Biohybrid cochlear implants in human neurosensory restoration

Abstract

Background: The success of cochlear implantation may be further improved by minimizing implantation trauma. The physical trauma of implantation and subsequent immunological sequelae can affect residual hearing and the viability of the spiral ganglion. An ideal electrode should therefore decrease post-implantation trauma and provide support to the residual spiral ganglion population. Combining a flexible electrode with cells producing and releasing protective factors could present a potential means to achieve this. Mononuclear cells obtained from bone marrow (BM-MNC) consist of mesenchymal and hematopoietic progenitor cells. They possess the innate capacity to induce repair of traumatized tissue and to modulate immunological reactions.

Methods: Human bone marrow was obtained from the patients that received treatment with biohybrid electrodes. Autologous mononuclear cells were isolated from bone marrow (BM-MNC) by centrifugation using the Regenlab™ THT-centrifugation tubes. Isolated BM-MNC were characterised using flow cytometry. In addition, the release of cytokines was analysed and their biological effect tested on spiral ganglion neurons isolated from neonatal rats. Fibrin adhesive (Tisseal™) was used for the coating of silicone-based cochlear implant electrode arrays for human use in order to generate biohybrid electrodes. Toxicity of the fibrin adhesive and influence on insertion, as well on the cell coating, was investigated. Furthermore, biohybrid electrodes were implanted in three patients.

Results: Human BM-MNC release cytokines, chemokines, and growth factors that exert anti-inflammatory and neuroprotective effects. Using fibrin adhesive as a carrier for BM-MNC, a simple and effective cell coating procedure for cochlear implant electrodes was developed that can be utilised on-site in the operating room for the generation of biohybrid electrodes for intracochlear cell-based drug delivery. A safety study demonstrated the feasibility of autologous progenitor cell transplantation in humans as an adjuvant to cochlear implantation for neurosensory restoration.

Conclusion: This is the first report of the use of autologous cell transplantation to the human inner ear. Due to the simplicity of this procedure, we hope to initiate its widespread utilization in various fields.

Keywords: Biohybrid electrode; Bone marrow-derived mononuclear cells; Cochlear implants; Hearing loss.

Fig. 1

Gating strategy of depleted bone marrow samples for the detection of mesenchymal stem cells. a Flow cytometry-based characterisation of viable CD45^– cells were discriminated from CD45⁺ leukocytes in the depleted bone marrow (left histogram). The middle histogram is defined as a region to exclude the 7-aminoactinomycin D (7AAD)⁺ dead/apoptotic cells from 7AAD^– viable cells. The 7AAD⁺ cells are not further presented in the right dot plot, which displays an overview of all scatter event properties and shows differentiations (see CHECK CD45 ^neg) to non-specifically stained debris by low side (SS) and forward scatter (FS) signals. b In order to detect putative progenitor cells among viable CD45^– cells, singularized cells were stained with CD73, CD105, CD166, and CD90 (positive discrimination markers) and CD3, CD14, and CD34 (negative discrimination markers)

Fig. 2

Demonstration of the immunomodulating and neuroprotective secretome of BM-MNC in vitro and generation of biohybrid electrode arrays. a Proteomic analysis of bone marrow supernatant. Immediately after centrifugation of the bone marrow, the BM-MNC-containing plasma supernatant was obtained and stored ice-cold until proteomic analysis. Different cytokines, chemokines, and growth factors were present at biologically relevant concentrations in the supernatant. Among these, factors that promote wound healing and modulate and control neuroinflammation were identified. b Neuroprotective effect of BM-MNC. Surviving spiral ganglion neurons (SGN) were quantified for each treatment condition and compared to the medium (serum-free culture medium; neg. control) and to the positive control (medium supplemented with BDNF; pos. control). Supernatant was obtained immediately after centrifugation of bone marrow using the RegenKit-THT tubes from each patient. The term BM-MNC denotes the mononuclear cell fraction re-suspended in supernatant. Conditioned medium was obtained after 24 or 48 h (cond. med. 24/48) cultivation of the BM-MNC solution in serum-free culture medium. When compared to the negative control, significantly increased survival was determined in the positive control as well as after treatment with conditioned medium. c, d Biocompatibility of fibrin adhesive tested in well-established in vitro culture assays. Human MPC and SGN isolated from rodents were treated with fibrin adhesive and the survival was compared to the positive control (medium supplemented with BDNF) as well as to cultivation in medium without supplements (neg. control). Cell survival was not altered in the presence of fibrin adhesive when compared to the medium control. Values are presented as the mean with standard deviation. *p < 0.1, **p < 0.01, ***p < 0.001. FGF fibroblast growth factor, MNC mononuclear cells, MSC mesenchymal stem cells, ns not significant

Fig. 3

Determination of insertion forces in a cochlear model. a Electrode insertion for the measurement of force development due to implantation of a cell-coated electrode compared to uncoated electrode. Human-adjusted cochlear models were used. Electrodes were pre-inserted for 7 mm prior to the start of the measurements. This depth denoting the actual start of the registration of the insertion forces was defined as 0 mm. b At the maximum insertion depth of 12 mm, the electrode is regularly jollied. A total of five electrodes (two non-coated (CI) and three coated (biohybrid) cochlear implant electrodes) were used for the measurement of insertion forces. Each electrode was inserted repeatedly five times. c The insertion force was measured from 0 to the 12-mm insertion depth. Here, the mean forces measured at first insertion of each electrode without coating (blue) and with coating (red) are shown. There was no difference in the insertion forces between the coated and the non-coated electrodes. d One electrode was analysed for its stiffness compared to the coating, performing five measures without coating. Then, after coating of the same electrode with the fibrin cell solution, another measurement of insertion forces was performed showing that there are no increases in insertion forces due to the fibrin coating. e The first insertion of each biohybrid electrode (biohybrid) is depicted here (No.1–3) as well as the mean and standard deviation of all first biohybrid insertion forces. f The first insertion of each regular cochlear implant electrode (CI) is depicted here (No.1 and 2) as well as the mean and standard deviation of all first insertion forces. None of the electrodes showed increased insertion forces. g Each electrode is depicted here for direct comparison of the behaviour of uncoated and coated electrode. h The mean forces (including standard deviation) of each repeated insertion of biohybrid, as well as CI, electrodes show that there is no difference between the coated and the uncoated electrodes

Fig. 4

Live-staining of biohybrid electrodes. Light microscopy of the electrodes (I) depicting the fibrin layer. Live staining allowed the visualisation of single cells entrapped in the fibrin layer (II and III). Cells survived on the surface of the electrode without any migration as shown in the micrographs of the same area of the electrode array 3 (a), 7 (b), and 10 (c) days after coating

Fig. 5

Generation of a biohybrid electrode for human implantation and comparison of performance with standard (CI) and cell-coated electrodes (biohybrid). a Dipping procedure for the intraoperative coating of the electrode with a fibrin cell layer prior to implantation. The BM-MNC solution was mixed with the fibrinogen solution and used for the dip-coating of the electrode. A second dipping into the thrombin solution allowed the stabilization of the coating by conversion of the fibrinogen to fibrin. b Cone-beam computed tomography was utilized to check the electrode position immediately after insertion in the operating room. The micrograph shows the correct intracochlear position of the electrode array without any displacement. c (Left panels) Impedances and speech perception (monosyllables and numbers) of patient one. The patient received a MedEl Synchrony Standard electrode and showed an overall good performance; slightly impaired results on the side with long-term deafness that was treated with the biohybrid electrode are evident. (Middle panels) Impedances and speech perception (monosyllables and HSM sentence test) of patient two. The patient received a Cochlear Nucleus CI512 Profile. The results compared favourably to the contralateral side. (Right panels) Impedances and speech perception (monosyllables and numbers) of patient three. The patient received a Cochlear Nucleus CI512 Profile electrode. His results with the biohybrid electrode exceeded expectations, taking into consideration the presence of peri/prelingual idiopathic deafness on the side that was treated with the biohybrid electrode. OP operation, 2D initial test (only impedances) on the second day after surgery, 5 W first fitting week performed 5 weeks after operation, 5 M control testing 5 months after operation, respectively about three months after the first fitting