Autoimmune inner ear disease.

Autoimmune inner ear disease (AIED) is a poorly defined condition, that may be primary, when the pathophysiology is limited to the ear itself. However it is estimated that 15–30% of AIED is not ear-specific and is instead part of a systemic autoimmune disease such as rheumatoid arthritis, systemic lupus erythematosus or Cogan syndrome [10]. There are likely many underlying pathologies, and it is worth noting that many cases do not respond to immunosuppression: in one study 65% of ears saw no change in their pure tone audiogram (PTA) (±5 dB) from baseline with steroid treatment [11]. To date no serological test has proven of significant value in defining AIED patients, and clinical diagnosis is typically made on the basis of a beneficial response to immunosuppressive therapy, and usually corticosteroids [12, 13].

There are several comprehensive reviews published that summarise the immunosuppressive agents that have been used for AIED [12–15]. Most data relate to small case series, and the variable populations and reported outcomes greatly limit the extent to which immunosuppression can be compared between agent or regimens.

The immunosuppressive agents used in identified studies for AIED are summarised in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Immunosuppressive agents in AIED.

https://doi.org/10.1371/journal.pone.0318165.t001

Steroids. Systemic steroids are often used as first line treatment for AIED, and have different activities that may benefit patients with AIED: anti-inflammatory, immunosuppressive, and anti-oedema effects. Immunosuppression reduces the formation of immune complexes, while anti-inflammatory and anti-oedema actions restore the normal calibre of affected vessels. Not all effects of steroids in AIED may therefore be relevant to immunosuppression within the labyrinth. The efficacy of oral steroids in improving hearing is reported to range from 50–70% [14] in initial treatment of AIED, though most patients become poorly responsive to steroid over time.

Steroids are commonly given by both systemic and intratympanic routes in clinical practice. However, the evidence for differences in route of administration is weak, and surpassed by work done in the field of sudden sensorineural hearing loss (detailed below).

Disease-modifying antirheumatic drugs (DMARDs). Various DMARDs are used in the treatment of AIED, with cyclophosphamide and methotrexate most commonly used, and azathioprine and mycophenolate mofetil also trialled. The low number of studies for each intervention and significant heterogeneity of population, intervention and outcomes, has meant that meta analysis has been of limited benefit. Gordis et al. included 12 retrospective and prospective studies in a metanalysis of a disease-modifying antirheumatic drugs as treatment of autoimmune inner ear disease [16]. Treatment included 187 patients receiving heterogenous mix of interventions including methotrexate (5 studies) and biologics (4 studies). DMARDs demonstrated statistically, though not clinically significant improvement in pure tone thresholds and a moderate improvement in speech discrimination thresholds.

Little high-level evidence has been generated. In an RCT setting, methotrexate was no more effective than placebo in maintaining the hearing improvement attained with prednisone treatment in patients with inner ear autoimmune disease [17]. Similarly, in the only other identified RCT for AIED (a pilot study), etanercept was not found to be superior to placebo in improving pure tone average [18].

Intratympanic methotrexate is suggested to be safe, with middle ear mucosal oedema the only change identified in a rat model [19].

Biologics. A 2022 systematic review [20] summarised the evidence for biologics in inner ear autoimmune disease. The review identified data relating to agents targeting three molecular targets, tumour necrosis factor-alpha (TNF-α, thought to be produced by resident immune cells in the cochlea), CD20, and interleukin-1 (IL-1, a key regulator of local inflammation in the cochlea). These mediators are involved in the recruitment of immunocompetent cells to the cochlear [21]. Of 14 studies included, all had fewer than 23 participants, and with considerable heterogeneity in study population, interventions and outcomes, the effect of biologics in included studies were highly variable [20].

Animal models studies have been supportive of etanercept, a TNF blocker, in reducing inner ear inflammation. Satoh et al. found TNFα and IL-1b generated within the cochlea during surgical trauma and inflammatory insult [22]. Systemic etanercept reduced both the number of inflammatory infiltrating cells and cochlear fibrosis. Wang et al. employed intratympanic (via osmotic pump remote to the ear) and systemic etanercept in a guinea pig model of immune-mediated hearing loss [23]. Regardless of delivery route there was significantly less histologic evidence of inflammation in the cochlea from etanercept-treated animals, and reduced hearing loss. Etanercept was found not to be ototoxic, even with direct cochlea injection. Etanercept has also showed its equivalence to steroids in an induced labyrinthitis model [24].

Looking to the wider field of transplantation, monoclonal antibody (mAb)-based therapies have become a staple of induction and acute rejection therapy in solid organ transplant patients [25]. A series of mAbs have been developed to prevent transplant rejection. The mechanisms of these mAbs are diverse, but all target specific cluster designation (CD) proteins on the T or B cell surface. These include mAbs against CD3, CD25 and CD52.

Route of immunosuppression administration in AIED. Currently no clinical study has compared the clinical efficacy of oral and intratympanic steroids in AIED, and individual studies have significant heterogeneity of population and outcomes.

In a guinea pig model of keyhole limpet hemocyanin (KLH)- induced labyrinthitis, dexamethasone, ciclosporin, prednisolone acetate, fluorouracil and tacrolimus were delivered to the round window membrane via intratympanic injection or osmotic minipumps [26]. None of the immunosuppressants were effective at reducing hearing loss or inflammation, with no significant differenced found in histological analysis. This is in contrast to the above work by Wang et al. using a similar model with etanercept [23].

Local (e.g. intratympanic) administration of immunosuppression may be less relevant to AIED than when considering stem cell transplants due to extra-cochlear effects of the condition. As an example, with SLE several mechanisms for hearing loss are identified, with some occurring within the labyrinth (antibody/ antigen direct reactions, cytotoxic action) and others extra-cochlear such as immune complex deposition in auditory artery leading to hypoxia [27].

Sudden sensorineural hearing loss.

The aetiology of sudden sensorineural hearing loss (SSNHL) is poorly understood, but inflammation is considered an important part of the pathophysiology in many cases, and the commonly used treatments for SSNHL in current clinical practice are designed to reduce inflammation. Although not a common condition, treatment outcomes for SSNHL have been well studied. Care must be taken when extrapolating clinical results as a marker of inner ear immunosuppression, as both the aetiology and the site of lesion in SSNHL is poorly understood, and some cases may represent lesions of the auditory nerve or other structures in the auditory pathway.

A Cochrane review explored the effect of intratympanic steroid on SSNHL [28]. Comparing intratympanic corticosteroids versus no treatment or placebo as secondary therapy, they found that intratympanic therapy may have a small effect on the change in hearing threshold, but appeared to be associated with a much higher proportion of participants whose hearing was improved (RR 5.55, 6 studies, low-certainty), and more favourable final hearing thresholds (mean difference -11.09 dB, 5 studies, low-certainty).

When Plontke et al. compared intratympanic corticosteroids versus systemic corticosteroids as primary therapy in their Cochrane review, they found no significant difference in any of; change in hearing threshold, the proportion of participants whose hearing was improved, or final hearing threshold (based on 7–14 studies, all conclusions of low-certainty).

The same Cochrane review compared intratympanic plus systemic corticosteroids (combined therapy) versus systemic corticosteroids alone as primary therapy. They found that combined therapy may have a small effect on the change in hearing threshold (mean difference -8.55 dB, low-certainty), but other benefits over monotherapy were very uncertain.

A systematic review and network meta-analysis published in 2019 [29] explored the outcome of different steroid therapies in SSNHL. They found that combination therapy, IT plus systemic steroids, demonstrated the largest difference in PTA improvement compared to placebo (25.85 dB, 95% CrI 7.18–40.58), followed by IV plus PO steroids (22.06 dB, 95% CrI 1.24–39.17), IT steroids (18.24 dB, 95% CrI 3.00–29.81) and finally PO steroids (14.7 dB, 95% CI 0.8–27.0). It was noted that many trials had unclear to high risk of bias.

Two systematic review-based studies by the same team have analysed several questions relevant to the technique and regimen for inner ear immunosuppression, exploring outcomes from primary [30] and secondary [31] treatment of SSNHL, utilising data from controlled and uncontrolled studies and mathematical modelling of cochlear drug levels. Intracochlear drug concentrations were calculated by a validated computer model of drug dispersion in the inner ear fluids based on the concentration and volume of glucocorticoids applied, the time drug remained in the middle ear, and on the specific timing of injections. 25 studies (28 treatment groups) were included in the secondary therapy review, 22 using dexamethasone, 6 using methylprednisolone. 30 studies (33 treatment groups) were included in the secondary therapy review, 19 using dexamethasone, 14 using methylprednisolone.

For primary and secondary intratympanic or combined therapy of ISSHL there was no dependence of change in hearing threshold (or the average final hearing threshold) on the drug used (dexamethasone or methylprednisonone), the concentration used, the number of injections, the frequency of injections, the estimated time of drug in the middle ear or the total duration of treatment.

Various explanations may account for these findings, including that steroid therapy may be ineffective for treating ISSHL, and spontaneous recovery is likely to mask some effects, thus highlighting the uncertainty with which the extent of immunosuppression can be extrapolated from outcomes in SSNHL data.

The agents, routes and doses of immunosuppression used in identified SSNHL studies and reviews is summarised in Table 2. No study used a non-steroid immunosuppression, though in some a non-immunosuppressant adjuvant was employed.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Immunosuppressive agents in SSNHL.

https://doi.org/10.1371/journal.pone.0318165.t002

Meniere’s disease. The aetiology of Menieres is not well understood, though steroids have proven efficiacy [32, 33]. Clinically employed immunosuppression for Menieres disease mirrors that of SSNHL, except where AIED is suspected. Only one study has directly compared local immunosuppressive treatments for Menieres disease in a randomised controlled trial [34], finding intratympanic methylprednisolone superior to intratympanic dexamethasone in improving hearing level, but not vertigo symptoms.

Cochlear implantation.

Inflammation following cochlear implant electrode insertion trauma can be associated with poorer outcomes in patients, and has been a focus of translational and clinical research. The evidence in this field provides a potential marker of inner ear immunosuppression efficacy, and implant electrodes also provide a mechanism for intracochlear drug delivery. Studies may also be relevant as stem cell transplantation may necessitate low-level trauma from cochleostomy or round window injection.

As noted earlier in this report, Gay et al. have summarised preclinical and clinical evidence for reducing inflammation associated with CI, concluding that the use of glucocorticoids leads to reduced fibrosis within the cochlea, and reduced electrode electrical impedance following cochlear implantation surgery [35].

A recent randomised human study of high dose methylprednisolone vs placebo has suggested that there is no additional benefit with respect to hearing preservation, though all patients also received intravenous dexamethasone at induction of anaesthesia, which may explain this effect [36]. A further human study of a single high dose of systemic prednisolone did not provide significant reductions in post procedure impedances, when compared to controls. The same group performed a further human study in which triamcinolone was injected via an intracochlear catheter prior to cochlear implantation, with demonstrated protection of residual hearing, suggesting that this may be a suitable route for the administration of other drugs [37].

Directly comparing administration routes, a small randomised controlled trial by Kuthubutheen et al. compared perioperative oral and intratympanic steroid delivery to a control group [38]. Subjects receiving transtympanic steroid had better PTA thresholds (125 to 8000 Hz) at three and twelve months compared with the control and oral GC group, but no differences were found in speech discrimination.

Many of the issues around periprocedural administration of glucocorticoids are discussed further in the review by Fuentes et al. [39]. This summarises several systematic reviews and trials, with the conclusion that there appears to be short and possibly long-term benefits to the use of glucocorticoids at the time of implant surgery, but there is not clear superiority of oral, intratympanic or combined routes of delivery, in large part due to a lack of adequately designed trials to demonstrate this.

Choice of immunosuppressive agent.

Ciclosporin has been the immunosuppressive agent of choice for cochlea stem cell transplant studies utilising immunosuppression, though the rationale for this choice is not elaborated in the identified clinical studies. Interestingly ciclosporin is one agent that has not previously been employed clinically for otological disease, and also has not been employed in any of the stem cell studies in allied areas.

Ciclosporin is a calcineurin inhibitor, supressing cell-mediated immune response through inhibiting production and release of lymphokines. It is clinically used to prevent allograft rejection, and is widely administered following solid organ transplant (kidney, liver, and heart). It has also attained common use in certain autoimmune conditions including psoriasis and rheumatoid arthritis. Ciclosporin is available for oral, intravenous or topical application (as eye drops). Monitoring is required (full blood count, renal function, liver function). In the short term mild adverse effects are common, and longer-term use is associated with increased risk of cardiovascular events and malignancies. Ciclosporin and other calcineurin inhibitors have been shown to exhibit minimal ototoxicity, though evidence is limited [51, 52].

Studies without immunosuppression.

Five intracochlear stem cell studies did not use immunosuppression and investigated intracochlear inflammation [40, 42, 45, 49, 50]. Several authors examined cochlea histologically and found no evidence of inflammatory cell infiltrate [40, 42, 45]. Chang et al. noted fibrotic tissue infiltration in the scala vestibuli in some cochlea, and lymphocyte aggregation and dilated blood vessels in tissue adjacent to the transplanted cochlea, but there was minimal cell response within the membranous labyrinth [50].

Eshraghi et al. describe the most thorough evaluation of inflammatory response to stem cell transplantation [49], relating to the use of rat mesenchymal stem cells transplanted via an intratympanic injection. Comparing stem-cell injected cochlear to controls, they found no evidence of enhanced intracochlear oxidative stress (determined by 8-isoprostane immunostaining), no increased activation of the caspase3 pathway, and no increase in the generation of TNF-α, IL-1β, IL-6 and IL-12. It should be noted that cell type and route of delivery notably different to other studies and are potentially less likely to trigger an immune response. Notably, mesenchymal stem cells (MSCs) present low immunogenicity, and undifferentiated MSCs express low levels of human leukocyte antigen (HLA) class I and II molecules [53]. MSCs hinder T cells from the contact with antigen presenting cells [54] and fail to stimulate allogeneic peripheral blood mononuclear cells or T-cell proliferation [55].

No studies utilised autologously harvested mesenchymal cells, although this has been proposed as an approach to eliminate the need for immunosuppression to prevent rejection. Overall study heterogeneity was high, and there was no scope for meaningful comparisons of outcomes between studies grouped by use of immunosuppression.

Experimental studies on the role of immunosuppression.

Only two studies have included methodology to experimentally assess the impact of immunosuppression in stem cell transplantation for hearing loss, and to date one has only been reported as a conference abstract. Lopez-Juarez et al. found that human otic progenitor cell (OPC) transplantation triggered an immune response in amikacin-exposed guinea pig cochlea [48]. They describe a moderate and limited inflammatory response after the injection human OPCs assessing cochlea up to 14 days post-graft. This reaction was assessed by the expression of IBA1, a marker of macrophages/microglia. IBA1-immunopositive cells appeared at day 4 post-transplantation in ototoxic-grafted cochleae, although the authors also note that IBA1-positive cells were upregulated in response to cochlear damage induced by ototoxicity, but this was more limited and focussed at the modiolus. In addition, double labelling revealed that some stem cells were immunopositive for IBA1, suggesting possible initiation of an immune response following OPC injection.

Lopez-Juarez et al. also found that the number of human OPCs and their migration increased in ciclosporin-pre-treated animals. In ciclosporin-treated animals, the distribution of labelled stem cells showed a gradient from base-to-apex with a higher number of labelled cells found at the basal and second cochlear turns in the ciclosporin treated. Labelled stem cells were also observed to migrate further from the basilar membrane, closer to the reticular lamina in the ciclosporin treated group. The authors conclude that transient and moderate immunosuppression may enhance the number of migrated human OPCs within the sensory epithelia.

Perhaps the most interesting study, which is only reported in abstract form, is by Hartmann et al. [56]. In this study the group appear to have transplanted otic progenitor cells to gerbils treated with ouabain to experimentally reduce spiral ganglion cells. Four different methods of immunosuppression were assessed. Two groups of gerbils were treated with single-agent systemic immunosuppression, either ciclosporin 15mg/kg subcutaneously daily or dexamethasone 40mg/kg subcutaneously weekly. The third group was treated with a combination of subcutaneous injection of ciclosporin and dexamethasone daily. The fourth group were treated with local immunosuppression alone in the form of a Gelitta sponge containing ciclosporin (5mg/0.1 ml) and dexamethasone (4mg/0.5ml) placed in the round window niche after stem cell injection. The authors report that preliminary results indicate that systemic immunosuppression is necessary to ensure successful transplantation of OPC, but no further details are given. The study authors were contacted for any updated data but failed to respond.

Summary.

Several animal studies utilising stem cell treatment of the cochlea have not employed immunosuppression, and investigators did not detect evidence of an immune or inflammatory reaction within the cochlea. Many of these authors concluded that the cochlea displays immunological privilege that is sufficient to allow xenotransplantation [45, 49, 50]. Where immunosuppression has been used in vivo, systemic ciclosporin treatment has been used, and steroids, whether locally or systemically delivered, have not been investigated. In the only studies designed to explore the effect of immunosuppression, the study by Lopez-Juarez et al. and preliminary report by Hartmann et al., findings support a role for immunosuppression. Future work should explore the effect of different immunosuppressive agents, and care is required to isolate the inflammatory response promoted by the experimental model insult or stem cell introduction method from that of the injected cells themselves to determine the long-term requirement for immunosuppression.