characteristics of 22 patients with TRIM46-IgG

Fifteen patients (68%) presented with a subacute cerebellar syndrome, six in isolation, nine with additional signs: encephalopathy (four, one of whom additionally had autonomic dysfunction and one had opsoclonus-myoclonus), brainstem signs (two), parkinsonism and dystonia (one), myelopathy (two, one with lower extremity hyperreflexia and bladder dysfunction, and one with bilateral lower extremity weakness). Another three patients presented with limbic encephalitis. Two patients had myelopathy (one with bilateral lower extremity weakness; one with bilateral upper and lower extremity weakness and bladder dysfunction). Other individual neurological presentations included encephalopathy and seizures with encephalopathy.

Brain MRI was available for 16 patients and abnormal in 8 (figure 3A–C). Five revealed cerebellar atrophy (one additionally showed medullary and pulvinar T2 signal hyperintensities), and three medial temporal T2/FLAIR hyperintensities (two bilateral, one unilateral). CSF analysis was available for 14 patients and in 9 (64%) demonstrated inflammation (elevated leucocyte count or CSF-restricted supernumerary oligoclonal bands).

Figure 3

Characteristic brain images from three TRIM46-IgG seropositive patients. Brain images of three TRIM46-IgG seropositive patients. (A) MRI, sagittal T1 sequence, showing cerebellar atrophy in a patient who presented with subacute cerebellar ataxia, and was diagnosed with small cell lung cancer 4 years after neurological symptom onset. (B) Axial FLAIR-T2 sequences of a patient with endometrial carcinoma on treatment with pembrolizumab, who presented with a subacute cerebellar and brainstem syndrome. At onset of symptoms, a T2-hyperintense area in the anterior left medulla was observed (B1). Ten weeks later, she had developed cerebellar atrophy and T2 signal hyperintensity in the left pulvinar (B2). (C) Axial FLAIR-T2 sequence showing medial right temporal lobe T2-hyperintensity in a patient with lung adenocarcinoma treated with pembrolizumab, who presented with limbic encephalitis. TRIM46, tripartite motif-containing protein 46.

Oncological associations

Cancer was diagnosed in 18 patients (82%), 8 of whom had metastatic disease (none with CNS metastasis). Twelve cases had been diagnosed prior to neurological symptom onset (median 17 months prior, range 3 months–8 years). Neurological symptoms in three patients with prior cancer diagnosis developed after initiation of immune checkpoint inhibitor (ICI) therapy for cancer management. Cancer was found in six patients in the course of evaluation at the onset of neurological symptoms, and SCLC was diagnosed in one patient 4 years after the onset of neurological symptoms.

The detected neoplasms were of neuroendocrine lineage in half of the patients, including SCLC (four), and neuroendocrine tumours of the pancreas (three), oesophagus (one) and endometrium (one). Other cancers included lung adenocarcinoma (two), ovarian adenocarcinoma (two), sarcoma (two), endometrial adenocarcinoma (one), breast adenocarcinoma (one) and gastrointestinal tumour not otherwise specified (one) (online supplemental figure 1). Two additional patients had high suspicion for cancer based on chest CT findings, despite no histopathological confirmation: one patient had multiple pulmonary nodules, and the other, a smoker, had hilar lymphadenopathy. The two remaining patients had no evidence of cancer by CT imaging of the chest, abdomen and pelvis.

Post-ICI cases

Neurological symptoms in three patients developed 2–8 months after initiation of ICI for cancer management. The first patient had received pembrolizumab for endometrial adenocarcinoma (patient 20, online supplemental video 1), and two patients with lung adenocarcinoma had received pembrolizumab (patient 21), and combination therapy with ipilimumab and nivolumab (patient 22). Clinical presentation in two of them was cerebellar ataxia with additional signs (one myelopathy, one brainstem involvement). The third patient presented with limbic encephalitis.

Supplementary video

In all cases, ICI therapy was discontinued. One patient received oral prednisone without improvement, followed by plasmapheresis and intravenous cyclophosphamide, and another patient received intravenous methylprednisolone and plasmapheresis. Both had partial improvement of neurological symptoms after treatment (follow-up 2 and 3 months after onset). The third patient transitioned to hospice care and died shortly thereafter.

Treatment and outcomes

In addition to the post-ICI cases discussed above, 14 patients received immunosuppression: oral or intravenous methylprednisolone (eight), intravenous immunoglobulin (nine), plasmapheresis (three), rituximab (one) and cyclophosphamide (one). Of the 12 patients with available post-therapy follow-up data, none had significant neurological improvement. Two patients were stable for 6 and 24 months and lost to follow-up afterwards. Eight patients continued to deteriorate and died (five within 3–7 months; of three patients with available data, two died due to cancer progression and one due to neurological progression), and two transitioned to hospice care at 6 (due to severe neurological deficits and cancer diagnosis, no chemotherapy given due to poor performance status) and 18 months (due to neurological progression) after onset.

Four additional patients who did not receive immunosuppression continued to deteriorate, three deceased (two within 1 year from symptom onset; one with available information died due to cancer progression) and the other transitioned to hospice care. The remaining three patients were lost to follow-up.


TRIM46-IgG is a biomarker of paraneoplastic neurological autoimmunity. The neurological phenotypes were diverse but restricted to the CNS, as suggested by the initial report, which included encephalomyelitis, cerebellar ataxia and rapidly progressive dementia.1 3 In the 22 additional patients described here, cerebellar ataxia was the most common phenotype, either in isolation or with additional signs.

Although the most common neoplasm was SCLC as previously reported,1 3 we found additional neuroendocrine tumours and other tumours of diverse lineage. The TRIM46-IgG seronegativity of patients with SCLC, who lacked neurological disorders and other control subjects’ sera, supports the clinical specificity of this antibody for paraneoplastic neurological syndrome (PNS). Furthermore, we demonstrated the expression of TRIM46 in breast adenocarcinoma tissue, supporting that TRIM46 paraneoplastic autoimmunity is not limited to SCLC cases.

It is noteworthy that neurological symptoms in 3 of the 22 patients followed administration of ICI therapy. All post-ICI TRIM46-IgG seropositive cases had severe CNS neurological immune-related adverse events (nirAEs), requiring aggressive immunosuppression leading to clinical improvement in two cases, and the remaining patient who did not receive immunosuppression died shortly after nirAE development. These findings support the utility of neural-specific autoantibodies such as TRIM46-IgG as biomarkers aiding the diagnosis and selection of the most appropriate therapies for CNS nirAE, especially those resembling PNS phenotypes, as recommended by the recent consensus diagnostic criteria for PNS, and ICI nirAE.6 7

Our study’s identification of TRIM46 as the autoantigen target of a paraneoplastic IgG through PhIP-Seq technology supports the application of this methodology in discovery and characterisation of novel autoantigens.5 8 PhIP-Seq epitope mapping data also demonstrate a polyclonal response targeting different TRIM46 peptides, as reported by an earlier study.3

Consistent with the patients’ CNS-restricted clinical phenotypes, TRIM46 is a brain-restricted microtubule binding protein located in the AIS, which has an important role in the generation of action potential and neuronal polarity.9 It is required for axon formation and establishment of neuronal polarity during neuronal development. TRIM46 regulates microtubule organisation, forming parallel microtubule arrays that drive cargo trafficking in axons of differentiated neurons.2 Antibodies against other TRIM proteins have also been identified in paraneoplastic cerebellar degeneration. TRIM9 and TRIM67 were recently identified in two patients with subacute cerebellar ataxia and lung adenocarcinoma and one patient with melanoma.5 10

Given that TRIM46 is an intracellular protein located in the cytoplasm, the antibodies are likely not pathogenic. It is more plausible that T-cell cytotoxicity is responsible for neurological dysfunction. Similar to other PNS associated with antibodies directed against intracellular antigens, the prognosis is poor.11

This study supports TRIM46-IgG as a biomarker of PNS and extends the associated clinical phenotypes and malignancies. Having now been identified as a biomarker of subacute onset CNS syndrome with a strong oncological association (>70% of cases) by two independent laboratories, TRIM46-IgG should qualify as high-risk paraneoplastic autoantibody validated to be strongly predictive of underlying cancer.6 Testing for TRIM46-IgG should be considered in patients presenting with intermediate or high-risk PNS phenotypes,6 especially subacutely progressive cerebellar ataxia.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by Mayo Clinic Institutional Review Board (IRB) numbers 08-006647 and 08-007846. Retrospective review of the patients’ charts and human specimen acquisition was approved by the Mayo Clinic Institutional Review Board.


The authors would like to thank Dr Cherry Yu and Dr Baber Khan for the assistance in recording the online supplemental video), Dr Yajue Huang for the assistance with breast adenocarcinoma tissue TRIM46 staining review, and Sara Vinje for administrative support.


Supplementary materials


  • Contributors CV-S contributed to acquisition and analysis of data, drafting the manuscript and figures. AMK, MBH, JRM, TJK, ALP, AA, MH, CFL, VAL, AM and SJP contributed to acquisition and analysis of data. YG contributed to acquisition and analysis of data and drafting figure 3. DD contributed to acquisition and analysis of data, conception, design and study supervision. All authors revised the manuscript critically for intellectual content and approved the final version.

  • Funding This work was supported by Mayo Clinic Center of Individualized Medicine Welch Award.

  • Competing interests CV-S, AMK, MBH, YG, JRM, AA and MH have no competing interests to disclose. TJK has a patent AQP4-IgG with royalties paid, a patent KLHL11, septin 5 and MAP1B IgG pending. ALP reports grants from University of Colorado, grants from Rocky Mountain MS Center, personal fees from Genentech/Roche, personal fees from Alexion. CFL received grants from National Institute of Health, National Multiple Sclerosis Society, National Institute of Neurological Disorders and Stroke, Kingsland Foundation, Biogen Idec. VAL has a patent AQP4-IgG with royalties paid, a patent KLHL11, septin 5 and MAP1B IgG pending. AM has patent pending for KLHL11, Septin 5, and MAP1B and GFAP IgGs as markers of neurological autoimmunity and paraneoplastic disorders. He has received research support from Alexion, Grifols, and Euroimmun but has not received personal compensation. SJP has a patent # 8889102 (Application # 12-678350) -Neuromyelitis Optica Autoantibodies as a Marker for Neoplasia issued, and a patent # 9891219B2 (Application # 12-573942) Methods for Treating Neuromyelitis Optica (NMO) by Administration of Eculizumab to an individual that is Aquaporin-4 (AQP4)-IgG Autoantibody positive issued. He has a patent pending for GFAP, Septin 5, MAP1B, KLHL11 and PDE10A IgGs as markers of neurological autoimmunity and paraneoplastic disorders. He has consulted for Alexion, Euroimmune, Medimmune, Astellas, Genetech, Sage Therapeutics, Prime Therapeutics. He has received research support from Grifols and Alexion. He has received research support from NIH, Guthy Jackson Charitable Foundation, Autoimmune Encephalitis Alliance. All compensation for consulting activities is paid directly to Mayo Clinic. DD has received research support from Center of Multiple Sclerosis and Autoimmune Neurology, Center of Individualized Medcine and Grifols pharmaceuticals. He has consulted for UCB and Astellas pharmaceuticals. All compensation for consulting activities is paid directly to Mayo Clinic. He has a patents pending for KLHL11-IgG and LUZP4-IgG as markers of testicular cancer and neurological autoimmunity.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.