Background: LGI1(leucine-rich-anti-glioma1protein 1) is one of the latest identified voltage-gated potassium channels related antigens. Anti-LGI1 encephalitis is one of the autoimmune encephalitis and characterized with neurological symptoms and hyponatremia. The anti-LGI1 encephalitis was thought to be non-malignancy related. And the hyponatremia of anti-LGI1 encephalitis was considered as inappropriate secretion of antidiuretic hormone. 

Case description: Here, we firstly report a 71-year-old female diagnosed with anti-LGI1 encephalitis with renal clear cell carcinoma and renal salt wasting. The old female was discovered a renal mass identified as renal clear cell carcinoma. Two months after the renal carcinoma resection, the patient fell frequently and showed discontinuous neuro-psychiatric disorders manifested as memory decline, disorientation, and unintentional upper limb movements as typical faciobrachial dystonic seizure episode. She also had hyponatremia. The combination of her clinical presentations and laboratory assessments supported a diagnosis of anti-LGI1 encephalitis and renal salt wasting. The Immunohistochemistry studies of the kidney resection indicated circulating LGI1 antibodies in sera might be binding to extracellular LGI1 predominantly in the proximal tubule where the major defect in solute transport exists in renal salt wasting. 

Conclusion: Anti-LGI1 encephalitis with renal carcinoma indicated its paralimbic pathology origin. The early diagnosis and immune-modulation therapy could lead to a good outcome. The hyponatremia in anti-LGI1 encephalitis was renal salt wasting instead of syndrome of inappropriate secretion of antidiuretic hormone. 

Anti-LGI1 encephalitis affects the limbic systems and presents with subacute onset of progressive neurological, cognitive, psychiatric disturbance and obstinate hyponatremia [1, 2]. Anti-LGI1 encephalitis is mostly considered to be non-malignancy related [3, 4]. We reported a case of anti-LGI1 encephalitis presented cognitive disorder and FBDS (faciobrachial dystonic seizure) after resection of her renal carcinoma for the first time. The unique relationship between serum sodium and fractional excretion of uric acid FEurate (fractional excretion of uric acid) determined that the hyponatremia was due to CRSW (cerebral/renal salt wasting) or more appropriately RSW (renal salt wasting) instead of the SIADH (syndrome of inappropriate secretion of antidiuretic hormone). The IHC (Immunohistochemistry) studies of the kidney indicated that circulating LGI1 antibodies in sera might be binding to extracellular LGI1 predominantly in the proximal tubule where the primary defect in solute transport exists in RSW. 

Clinical manifestation

A 71 years old female had a renal mass surgically removed, which was identified as renal clear cell carcinoma by pathological and immunological evaluation.  Almost two months after surgery, the patient complained about fatigue, fell frequently, unstable gait, and reported unintentional upper limb movements. Her family also noticed cognitive defects as short term memory, inability to recognize family members, and being unaware of home circumstances or time in the patient.  There was no family history of psychiatric disorders or dementias. She did not take medicines and denied the use of tobacco or alcohol.  The symptom evolution of the patient presented typical neurophysiological symptoms during the development instead of at the very beginning of the disease. She also reported FBDS (faciobrachial dystonic seizure) episode (supplementary file1). Her MMSE (minimental state examination) was 3/30 (orientation 1, memory 1, language 1). Detailed neurological examination showed a lack of coordination, including normal cranial nerves, predicted IV symmetric strength throughout, active reflexes, incorporated finger-to-nose testing. On physical examinations, she had a normal cardiovascular, respiratory, abdominal, and pulmonary examination during the reported medical history.

Laboratory results

Serum and CSF from the patient was qualitatively tested for neuropil antibodies associated with autoimmune encephalitis including antibodies to glutamate receptors type NMDA, type AMPA1 and type AMPA2, LGI1, CASPR2 and GABARB1/B2using the indirect immunofluorescence test (IIFT) (EUROIMMUN, FA 112d-1005-1, Germany). Cellbased assays (CBAs) for those antibodies were performed using EU90 cells (EUROIMMUN) transfected with cDNAs encoding the relevant proteins. Combinations of substrates were incubated with patient serum (1:10 dilution) or undiluted CSF sample. In a second step, the attached antibodies were stained with fluorescein-labelled anti-human antibodies (EUROIMMUN) and made visible with a fluorescence microscope. Fluorescence intensity level was used to describe the intensity of the specific fluorescence as a numeric value, reaching from “0” or “−” (no specific fluorescence) to “5” or “+++++” (extremely strong specific fluorescence). The deviation in the fluorescence intensity of the IIFT amounted to no more than±1 fluorescence intensity level for all samples[36]. A cell-based assay showed serum VGKC (voltage-gated potassium channels) complex proteins (EUROIMMUN, Germany) serum LGI1-Ab to be positive+++(1:100) while the CSF LGI1-Ab was negative; other autoimmune encephalitis antibodies including AMPA1 (glutamate receptor, ionotropic, alpha1), AMPA2 (glutamate receptor, ionotropic, alpha2), Casp2, cerebellum-1, GABAR (gamma-aminobutyric acid receptors), NMDAR (N-methyl-D-aspartate receptors) in serum and CSF were all negative, which established the diagnosis of anti-LGI1 encephalitis.  CSF (Cerebral spinal fluid) tests for protein, cells, glucose, chloride, and culture were normal. Serum sodium ranged between 118-148mmol/L during the reported medical history. The cortisol, FT4 and TSH were normal, which excluded hypothyroidism and Addison’s disease. Autoimmune, infectious, endocrinologic, neoplastic and paraneoplastic screenings were unremarkable. FEurate was increased after correction of her hyponatremia (Supplementary Table 1), which is in consistent with RSW and not SIADH[5] (Supplementary Table 2). The increased aldosterone during the period of hyponatremia was in consistent with RSW;

Posm (plasma osmolality); Uosm (Urine osmolality); AII, (angiotensin II); Aldo (aldosterone); creat (creatinine); UA (uric acid); FRT4 (free T4); FRT3 (free T3); TSH (thyroid stimulating hormone); FEurate (fractional excretion of uric acid, FEurate=(serum creatine X urine UA(24hrs))/(serum UA X urine creatin(24hrs))X100%).

Supplementary Table 1: Patient's laboratory findings of hyponatremia correction.

Supplementary Table 2: Differentiation of SIADH from RSW

ECV, extracellular volume; RSW, renal salt wasting; SIADH, secretion of antidiuretic hormone; UNa, urinary sodium concentration.Table comparing laboratory expectations for RSW and SIADH. UNa can be normal or often＞20mmol/l; serum urate and FEurate are increased during hyponatremia in both RSW and SIADH but differ when serum is normal. Serum urate and FEurate remain abnormal in RSW and normalize in SIADH when serum sodium is normal.  Expression of LGI1 by patient’s carcinoma. The IHC staining of LGI1 protein expression in kidney, and renal cell carcinoma to explore the underlying LGI1 related RSW mechanism (Figure 2). Paraffin embedded tumor junction tissue sections of the patient’s renal clear cell carcinoma from a patient with serum LGI1 antibodies (A tumor junction, B normal tissue, C carcinoma tissue), another patient with only renal clear cell carcinoma without serum LGI1 antibodies(D tumor junction, E normal tissue, F carcinoma tissue),  a commercial antibody against LGI1 (rabbit polyclonal anti-LGI1, Abcam, ab137045, diluted 1:50) was used. Note the expression of LGI1 in normal renal tubule (B, E) and absent in carcinoma tissue (D, F), and the more robust  expression pattern in the patient without LGI1 serum antibody (B, E). ()  Note expression of LGI1 in normal renal tubule (C, D) and absence in renal carcinoma tissue (E, F). There was reduced expression of the LGI1 in the patient when the circulating LGI1 antibody was present as compared to the absence of LGI1 antibody(C, D). There was a higher expression of LGI1 in proximal as compared to distal convoluted tubules both in the patients with or without circulating LGI1 antibodies(C, D).

Imaging results

Cerebral MRI (magnetic resonance imaging) (Figure 1 A, B) scans, including T2 flare (Figure 1 A), DWI(Figure 1 B) were normal. EEG (electroencephalogram) showed accidental sharp and slow wave complex in bifrontal and biparietal leads (Figure 1 E) with diffuse theta and delta wave background during FBDS. sLORETA (Standardized low-resolution brain electromagnetic tomography) spike source analysis by ASA 4.9 software showed the sharp and slow wave complex were sourced in frontal lobe (Figure 1 C, D).

Figure 1: MRI, EEG and spike sLORETA analysis presentations of the patient. The patient presented normal MRI in T2 flare and DWI sans(A, B); slow and accidental sharp and slow wave complex inbifrontal and biparietal leads with diffuse theta and delta wave background during FBDS (E), sLORETA (Standardized low-resolution brain electromagnetic tomography) spike source analysis by ASA 4.9 software showed the sharp and slow wave complex were sourced in frontal lobe (C, D).

Figure 2: Expression of LGI1 by patient’s carcinoma. Tumor junction tissue sections of the patient’s renal clear cell carcinoma from a patient with serum LGI1 antibodies (A, C, E), another patient with only renal clear cell carcinoma without serum LGI1 antibodies (B, D, F). A commercial antibody against LGI1 (rabbit polyclonal anti-LGI1, abcam, ab137045, diluted 1:50) was used for IHC. Note the expression of LGI1 in normal renal tubule (C,D) and absent in carcinoma tissue (E,F), and the reduced expression pattern in the patient with LGI1 serum antibody (C,D), and higher expression in proximal than distal convoluted tubules (C,D). (The bar presented 20mm, 10mm, 10mm for A,C,E,respectively).

Figure 3: Sodium level with according treatment. IVMP intravenous methylprednisolone; IVIG intravenous immunoglobulin.

Diagnosis, treatment and prognosis 

The treatments were given and modulated according to the evolution of diagnosis and evaluation of the main symptoms (Supplementary Figure 1). When the LGI1 antibody encephalitis diagnosis was established, treatment with IVMP (intravenous methylprednisolone) 1.0/day for 5 days, 0.5/day for 3 days, 0.125/day for 5 days, followed with IVIG (intravenous immunoglobulin) at a dose of0.4 g/kg/day for 5 days 0.4 g/kg/day for 5 days and continued oral prednisolone 40 mg/day treatment for a month declined to 8mg/day maintained. VPA (valproic acid sodium)0.5 oral bid combined OXC (oxcarbazepine) 0.3 oral bid were administered to prevent epilepsy recurrence. The patient’s neurological dysfunction responded well to immunoglobulin with diminution of facial muscle jerk as FBDS and improved MMSE scores during the IVIG and consisted afterward, which increased to 15/30(orientation 6, memory and recall ability 2, attention and calculation 2, language 3, executive function 2, visual spatial 0) after IVIG therapy, and jump to 22/30(orientation 8, memory and recall ability 4, attention and calculation 4, language 3, executive function 3, visual spatial 0) at 6-month follow up, 27/30(orientation 10, memory and recall ability 6, attention and calculation 4, language 3, executive function 3, visual spatial 1) at 2-year follow up. For hyponatremia, the patient received saline supplementary as 10% sodium chloride solution intravenous1.5% sodium chloride 250ml qd-bid plus 20ml oral bid-qid according to the serum sodium level. Hypersaline supplementary and toveptan was given in a combination of immunomodulation therapy, and the relationship of sodium level between the therapy showed good effectiveness of IVMP and hyper saline supplementary treatment (Figure 3). The elevated FEurate indicated the diagnosis of RSW instead of SIADH. No hyponatremia and other complaints were observed at the 6 months and 2 years follow up.

Supplementary Figure 1: Evolution of diagnosis and evaluation of the main symptoms.

The diagnosis of encephalitis was not initially entertained when the patient presented with fatigue and found to have hyponatremia, but the development of more obvious symptoms such as repeated falls and FBDS episodes were highly suggestive of a paraneoplastic condition associated with LGI1 encephalitis. There has been a growing interest and understanding of paraneoplastic encephalitis and autoimmune encephalitis. Since the 1980s, there has been investigations which identified two major groups of antigens that categorize paraneoplastic autoimmune encephalitis, the classic onco-neuronal antigens that include Yo, Hu, Ri, Tr, CRMPs (Amphiphysin, collapsin response mediator proteins), Recoverin, Ma; autoimmune synaptic antigens including NMDAR, AMPAR, LGI1, Casp2, GABAR and unknown antigens[6].Paraneoplastic autoimmune encephalitis has been reported in patients with lung small cell carcinoma, adenocarcinoma of breast, ovarian teratoma, lung cancer etc.[3]. We present a rare case of renal carcinoma related paraneoplastic encephalitis, which has been reported previously in 3 cases over the last two decades, the diagnosis being hampered in most by a lack of specific antigens[7-9]. For example, a hallucinating 66-year-old with renal cell carcinoma had a diagnosis of limbic encephalitis made based on tests for intracellular autoimmune encephalitis that included Hu, Yo antibody but failed to identify antibodies in serum or CSF extracellular antibodies including NMDAR, AMPAR, LGI1, Casp2, GABAR[7], which would be possible autoimmune encephalitis antigens. On the other hand, the anti-IGI1 antibody was first reported in 2010 as an extracellular autoimmune factor that targeted the nervous system playing a role in paraneoplastic autoimmune encephalitis[10]. In time LGI1 paraneoplastic encephalitis or anti-LGI1 encephalitis have been reported in lung cancer[11], and thymoma[12]. The diagnosis of this rare condition can thus be challenging for primary care physicians, and we hope that this report will raise awareness and highlight the unique presentation of patients with paraneoplastic anti LGI1 encephalitis.  Despite the lack of a clear line to distinguish the paraneoplastic encephalitis and anti-LGI1 encephalitis, the main characteristic presentations of anti-LGI1 encephalitis could be summarized as follows:1. Serum or CSF LGI1 antibody; 2. Neuropsychic-disorder as memory decline, disorientation or hallucination; 3. Tonic seizure as FBDS; 4.Hyponatremia related symptoms. The LGI1 gene was discovered in the 1980s, which encodes a protein LRR (leucine-rich repeats) with conserved flanking sequences. In the LRR domain, LGI1 shares the highest homology with many transmembrane and extracellular proteins, and these proteins act as receptors and adhesion proteins[13]. LGI1 primarily expresses in neural tissues, particularly in the brain, which reduced in low-grade brain tumors as malignant gliomas[14, 15]. LGI1 modulates ADAM22 (a disintegrin and metalloproteinase domain-containing protein 22) or ADAM23 as a secretion protein[1]. LGI1 micro-rearrangements were observed in a collection of ADLTE (autosomal dominant lateral temporal epilepsy) families and sporadic LTE (lateral temporal epilepsy) patients and investigated novel ADLTE and LTE patients[16]. Anti-LGI1 limbic encephalitis was distinguished from VGKC antibody group encephalitis (anti-LGI1, Caspr2, or VGKC positive groups) with hyponatremia and typical FBDS[17-19]. In this report, the patient failed to show obvious abnormalities in MRI scans, including T1, T2, DWI scans(Figure 1), thus lacked the characteristic imaging presentations in the hippocampus[20], striatum[21] and mesial temporal lobes[22]. However, EEG based studies showed hippocampal functional dynamics changes beyond structural abnormalities[19, 23]. In this report, the sLORETA based source analysis showed frontal lobe sourced epileptic foci (Figure 1 C, D), which could shed light on the positioning diagnosis. There is no congruous treatment strategy for autoimmune encephalitis; the indicated and accepted therapies are immuno-therapies as steroids and sequenced immunoglobulins [24, 25]]. . In our report, the patient presented with typical features of anti-LGI1 limbic encephalitis, including cognitive defects, FBDS, hyponatremia, and had an excellent response to steroids and IVIG therapy [26, 27]. Hyponatremia, another characteristic symptom of LGI1 encephalitis patients, defined as serum sodium<135mmol> 30mmol/L, increased FEurate with normal renal, adrenal and thyroid function [29, 30]. There is accumulating evidence to utilize a new algorithm where FEurate is central to our evaluation of hyponatremia. In this algorithm, FEurate increases to >11% during hyponatremia in both SIADH and RSW. However, after the correction of hyponatremia, FEurate returns to normal in SIADH but remains increased in RSW(Supplementary Table 1). Our patient with paraneoplastic LGI1 encephalitis had increased FEurate after correction of hyponatremia when serum sodium was 141 mmol/L to meet the criteria for RSW where treatment with saline is most appropriate[30]. The etiology of RSW is most likely due to the presence of a circulating natriuretic peptide that is somehow up-regulated in diverse clinical conditions, and is not confined to those with cerebral disease. The natriuretic factor and increased FEurite  has its major effect in the proximal tubule where uric acid is exclusively transported[31-33]. In this patient, we traced the LGI1 origin in renal tissues for the clues of this phenomenon. Even there was no reported LGI1 expressions in the literature research, it is in accordance with previously reported mice LGI1 expressions in renal tubule in mice[34] It would be interesting to speculate that LGI1 may play a role in the development of the urology carcinoma, as it has been shown for the development of prostate cancer[35]. According to IHC staining studies of the kidney, LGI1 appears to exist more abundantly in the proximal tubule as compared to the distal tubule (Figure 2). The intensity of the staining is reduced when there are circulating antibodies in serum(Figure 2, A, C) as compared to the absence of circulating antibodies to LGI1(Figure 2, B, D). It is interesting that there was absence of staining in renal carcinoma cells with or without the presence of circulating antibodies (Figure 2 E, F). These data indicate that circulating antibodies in sera might be binding to extracellular LGI1 predominantly in the proximal tubule where the major defect in solute transport exists in RSW. 

Anti-LGI1 encephalitis with FBDS should be assessed not only structural but also functional measures as EEG based analysis. The sLORETA analysis could provide insight into the Anti-LGI1 encephalitis. Also, LGI1 encephalitis is a multisystem disorder that includes the renal tubule, which is manifested as a renal salt wasting syndrome due to a circulating natriuretic peptide and must be differentiated from SIADH because of opposed therapeutic goals. Anti-LGI1 encephalitis should be a syndrome that is not only confined to patients with encephalitis. However, this case report did not obtain FEurate before hyponatremia correction, the etiology of hyponatremia as RSW could be evaluated in a large sample of anti-LGI1 encephalitis with or without renal carcinoma, thus lead to the conclusion besides rare circumstances. 

LGI1 leucine-richanti-glioma1protein; FBDS faciobrachial dystonic seizure; RSW  renal Salt Wasting; CRSW cerebral/renal salt wasting; FEurate  fractional excretion of uric acid; SIADH  syndrome of inappropriate secretion of antidiuretic hormone; IHC Immunohistochemistry; MMSE minimental state examination; CSF cerebral spinal fluid; MRI magnetic resonance imaging; VGKC voltage-gated potassium channels; AMPA1 glutamate receptor, ionotropic, alpha 1; AMPA2 glutamate receptor, ionotropic, alpha 2; GABAR gamma aminobutyric acid receptors; NMDAR N-methyl-D-aspartate receptors; IVMP intravenous methylprednisolone; IVIG intravenous immunoglobulin; VPA valproic acid sodium; OXC oxcarbazepine; CRMPs amphiphysin, collapsin response mediator proteins; LRR leucine-rich repeats; ADAM22 a disintegrin and metalloproteinase domain-containing protein 22; ADLTE autosomal dominant lateral temporal epilepsy; LTE lateral temporal epilepsy; HLA human leukocyte antigen; SAH subarachnoid hemorrhage.

Ethics approval and consent to participate

Written Informed consent and ethical approval were obtained from the subject in accordance with the Helsinki Declaration.

Written Informed consent to participate and publication including clinical data, images and videos were obtained from the patient at 6-month follow up, when she had no orientation defections and estimated by 2 attending physicians in neurology department and her family to for the acceptable capacity to make decisions.

Availability of data and material

All data generated or analyzed during this study are included in this published article [and its supplementary information files]. The raw datasets used and/or analyzed during the current study are also available from the corresponding author on reasonable request.

Competing interests

All of the authors declare no conflict of competing interests.

Funding

This work was supported in part by grants from the Key Science and Technology Project of Hainan Province (ZDXM2015070) and (ZDYF2018141). The funding supported the collection, analysis, and interpretation of data.

Authors' contributions 

All authors listed have contributed sufficiently to the project to be included as authors, and all those who are qualified to be authors are listed in the author byline. Dr. YL M.D.& P.H.D. -acquisition of data, analysis and interpretation, critical revision of the manuscript for important intellectual content. Dr. ZCL M.D.-study concept and design, analysis and interpretation of data analysis and interpretation of data analysis and interpretation of data. Pro. JM. -conceptual guidance, anaylysis and interpretation, critical revision of the manuscript for important intellectual content. Dr. YLH M.D.- analysis and interpretation. Dr. XLX M.D.& M.S-acquisition of data. Pro. GQW M.D.- analysis and interpretation. Dr. QQ M.D.& P.H.D. -study concept, analysis and interpretation, critical revision of the manuscript for important intellectual content. Dr. TL M.D.& P.H.D.-study concept and design, analysis and interpretation of data. analysis and interpretation of data analysis and interpretation of data, critical revision of the manuscript and supervision of the study. Dr. SXH M.D.& P.H.D. -study concept, acquisition of data, analysis and interpretation, critical revision of the manuscript for important intellectual content, study supervision.

This study was supported by mutli-department cooperation. Including Dr. HB Quan, Dr. DH Cai, Dr. DJ Chen for their clinical care of the patient; Dr. Hongwen Zhu; Prof. Wei Qiu; Dr. Yun Zhong for the manuscript preparation.