WO2024160819A1

WO2024160819A1 - Nanoparticles with antibodies for ocular treatment

Info

Publication number: WO2024160819A1
Application number: PCT/EP2024/052232
Authority: WO
Inventors: Regina RODRIGO NICOLÁS; José Luis Pedraz Muñoz; Idoia GALLEGO GARRIDO; Gustavo PURAS OCHOA; Lorena OLIVARES GONZÁLEZ; Sheyla VELASCO GOMÁRIZ
Original assignee: Consorcio Centro De Investigación Biomédica En Red; Fundación De La Comunidad Valenciana Centro De Investigación Príncipe Felipe; Universidad Del Pais Vasco / Euskal Herriko Unibertsitatea
Priority date: 2023-01-30
Filing date: 2024-01-30
Publication date: 2024-08-08

Abstract

The invention refers to lipid nanoparticles in the form of nanostructured lipid carriers which comprise a matrix with at least a lipid which is solid at room temperature and at least a lipid which is liquid a room temperature, at least a surfactant and an antibody such as Adalimumab entrapped in the nanoparticle. The use of such nanoparticles for the treatment of an inherited retinal dystrophy, such as retinitis pigmentosa, allows the administration of antibodies against TNFα such as Adalimumab or biosimilars protected by nanoparticles which reach the posterior part of the eye, so that the nanoparticles of the present invention enable the use of the ocular topical route for the administration of the nanoparticles. Such administration reduces signs and symptoms of retinitis pigmentosa such as retinal inflammation and vision loss, ameliorating retinal function.

Description

NANOPARTICLES WITH ANTIBODIES FOR OCULAR TREATMENT

FIELD OF THE INVENTION

The present invention relates to the field of nanoparticles for clinical use. More particularly, the invention relates to lipid nanoparticles loaded with antibodies against TN Fa for ocular treatment.

BACKGROUND OF THE INVENTION

Inherited retinal dystrophies (IRDs) are a group of diseases which affect retina (the light sensitive layer of tissue in the back of the eye) and that cause progressive loss of vision. Retinitis pigmentosa (RP) is the main I RD. RP is the main genetic cause of blindness in developed countries, affecting about two million people. It is a rare disease (1/4,000), mainly associated with photoreceptor (PR, rods and cones) dysfunction and their loss that, eventually leads to blindness. RP has a high genetic (more than 125 genes) and clinical heterogeneity which makes difficult it to find a correct treatment. To date, there is only one treatment for RP, Luxturna®. Luxturna® is a gene therapy product for patients with mutations in both copies of the RPE65 gene, which provides a working RPE65 gene which allows those cells in the retina can produce RPE65 protein, thus having the potential to make the visual cycle work properly again.

Several studies indicate that chronic and sustained inflammation may contribute to photoreceptor cell death and, consequently, to visual impairment. In patients with RP and murine models of RP, the activation of microglia, a marker of neuroinflammation, as well as the upregulation of different inflammatory markers such as tumour necrosis factor alpha (TNFa), interleukin 6 (IL-6), IL-1 , have been reported.

In previous studies, the group of the present inventors reported an increase of TNFa in the aqueous humour (Martinez-Fernandez de la Camara et al., J of Neuroinflammation, 2014, 11 :172) and blood (Olivares-Gonzalez et al., Frontiers in Nutrition, 2022, 9:847910) of patients, as well as in the retina of RP model mice (rd10 mice) both at short postnatal times (P13-P20) (Martinez-Fernandez de la Camara et al., Scientific Report, 2015, 5:11764) and longer times (Olivares-Gonzalez et al., The FASEB Journal, 2018 vol. 32(5):2438-2451). TNFa is a pleiotropic cytokine which binds to its TNFR1 or TNFR2 receptors, triggering various cellular processes such as inflammation, differentiation, proliferation, survival and cell death. After binding to its receptors, depending on the activation of one or the other signalling complexes, different cellular pathways can be activated. Excepting the specific gene therapy treatment above mentioned, there are no standard treatments for RP. Since TNFa has an important implication in this disease, anti-TNFa biological products ("biologies") such as monoclonal antibodies have been studied, in order to prevent the binding of TNFa to its TNFR1 and TNFR2 receptors. Antibodies that prevent the binding of TNFa to its TNFR1 and TNFR2 receptors and their subsequent biological actions will be referred hereinafter as TNFa antibodies.

One of said TNFa antibody is infliximab, a chimeric monoclonal antibody which has been shown to reduce retinal degeneration in cultures of porcine retina (Martinez-Fernandez de la Camara et al., J of Neuroinflammation, 2014, 11 :172). Another TNFa monoclonal antibody is adalimumab (ADA). ADA is a TNFa antibody that alters the immune response by reducing the ability of cells to recognise the cytokine TNFa. Its therapeutical use was approved by the FDA in December 2002. ADA is indicated for reducing the symptoms of various chronic human inflammatory diseases such as rheumatoid arthritis, psoriatic arthritis, Crohn's disease, ankylosing spondylitis, etc. In ophthalmology, several preclinical studies and clinical trials show its effectiveness in the treatment of severe ocular inflammation such as Behcet's disease, posterior uveitis and panuveitis.

The use of infliximab and adalimumab, as happens with other monoclonal antibodies, is limited, in general, by the high doses required to achieve a therapeutic effect and the risk of associated secondary effects, particularly when the systemic route is used. Infliximab and adalimumab, for instance, are immunosuppressant agents whose use increases the risk of suffering from latent diseases such as tuberculosis. Even though adalimumab is often administered by subcutaneous injection when used, for instance, against rheumatoid arthritis or uveitis, it is advisable to discard a latent tuberculosis before beginning the administration and avoid vaccination with live vaccines during treatment.

In animal models of RP, intraperitoneal or intravitreal administration of ADA has been reported to slow down retinal degeneration in the murine model of RP, the rd10 mice (Martinez-Fernandez de la Camara et al. Sci. Rep 2015, 5:11764; Olivares-Gonzalez et al., The FASEB Journal, 2020, 00:1-23). A clinical trial is ongoing to assess safety and effectiveness of repeated intravitreal injections of ADA in RP patients (ADARET study).

Notwithstanding, it is well known that repeated intravitreal administrations of drugs may cause side effects including endophthalmitis, patient discomfort, etc., what makes advisable to find alternatively routes of administration. The use of the ocular topical route (also known as ophthalmic route), however, is limited by the low bioavailability of the drugs administered by this route, as discussed below. This is particularly the case of monoclonal antibodies, which require high doses to achieve a therapeutic effect, due to their relatively high size and hydrophilic nature, which make difficult for them to reach the target tissue, especially when it is located in the posterior part of the eye. All this leads to particularly low availability and tissue distribution of antibodies administered by the ocular topical (ophthalmic) administration route is used. And high doses predispose to the appearance of hypersensitivity reaction, common side effects of monoclonal antibody therapies. Therefore, there is a need of developing alternative forms of antibody administration for the treatment of ocular diseases by using the most desirable administration route, the topical route, which is the easiest-to-use and most comfortable route for the patient but, also, the one that poses more challenges to achieve that enough amount of antibodies (or other drugs) reaches the desired tissue, especially if it is located in the posterior part of the eye.

Drug Delivery Systems (DDS) are having a huge impact on medical technology by improving the mode of drug delivery. DDS protect drugs from degradation, improve penetration across biological barriers, increase circulating half-life and stability, and allow sustained and controlled drug release. In recent years, DDS have emerged as an alternative to conventional dosage forms and have proven to be promising candidates for drug encapsulation and application in ocular diseases. Among the encapsulating systems are of great importance, particularly for ocular treatment, nanotechnological systems, those systems where the matter is at the level of nanoscale, that is, those systems with matter of dimensions with values between 0.1 and 1000 nanometers approximately. Nanoparticles in particular are a type of particulate system having a size usually ranging from 10 to 1000 nm.

There are different types of DDS, most of them being based on nanoparticles of different matters or compounds which are linked to and/or encapsulate molecules of interest, such are drugs. Nanoparticles are capable of transporting drugs to tissues based on their physicochemical characteristics. Nanoencapsulation of monoclonal antibodies, for instance, represents an effective strategy against their degradation and to increase their half-life. The controlled release of monoclonal antibodies from these nanodevices also leads to decrease the effective dose and, in turns, toxicity and costs of the treatment.

Drug therapies directed to the eye must take into account the anatomy of the eye. In order to design effective therapies, particularly when they are intended to avoid nuisance to the patients caused by injections (as occurs in the intravitreal and subretinal routes) by using the direct ophthalmic administration (also called topical route), it must be taken into account that one of the biggest challenges in ophthalmic treatment is achieving that enough amount of drug arrives to the eye area where the drug should take effect even though an important amount of the drug administered by the direct ophthalmic (topical) route is usually drained through the tear duct towards the back of the nose for elimination. Also, an important point is that the delivery system must facilitate contact with the ocular surface and an appropriate release of the drug so that it can arrive and act in the appropriate area of the eye. For that, it is meaningful that the thickness of the cornea is around 550 pm in the center and 600 pm in the periphery and that it is a tissue composed of 5 layers: epithelium, Bowman's layer, stroma, Descemet membrane and the endothelium, the epithelium and the stroma being the rate limiting layers for transcorneal permeation of drugs because of the tight junctions of the epithelium (barrier to hydrophilic molecules) and the collagen fibers of the stroma (barrier to lipophilic molecules).

Nanoencapsulation strategies for ocular delivery of monoclonal antibodies have been proposed as a way of increasing antibody availability, and controlling their release kinetics, in general, and as a way of overcoming limitations related to ocular delivery, since a sustained release would decrease the number of intravitreal and sub-retinal administrations, improving the patient's quality of life.

Polymeric nanoparticles with encapsulated adalimumab, mixed with rituximab and trastuzumab have been developed (He et al., J Biomed Nanotechnol. 2020 Aug 1 ; 16(8): 1254-1266, doi: 10.1166/jbn.2020.2966, PMID: 33397555) and have been assayed in vivo to evaluate their potential use for rheumatoid arthritis treatment, a study where tissue biodistribution and pharmacokinetics were determined as the proof of concept of their therapeutic efficacy. The study concluded that the mentioned particles have an increased therapeutic effect than the conventional drug. However, it has been reported that this type of nanoparticles is characterized by possible toxicity and aggregation (Mitchell MJ et al., Nat Rev Drug Discov. 2021 Feb;20(2):101-124. doi: 10.1038/s41573-020-0090-8, PMID: 33277608).

The encapsulation of antibodies in liposomes for eye disease treatment can be found described or suggested in some publications, although there is usually another compound or structure that facilitates the delivery to and release at the posterior part of the eye and/or another therapeutic agent which is considered a better therapeutic agent for such purpose, for which the suitable formulation is described. For instance, in the international patent application WO 2022/187607 A1 , pharmaceutical compositions for the delivery of an active agent to an eye are disclosed. The compositions comprise a lipid mixture and a biocompatible hydrogel. The lipid mixture is said to be a mixture of one or more phospholipids and, optionally, a sterol like cholesterol. The lipid mixture can be in the form of liposomes (which, together with micelles, are the only kind of lipid nanoparticles mentioned in the application WO 2022/187607) or can be directly prepared as a lipid mixture.

The main object of the invention disclosed in WO 2022/187607 is avoiding the problems of blurred vision and turbidity observed when compositions comprising large unilamellar vesicles (unilamellar liposomes) with high encapsulation efficiency are administered to the vitreous humour. Moreover, it is mentioned, as something well known for those skilled in the art, that liposomes may modify the pharmacokinetics of the active agents, considering that a major concern of safety when developing a liposomal drug, which is tried to be avoided.

In order to reduce all such problems, the pharmaceutical composition of WO 2022/187607 must comprise, additionally to the lipid mixture and the active agent, a biocompatible hydrogel, which hydrogel can comprise at least one polysaccharide. The pharmaceutical compositions of WO 2022/187607, thanks to the presence of the hydrogel, are said to solve the undesired side effects linked to turbidity in the vitreous humour, exhibiting sustained release of the active agent and maintaining prolonged efficacy without undermining safety. Among the possible active agents that can be delivered with the pharmaceutical compositions of the invention of WO 2022/187607, adalimumab is mentioned among many others compounds, not indicating or suggesting the lipid composition, suitable proportion of active agent in the formulation or the preparation method that could be used for adalimumab delivery to the eye, nor even for intravitreal administration, which is the administration route towards which all the invention is oriented. Other possible administration routes, including topical administration, are listed in one paragraph, but no mention of the particular embodiments of the invention that could be more appropriate for such route can be found thereinafter.

The pharmaceutical compositions of WO 2022/187607 are for use in the treatment of ophthalmic diseases or disorders, retinitis being one of them.

According to WO 2022/187607, the method of producing the lipid nanoparticle formulation can influence and/or dictate distribution of certain components within the lipid nanoparticles and that distribution can influence and/or dictate physical (e.g. stability) and/or biological (e.g. efficacy, intracellular delivery, immunogenicity) properties of the lipid nanoparticles.

For the preparation of liposomes, the use of techniques considered conventional is indicated, such as the ether injection method, the surfactant method, the freeze-thaw method, the reverse phase evaporation method, the ultrasonic treatment method, the ethanol injection method, the extrusion method, the French press method, and others. After sterilization, the liposomes can be lyophilized to form a powder or a cake. In preparatory Example 1 , a mixture of the phospholipids DOPC (1 ,2-dioleoyl-sn-glycero-3- phosphatidylcholine) and DOPG (dimyristoylphosphatidylcholine) and also cholesterol is dissolved in absolute ethanol, in an ultrasonic bath, induced to form liposomes by adding sodium phosphate solution, extruded to form a liposome suspension, dialyzed and concentrated by filtration and sterilized also by filtration. After adding cryoprotectant, a lipid cake is obtained by lyophilization.

Liposome-drug compositions are prepared in the Examples of WO2022/187607 from a lipid cake including the phospholipids DOPC and DOPG and also cholesterol and subsequent reconstitution with an active agent. Dexamethasone sodium phosphate (DSP) and bevacizumab are presented as the exemplary active agent, although bevacizumab is only used in a preparation Example, and the tests of viscosity determination, dispersion behaviour, turbidity analysis and in vitro release of the active agent are carried out comparing liposome-DSP compositions with HA/liposome DSP compositions (compositions with hyaluronic acid as a component of the hydrogel). All assays are in vitro assays, not in vivo assays with any animal model or even samples taken from said animal.

In other approaches, the presence of a cell penetrating peptide (CPP) is a necessary element of drug-delivery systems designed for treating eye disorders. For instance, United States application US2019/0015521 discloses a transepithelial, transmembrane or transmucosal drug-delivery systems designed for treating eye disorders (particularly age- related macular disease), comprising a therapeutic agent and a CPP. Several possible ways of administering the CPP and a therapeutic agent are proposed (for instance, mixed, non- covalently associated or covalently bonded to a CPP), one of them being administering the therapeutic agent encapsulated in CPP-conjugated nanoparticles, micelles or liposomes. Regarding their composition, it is mentioned that, in certain embodiments, micelles or liposomes are composed of phospholipids, comment that is followed by a list of examples of phospholipids. However, none of the 17 Examples of the US application includes the use of any lipid nanoparticle (micelle, liposome or any other one) or its possible specific composition.

Eye drops are mentioned in US20190015521 as one of the possible ways of administering the system, which system is considered to be capable of delivering the therapeutic agent into the posterior segment of the eye. But only the assays of Examples 2 -4 are carried out with eyes (in living animals or having been resected from dead animals) receiving eye drops; the drops contain CPPs mixed, non-covalently or covalently associated to a therapeutic agent, but not included in particles. Adalimumab is mentioned in US20190015521 as one of the many therapeutic agents that could be used as a part of the delivery system, as an example of inhibitors of TNF-a. Example 11 describes the CPP-assisted delivery of adalimumab across an egg shell membrane, as a model of topical delivery to an eye, where adalimumab is administered alone or mixed with hexa-arginine as CPP, but not included in any kind of nanoparticle. The results indicate that more adalimumab crosses the egg shell membrane when it is mixed with hexa-arginine, but no additional assay is carried out to verify any possible therapeutic effect of the amount having crossed the egg shell membrane.

Other approaches, like the one disclosed in WO2017/184427 have addressed the problems associated with delivery of nucleic acids to a subject as therapeutic agent, trying to avoid complex lipid formulations, association to charged peptides or polymers or the use of solid structures and the problems associated with such delivery formulations. For that, as disclosed in WO2017/1844227, the active agent can be provided linked to the core of a spherical nucleic acid (SNA) (which is the structure which facilitates the delivery to the back part of the eye) or as a stable self-assembling three-dimensional nanostructure formed spontaneously by a single oligonucleotide with therapeutic effect, for instance an antisense oligonucleotide (optionally with chemical modifications) which is an inhibitor of TNF-a; the latter alternative does not require further components to be an effective therapeutic delivery formulation and can be free of lipids, polymers or solid cores. Also suggested in the above mentioned international application is the association of an antisense oligonucleotide to a solid or a hollow core, such as a liposomal core or a niosome, Liposomes are defined in WO2017/184427 as artificial, self-closed vesicular structures of various sizes and structures, where one or several membranes encapsulate an aqueous core; niosomes are defined as vesicles formed from non-ionic surfactants, rather than lipids, oriented in a bilayer. Adalimumab is mentioned only once, among other antibodies, as an example of the antibodies which can be included in a liposome, but no specific example of the composition of a liposome appropriate for the delivery of such antibody is included in the above patent application. The only Example of application WO2017/184427 shows the administration to rabbits of eye drops containing TNF-a antisense oligonucleotide SNAs lacking any lipid apart from a cholesterol moiety linked to the 3’-end of the oligonucleotides.

Nanostructured lipid carriers (NLCs), are a specific type of DDS that have emerged as a second generation of lipid nanoparticles, that derive from the first-generation ones, known as solid lipid nanoparticles (SLN). NLCs consist of a lipid core formed by a mixture of liquid and solid lipids that are stabilised by surfactant compounds, which mainly favour the possibility of encapsulating lipophilic and hydrophilic drugs. In addition, NLCs have further advantages such as long-term stability, sustained and controlled drug release, higher drug loading, and biocompatibility.

Encapsulation of monoclonal antibodies, mainly bevacizumab (an anti-angiogenic, specifically anti-VEGF, antibody), has been described using mainly polymeric, inorganic and liposomal nanoparticles. In one study (Chirio et al., Pharmaceutics. 2021 Apr 15;13(4):560, doi: 10.3390/pharmaceuticsl 3040560, PMID: 33921167; PMCID: PMC8071554) lipid nanoparticles were used to encapsulate monoclonal antibodies, in particular Bevacizumab. However, this study failed to test its effectiveness either in vivo or in vitro.

With this background, an effective treatment for RP, which could be administered directly by the ocular topical (ophthalmic) route, in the form of drops, without having to use the intraperitoneal, subcutaneous or intravitreal route, avoiding nuisance, undesired side effects and risks associated to said routes, and also avoiding the need of administering high amounts of drug to be effective, particularly when the drug is an antibody, is still pending to be found.

The present invention provides a solution to said problem.

SUMMARY OF THE INVENTION

The present invention, in a first aspect, relates to a lipid nanoparticle comprising: a) a lipid matrix comprising a mixture of lipids and/or molecules of lipophilic nature comprising a.1) at least one lipid whose melting point is equal or greater than 25°C, a.2) at least one lipid whose melting point is less than 25°C, b) a TN Fa monoclonal antibody entrapped in the lipid matrix, c) at least one surfactant.

The above disclosed formulation is the formulation of a nanoparticle with the structure of a nanostructured lipid carrier with an active agent, a TNFa antibody, entrapped in the lipid matrix. Thus, in the first aspect of the invention, it relates to nanoparticles which have the structure of nanostructured lipid carriers with TNFa antibodies entrapped in the lipid matrix.

In a second aspect, the invention also relates to a pharmaceutical composition comprising at least one lipid nanoparticle of the present invention, which composition is in the form of a powder or as an aqueous dispersion. In a third aspect, the invention additionally relates to a method for the preparation of a lipid nanoparticle of the first aspect of the present invention, which comprises the following steps:

(i) preparing a mixture of lipids whose melting point is equal or greater than 25°C, lipids whose melting point is less than 25°C and the freeze-dried antibody adalimumab by heating to a temperature slightly above the melting point of the lipid whose melting point is equal or greater than 25°C;

(ii) preparing an aqueous solution comprising one or more surfactants,

(iii) heating the aqueous solution obtained in (ii) up to the same temperature than the lipophilic solution with adalimumab (ADA) prepared in (i),

(iv) adding the aqueous solution obtained after performing step (iii) to the lipophilic solution prepared in (i), subjecting the resulting mixture to sonication at 20W to 50 W and for 10 seconds to 45 seconds, to obtain an emulsion,

(v) cooling down the emulsion obtained in step (iv) firstly at 15°C-25°C for at least 1 hour under magnetic stirring and, subsequently, at refrigeration temperature to allow lipid recrystallization and nanoparticle formation,

(vi) washing the nanoparticles obtained in step (v) with water combined with centrifugal ultrafiltration for recovering the nanoparticles,

(vii) lyophilizing the nanoparticles with a cryoprotectant.

Finally, in an additional aspect, the invention also refers to a nanoparticle of the present invention or a pharmaceutical composition of the present invention for use in the treatment of a retinal degenerative disease by the ophthalmic route.

DESCRIPTION OF THE FIGURES

Fig. 1 shows schematic representations of a liposome (panel A), a SLN (panel B) and a NLC (panel C).

Fig. 2 refers to the characterization of the nanoparticles. Microphotographs obtained by transmission electron microscopy (TEM) of empty NLCs (left side) and NLCs where ADA is entrapped NLC-ADA, where the bars on the bottom right part represent 500.0 nm length (a). Mean diameter (left y-axis) and Zeta Potential (right Y-axis) of the same nanoparticles, empty NLCs (right, white-filled bar) and NLC-ADA (left, grey-filled bar) (b).

Fig. 3 shows the results of ADA release from NLC-ADA nanoparticles along time.

Fig. 4 shows charts and photographs corresponding to different in vitro studies with 661 W cells: charts related to studies of cell toxicity (a), protection against TNFa of ADA-loaded NLC (NLC-ADA particles) (b) and cellular uptake of coumarin6-loaded NLC (c) and photographs of the latter study (d). Fig. 4a shows the percentage of cell viability over time after treatment with NLC-ADA at different concentrations as indicated in the panel, where the upper line (completely filled circles) is the one corresponding to NLC-ADA at 12.5 pg/mL and the lower line (dotted line with partially filled circles) the one corresponding to NLC- ADA at 125 mg/mL, while the intermediate line (white-filled circles) corresponding to NLC- ADA at 25 mg/mL Fig. 4b shows the percentage of cell viability after 6 hours of treatment with TNFa, a cell inducer, and NLC-ADA, calculated with regard to the control, "C" (value of 100%, represented by the darker bar of the left); One-way ANOVA test followed by Tukey's multiple comparisons test were used to compare all groups ****p<0.0001 for differences between control and treated cells; #^<0.01 ; ^###p<0.001 for differences between TNFa-treated cells and NLC-ADA-treated cells.. Fig. 4c shows the percentage of cells having internalized Coumarin6-loaded NLC (NLC-Cou6) after incubation at different nanoparticles concentrations, as indicated in the panel, at different time points. Kruskal- Wallis test followed by Dunn's multiple comparisons tests were used to compare control cells vs treated cells **p <0.01 ; ***p<0.001. Data were presented as mean ± standard error of the mean (SEM). Fig. 4d shows significant images of fluorescence analysis of cells (scale bar, visible in the last photograph of each group: 10 pm) having been submitted (or not, as in the photographs of the control samples of the first column on the left) ) to incubation with different concentrations of NLC-Cou6 (Coumarin 6 fluorescence shown in the bottom group of photographs, labelled "NLC-Cou6) during different times after DAPI counterstaining (DAPI fluorescence shown in the upper group of photographs, labelled "Nuclei".

Fig. 5 illustrates the effect of different concentrations of eNLCs or NLC-ADA on the haemolysis of erythrocytes isolated from porcine blood, as it is schematically represented in Fig. 5a. Fig. 5b shows values of haemolysis relative to positive control (distilled water) and Fig. 5c shows denaturation index (DI) of haemoglobin relative to positive control (SDS, sodium dodecyl sulphate). Data were presented as mean ± standard error of the mean (SEM). One-way ANOVA test followed by Tukey's multiple comparisons test for haemolysis and Kruskal-Wallis test followed by Dunn's multiple comparisons tests for DI were used to compare control cells vs treated cells *p < 0.05; **p < 0.01 ; ****p < 0.0001 for differences between control and treated cells; ##p<0.01 for differences between SDS-treated cells and NLC-ADA-treated cells.

Fig. 6 shows representative photomicrographs of retina sections showing TUNEL-staining (lower row of images, showing green signals in the original photomicrographs) and DAPI- counterstaining (upper row of images, showing blue signals in the original photomicrographs) in organotypic cultures of porcine retinas exposed to different NLC-ADA concentrations (as indicated in the lower part of each image) for 24h (scale bar: 50 pm).

Fig. 7 illustrates the assays related to corneal toxicity of NLC-ADA in corneas from C57BL/6J mice through fluorescein staining and detection of died cells by the TUN EL technique. Fig. 7a shows representative photographs (green colour in the original) of the mice corneas with fluorescein staining at different times (indicated over the photographs) after PBS (first row of photographs) or NLC-ADA (second row of photographs, labelled INS NLC-ADA on the left side) after instillation, while Fig. 7b shows a line chart showing the evolution of the score of corneal staining for each treatment. Fig 7c shows representative photomicrographs of corneal sections showing TUNEL-stained (green colour in the original) and DAPI-counterstained (blue colour in the original) sections, scale bar: 50 pm). Data were presented as mean ± standard error of the mean (SEM). Unpaired t- test or Mann-Whitney test was used to compare PBS vs NLC-ADA corneas at different times. *p < 0.05. At least six mice were analysed for each group.

Fig. 8 illustrates ex vivo experiments in porcine corneas, where corneal permeation of Coumarin 6-loaded NLC (NLC-Cou6) has been determined. Fig. 8a is a schematic representation of a section of a pig eye and how the dispersion with the NLC-Cou6 were administered to the extracted pig eyes by instillation (ocular topical administration), with augmented representation of the NLC-Cou6 (labelled as NLC-fluo, referring to the fluorescence associated to Coumarin6) contained in the drops. Fig.8b shows microscopy images of corneal tissue cryosections fixed after incubation with NLC-Cou6 and counterstained with DAPI (blue colour in the original), where distribution of the NLC-Cou6 (green colour in the original) to the epithelium and the stroma can be observed in all NLC- Cou6 treated samples, in higher amounts with longer incubation times, presence in the stroma being particularly evident tin the samples corresponding to 120 minutes of incubation (lower three photographs) (scale bar: 100 pm). Fig. 5c shows the percentage of coumarin6 fluorescence detection in aqueous humour samples with regard to NLC-Cou6 exposure time.

Fig. 9 shows the ocular distribution of Coumarin 6-loaded NLC (NLC-Cou6) in whole eyes (group of photographs on the left) (scale bar: 300 pm), cornea (intermediate group of photographs) (scale bar: 30 pm), and retina (group of photographs on the right) (scale bar: 30 pm), after in vivo studies in control (C57BL/6J mice). With different time exposures (as indicated in the images themselves) to NLC-Cou6 administered by the ophthalmic route (instillation). Fig. 10 shows dot blot assay with anti_Adalimumab (ADA) antibody from retinal homogenates incubated with or without protein A-sepharose (a). PBS, 107 mg/mL NLC- ADA and retina without ADA/NLC-ADA were used as controls (b). Spots indicated positive signals for ADA detection in retinal samples after intravitreal administration of ADA alone at two concentrations (0.375 and 4.61 g/uL) and after ophthalmic administration of 0.390 NLC-ADA.

Fig. 11 shows the ERG results obtained after topical application of NLC-ADA improved visual function in rd10 mice at postnatal day (P) 18. Amplitudes of ERG b-wave are represented in panels a and b, where the lower curves of data correspond to untreated mice, the curves represented over those ones correspond to mice treated with NLC-ADA by instillation and the upper curve of data of panel a corresponds to control mice); a- and b- wave implicit time or latencies recorded are represented in panels c and d, where the bar on the left of each group of three correspond to the data obtained from dark-adapted control mice, the intermediate ones correspond to untreated rd10 mice and the bars on the right of each group of three correspond to NLC-treated rd10 mice (INS NLC-ADA-treated rd1O) at different intensities of light stimuli. Rd10 mice were every two days treated with NLC-ADA from P12 to P18. Mann-Whitney test were used to compare rd10 vs INS NLC-ADA-treated rd10 mice. ^#p <0.05; #^<0.01 ; ^###p<0.001 ; ^####p <0.0001. Data were presented as mean ± standard error of the mean (SEM). At least nine mice were analysed for each group.

Fig. 12 shows how topical application of NLC-ADA improved visual function in rd10 mice at postnatal day (P) 18. Light/dark box test was performed to examine the light aversion of mice, using a light/dark box as represented in Fig. 11a. rd10 mice were every two days treated with NLC-ADA from P12 to P18. Data were presented in Fig. 11 b as mean ± standard error of the mean (SEM), the first bar of each group (light or dark) corresponding to control mice (C57BL/6J mice), the second to rd10 mice at P18 and the last (right) one to the rd10 mice having received NLC-ADA by instillation (group INS NLC-treated rd10 P18). At least nine mice were analysed for each group light/dark box test for these three groups. One-way ANOVA test followed by Tukey's multiple comparisons tests were used to compare three groups **** p <0.0001 for differences between control and rd10 mice or INS NLC-ADA-treated rd10 mice ; ^### p< 0.001 for differences between rd10 and rd10+ NLC- ADA mice.

Fig. 13 shows how topical application of NLC-ADA ameliorated photoreceptor degeneration in rd10 mice at postnatal day (P) 18. Quantification of number of rows of nuclei at ONL (panel a) and representative photomicrographs of retinal sections showing DAPI staining (panel b); Quantification of PAR-positive cells (panel c) and representative photomicrographs of retinal sections showing PAR staining (panel d); Gene expression of Ripkl, Nlrp3, 1118 and Asc in retinal homogenates (panel e) from control mice, untreated rd10 mice and NLC-treated rd10 mice (INS NLC-ADA-treated rd10) are shown. Scale bar: 50 pm. ONL: outer nuclear layer. One-way ANOVA followed by Tukey's multiple comparisons test, *p <0.05; **p<0.01 ; ***p<0.001 ; **** p <0.0001 for differences between control and rd10 mice or INS NLC-ADA-treated rd10 mice; ^#p< 0.05; #*p < 0.01 ; ^###p<0.001 for differences between rd10 and INS NLC-ADA-treated rd10 mice. For the analysis of Nlrp3 and 1118 gene expression Kruskal-Wallis test followed by Dunn's multiple comparisons tests was used. Data were presented as mean ± standard error of the mean (SEM). At least nine mice were analysed for each group.

Fig. 14 shows that topical application of NLC-ADA ameliorated retinal inflammation in rd10 mice at postnatal day (P) 18. Quantification of relative fluorescence of GFAP (panel a); representative photomicrographs of retinal sections showing GFAP labelling (red signals in the original photographs) (panel b); quantification of migration index of microglia cells (panel c) and representative photomicrographs of retinal sections showing I ba1 positive cells (red signals in the original photographs) (panel d); gene expression of Gfap in retinal homogenates (panel e) are shown from control mice (C), untreated rd10 mice and NLC- treated rd10 mice (indicated as INS NLC-ADA rd10, INS NLC-ADA-treated rd10 P18, or simply NLC ADA rd10 in the images) are shown. Scale bar: 50 pm. ONL: outer nuclear layer. One-way ANOVA followed by Tukey's multiple comparisons test, *p <0 .05; **p<0.01 ; ***p<0.001 ; ****p <0.0001 for differences between control and rd10 mice or INS NLC-ADA- treated rd10 mice; ^#p< 0.05; #*p < 0.01 ; ^###p<0.001 for differences between rd10 and INS NLC-ADA-treated rd10 mice. Data were presented as mean ± standard error of the mean (SEM). At least nine mice were analysed for each group.

Fig. 15 shows that topical application of NLC-ADA affected the number of microglia, infiltrated macrophages and lymphocytes in rd10 mice at postnatal day (P) 18. Fig. 14a shows the percentage in live cells of microglia (CD11b⁺/CD45^low), macrophages (CD11 b⁺/CD45^hi9h) and lymphocytes (CD11b7CD45⁺) cells in untreated rd10 mice (rd10) and NLC-ADA-treated

mice (INS NLC-ADA-treated rd10) compared to control mice (C). Fig. 14b and 14c shows the fluorescence intensity of the subpopulation of microglia (Fig. 14b) and macrophages (Fig. 14c) which are expressing the markers CD86, CD68 or TREM2. One-way ANOVA or Kruskal-Wallis test followed by Tukey's or Dunn’s multiple comparisons test was used to compare three groups, *p > 0.05; **p<0.01 ; ***p<0.001 ; ****p <0.0001 for differences between control and rd10 mice or INS NLC-ADA-treated rd10 mice; #*p < 0.01 for differences between rd10 and INS NLC-ADA-treated rd10 mice. Data were presented as mean ± standard error of the mean (SEM). At least seven mice were analysed for each group.

Fig. 16 shows that topical application of NLC-ADA reduced the gene expression of some M1 -phenotype genes (pro-inflammatory genes) in rd10 mice at postnatal day (P) 18. The gene expression of several M1- and M2-phenotype genes in untreated rd10 mice (rd10) and NLC-ADA-treated

mice (INS NLC-ADA-treated rd10) compared to control mice (C) are shown. RNA was extracted from frozen retinas and real-time quantitative PCR was performed to evaluate gene expression. One-way ANOVA or Kruskal-Wallis test followed by Tukey’s or Dunn’s multiple comparisons test was carried out to compare three groups, **p<0.01 ; ***p<0.001 ; ****p <0.0001 for differences between control and rd10 mice or INS NLC-ADA-treated rd10 mice; ^#p > 0.05; #*p < 0.01 for differences between rd10 and INS NLC-ADA-treated rd10 mice. Data were presented as mean ± standard error of the mean (SEM). At least eight mice were analysed for each group.

Fig. 17 shows the ERG results obtained after intravitreal application of NLC-ADA improved visual function in rd10 mice at postnatal day (P) 23. Amplitudes of ERG b-wave are represented in panels a and b, where the lower curves of data correspond to untreated mice, the curves represented over those ones correspond to mice treated with NLC-ADA by intravitreal injection and the upper curve of data of panel a corresponds to control mice); a- and b- wave implicit time or latencies recorded are represented in panels c and d, where the bar on the left of each group of three correspond to the data obtained from dark-adapted control mice, the intermediate ones correspond to untreated rd10 mice and the bars on the right of each group of three correspond to NLC-ADA-treated rd10 mice (IV 0.39 NLC-ADA- treated rd10 , ) at different intensities of light Stimuli. rd10 mice received a single intravitreal injection of NLC-ADA at P12. Mann-Whitney test was used to compare rd10 vs rd10 + IV 0.39 NLC-ADA-treated rd10; . ^#p <0 ,05; ^##p<0.01 ; “p<0.001 (Fig. 16b). One-way ANOVA test followed by Tukey's multiple comparisons test was used to compare three groups (Fig. 16c and 16d), **p<0.01 ; ***p<0.001 ; ****p <0.0001 for differences between control and rd10 mice or rd10 + NLC-ADA mice; #*p < 0.01 ; ^###p<0.001 ; ^####p<0.0001 for differences between rd10 and IV 0.39 NLC-ADA-treated rd10 mice. Data were presented as mean ± standard error of the mean (SEM). At least seven mice were analysed for each group.

Fig. 18 shows the comparison of ERG results obtained after intravitreal application of 0.39 NLC-ADA or 0.375 ADA alone showing that NLC-ADA was slightly more effective in improving visual function than ADA alone (at high light intensities) in rd10 mice at postnatal day (P) 23. Amplitudes of ERG b-wave are represented, where the lower curves of data correspond to untreated mice, the curves represented over those ones correspond to mice treated with NLC-ADA by intravitreal injection and the other to mice treated with ADA alone; rd10 mice received a single intravitreal injection of NLC-ADA or ADA alone at P12. Oneway ANOVA test followed by T ukey's multiple comparisons test was used to compare three groups ⁶p<0.05 for differences between rd10 and NLC-ADA-treated rd10 mice (IV 0.39 NLC-ADA-treated rd10 , ^#p <0.5; #^<0.01 ; ^###p<0.001 for differences between rd10 and IV 0.39 NLC-ADA-treated rd10 mice. Data were presented as mean ± standard error of the mean (SEM). At least seven mice were analysed for each group.

Fig. 19 shows how intravitreal application of 0.39 NLC-ADA or 0.375 ADA alone did not significantly improve visual function in rd10 mice at postnatal day (P) 23. Light/dark box test was performed to examine the light aversion of mice, using a light/dark box as represented in Fig. 13a. rd10 mice received a single intravitreal injection of NLC-ADA or ADA alone at P12. Data were presented as mean ± standard error of the mean (SEM), the first bar of each group (light or dark) corresponding to control (C) mice, the second to rd10 at P23, the third to rd10 mice having received NLC-ADA (IV 0.39 NLC-ADA-treated rd1O) and the last (right) one to the rd10 mice having received ADA alone. At least seven mice were analysed for each group light/dark box test for these four groups. One-way ANOVA test followed by Tukey's multiple comparisons test was used to compare three groups *p<0.05; **p <0.01 ; ***p <0.001.

Fig. 20 shows how intravitreal application of NLC-ADA ameliorated photoreceptor degeneration and retinal inflammation in rd10 mice at postnatal day (P) 23. Quantification of number of rows of nuclei at ONL (outer nuclear layer) (a) and representative photomicrographs of retinal sections showing DAPI staining (b); Quantification of relative fluorescence of GFAP (c); Representative photomicrographs of retinal sections showing GFAP labelling (d); quantification of migration index of microglia cells (e) and representative photomicrographs of retinal sections showing Iba1 positive cells (f) from control mice, untreated rd10 mice, and NLC-ADA-treated rd10 mice (and IV 0.39 NLC-ADA-treated rd1O). Scale bar: 50 pm. ONL: outer nuclear layer. One-way ANOVA followed by Tukey's multiple comparisons test, *p <0 .05; **p<0.01 ; ***p<0.001 ; ****p <0.0001 for differences between control and rd10 mice or rd10 + and IV NLC-ADA-treated rd10 mice; ^#p< 0.05; #*p < 0.01 ; ^###p<0.001 for differences between rd10 and rd70+NLC-ADA mice. Data were presented as mean ± standard error of the mean (SEM). At least seven mice for each group were analysed. DETAILED DESCRIPTION OF THE INVENTION

The present invention refers to nanostructured lipid nanoparticles comprising a monoclonal antibody directed against tumour necrosis factor alpha (TNFa), such as Adalimumab (ADA), as well as to pharmaceutical compositions comprising such lipid nanoparticles in powder or aqueous dispersion form and, additionally, to such lipid nanoparticles and/or pharmaceutical compositions comprising them for use in the treatment of a retinal degenerative disease by the ocular topical route (which is also known as ophthalmic route). Also comprised within the scope of the present invention is a procedure for preparing such lipid nanoparticles.

The invention is based in the characteristics of the lipid nanoparticles developed by the present inventors, which are nanostructured lipid nanoparticles containing ADA and which are capable of transporting this drug to the retina by different routes of administration, including the ocular topical route, as they can adhere to the cell surface, increasing the retention time in the cornea, internalise in the ocular tissues and diffuse rapidly until they reach the target tissue, the retina. The nanoparticles of the invention are protected against premature degradation and show sustained release of the antibody after ocular topical administration by instillation, that is, after being administered in the form of an aqueous dispersion falling down drop by drop directly onto the eye surface (as can be seen represented in Fig. 8a). As can be seen in Examples of the present application, the nanoparticles of the invention are also appropriate for intravitreal (injection) administration.

The lipid nanoparticles of the present invention, which are ADA-containing nanostructured nanoparticles as those represented in Fig. 1c, are different from other nanoparticles designed for eye disorder treatment like, for example, those described in international application WO2022/187607. In said application, micelles and liposomes are the only kind of lipid nanoparticles mentioned, being composed by, instead of a surfactant and a mixture of solid and liquid lipids like those of the present invention, by a mixture of phospholipids and, optionally, a sterol like cholesterol (a typical composition of liposomes). The nanoparticles of WO2022/187607 are designed for the delivery of therapeutic agents by the intravitreal route. In spite of that, the pharmaceutical compositions containing the lipid nanoparticles of WO2022/187607 need an additional component, a hydrogel that can comprise at least one polymer, such as a polysaccharide, which component is added to avoid the problems of blurred vision and turbidity often caused by liposome in the vitreous humour, such hydrogel being a component that is not necessarily included in the compositions of the present invention, which compositions can lack any hydrogel, as those used for the assays of the present application. Moreover, the Examples of the present application show that ADA-containing nanoparticles of the present invention, administered to the eye by the topical route, are suitable to deliver adalimumab to the posterior part of the eye and ameliorate retinitis symptoms, while adalimumab is mentioned in WO2022/187607 as a therapeutic agent that could be used to treat eye diseases, but no specific nanoparticle containing it is specifically described in the mentioned application and no evidence of the effective capability of the nanoparticles of WO2022/187607 to deliver ADA to the posterior part of the eye, or to obtain any effect with ADA, is provided in said application.

The ADA-containing nanoparticles of the present invention, contrary to those described in patent application US2019/0015521 , do not need the presence of a cell penetrating peptide (CPP) in the nanoparticle composition, associated to the therapeutic agent, to be able to deliver ADA to the posterior part of the eye, even after administration as eye drops, so that it can exert the desired therapeutic effects.

In the nanostructured lipid nanoparticles of the present invention, as it is schematically represented in Fig. 1c, the therapeutic agent, ADA, is entrapped in the lipid matrix, not in an aqueous hollow core like those of liposomes, as it is suggested in WO2017/184427 where, besides, no specific nanoparticle which such a structure is described (particularly none containing ADA) and no evidence of their capability for delivering any therapeutic agent to the posterior part of the eye after ocular topical administration is provided; instead of that, the only experiments provided are experiments of ocular distribution following topical administration of spherical nucleic acid structures, formed by an antisense oligonucleotide which is a TNF-a inhibitor.

Thus, unexpectedly from the content of the prior art referring to lipid nanoparticles where ADA is mentioned as therapeutic agent for eye diseases like retinitis or other disorders of the posterior part of the eye, the present invention combines the advantages of enabling the delivery of a monoclonal antibody such as ADA encapsulated in a nanostructured nanoparticle lacking peptides or nucleic acids in it is composition (which represents an effective strategy against its degradation, increasing its half-life, and also allows a decrease of the effective dose thanks to its controlled release from these nanosystem decreasing as well toxicity and costs of the treatment) and also providing a suitable antibody delivery system capable of enabling that an antibody effective dose reaches the retina even after ocular topical administration.

Then, the use of the lipid nanoparticles with ADA of the present invention for retinal therapies administered by the ocular topical route means a huge advantage over other therapies administered intravitreally or subretinally since, for example, recurrent use of intravitreal injections increases the risk of endophthalmitis and, additionally, the nanoparticles for use according to the present invention, which comprise a TNFa antibody such as ADA or infliximab, target an inflammatory molecule, TNFa, which is increased in RP patients regardless of the genetic defect causing RP.

The lipid nanoparticles of the present invention are nanoparticles comprising nanostructured lipid carriers (NLCs), that is, nanoparticles comprising antibodies included (entrapped) in a matrix comprising a mixture of one or more solid lipids at room temperature (25°C), whose melting point is equal to or greater than 25°C, and also one or more liquid or semi-solid lipids at room temperature, whose melting point is less than 25°C, and also comprising at least one surfactant. No other component is necessary for the purposes of the present invention, so that the nanoparticles of the present invention can consist of a mixture of one or more than one different lipid compounds whose melting point is equal to or greater than 25°C and one or more than one different lipid compounds whose melting point is lower than 25°C, at least one surfactant compound and a TNFa antibody entrapped in the mixture of lipids, which antibody preferably is adalimumab of a biosimilar thereof. Thus, the lipid nanoparticles of the present invention differ from the nanoparticles called solid lipid nanoparticles (SLN) in that NLCs comprise lipids which are liquid at room temperature additionally to lipids which are solid at room temperature. Preferably, the mixture of solid and liquid lipids varies in a ratio of 70 : 30 up to a ratio of 99.5 : 0.5 w/w respectively but, regardless of the presence of liquid lipids, the NLCs are usually solid at room temperature. The mixture (blend) of solid and liquid lipids gives rise to an unstructured matrix with more imperfections that the SLN and with higher entrapment efficiency, which allows NLCs to hold a greater number of drug molecules than SLN and also to give rise to lower toxicity risk, better drug protection and more stability upon storage. Both NLCs and SLN share with liposomes the characteristics of being lipid nanoformulations that can act as drug delivery tools that can transport drugs with the protective outer layer of lipids but, as can be seen in Fig. 1 , neither SLN nor NLCs showed the continuous lipid bilayer of phospholipids surrounding an aqueous pocket that characterises liposomes and, additionally, their drug leakage is lower than that of liposomes, their stability during storage is better and their drug release takes place during a more prolonged time.

As used in the present invention, the term "lipid" encompasses any molecule of lipophilic nature. In a possible embodiment, compatible with any other, the lipid nanoparticle lacks phospholipids, so that, for that specific embodiment, the term lipid does not encompass any phospholipid or any amphiphilic molecule which comprises a polar head and two hydrophobic tails (understanding as such compounds which comprise a hydrophilic “head”, which contains a phosphate group in the case of phospholipids, to which other functional groups, such as choline or ethanolamine frequently in the case of phospholipids, may be linked, and two hydrophobic “tails”, usually derived from fatty acids, so that, in the specific case of phospholipids, the phosphate group and the tails are each of them joined to an alcohol group which usually is part of a glycerol residue). In another possible embodiment, analogously compatible with any other embodiment, the lipid nanoparticle lacks any conjugated (joined, especially by a covalent bond, but also by van der Waals forces, hydrophobic forces, ionic forces and so on) peptide (understanding as peptides such polypeptides shorter than proteins, which contain no more than 50, preferably no more than 40 and preferably no more than 30 amino acids), in particular any cell penetrating peptide (understanding as such those explicitly mentioned in US 2019/015521), thus, for this embodiment, the term lipid does not encompass any compound of lipidic nature which is conjugated to a peptide). In another possible embodiment, also compatible with any other one, the nanoparticle lacks any nucleic acid, so that, for instance, none of the lipids of the nanoparticle is linked to a nucleotide or a nucleic acid molecule.

A surfactant, in turn, is a surface-active agent that decreases the surface tension between two phases which, in the present case, are an aqueous phase and a lipid phase, so that, for the purposes of the present invention, they will comprise at least hydrophobic tails and hydrophilic heads that will be in contact with the surrounding liquid, which liquid, when the nanoparticles are administered by the ophthalmic route, will be an aqueous solution: the tear film that surrounds the ocular surface.

With regard to the antibody entrapped in the NLCs, it can be any antibody against TNFa, since, in accordance with the definition of TNFa antibodies given hereinbefore, they will prevent the binding of TNFa to its TNFR1 and TNFR2 receptors and their subsequent biological actions on the cells, what will help to diminish the ocular inflammation and, also, other related symptoms such as reactive gliosis, microglia migration, or increase of gene expression of pro-inflammatory molecules (as shown in Example 10). Besides, they contribute to ameliorate loss of vision including an improvement in light perception, a better retinal electrical response to light stimuli and a reduction in photoreceptor degeneration, justifying the nanoparticles of the present invention can be for use in the treatment of a retinal degenerative disease even by the ocular topical route, as the nature of the nanoparticle of the invention makes possible that enough amount of the TNFa antibody arrives at the retina to be effective. Preferably, the antibody will be a monoclonal antibody, such as Adalimumab (the monoclonal antibody present as active ingredient in the medicament commercialized under the brand name Humira® by AbbVie). Instead of ADA as such, it is possible to use a biosimilar. A biosimilar, in general, is a biologic medical product that is almost an identical copy of an original product that is manufactured by a different company once the original product's patent expires. For the purposes of the present invention, biosimilars must also fulfil the requirements of having no clinical differences in their mechanism of action and functionality with adalimumab, binding to the same epitope that adalimumab, preferably, they should have been approved for commercialization with therapeutic purposes, preferably the same as adalimumab. Biosimilars such as adalimumab-adaz (commercialized as Hyrimoz by Sandoz A/S), adalimumab-bwwd (commercialized as Hadlima by Merck Sharp & Dohme Corp), adalimumab-afzb (commercialized as Abrilada by Pfizer), adalimumab-fkjp (commercialized as Hulio by Viatris) or any other available adalimumab biosimilar are encompassed by the scope of the present invention. Thus, it is a possible embodiment of the invention, compatible with any other, the one where the TNFa antibody is ADA or a biosimilar thereof. Adalimumab-adaz, adalimumab-bwwd, adalimumab-afzb and adalimumab-fkjp are possible examples of biosimilars compatible with the present invention.

In one embodiment of the present invention, compatible with any other, the weightweight (w:w) proportion of solid lipid(s): liquid lipid(s) varies in the range of 70:30 to 99.5:0.5. In a specific embodiment of this one, the relation solid lipid(s):liquid lipid(s) is specifically, 10:1 w/w).

In another embodiment, compatible with the previous ones, the solid lipid is glyceryl palmitostearate, also known as precirol (due to its commercial name). In another embodiment, compatible with the previous ones, the liquid lipid is a low-viscosity hydrophilic ester oil, preferably glycerol octanoate decanoate, also named miglyol 812N, due to its commercial name). In a particular preferred embodiment, the solid lipid is glyceryl palmitostearate, the liquid lipid is glycerol octanoate decanoate and their relative relation is 10:1 w/w glyceryl palmitostearate : glycerol octanoate decanoate.

In another embodiment, compatible with all the previous ones, adalimumab (ADA) is present in the final nanoparticle in a weight: weight relation range with regard to the final nanoparticle weight of 1 :35 to 1 :15. In a more specific embodiment, ADA is present in the nanoparticle in a weightweight relation with regard to the final nanoparticle weight of 1 : 25 ± 5 w/w, more specifically 1 : 28.4 w/w ADA : NLC.

With regard to the surfactant or surfactants, in an embodiment at least a non-ionic surfactant is present, because they combine uncharged hydrophilic and hydrophobic groups that make them effective in wetting and spreading and as emulsifiers and foaming agents. In a preferred more specific embodiment, at least the nonionic surfactant polysorbate 80 (polyoxyethylene (20) sorbitan monooleate, very often referred by one of the brand names under which it is commercialized, Tween® 80) is present.

In another embodiment, compatible with any previous one, at least one surfactant being a triblock copolymer composed of a central hydrophobic chain flanked by two hydrophilic chains is present in the nanoparticles, which surfactant will preferably be a non-ionic surfactant. Preferably, such triblock copolymer is of the type of the ethylene oxide/propylene oxide copolymers known as poloxamers, that is, a triblock copolymer composed of a central hydrophobic chain of polyoxypropylene (polypropylene oxide), abbreviated PPO) flanked by two hydrophilic chains of polyoxyethylene (polyethylene oxide), abbreviated PEO). It is specially preferred that the triblock copolymer is the FDA-approved biocompatible block copolymer poloxamer 188 that, additionally to its amphiphilic nature, exhibits a high hydrophile-lipophile balance (HLB) value of 29, being used as a stabilizer/emulsifier in many cosmetic and pharmaceutical preparations.

In a preferred embodiment, compatible with all the previous ones, the lipid nanoparticles comprise at least a non-ionic surfactant, and at least a triblock copolymer composed of a central hydrophobic chain flanked by two hydrophilic chains. The triblock copolymer can be a non-ionic surfactant as such but , preferably, another nonionic surfactant different from a triblock copolymer will be also additionally present. In a more particular embodiment of the previous one, the nanoparticles comprise polysorbate 80 and poloxamer 188. In an even more preferred specific embodiment, the weightweight proportion of polysorbate 80 : poloxamer 188 is 2:1 w/w (1 :0.5, for instance, as in the nanoparticles prepared and assayed in the Examples shown below).

In another embodiment, compatible with all the previous ones, the weightweight relation of the total surfactant amount with regard to the total weight of the complete nanoparticle is.3: 14 w/w.

In a particularly preferred embodiment, the lipid nanoparticles comprise: a) a lipid matrix comprising a mixture of lipids and/or molecules of lipophilic nature comprising: a.1) one lipid whose melting point is less than 25°C which is glyceryl palmitostearate, a.2) one lipid whose melting point is equal or greater than 25°C, which is glycerol octanoate decanoate, wherein the ratio weightweight glyceryl palmitostearate: glycerol octanoate decanoate is 10:1 w/w; b) the monoclonal antibody adalimumab entrapped in the lipid matrix at a ratio weightweight with regard to the total weight of the nanoparticle of 1 ± 0.5 : 25; c) the surfactants polysorbate 80 and poloxamer 188 at a ratio weigh weight polysorbate 80: poloxamer 188 2:1 w/w and wherein the weightweight relation of the total surfactant amount with regard to the total weight of the complete nanoparticle is 3:14.

In an even more preferred embodiment of the previously mentioned embodiment, the nanoparticles fulfilling conditions a), b) and c) as indicated just above have a diameter from 150 to 485 nm, as those prepared in Example 1 of the present invention (see Example 2 for characterization). Most preferred is that the diameter is from 189 to 482 nm.

For the preparation of the lipid nanoparticles for use according to the present invention the heat fusion homogenisation technique can be used. As described in detail in Example 1 of the present application, this technique comprises the following steps:

(i) Preparing a mixture of the lipids and freeze-dried antibody by heating to a temperature slightly above the melting point of the solid lipid(s),

(ii) Preparing an aqueous solution comprising one or more surfactants,

(iii) Heating the aqueous solution obtained in (ii) up to the same temperature than the lipophilic solution with ADA prepared in (i),

(iv) Adding the aqueous solution obtained after performing step (iii) to the lipophilic solution prepared in (i), subjecting the resulting mixture to sonication to obtain an emulsion,

(v) Cooling down the emulsion obtained in step (iv) at 15°C-25°C under magnetic stirring and, subsequently, at refrigeration temperature to allow lipid recrystallization and nanoparticle formation,

(vi) Washing the nanoparticles obtained in step (v) with water combined with centrifugal ultrafiltration,

(vii) Lyophilizing the nanoparticles with a cryoprotectant.

The size, poly-dispersion rate and encapsulation efficiency of the nanoparticles depend on the sonication power and sonication time, so that the selection of said conditions is important for step (iv). For the purposes of the present invention, preferably it is sonicated between 10W and 60W and most preferably between 20W and 50W, for between 10 seconds and 45 seconds, preferably between 15 seconds and 31 seconds and very preferably between 29 seconds and 30 seconds. Such conditions give rise to nanoparticles with size in nanometer range from 150 to 485 nm, approximately (as can be seen in Example 2), slightly negative superficial charge and homogeneous poly-dispersion. With regard to the nanoparticle formation step (v) (the cooling step necessary for the resolidification of the lipids of higher melting point that results in nanoparticle formation), the emulsion obtained in step (iv) is preferably stored at refrigeration temperature for at least between 10 hours and 30 hours, more preferably between 10 hours and 20 hours and very preferably between 11 hours and 13 hours, 12 hours being the most preferred time, the refrigeration temperature being a temperature selected from the range between 1°C and 10°C, preferably between 2°C and 6°C and very preferably between 3°C and 5°C. As it is preferable to cool down progressively, it is advisable to maintain the emulsion obtained in step (iv) at room temperature or a temperature lower but closer to room temperature, for instance one selected from the range from 15°C to 25°C such as 21 °C, for at least 1 hour, previously to maintaining it at refrigeration temperature (such as from 2°C to 6°C).

As the nanoparticles of the present invention are intended to be used for therapeutic purposes, particularly to be administered to the eye, the washing step (vi) previous to lyophilization is carried out, preferably, with type I (or ultrapure) or type II (or pure) water, the kind of water that is usually referred to by the term "M ill iQ water", because it can be obtained with apparatuses commercialized as Milli-Q™ Water Purification Systems (MilliporeSigma™, commercialized by Fisher Scientific in the United States and Europe). The washing step is carried out several times, for instance three, combining the treatment with water with ultrafiltration in a centrifugal ultrafiltration device (a device where ultrafiltration is favoured by subjecting the sample to centrifugation), replenishing with water the volume eluted after each centrifugation step and recovering finally the portion located above the filter (the desired nanoparticles), but also, if desired, the eluted supernatant, the latter being recovered because it contains surfactant not included in the nanoparticles and non-entrapped (encapsulated) ADA, so that it can be used to assess the ADA encapsulation (entrapment) efficiency by indirect calculation. Finally, the freeze-dried step (vii) (lyophilization), as indicated above, is carried out using a cryoprotectant, that is, a substance that protects biological tissues and/or structures similar to biological structures from freezing damage. Known cryoprotectants such as sugars and polyols can be used, as for instance glycerol, sucrose, trehalose, propylene glycol or ethylene glycol but, for the purposes of the present invention, the cryoprotectant is preferably 15% trehalose.

The lyophilized nanoparticles (which, after lyophilization, are in the form of powder), can be stored at refrigeration temperature (for instance, 4°C). This lyophilized powder, which comprises nanoparticles and the added cryoprotectant, can be the commercialization form of the nanoparticles. The commercialization form can also be the form ready to be administrated prepared from that powder, which will be a nanoparticle dispersion in an aqueous solution where the nanoparticles are homogenized and dispersed, which aqueous solution can be, for instance, saline solution, phosphate buffer saline (PBS), or specific buffers for ocular administration purposes including citrate, Tris-HCI or borate buffers. Both forms of compositions are encompassed with the scope of the present invention. Thus, another aspect of the invention is a pharmaceutical composition comprising nanoparticles of the present invention as a powder or as an aqueous dispersion. Additionally, to the cryoprotectant added before lyophilisation (and the aqueous solution where the nanoparticles are dispersed, if the pharmaceutical composition is in the form of an aqueous solution), the pharmaceutical composition of the present invention can also comprise any pharmaceutically acceptable excipient, such as antioxidant or microbial preservative agents.

In order to obtain the above mentioned ready-to-use form, it is necessary to reconstitute the lyophilized nanoparticle composition, by homogenizing and dispersing the nanoparticles in an aqueous solution. Thus, it is also an aspect of the invention a procedure for preparing a pharmaceutical composition of the present invention which comprises the steps of (i) to (vii) of the above mentioned procedure for the preparation of nanoparticles of the present invention and which, additionally, comprises at least one step where a pharmaceutical composition of the present invention, in the form of an aqueous dispersion, is prepared. The preparation of the mentioned aqueous dispersion from nanoparticles of the present invention, either in lyophilized form or in any other form, is also encompassed within the scope of the present invention. When the aqueous composition is prepared from the lyophilized nanoparticles of the invention, the composition will be prepared by adding an aqueous solution to the lyophilized nanoparticles and homogenizing and dispersing the nanoparticles in such solution by agitation and, preferably, also by sonication. In possible embodiments of this aspect of the invention, the agitation is effected by vortexing for at least 10 seconds the mixture of nanoparticles and aqueous solution and, if the sonication is also carried out, it can take place in an ultrasonic bath: between 30 Hz and 60 Hz and most preferably between 40 Hz and 50 Hz, for between 10 minutes and 45 minutes, preferably between 15 minutes and 31 minutes and very preferably during 20 minutes; an additional vortexing substep, for instance during five seconds, is advisable after sonication. The specific resuspension, agitation and sonication conditions might be those used to obtain the instillation dispersion administered in Examples 10 and 11 : adding PBS at 107 mg/mL, vortexing for 10 seconds, sonication in an ultrasonic bath at 40 Hz during 20 minutes and additional vortexing for 5 seconds. One of the advantages of this preparation method is that no organic solvents are used, thus avoiding the need to carry out tests for trace organic solvents prior to commercialization of the nanoparticles for human use.

Once prepared, the assays set forth below in the Example section of the present application show that the present inventors have achieved to obtain nanoparticles with suitable properties to be administered to the eye by the ocular topical route so that enough amount of the active principle, adalimumab, arrives at the retina and act there. For instance, the lipid nanoparticles of the present invention exhibit the ability to:

- be stable lipid nanoparticles with sustained and/or regulated release effect of the antibody

- protect the antibody from premature degradation

- penetrate through the cornea and reach the retinal cells even when they are administrated by the ophthalmic route and consequently

- be administered directly into the eyeball (ophthalmic route), avoiding systemic side effects such as those associated to the intravitreal route

- obtain better minimum effective concentration values than free antibody,

- result in fewer toxic effects due to the antibody, such as a low corneal (similar to that obtained with an aqueous buffer solution such as PBS lacking NLCs) and retinal toxicity at low concentrations having therapeutic effect.

For all the above, as commented before, the present invention refers to NLCs loaded with ADA as well as to a pharmaceutical composition comprising the mentioned NLCs and additionally, the invention relates to NLCs of the invention loaded with ADA and/or pharmaceutical compositions comprising them for use in the treatment of a retinal degenerative disease by the ocular topical route. The nanoparticles of the present invention not only show the properties of allowing the administration of ADA included in NLCs but also the possibility of using the ocular topical route for such administration, that is, they not only confer to the invention the advantages related to the use of nanoencapsulation strategies for ocular delivery (increasing drug availability, controlling its release kinetics, decreasing the number of intravitreal and sub-retinal administrations), but the particular entrapment in NLCs, specifically in the NLCs of the present invention, also permits to avoid both the side effects associated to the intravitreal administration of drugs (endophthalmitis, patient discomfort, etc.) and the high doses of monoclonal antibodies required to achieve a therapeutic effect when they are administered subcutaneously without being included in a particle. As can be seen in Example 10, some drops of few more than 3.5 pg/pL of ADA (namely, 2 pL of a dispersion of NLC-ADA with 3.54 pg/pL of ADA), administered each two days for only seven days, are enough to achieve an amelioration of symptoms of RP in an animal model. And the ameliorated symptoms are not only limited to an amelioration of inflammation and the symptoms associated with it (including the reduction of reactive gliosis, microglia migration, and gene expression of pro-inflammatory molecules such TNFa itself, TNFR1 , IL6 and I L-1 P), as could be expected from ADA or any other TNFa antibody, but also others such as partial restoration of light perception, retinal response to light stimuli or reduction of cell death.

As commented above, it is well known that repeated intravitreal administrations of drugs may cause side effects including endophthalmitis, patient discomfort, etc. Besides, the use of monoclonal antibodies is limited by the high doses required to achieve a therapeutic effect. As shown in the present application, NLCs loaded with ADA, when used for the treatment of a retinal degenerative disease, decrease the risk of side effects and improves the patient's quality of life, because they enable the use not only of the ocular topical route, but also the administration of the NLCs through pharmaceutical compositions comprising the NLCs of the present invention where the NLCs are dispersed in an aqueous solution, thus decreasing the possibility of experimenting undesired adverse effects associated to liposome carriers such as temporal blurred vision or other discomfort manifestations. What is more, applying the conversion factor use by the present inventors in previous studies related to ADA intravitreal administration for the treatment of retinopathies (Olivares- Gonzalez et al., The FASEB Journal, 2018 vol. 32(5):2438-2451), if an equivalent dose of NLC-ADAs of the present invention (0.390 mg/mL for mice) had been administered intravitreally to a human being, the amount of ADA administered would have been approximately 0.127 mg and the resulting concentration in the human vitreous humour would have been 0.0317 mg/mL, almost 15 times lower that the concentration that is used and considered safe for human beings (2 mg, 0.5 mg/mL) for intravitreal administration. Although the same conversion factor cannot be applied for ocular topical administration and any estimation can be considered only speculative, the present inventors expect that the calculated dose (3.54 mg/mL x 2 pL = 7.08 pg ~ 7.1 pg) would be low but effective. NLCs are lipophilic compounds which improve absorption through the cornea and conjunctiva. Previous studies demonstrated relevant concentrations of bevacizumab, an antibody, after ocular topical administration of liposomes (13 mg/mL or 25 mg/mL of antibody) in the posterior segment of the eye in rats (127 ng/g) and rabbits (18 ng/g) (Davis B., et al. Small. 2014;10:1575-1584. doi: 10.1002/smll.201303433).

Thus, also an aspect of the invention is a nanoparticle of the present invention or a pharmaceutical composition of the present invention for use in the treatment of a retinal degenerative disease by the ocular topical route. The intravitreal route can also be used, but the ocular topical route is preferred. Defining this aspect of the invention in that way can be considered equivalent to defining it as use of a nanoparticle of the present invention or a pharmaceutical composition of the present invention for preparing a medicament for the treatment of a retinal degenerative disease to be administered by the ocular topical route. It can be considered also analogous defining it as a method for the treatment of a retinal degenerative disease in a subject in need thereof by the administration by the ocular topical route of a therapeutically effective amount of nanoparticles of the present invention.

As used in the present application, the term topical route, when applied in the context of ocular treatment, will be considered a synonym of ophthalmic route and implies the direct administration of a liquid composition, preferably in the form of drops, to the eye, an administration way which is also called "instillation". And, in order to use the topical route, but also when intravitreal administration is desired, the lyophilized NLCs prepared by the method of the present invention are reconstituted in an aqueous solution, which must be well mixed by vigorous agitation to obtain a homogenous dispersion.

The retinal degenerative disease can be any one with an inflammatory component. Preferably, the retinal degenerative disease is an inherited retinal dystrophy, more preferably retinitis pigmentosa. The symptoms to treat can be any one associated to an inherited retinal dystrophy, more particularly retinitis pigmentosa, particularly inflammation, but also others such as ocular inflammation, reactive gliosis, microglia migration, increased gene expression of pro-inflammatory molecules, reduced vision, reduced light perception, reduced retinal electrical response to light stimuli and/or photoreceptor degeneration.

The treated subject can be any mammal, such as a pig or a mouse as in the Examples of the present application, but it will be preferably a human being.

The invention will be now explained in more detail by the Figures and Examples below.

EXAMPLES

Example 1.- Lipid Nanoparticle Preparation

The lipid nanoparticles for use according to the present invention were prepared by the heat fusion homogenisation technique.

In particular, the lipid nanoparticles used in the following Examples of the present application were formed by 6.6% (w/v) solid lipid Precirol® ATO5 (Gattefosse Espana, Madrid, Spain) and 0.6% (w/v) liquid lipid Miglyol® 812N (IOI Oleochemicals, Hamburg, Germany) (which means a mass proportion of 10:1 solid lipid : liquid lipid), and an aqueous component, which contains the surfactants Tween® 80 (Panreac Chemicals, Barcelona, Spain) and Poloxamer 188 (purchased as Kolliphor® P 188 Bio from BASF Corporation, Ludwigshafen, Germany), diluted in MilliQ water at 1.3% (w/v) and 0.6% (w/v), respectively (which mean mass proportions of 2:10 and 1 :10, respectively, in relation to the solid lipid).

Then, the formula of the nanostructured lipid carriers as such, which act as ADA carriers when ADA is added, is: 1000 mg solid lipid Precirol® ATO5, 100 mg liquid lipid Miglyol® 812 N, 200 mg Tween® 80, 100 mg Poloxamer 188, so that the proportion of each component with regard to the total weight, 1400 mg, is: 71.43%, 7.14%, 14.29%, 7.14%, respectively.

The freeze-dried antibody, adalimumab, was obtained from the commercial medicinal form. 40 mg or 80 mg of Adalimumab (injectable Humira®, Abbvie, North Chicago, Illinois, UE) (dose of 40 mg/0.4 mL or 80 mg/0.8 mL) were frozen in Eppendorf tubes for 24 hours at - 80°C. Then frozen Adalimumab was lyophilized at 100 mg/mL in water, mannitol and Polysorbate 80 (the components comprised in the Humira® injectable solution) for 6-10 hours. The freeze-dried adalimumab was added to the lipid phase to obtain NLC-ADA particles at a ratio of ADA:total NLC-ADA 1 :28.4 (w/w).

In order to prepare the nanoparticles, the following steps were taken:

(i) preparing a mixture of the lipids and the freeze-dried antibody by heating to a temperature of 65°C for 3 minutes;

(ii) preparing an aqueous solution comprising Tween® 80 and Kolliphor® P 188 Bio in MilliQ® water;

(iii) heating the aqueous solution obtained in (ii) up to 65°C, the same temperature used in step (i);

(iv) once both phases had reached the same temperature, adding the aqueous solution obtained after performing step (iii) on top of the lipophilic solution prepared in (i), subjecting the resulting mixture to sonication to obtain an emulsion, at 50 W for 30 seconds;

(v) cooling down gradually the emulsion obtained in step (iv), maintaining it at 21°C for 1 hour, followed by 5°C ± 3°C to allow lipid recrystallization and nanoparticle formation, between 11 hours and 13 hours (for instance, overnight) and at a temperature between 3°C and 5°C, (specifically 4°C in this case);

(vi) washing the nanoparticles obtained in step (v) with MilliQ® water three times combined with ultrafiltration in centrifugal filtration devices (Amicon®, Life Science Research, Sigma Aldrich) at 2500 rpm for 15, 10 and 10 minutes, replenishing with MilliQ® water the same volume eluted from the centrifugation steps, recovering the eluted supernatant (to assess the encapsulation efficiency by indirect calculation), and repeating the washing and centrifugation for three times, recovering the nanoparticles from the portion situated over the filter

(vii) adding trehalose (15% w/w with regard to the solid lipid, that is, 50 mg trehalose dissolved in 500 L milliQ® water for each 5 mL of NLC suspension, which contains 333 mg of solid lipid) to the nanoparticles of step (vi) as cryoprotectant, and lyophilizing them (Telstar Lyobeta freeze-dryer, Terrassa, Spain).

The lyophilized nanoparticles were stored at 4°C until use, although storing at room temperature could also be possible. Before being administered, they were reconstituted in an aqueous solution, as described below.

Example 2.- Lipid Nanoparticle Characterization

Freeze-dried (lyophilized) nanoparticles were resuspended in MilliQ water (firstly at 6.7 mg/mL and them and additional 1 :100 dilution of the before mentioned suspension was carried out). These diluted nanoparticles were analysed by transmission electron microscopy (TEM) (JEQL-4000, Tokyo, Japan) to analyse their shape and morphology. Particle size and polydispersity index (PDI) were determined by cumulative analysis of the dynamic light scattering (DLS), while zeta potential was measured by laser Doppler velocimetry applying the Smoluchowski approximation (Smoluchowski M.V., Physik. Zeit. , vol. 17, p. 557-585 (1916)). For these measurements a Zetasizer Nano ZS (Malvern Instruments, United Kingdom) was employed.

Significant photographs of empty NLC (without entrapped ADA) and NLC-ADA (NLCs comprising entrapped ADA) can be seen in Fig. 2a. Panel 2b shows a bar chart representing the obtained data for mean particle size and zeta potential, which are also summarized in Table 1 below, where can be seen that the NLCs obtained by the method of the present invention, particularly those obtained with the specific conditions used in Example 1 , shows mean diameters from approximately 150 nm to approximately 500 nm (approximately 189 to approximately 500 nm in the case of NLC-ADA).

Table 1

Their nanometer size and homogeneous dispersion (PDI < 0.4, which is under acceptance for drug delivery purposes) made these nanoparticles suitable for drug delivery, avoiding the ocular irritation associated with particles greater than 10 pm. As expected, ADA entrapment into NLCs caused a slight increase in nanoparticle size. Zeta potential values were negative for empty and ADA-loaded NLCs. Slight variations in the superficial charge were found when loading ADA, associated with the nature of the loaded monoclonal antibody.

Example 3.- Lipid Nanoparticle Drug Loading Efficacy and Release Profile

3.1.

Adalimumab entrapment efficiency was indirectly assessed by ELISA, quantifying the amount of the monoclonal antibody at the collected supernatant in the wash steps after centrifugation of formulations within Amicon® ultrafiltration devices. The drug loading efficiency of NLC:ADA was 93.26 ± 0.28%, which means a relation NLC:ADA 28.4:1 w/w. For reasons of easiness of reading, this relation will be referred hereinafter (excepting when the exact value is important) as NLC:ADA 25:1 w/w. Nominal ADA concentration refers to the theoretical ADA concentration in NLCs. Real ADA concentration refers to the ADA concentration after considering loading efficiency.

3.2. In vitro Release Studies

Freeze-dried nanoparticles were resuspended in PBS 0.02M (pH 7.4) and maintained at 37°C in a rotatory shaker. At established timepoints, samples were centrifuged at 20.000g for 15 minutes. The resulting supernatant (1 mL) was collected and 1 mL of fresh PBS 0.02M (pH 7.4) was added to the pellet containing the NLCs for Adalimumab determination by Enzyme-Linked Immunosorbant Assay (ELISA) (#EL-1611-011 , Affinitylmmuno, ClinicSciences) following manufacturer's instructions. The obtained results can be found in Table 2 below and are represented in Fig. 3, where it can be seen that the results suggested that approximately 50% of ADA is released after 30 minutes, showing a constant, but not so quickly released until 75 minutes and maintaining and slow-release rate until the final timepoint. Table 2

Example 4.- Cell Proliferation and Cellular Uptake of Nanoparticles

4.1. Effect of NLC-ADA on Cell Proliferation

In order to carry out assays about the effects of NLC-ADA on cell proliferation, 661 W photoreceptor cells (retinal cells) were used. The 661W photoreceptor cells were provided by Dr. Muayyad Al-Ubaidi (University of Oklahoma Health Sciences Center). Cells were grown in Dulbecco’s Modified Eagle’s Medium/Nutrient Mixture F-12 Ham (#11320033 DMEM:F12; Gibco Thermo Fisher Scientific, Madrid, Spain) supplemented with fetal bovine serum (#26140079, Gibco, Thermo Fisher Scientific, Madrid, Spain) and 100 units/mL penicillin/100 pg/mL streptomycin (#15140122, Gibco, Thermo Fisher Scientific, Madrid, Spain), 0.004% [3-mercaptoethanol (#M7154, Sigma-Aldrich, Madrid, Spain) and incubated at 37 °C in 5% CO2 humidified atmosphere (#3121 , Thermo Electron Corporation, Waltham, Massachusetts, US). For cytotoxicity assays 661 W were seeded in 96 well plates and for cytometry or fluorescence assays 661W were seeded in 24 well plates. In all cases, the concentration of 661 W cells was 150.000 cells/ml.

To test the effects of empty NLC (eNLCs) or ADA-loaded NLC (NLC-ADA) on cell viability, 661W cells were exposed to eNLCs or NLC-ADA at different times (from 4 to 48 hours). The NLC-ADA selected for carrying out these assays, NLC-ADA with a relation NLC-ADA 28.4:1 , were diluted in culture medium so that NLC concentrations from 12.5 to 125 pg/mL and consequently different concentrations of ADA were used: 0.410, 0.820 and 4.100 pg/mL, which concentrations would be, nominally, 0.440, 0.88 and 4.400 pg/mL if the encapsulation efficacy would be 100%, for which sample denomination has been simplified as ADA0.5, ADA1 and ADA5. The NLC-ADA and nominal and real ADA concentrations used, as well as the specific denominations for each sample are summarized in Table 3 below. Table 3

To test cell viability, an MTT assay was carried out. For that, the 661 W cells were grown to 80% confluency and then treated with different concentrations of nanoparticles (empty- NLCs or the samples of NLC-ADA indicated in Table 3) for different times (6, 15, 24, 48 hours), 100 pL of 5 mg/mL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (#M5655, MTT; Sigma-Aldrich, Madrid, Spain) diluted in 0.1 M PBS was added to the wells, and the cells were incubated for two hours at 37°C. Mitochondrial succinate dehydrogenase (EC1.3.9.9) from viable cells reduces MTT to form a dark blue formazan crystal. Then, the media were removed the formazan precipitate was dissolved in 100 pL dimethyl sulfoxide (DMSO) (#D2650, Sigma-Aldrich, Madrid, Spain), and the absorbance was read at 550nm in a microtiter plate reader Multiskan Sky High microplates (Thermo Fisher Scientific, Madrid, Spain). The viability of the cells was expressed as the percentage of untreated cells (100% viability). At least five-six experiments (different cultures with several replicas) were used for each determination.

The results are represented in Fig. 4a, where it can be seen that NLC-ADA decreased cell viability in a dose- and time-dependent manner, especially at 125 pg/mL. After six hours, NLC-ADA did not exhibit cytotoxicity at a concentration of 12.5 and 25 pg/mL.

Thereafter, to test the effects of ADA-loaded NLC (NLC-ADA) on TNFa-induced cell death, it was first examined whether it was toxic in retinal cells. 661W cells were exposed to exogenous cytotoxic agent, TN Fa (100 ng/mL) for six hours. The cells were also treated with NLC-ADA. The NLC-ADA selected for carrying out these assays were ADA0.5 and ADA1 , those that were not toxic at 6 hours in the previous assay. To assess whether these concentrations of NLC-ADA prevented from TNFa-induced cell death after six hours, the same viability test (MTT assays) above described was carried out. In this case, results were expressed with regard to control cells. As can be seen in Fig.4b, after six hours, 25 pg/mL NLC-ADA reduced cell toxicity induced by TNFa.

4.2. Cellular Uptake of NLC Formulations

For internalisation assays, 661W cells were seeded at 150.000 cells/ml in 24 well plates. In order to visualize the NLCs inside the cells, Coumarin6-loaded NLC were prepared, following the same method used for the preparation of NLC-ADA. Cells were treated with different concentration of Coumarin6-loaded NLC (from 12.5 to 125 pg/mL) at different times (0, 5, 15, 30 and 60 minutes). After exposure, cells were collected or fixed for further analysis of internalisation of nanoparticles. Each experiment was performed four times with 3 replicas for each treatment.

For the fluorescence analysis, treated cells were fixed with 4% filtered paraformaldehyde (PFA) (#158127, Sigma-Aldrich, Madrid, Spain) in PBS during 15 min. Thereafter, cells were counterstained with DAPI, mounted with Fluoromount G (#0100-01 , Southern Biotechnology, Birmingham, AL, USA), and imaged using a fluorescence microscope (100X magnification, LEICA, Dmi8, Leica Microsystems CMS GmbH, Mannheim, Germany).

For the flow cytometry analysis, cells treated with different concentration of Coumarin6-NLC during 0, 30, 60 and 120 min, were recollected with trypsin (Gibco, Thermo Fisher Scientific, Madrid, Spain) and centrifugated at 300g for 3 minutes. Then, the pellet was resuspended in 200 L of culture medium with PI (propidium iodide) at 1 pg/mL. Flow cytometry of samples was performed using a CytoFLEX S (Beckman Coulter Life Sciences, Indianapolis, IN, USA). Coumarin 6 fluorescence was collected through a 488 band-pass filter. PI fluorescence was collected through a 610/620 band-pass filter. Data acquisition was performed using the CytExpert 2.1 software (Beckman Coulter Life Sciences, Indianapolis, IN, USA). Three types of cell populations (%) were detected in a dot plot: live cells (PI-), Coumarin6 positive live cells (C6+/PI-) and necrotic cells (PI+). Each experiment was performed four times with 3 replicas for each treatment.

Fig. 4c shows the percentage that C6+/PI- cells (indicated in the bar chart as NLC+ cells) with regard to the total number of cells. Accordingly, flow cytometry studies revealed a dose- and time-dependent cellular uptake, that was also visualized by microscopy as can be seen in Fig. 4d, where significant images obtained after the fluorescence analysis are shown.

Example 5.- Haemolytic activity of nanoparticles

The research of NLCs on blood interaction shows that NLCs present reasonably good blood compatibility with low haemolytic activity (Guo S et al. Asian J Pharm Sci. 2021 Sep;16(5):551-576). To evaluate the safety of eNLC and NLC-ADA perse, the haemolytic activity was determined (Fig. 5a). The lysis of erythrocytes (haemolysis) and denaturation index (DI) of released oxyhaemoglobin were measured according to DB-ALM protocol n° 37 (Pape WJ et al. Toxicol In Vitro. 1999;13(2):343-54; Gerecke C et al. Nanotoxicology. 2017;11(2):267-277).

Procedure Fresh swine blood samples were kindly provided by the veterinary of the IIS La Fe (Health Research Institute Hospital La Fe, Valencia, Spain). Blood samples were collected into citrate/citric acid tubes and centrifuged at 1500 x g for 20 minutes at room temperature, in a swinging bucket rotor, within the first 60 min after collection. Supernatant (plasma) was carefully aspirated and red blood cells (RBC) were washed three times with isotonic PBS pH 7.4 containing 10 mM glucose. RBC diluted with isotonic PBS to 8 x 10⁹ cells/mL was added to eNLC or NLC-ADA in PBS (from 1.25 up to 500 pg/mL final concentration) and incubated for ten minutes. Samples were centrifuged at 1000 x g for one minute. The absorbance of the released oxyhaemoglobin at 560 nm was measured from the supernatant and the relation between effective concentration of 50% haemolysis and protein DI was calculated to evaluate acute eye irritancy potential. De-ionized water served as positive control, PBS as negative control and sodium dodecyl sulphate (SDS) solutions ranging in concentrations from 0 to 80 pg/mL as internal standard for calculation of the H50 value and the DI standard haemolysis and denaturation. For DI, RBC were incubated with eNLC or NLC-ADA at 1 mg/mL. DI was calculated as [(R1-Ri)/(R1-R2)]x100, where R1 , R2 and Ri were the ratios between the absorbance readings of the haemoglobin released from the erythrocytes when in contact with: (i) distilled water; (ii) SDS at 1 mg/mL and (iii) eNLC or NLC-ADA, respectively. Haemolysis was calculated as the absorbance of an erythrocyte suspension incubated with each formulation, relative to that of a completely haemolysed control (100%) at 540nm. Each experiment was performed five times with three replicas for each experiment.

Fig. 5b shows the percentage of erythrocyte haemolysis at different concentrations of eNLC or NLC-ADA with regard to the haemolysis induced by distilled water. As can be seen in Fig.5c, eNLC or NLC-ADA induced significant haemolysis at the highest concentration analysed, but moderate and similar to that of the control until the concentration of 125 pg/mL, where starts to increase clearly. It was a dose-dependent haemolysis. 1000 pg/mL NLC-ADA or eNLC increased erythrocyte haemolysis by 28% and 22%, respectively. DI of released haemoglobin increased up to 27% for both nanoparticles with regard to the DI induced by SDS. Thus, the toxicity of the NLCs of the present invention is acceptable at the doses where they are expected to be used.

Example 6.- Ex vivo retinal toxicity

Ex vivo retinal toxicity studies of NLC-ADA were carried out in organotypic retinal explant cultures obtained from porcine eyes. These studies included the labelling with the TUNEL technique and the measurement of the caspase 3 activity (Fig. 6). 6.1. Experimental Procedure

The porcine eyes were collected from the local slaughterhouse (MercaValencia), and organotypic retinal explant cultures were prepared as previously described (Martinez- Fernandez de la Camara C et al. Exp Eye Res. 2013;111 :122-33. doi: 10.1016/j.exer.2013.03.015).

It is considered that this ocular bioavailability (determined from aqueous humour) is typically 1-4% due to the rapid drainage of eye drops, corneal barriers, efflux pumps, etc. and even less in the vitreous humour and retina (by 1 % of instilled drug) (Gaudana R, et al. AAPS J. 2010;12(3):348-60. doi: 10.1208/s12248-010-9183-3). The equivalent amount of PBS was added to the culture medium of controls. Based on the initial concentration of NLC-ADA instilled (2 pL of NLC-ADA 107 mg/mL) and the proposed ocular bioavailability of drugs after instillation (by 1%), different concentrations of NLC-ADA were prepared for a range of ocular bioavailability from 0.25 to 2%) (Table 4). NLC-ADA was added the day of the culture to the culture medium and maintained for 24 hours. Four different organotypic retinal explant cultures were performed with three replicas for each treatment.

Table 4

6.2. Histology and Microscopy

After 24h, some retinal explants were fixed in 4% filtered PFA (Sigma-Aldrich, Madrid, Spain) in 0.1 M PBS (pH 7.4) and cryoprotected in a saccharose gradient (15-20-30%) (Panreac Quimica, Barcelona, Spain). Samples were frozen embedded in Tissue-Tek® OCT™ Compound (Sakura Finetek Europe BV, Zoeterwoude, The Netherlands).

After this, retinal explants were embedded in OCT and 10 pm sections were cut in a cryostat (Leica CM 1900, Nussloch, Germany). To evaluate cell death, the terminal deoxynucleotidyl transferase dUTP nick and labelling (TUNEL) assay (#G3250, Promega Corporation, Madison, USA) was used following the manufacturer instructions. Cryosections were stained with DAPI (#D9542, Sigma-Aldrich, Madrid, Spain). Cryosections were visualized under a fluorescence microscope (40X magnification, LEICA, Dmi8, Leica Microsystems CMS GmbH, Mannheim, Germany). 6.3. Caspase 3 activity

After 24h, some retinal explants were homogenized in lysis buffer to measure caspase 3 activity by a colorimetric assay kit following the manufacturer's instructions (Bio-Vision, Mountain View, CA, USA). Total retinal protein was measured by the PIERCE bicinchoninic acid (BCA) protein assay kit (#23225, Thermo Fisher Scientific, Madrid, Spain).

6.4. Results

Some significant images obtained from untreated explants, and explants treated with different NLC-ADA concentrations are shown in Fig. 6, where it can be seen that NLC-ADA did not significantly increased TUNEL-positive cells, thus indicating that NLC-ADA nanoparticles did not induce significant cell death. Only the highest concentration of NLC- ADA (0.8 pg/mL) tended to increase TUNEL-positive cells. Direct counting of the number of TUNEL-positive cells was carried out using the Imaged open-source Software. Number of positive cells are expressed as percentage of positive cells in PBS-treated explants. Quantification of these cells is shown in Table 5. NLC-ADA at low concentrations would not only be non-toxic but could even prevent from the inherent degeneration of this model, as these retinal explants undergo a degenerative process due to a retinal detachment. The percentage of TUNEL-positive cells was lower than that observed with PBS without nanoparticles, indicating lower cell death excepting at the higher concentration of NLC-ADA (0.8 pg/mL). However, this last effect was not significant.

Caspase 3 activity was expressed as arbitrary units (au)/mg of protein and normalized to the PBS-treated explants (Table 5). After NLC-ADA incubation, no significant increase in caspase 3 activity was observed in retinal explants. The ex vivo studies indicated that NLC- ADA were not toxic to the retina at the selected concentrations.

Table 5

Example 7-. In vivo corneal toxicity

In vivo corneal toxicity studies were carried out through the analysis of fluorescein staining and the detection of died cells by the TUNEL technique in wild-type C57BL/6J mice, showing that topical administration of NLC-ADA did not induce a significant toxicity in the mice corneas.

7.1. Animals

Wild-type C57BL/6J mice (Charles River, Cerdanyola del Valles, Barcelona, Spain) with the same genetic background as rd10 mice were used as a control group for ocular distribution studies. Mice were kept under a 12-hour light/dark cycle, humidity and temperature controlled and with food and water supplied ad libitum. Mice were housed in the Animal Facility of Research Center Principe Felipe (CIPF) of Valencia. This study was carried out in accordance with the European Union Guidelines for the Care (European Union Directive (2010/63/EU) and the guidelines for the Use of Laboratory Animals. All animal procedures and protocols were approved and monitored by the Committee of Ethics in Research Center Principe Felipe and the local government (Conselleria de Agricultura, Desarrollo Rural, Emergencia Climatica y Transition Ecologica, Generalitat Valencia) with reference 2020/VSC/PEA/0124 type 2. At least six mice were used.

7.2. Nanoparticles reconstitution before instillation:

Freeze-dried NLCs were weighed and resuspended in PBS at 107 mg/mL, mixed in vortex for 10 seconds, sonicated in an ultrasonic bath for 20 min at 40 Hz and mixed again by vortexing for 5 seconds.

Procedure A study was conducted to determine the presence of corneal epithelial defects after topical use of NLC-ADA in the eyes of control mice in vivo.

Fluorescein staining serves as indicator of an increased epithelial permeability of the cornea or conjunctiva by staining devitalized areas of the ocular surface. The eyes received an NLC-ADA instillation of 7.08 pg of ADA or an PBS instillation at different times (10 min, 24h, 48h, 96h and 144h) from postnatal day (P) P12 to P18. (2 pL of ADA-loaded NLC: [NLC]: 107 mg/mL; Nominal [ADA]: 3.7676 mg/mL; Real [ADA]: 3.54 mg/mL (3.54 pg/pL), Real ADA amount administered each time: 3.54 pg/pL x 2 pL = 7.08 pg ~ 7.1 pg). Mice were euthanized at P18. At least six mice were used.

7.3. Fluorescein staining:

Fluorescein staining was performed by applying 1 pL of 0.25% fluorescein solution onto the conjunctiva sac after corneal instillation of NLC-ADA or PBS at different times (10 min, 24h, 48h, 96h and 144h) in C57BL/6J mice. The procedure started at P12 and finished at P18. The ocular surface was examined using a yellow barrier-filter with the cobalt blue light of a slit-lamp (Haag-Streit Diagnostics, Wedel, Germany) and divided into five areas that were separately scored using a previously validated grading system for mice in a blinded fashion on a scale from 0 to 3 according to 0 = no staining, 1 = slight punctate staining, 2 = distinct punctate or slight coalescent staining, 3 = distinct patchy staining. Values in all sectors were summed to generate a score out of 15, then averaged across all observers: 0 = No effect 1-5 = Mild effect

5-10 = Moderate effect

10-15 = Severe effect

7.4. Histology and

After six days receiving NLC-ADA or PBS from P12 to P18, control mice were euthanized to analyse the ocular surface by histology. Briefly, eyes were enucleated, corneal tissue was collected, fixed in 4% filtered PFA for two hours at room temperature and cryoprotected in a sucrose gradient (15-20%-30%). The corneas were frozen, embedded in OCT and 10 pm sections were cut in a cryostat (Leica CM1900, Nussloch, Germany). To evaluate cell death, the terminal deoxynucleotidyl transferase dUTP nick and labelling (TUN EL) assay (#G3250, Promega Corporation, Madison, USA) was used following the manufacturer instructions. Cryosections were stained with DAPI (#D9542, Sigma-Aldrich, Madrid, Spain). Cryosections were visualized under a fluorescence microscope (40X magnification, LEICA, Dmi8, Leica Microsystems CMS GmbH, Mannheim, Germany).

Results Some representative corneal fluorescein staining at different times (Fig.7a) and its analysis (Fig. 7b) from control mice treated with PBS or NLC-ADA are shown, where it can be seen that NLC-ADA has no effect or a mild effect on corneal surface.

Concerning TUNEL analysis, some significant images obtained from control mice treated with PBS or with NLC-ADA are shown in Fig. 7c, where very sparse TUNEL-positive cells are observed. Direct counting of the number of TUNEL-positive cells (manual cell counting), was carried out using the Imaged open-source Software. There was no significant effect of NLC-ADA (34 ± 8%, p = 0.41 , unpaired t-test analysis) on the number of TUNEL-positive cells compared to PBS (24 ± 9 %).

It can be concluded that topical administration of NLC-ADA did not induce a significant toxicity in the mice corneas. Example 8.- Ex Vivo Transport Studies

Ex vivo corneal transport studies were carried out with corneal tissues obtained from porcine eyes from the local slaughterhouse (MercaValencia). Corneal permeation of Coumarin6-loaded NLC was evaluated with such control eye samples at 115 mg/mL (as schematized in Fig. 8a).

8.1. Experimental procedure

Briefly, the whole eye was placed in the support with the outer surface of the cornea facing the donor side with 0.8 cm² contact area. 0.05 ml PBS containing 115 mg/mL. Coumarin6- loaded NLC was added on the contact area by installation as shown in Fig. 3a. Corneas and aqueous humour were collected from at 0 (Control), 15, 30 and 60 min, carrying with them microscopy or aqueous humour studies, as explained below. Each eye was used for one time. During ex vivo experiments, no tear turnover participated in loss of Coumarin6- NLC.

8.2. Histology and Microscopy

Corneal tissue was fixed with 4% PFA during two hours at room temperature and cryoprotected in a sucrose gradient (15-20%-30%) and was immersed in OCT freezing media, and snap frozen in isopentane cooled by liquid nitrogen. Corneas were cryosectioned (Thermo Shandon Cryotome E, Cambridge, UK) at 20 pm. Cryosections were counterstained with DAPI, mounted in Fluoromount-G (#0100-01 , Southern Biotechnology, Birmingham, AL, USA). Cryosections were visualized under a fluorescence microscope (100X magnification, LEICA, Dmi8, Leica Microsystems CMS GmbH, Mannheim, Germany). Some significant images obtained with the control and after different incubation times with Coumarin6-NLC particles are shown in Fig. 8b, where distribution of the nanoparticles to the epithelium and stroma can be observed.

8.3. Aqueous humour studies

Collected aqueous humour was used to quantify amount of Coumarin6-NLC that had completely passed through the cornea where the limiting layers of trans-corneal permeation, as explained in the "Background of the Invention" section, are the epithelium and the stroma. The fluorophore Coumarine 6 was detected in aqueous humour using plate reader Multiskan Sky High microplates (Thermo Fisher Scientific, Madrid, Spain).

Results In spite of the pore size of the corneal epithelium (2 nm) and its size (17 times smaller than the conjunctiva), it was observed that Coumarin6-NLC were able to penetrate the cornea and enhanced topical drug delivery. The results are represented in Fig. 8c, which shows that distribution to the aqueous humour was also observed.

Example 9. In Vivo Transport and Ocular Distribution Studies

In order to complement the assays of Examples 7 and 8, in vivo transport and ocular distribution studies were carried out. For in vivo experiments, it is necessary to consider some factors that reduce the amount of NLCs or ADA reaching the retina including loss of nanoparticles to the tear film, adsorption of tear proteins, rapid clearance via systemic and lymphatic circulation. Then, the assays were carried out as follows:

9.1. Animals

Wild-type C57BL/6J mice (Charles River Spain (Cerdanyola del Valles, Barcelona, Spain) with the same genetic background as rd10 mice were used as a control group for ocular distribution studies. Mice were kept under a 12-hour light/dark cycle, humidity and temperature controlled and with food and water supplied ad libitum. All cages were placed on the lower shelf of an IVC rack with light illuminance of 115 ± 7 lux (95% Cl: 98-131). Mice were housed in the Animal Facility of Research Center Principe Felipe (Cl PF) of Valencia. This study was carried out in accordance with the European Union Guidelines for the Care (European Union Directive (2010/63/EU) and the guidelines for the Use of Laboratory Animals. All animal procedures and protocols were approved and monitored by the Committee of Ethics in Research Center Principe Felipe and the local government (Conselleria de Agricultura, Desarrollo Rural, Emergencia Climatica y Transition Ecologica, Generalitat Valencia) with reference 2020/VSC/PEA/0124 type 2. At least six mice were used.

9.2. Experimental Procedure

A study was conducted to measure the biodistribution of fluorescent NLCs in the eyes of mice in vivo. The treatment was prepared by suspending dry Coumarin6-loaded NLCs in PBS at 107 mg/mL. One drop of either the NLCs suspension or PBS (approximately 2 L) was applied to both eyes (NLCs left eye and PBS right eye) of each animal. The mice were kept separate during the incubation period to prevent cross contamination of treatments between mice. Ocular biodistribution was analysed at five timepoints post-application: 0 min, 30 min, 60 min, 120 min and 24 hours, collecting eyes after the mentioned times.

9.3. Histology and Microscopy

To obtain retinal sections, the eyes were rapidly removed and fixed in 4% filtered paraformaldehyde for two hours at room temperature and cryoprotected in a sucrose gradient (15-20%-30%). The eyes were frozen, embedded in OCT and 10 m sections were cut in a cryostat (Leica CM1900, Nussloch, Germany). Immunofluorescent staining procedures were performed in 10 pm cryosections. Sections were post-fixed in 4% filtered PFA in 0.1 M phosphate buffer pH 7.4 for 15 minutes at room temperature. After sections were counterstaining with DAPI and mounted with Fluoromount G (#0100-01 , Southern Biotechnology, Birmingham, AL, USA), and imaged using a fluorescence microscope (100X magnification, LEICA, Dmi8, Leica Microsystems CMS GmbH, Mannheim, Germany).

Results Some of the images obtained for eye, cornea and retina at 0, 30, 60 and 120 minutes are shown in Fig. 9. As can be seen in said images, 30 minutes after instillation, NLCs were distributed throughout the retina in C57BL/6J mice.

9.4. Experimental Procedure

A study was conducted to detect the presence of ADA in the retina of control mice after topical or intravitreal administration of ADA-loaded NLCs or ADA alone. The treatment was prepared by suspending dry ADA-loaded NLCs in PBS at 107 mg/mL. For ophthalmic use, one drop of NLC-ADA suspension (approximately 2 pL) was applied to one eye of each animal. For intravitreal use, an intravitreal injection of NLC-ADA or ADA alone (0.5 pL) was administered in one eye of each animal. Eyes treated with PBS serve as controls. The mice were kept separate during the incubation period to prevent cross contamination of treatments between mice. The presence of ADA in the retina was analysed at two timepoints post-application: one and three hours, collecting eyes after the mentioned times.

9.5. Isolation of IgG anti-Adalimumab antibodies

To detect ADA that reached the retina after intravitreal or ophthalmic application, protein A expressed in Staphylococcus aureus Sp. coupled to an agarose size exclusion chromatography base matrix (called Sepharose-CL 4B) was used (#P-3391 , Sigma-Aldrich, Madrid, Spain). Protein A contains five regions that bind to the Fc region of IgG. In addition, Protein A has affinity for certain variants of the Fab region. Protein A is covalently attached to the affinity resin Sepharose-CL 4B, making them suitable for low-pressure antibody isolation, e.g., Adalimumab. Protein A lacks of affinity for mouse IgG 1 but it has affinity for human lgG1 such as Adalimumab, at physiological pH. The typical binding capacity of Protein A is anywhere from 15 to 35 mg of human IgG per milliliter of resin (Sepharose) (Fishman JB et al. Cold Spring Harb Protoc. 2019 Jan 2;2019(1). doi: 10.1101/pdb.prot 099143. PMID: 30602558). Briefly, the eyes were rapidly removed and retinas were homogenized in 50 pL 0.1M PBS pH 7.2 and incubated for two hours at 37°C to allow the release of ADA from NLCs. Protein A-Sepharose was prepared considering the swelling factor (1 g swells to 4 to 5 mL) supplied by the manufacturer Sigma-Aldrich. Fifty microliters of protein A-sepharose CL-4B (which corresponds to 11 mg of dry protein A-sepharose, called beads) were used for each retinal homogenate. Ten microliters of retinal homogenate were incubated with 50 L of protein A-sepharose in a tube rotator (ELMI Intelli-Mixer™ RM-2L, ELMI SIA, Riga, Latvia) for 30 min at room temperature and 26 rpm. After incubation, samples were centrifuged at 300 g for one minute. The supernatants were saved. The pellets (beads) were washed three times with 0.1 M PBS pH 7.2 containing 0.25 % Triton. Each supernatant was saved until the final product quality and yield was determined. After last wash, ADA was eluted from beads using 50 L of 0.1 M glycine (pH 3) and neutralized with Tris 1 M pH 8.8. The eluates were used to perform a dot blot against Adalimumab. Two microliters of the eluate were applied directly onto a PVDF membrane (# 10600023, Amersham Hybond, Ge Healthcare Life Science, Germany), which was then incubated with a specific antibody against Adalimumab (dilution 1 :2000, #mab9616 Bio- Techne R&D Systems, s.I.u, Madrid, Spain). After washing, the membrane was incubated with a Horseradish peroxidase (HRP)-conjugated secondary antibody and a HRP substrate (# orb1147872, Superkine weste Femto, Biorbyt, CliniSciences group, Madrid, Spain). PBS and 107 mg/mL NLC-ADA solutions and untreated retinas were used as controls. ADA was detected by chemiluminescence using the chemiluminescence and epifluorescence imaging system Alliance Q9 from UVITEC ((UVITEC Cambridge, United Kingdom).

Results A image of a blot dot showing eluates from retinal homogenates of eyes after instillation of NLC-ADA or intravitreal administration ADA are shown in Fig. 10. As can be seen in said images, ADA was detected in samples after three hours of intravitreal or ophthalmic administration in C57BL/6J mice. This study confirmed that after the instillation of NLC-ADA, ADA reached the retina and it was detected by dot blot.

Example 10.- In Vivo Effect of topical treatment with ADA-loaded NLC

Once confirmed retinal distribution of NLCs after topical application, the effect of ADA- loaded NLC on rd10 mice, a model of autosomal recessive RP, was assessed. As previously described, chronic inflammation (e.g., upregulation of TNFa) is observed during RP progression and it could exacerbate retinal degeneration. The group of the present inventors previously described that intraperitoneal injections of ADA or a single intravitreal injection of ADA ameliorated retinal degeneration at P18 and P23, respectively. Here, it was assessed whether NLC-ADA is capable to delayed retinal degeneration at P18, when a significant retinal dysfunction and a peal of photoreceptor degeneration was observed. The following specific assays were carried out in the conditions set forth below: 10.1. Animals

Rd10 mice were used as a murine model of autosomal recessive RP. Wild-type C57BL/6J mice with the same genetic background as rd10 mice were used as a control group. Both strains were obtained from Charles River, Spain (Cerdanyola del Valles, Barcelona, Spain). Mice were kept under a 12-hour light/dark cycle, humidity and temperature controlled and with food and water supplied ad libitum. All cages were placed on the lower shelf of an IVC rack with light illuminance of 115 ± 7 lux (95% Cl: 98-131). Mice were housed in the Animal Facility of Research Center Principe Felipe (CIPF) of Valencia. This study was carried out in accordance with the European Union Guidelines for the Care (European Union Directive (2010/63/EU) and the guidelines for the Use of Laboratory Animals. All animal procedures and protocols were approved and monitored by the Committee of Ethics in Research Center Principe Felipe and the local government (Conselleria de Agricultura, Desarrollo Rural, Emergencia Climatica y Transition Ecologica, Generalitat Valencia) with reference 2020/VSC/PEA/0124 type 2. At least six-eight mice were used for each group and analysis.

10.2. Nanoparticles reconstitution before instillation:

10.3. Experimental Procedure:

To measure the effectiveness of topical treatment with ADA-loaded NLCs, the eyes of rd10 mice received an NLC-ADA instillation of 7.08 pg of ADA, every two days from postnatal day (P) P12 to P18. (2 pL of ADA-loaded NLC: [NLC]: 107 mg/mL; Nominal [ADA]: 3.7676 mg/mL; Real [ADA]: 3.54 mg/mL (3.54 pg/pL), Real ADA amount administered every two days: 3.54 pg/pL x 2 pL = 7.08 pg ~ 7.1 pg). Mice were euthanized at P18.

10.4. Statistical analyses:

Statistical analyses were performed using GraphPad Software 9.0 (Prism; GraphPad Software, Inc, San Diego, CA). Normal distribution of data was analysed by Shapiro-Wilk and Kolmogorov-Smirnov tests. Comparisons between control, rd10 and rd10 + NLCs groups were performed using One-Way ANOVA followed by Tukey's multiple comparisons test or Kruskal Wallis followed by Dunn's multiple comparisons test depending on data distribution (parametric or non-parametric analysis). In some cases, comparisons between rd10 and rd10 + NLCs were performed using Mann- Whitney. At least eight mice/group were used for each analysis. A P value < 0.05 is considered statistically significant. The data were plotted using the Graph Pad Software 9.0. The data were presented as mean ± SEM.

10.5. ERG Recordings

Through a scotopic full-field electroretinogram (ERG), the global function of the retina was measured in treated and untreated rd10 mice in addition to wild-type mice (C57BL 16J) both at P18. The mice were adapted for 12h to the dark since the experiments were carried out in a dark room under a red light. The mice were placed on a temperature-controlled table at 38°C. The mice were inhaled anesthetized with the anaesthetic isoflurane. The pupil was then dilated with the topical application of 1% tropicamide I eye drops (Alcon Cusi, Barcelona, Spain). To obtain the electrophysiological signals, there are four electrodes, two electrodes were placed inside the inferior eyelids of the mice and two reference electrodes were inserted subcutaneously, one at the neck level, and the other, the ground electrode at the base of the tail. All electrodes were connected to a two-channel amplifier where the signals were amplified, averaged, and stored in a Reti-Scan-RetiPort electrophysiology unit (Roland Consult, Brandenburg an der Havel, Germany). The light stimulation device consisted of a Ganzfeld stimulator, which allowed full-field retinal stimulation (RETIport scan 121 , Roland Consult, Brandenburg and der Havel, Germany). Before ERG was recorded, impedance and baseline tests were performed, the latter of which evaluated the noise level in the environment.

Responses were collected simultaneously from both eyes. Scotopic that mainly reflect the function of the rods were caused by low intensity flashes (~ 0.01 cd-s I m2), on the other hand, the responses of the total retina (function of rods plus cones) were caused by flashes of high intensity (>3 cd-s I m2).

The ERG data were collected by the RETI scanning system amplifier with a sampling rate of 2 kHz and analysed with the RETIport software (Roland Consult, Brandenburg an der Havel, Germany). The response of ERG consisted of an initial negative component (a- wave) and a subsequent positive peak (b-wave) evoked by light stimulation.

The amplitude of the a-wave was determined by the distance between the baseline and the first negative peak. The distance between the peak of the a-wave and the next highest positive peak determined the amplitude of the b-wave. For the latencies (implicit time) of the a- and b-waves, the time from stimulus onset to the peak of each wave was measured, respectively. The electrical activity of retinal cells was represented as the mean ± standard error of the mean (SEM) of the b-wave amplitude for each flash of light under scotopic conditions. Results The results of the examination of the retinal function by measuring electrical activity of the retina in response to different light stimuli at P18 are shown in Fig. 11 , panels a to d. At this age, global ERG responses were considerably reduced in untreated rd10 mice (Fig.11a and 11 b). Under scotopic conditions, there was a significant decline in b-wave amplitudes at all flash intensities compared to control mice (Fig.10a). The a-wave was barely detectable. NLC-treated rd10 mice showed better ERG recordings than untreated rd10 mice. As shown in Fig. 10b, all scotopic b-wave amplitudes were significantly higher in NLC-treated rd10 mice than in untreated rd10 mice. The implicit time of a- and b-waves of untreated rd10 mice were significant longer than of control mice for multiple light intensities (Fig. 11c and 11d). NLC-ADA had a prominent effect in reducing a- but not fa- wave implicit times (Figs. 11c and 11d). Therefore, topical application of NLC-ADA partially restored retinal function in rd10 mice at P18.

10.6. Light Avoidance Test

To evaluate the visual response (to evaluate function), the innate aversion of mice to the brightly illuminated area can be used; thus, the behaviour of mice in an illuminated open field that contained a dark zone (light dark box) was examined. The light avoidance test was conducted using a light/dark box consisting of two equal sized compartments (20 cm x 40 cm), as represented in Fig. 12a. The compartments are connected by an aperture that allowed mice to transition between the chambers. The dark chamber is covered with a lid and the light chamber was kept illuminated by a LED light on the top of the chamber, emitting 1000 lux at the floor. Before every test, both chambers were cleaned with 70% ethanol. The mice were placed in the light chamber to explore the box for 5 min (300 s) (only one test per mouse) and their movement was tracked and recorded during that time, (including time spent in each zone, and number of entries in each zone) by computerized video tracking system using a camera in the centre of the box with AnyMaze 7.0 software. The percentage of time that the mouse spent in light zone and the transitions between the chambers were calculated.

Results The results are shown in Fig 12b, where the time that mice spent in light and dark zones is represented. Control mice spent less than 100 s in the light zone, whereas rd10 mice spent more time in the light zone (-200 s) than control mice. After NLC-ADA administration, the light zone significantly decreased (-120 s) and, in the dark zone increased compared to untreated rd10 mice. At P18, control mice had good light perception, and they preferred dark zone. However, rd10 mice showed poor light perception, but NLC- ADA partly restored the light perception. These results supported ERG studies and suggested topical application of NLC-ADA improved visual function of rd10 mice. 10.7. Histology and Microscopy

It was assessed whether NLC-ADA prevented from photoreceptor degeneration by histological evaluation.

To obtain retinal sections, the eyes were rapidly removed and fixed in 4% filtered paraformaldehyde for two hours at room temperature and cryoprotected in a sucrose gradient (15-20%-30%). The eyes were frozen, embedded in OCT and 10 m sections were cut in a cryostat (Leica CM1900, Nussloch, Germany). Immunofluorescent staining procedures were performed in 10 pm cryosections. Sections were post-fixed in4% filtered PFA in 0.1 M phosphate buffer pH 7.4 for 15 minutes at room temperature. Sections were pre-treated with citrate buffer pH 6.0 for epitope retrieval and incubated in the blocking solution containing 5% normal goat serum, 1% bovine serum albumin, and 0.25% Triton X- 100 (#A1388, PanReac AppliChem, Darmstadt, Germany) for one hour. They were later incubated with primary antibody against the microglia marker Iba1 (1 :1000, #019-19741 , Wako Pure Chemical Industries Ltd., Osaka, Japan), GFAP (1 :400, #G3893, Sigma-Aldrich, Madrid, Spain) or PAR (1 :400, #ALX-804-220, Enzo Life Science, Madrid, Spain) overnight at 4°C. Then these retinal sections were incubated with the fluorescence-conjugated secondary antibodies Alexa Fluor 488 or 647 (1 :400, #A-11001 , #A-21235, Invitrogen, Life Technologies, Madrid, Spain) for one hour at room temperature. After labelling with antibodies and counterstaining with DAPI, the sections were mounted in Mowiol, and observed under an SP5 confocal microscope (40X magnification, Leica TCS SP5 Confocal microscope, Leica Microsystems CMS GmbH, Mannheim, Germany).

The acquisition parameters for each fluorophore were adjusted (e.g., gain, smart offset and excitation energy) to get a proper image. In order to quantify the number of the remaining photoreceptors (number of rows of nuclei at outer nuclear layer, ONL), direct counting of PR nuclei in the outer nuclear layer (ONL), and Iba1-positive cells (manual cell counting) and the integrated density of GFAP was carried out using the Imaged open-source Software. Adobe Photoshop 10 software (Adobe Systems Inc., San Jose, CA, United States) was used to process the final images. Because the degenerative process in the rd10 model vary in different retinal locations, several measurements were performed across the entire retina (from the nasal to the temporal retina) for each mouse.

For the quantification of GFAP positive cells the following formula was used to calculate the corrected fluorescence (CF) for each cell layer:

CF= Integrated density of the selected area - (area of selected area X mean fluorescence of background) To evaluate microglial activation, the migration index (M.l) was measured, defined as the number of Iba1-positive cells weighed according to the retinal layer where they are located (M.l = ^(number of Iba1-positive cells in each layer x layer weighted factor)/total number of Iba1-positive cells in the section). The layer weighed factor was 1 for ONL (outer nuclear layer), 0.5 for outer plexiform layer (OPL) and 0.25 for INL (inner nuclear layer). In addition, microglia polarization from MO (quiescent microglia) to activated microglia M1 (pro- inflammatory) or M2 (anti-inflammatory) was analysed by flow cytometry and gene expression (sections 10.8 and 10.9).

Additionally, expression of some components of the NLRP3 inflammasome (ASC, NLRP3, IL18) and RIPK3 (which is involved in a form of cell death called necroptosis together with RIPK1 , and in NLRP3 inflammasome activation independently of the necroptotic pathway) was measured. To that aim, total RNA was extracted, complementary cDNA was synthesized therefrom and RT quantitative PCR was carried out using specific Taqman gene expression assays (comprising a specific pair of primers and a TaqMan probe labelled with the dye FAM or VIC at 5’ and a motif of binding to the minor groove (MGB) and non- fluorescent suppressor (NFQ) at 3’ (Applied Biosystems, Life Technologies Corporation, Carlsbad, California, USA). Specific information related to the chromosome location, genomic map, exon boundary or amplicon length for each TaqMan Gene Expression Assay (Table 6) is found in https://www.thermofisher.com/es/es/home/life-science/pcr/real-time- pcr/real-time-pcr-assays/taqman-gene-expression.html. The results obtained with an endogenous gene (TATA box binding protein or Tbp) were normalized to 1 and gene expression in rd10 or treated rd10 mice with regard to that control was calculated. The procedure is detailed in section 10.9.

The results obtained are shown in Fig 13. Thus, for instance, Fig 13a show the results of quantifying the number of the remaining photoreceptors (number of rows of nuclei at outer nuclear layer, ONL), complemented by some significant microphotographs of some analysed samples (Fig. 13b). At P18, NLC-ADA significantly reduced cell loss at ONL in rd10 mice compared to untreated rd10 mice (Fig. 13a and 13b).

In previous studies, the group of the present inventors observed an accumulation of polyADP polymers (PAR), the product of the polyADP ribose polymerase (PARP). Overactivation of PARP can induce cell death (e.g., PARthanatos). Besides, they observed an upregulation in a key molecule related to necroptosis RIPK3 and in some components of the NLRP3 inflammasome. Assembly of the NLRP3 inflammasome (ASC, NLRP3 and caspase 1) leads to caspase 1 -dependent release of the pro-inflammatory cytokines IL1 p and IL18, as well as to gasdermin D-mediated pyroptotic cell death. For that reason, it was herein assessed whether topical application of NLC-ADA affect these markers. As can be seen in Fig. 13c and 13d, NLC-ADA reduced the number of PAR-positive cells; it also reduced Ripk3, and Nlrp3 upregulation. However, NLC-ADA significantly increased Asc expression (Fig. 13e).

It is known that RP progression is accompanied by a sustained and chronic inflammation including reactive gliosis, microglia activation and migration, and upregulation of pro- inflammatory molecules like TNFa, IL6 or I L1 p. In order to analyse whether NLC-ADA was capable of reducing the inflammatory process at P18, additionally to the analysis of the expression of inflammation markers described in point 10.9, it was analysed whether NLC- ADA were capable of reducing the inflammatory process at P18 by histological analyses, complemented with the quantification of Gfap expression as described in section 10.9 below. As can be seen in Fig. 14a and 14b, a reduction in GFAP upregulation after NLC- ADA application in rd10 mice was observed. Besides, the present inventors confirmed its decrease analysing the gene expression in retinal homogenates (Fig. 13e). Microglia cells were labelled with anti-iba1 and microglial migration from the inner to the outer retina was quantified (Fig. 14c). For control mice (images marked with “c”), microglia were mainly limited to the inner retina layers with a horizontally ramified shape. For untreated rd10 mice, microglia migrated, infiltrated the ONL via radially oriented cellular projections and acquired a rounded and amoeboid morphology (Fig. 14d). NLC-ADA reduced microglia migration and, infiltration to ONL where degenerating photoreceptors are located (Fig. 14c and 14d).

10.8. Flow cytometry for microglia and other immune cell analysis

RP induces microglia activation toward M1 pro-inflammatory phenotype which is accompanied by several events such as migration, cytokine secretion (e.g., TNFa, I L1 p, or IL6), proliferation, chemotaxis, and phagocytosis. Alternatively, microglia could polarize into M2 anti-inflammatory phenotype (M2a, M2b, M2c or M2d), which are called classical activation and alternative activation, respectively (Zhang B. et al., 2018). M1 microglia induce inflammation and neurotoxicity, while M2 microglia promote inflammation resolution and restoration of homeostasis through anti-inflammatory factors such as I L10, TGFp and arginase (Arg1). induce anti-inflammatory and neuroprotection. In particular, M2a and M2b play an immunoregulatory role or promote M2 immune response, while M2c inhibits immune response and promote tissue remodelling and 2d stimulates angiogenesis. In addition, infiltration of immune cells (e.g., macrophages or lymphocytes) could occur and contribute to retinal degeneration. It was assessed whether NLC-ADA modulated microglia activation or cell infiltration by flow cytometry and gene expression analyses at P18. Mice were euthanized, and the retinas removed from the eyeballs. Retinas were placed in the RPMI medium and dissociated by gentle trituration (using pipette tips). Dissociated retinal cells were collected by centrifugation at 300* g for 5 min. Retinal cells were washed with PBS containing 1 % FBS and stained with the following antibodies against the corresponding antigens: CD45 Alexa 488 (2 pg/mL, #103121 , Biolegend, Amsterdam, The Netherlands), CD11 b APC (5 pg/mL, #101211 , Biolegend, Amsterdam, The Netherlands) and CD86 PE-Cy7 (1 pg/mL, #105115, Biolegend, Amsterdam, the Netherlands), CD68 (2.5 pg/mL, #566389 BD Biosciences, New Jersey, USA) and TREM2 (5 pg/mL ,#FAB17291S, Biotechne, Minneapolis, USA) for 30 min at 4°C in the dark. After staining, the cells were centrifuged at 300 g for 5 min; the supernatant was discarded and the pellets resuspended in 200 pL of PBS with propidium iodide (PI) at 1 pg/mL. Both retinas comprised a single sample; thus, each sample represented the entire population of immune cells in two retinas.

Flow cytometry of samples was performed using a CytoFLEX S (Beckman Coulter Life Sciences, Indianapolis, IN, USA), equipped with 4 lasers and 13 fluorescence detectors. Data acquisition was performed using the Cyto Expert 2.3 software (Beckman Coulter Life Sciences, Indianapolis, IN, USA). The gating strategy was defined to exclude the dead cells, defined as the cells stained with PI, and the aggregates, using the forward scatter area (FSC-A) and forward scatter height (FSC-H). After the selection of single and live cells, the double expression of CD45 and CD11 b was used to define the resident microglia (CD45low/CD11b+) or macrophages (CD45high/CD11b+). Lymphocytes was defined as positive expression of CD45 and negative expression of CD11 b (CD45+/CD11 b-). Selecting the CD45low/CD11b+ population, the concentration of these cells in each sample was quantified (events/pL) as well as their expression percentage of the markers CD68, CD86 and TREM2 (as %). Analysis was performed using at least seven mice (both retinas) for each experimental group.

Results The results obtained are shown in Fig 15. Thus, for instance, Fig. 15a show the results of quantifying the percentage of each cell population: microglia (CD45low/CD11b+), macrophages (CD45high/CD11b-) and lymphocytes (CD45+/CD11 b-). At P18, NLC-ADA significantly reduced the number of microglia and, lesser extent to infiltrated macrophages and lymphocytes in murine retinas (Fig. 15a). Fig. 15b and Fig 15c show subsets of microglia and macrophage expressing markers of M1 or M2-phenotype, respectively. At P18, untreated rd10 show a significant increase of CD86+, CD68+ cells (measured as fluorescence intensity), both in microglia and macrophages. TREM2+ cells are increased in the subset of macrophages but, they tend to decrease in microglia. NLC-ADA significantly decreased CD86+ and TREM2+ microglia cells (Fig. 15b). The effect of NLC-ADA is less evident in macrophages for these markers (Fig. 15c). These results indicated that ophthalmic use of NLC-ADA was capable of reducing the number of inflammatory cells (microglia and infiltrated macrophages), which would reduce the inflammatory process and contribute to improved retinal function in this model of RP.

10.9. Isolation of Total RNA and cDNA Synthesis

The gene expression of different markers of M1 and M2 microglia was analysed. The expression of other inflammatory components was analysed as well. Total RNA was isolated from frozen retinas at P18 using NZY Total RNA Isolation Kit (#MB13402, Nzytech, Lisboa, Portugal), following the manufacturer’s protocol. RNA concentration was determined by spectrophotometry on the Nanodrop 2000 (Thermo Fisher Scientific, Wilmington, DE, USA). Then, cDNA was synthesized starting from 0.1 pg of RNA by reverse transcription using the PrimeScriptTM rt Reagent Kit (Perfect Real Time) (#RR037A, Takara-Bio, Otsu, Japan), following the manufacturer’s instructions. The conditions of cycling consisted of reverse transcription at 37°C for 15 min and inactivation of reverse transcriptase at 85 °C for 5 seconds. The relative expression of Tnfa, Tnfrl, 111/3, 116, 1110, 1118, Asc, Ripk3, Nlrp3, inos, Gfap, Cd200r, Trem2, Ilr4, Cd64, Cd68, Cd86, Cd163, Ccr7, Cccl1, Cxcl13, Ym1, Vegf and Arg1 was measured in retinas by real-time PCR using thermal cycler (LightCycler® 480 System; Roche, Basel, Switzerland), TaqMan gene expression assays with specific commercial primers and TaqMan probes (Table 6, which contains each one of the codes of the used specific gene expression assays of Thermo Fisher, Madrid, Spain, https://www.thermofisher.com/es/es/home/life-science/pcr/real-time- pcr/real-time-pcr-assays/taqman-gene-expression.html) and Premix Ex Taq master mix for probe-based, real time PCR (#RR390A, Takara-Bio, Otsu, Japan). TATA box binding protein Tbp) gene was used as the housekeeping gene.

Table 6

Real-Time PCR was performed with one cycle of denaturation of 30 seconds at 95°C, continued by 40 cycles of 5 seconds denaturation at 95°C, 30 seconds annealing at 60°C, and one cycle of extension at 50°C for 30 seconds. Relative gene expression was normalized to 1 to determine the changes in the gene expression of untreated rd10 mice and NLC-ADA-treated rd10 mice.

Results: The results corresponding to the gene expression of the markers for different microglia phenotypes in retinas from control mice, untreated rd10 mice and NLC-ADA- treated rd10 mice, are shown in Fig. 16, where it can be seen that retinal degeneration induced a variable upregulation of different markers for M1 and M2 subtypes. RP has a significant impact on M1 , M2a and M2b markers (Tnfa, 111b, 116, Cd64, Cd68, Cd86, Cd200r and Arg 7). NLC-ADA significantly reduced gene expression of the markers of M1 and M2b phenotypes Tnfa, 111/3, 116, Cd68, Cd64, Cd86, and lesser extent of the markers of M2c phenotype Ilr4a, Cxcl13. NLC-ADA increased gene expression of the markers of M2a phenotype Cd200r, Arg1, Cd163 and Ym1. However, it was difficult to conclude the effect of RP or ADA on the microglia polarization at this age. Polarization of microglia is a dynamic and complex process where mixed populations of microglia could conflux at the same time. Some authors suggest that multiple activation states of microglia are not entirely consistent with the 1/M2 classification currently used. Microglia perform multiple functions through various phenotypes, each of which corresponds to a different phenotype (Wang J et al., Heliyon. 2023 Mar 21 ;9(4):e14713).

Example 11.- In Vivo Effect of intravitreal treatment with ADA-loaded NLC

The group of the present inventors previously described (Olivares-Gonzalez et al., The FASEB Journal, 2020 vol. 34(10):13839-13861) that a single intravitreal injection of 0.375 or 4.61 mg/mL ADA ameliorated retinal degeneration at P23. Here, it was assessed whether a single intravitreal injection of NLC-ADA (Real [ADA]: 0.39 mg/mL) was also capable to delayed retinal degeneration at P23. The following specific assays were carried out in the conditions set forth below.

Rd10 mice and control mice were used as in Example 7. At least six-eight mice were used for each group and analysis.

Procedure To measure the effectivity of intravitreal injection treatment with

ADA-loaded NLCs, the eyes of rd10 mice received a single intravitreal dose of ADA-loaded NLCs (reconstituted as described in Example 7) at P12. Intravitreal injection was administered in one eye (left eye), while the contralateral eye was injected with 0.1 M phosphate-buffered saline (PBS) (#L0615, Dulbecco’s Phosphate Buffered Saline w/o Calcium w/o Magnesium, Biowest, Nuaille, France) to serve as a control. Mice were anesthetized with isoflurane (AbbVie, Madrid, Spain) at P12. After their pupils were dilated with a drop of topical tropicamide (Alcon, Geneva, Switzerland), 0.5 pL of PBS or NLC-ADA (1.77 pg of ADA, 0.39 pg/pl final concentration of ADA in vitreous humour) were injected into the vitreous humour with a syringe (33G; Hamilton, Bonaduz, Switzerland) using a surgical microscope (#M320, Leica Microsystems SLU, L’Hospitalet de Llobregat, Spain). After intravitreal injection, a drop of tobramycin ophthalmic solution (Alcon, Geneva, Switzerland) was topically applied. Mice were euthanized by cervical dislocation at P23. (0.5 pL of ADA-loaded NLC: [NLC]: 107 mg/mL; Nominal [ADA]: 3.7676 mg/mL; Real [ADA]: 3.54 mg/mL (3.54 pg/pL), Real ADA amount administered: 3.54 pg/pL x 0.5 pL = 1.77 pg in 4.5 pL of vitreous humour, 0.39 pg/pL). Mice were euthanized at P23. In order to compare with intravitreal administration of ADA (alone), a group of rd10 mice was injected into the vitreous humour with 0.5 pL of PBS or ADA (1.6875 pg of ADA, 0.375 pg/pL final concentration of ADA in vitreous humour) at P12.

Statistical

: They were performed using GraphPad Software 9.0 (Prism; GraphPad

Software, Inc, San Diego, CA). Normal distribution of data was analysed by Shapiro-Wilk and Kolmogorov-Smirnov tests. Comparisons between control, rd10 and rd10 + NLCs groups were performed using One-Way ANOVA followed by Tukey's multiple comparisons test or Kruskal Wallis followed by Dunn's multiple comparisons test depending on data distribution (parametric or non-parametric analysis). In some cases, comparisons between rd10 and rd10 + NLCs were performed using Mann-Whitney. At least seven mice/group were used for each analysis. A P value <0.05 is considered statistically significant. The data were plotted using the Graph Pad Software 9.0. The data were presented as mean ± SEM. 11.1. ERG Recordings

ERG recordings were carried out following the same procedure as Example 7 but at P23. Besides, intravitreal administration of ADA alone or NLC-ADA was compared at this age.

The results of the examination of the retinal function by measuring electrical activity of the retina in response to different light stimuli at P23 are shown in Fig. 17, panels a to d. At this age, global ERG responses were considerably reduced in untreated rd10 mice (Fig.17a and 17b). Under scotopic conditions, there was a significant decline in b-wave amplitudes at all flash intensities compared to control mice (Fig. 17a). The a-wave was barely detectable. NLC-treated rd10 mice showed better ERG recordings than untreated rd10 mice. As shown in Fig. 12b, scotopic b-wave amplitudes were significantly higher in NLC-treated rd10 mice than in untreated rd10 mice for light from 0.03 cds/m2 (light intensity). The implicit time of a- and b-waves of untreated rd10 mice were significant longer than of control mice for the higher light intensities (Fig. 17c and 17d). NLC-ADA had a prominent effect in reducing a- but not b-wave implicit times at some light intensities (Figs. 17c and 17d). Therefore, intravitreal application of NLC-ADA partially restored retinal function in rd10 mice at P23. As shown in Fig. 18, intravitreal application of NLC-ADA seemed to be slightly more effective than ADA alone for higher light intensities at P23.

11 .2. Light Avoidance Test

The light avoidance test was carried out following the same procedure as Example 10 but at P23. The results are shown in Fig. 19, where the time that mice spent in light and dark zones is represented. Control mice spent less than 100 s in the light zone, whereas rd10 mice spent more time in the light zone (-150 s) than control mice. As previously observed in Example 10, control mice had good light perception, and they preferred dark zone. rd10 mice showed poor light perception. After NLC-ADA or ADA application, the time in the light zone tended to decrease (-130 s and 125 s, respectively) and, in the time in the dark zone tended to increase compared to untreated rd10 mice. Anyway, neither had a significant impact on light perception at P23.

11.3. Histological evaluation

It was assessed whether intravitreal application of NLC-ADA prevented from photoreceptor degeneration by histological evaluation. Retinal sections were processed as previously described in Example 10. Besides, intravitreal application of ADA alone or NLC-ADA was compared at P23.

The results obtained are shown in Fig. 20. Thus, for instance, Fig. 20a show the results of quantifying the number of the remaining photoreceptors (number of rows of nuclei at outer nuclear layer, ONL), complemented by some significant microphotographs of some analysed samples. At P23, NLC-ADA significantly reduced cell loss at ONL in rd10 mice compared to untreated rd10 mice (Fig. 20a and 20b). Similar effect was observed for ADA alone.

It was analysed whether NLC-ADA were capable of reducing the inflammatory process at P23 by histological analyses, complemented with the quantification of GFAP expression as described in section 8.4 below. As can be seen in Fig. 20c and 20d, a tendency but not significant downregulation of GFAP content NLC-ADA application in rd10 mice was observed. Similar effect was observed for ADA alone. Microglia cells with anti-lba1 were labelled and, quantified microglial migration from the inner to the outer retina (Fig. 20e and 20f). For control mice, microglia were mainly limited to the inner retina layers with a horizontally ramified shape. For untreated rd10 mice, microglia migrated, infiltrated the ONL via radially oriented cellular projections and acquired a rounded and amoeboid morphology (Fig. 20f). NLC-ADA tended to reduce microglia migration and, infiltration to ONL where degenerating photoreceptors are located (Fig. 20e and f). ADA application seemed to be more effective than NLC-ADA application for reducing microglia migration.

Maybe these findings were due to differences in the effective amount of ADA reached the retina after intravitreal administration of NLC-ADA or ADA alone (Example 11). Maybe ADA reached higher concentrations when it is administered loaded in NLCs than when it is administered alone. An excess of ADA could be harmful. On the other hand, a significant protective effect was observed after ophthalmic use of NLC-ADA (Example 10), whose retinal bioavailability could be around 1 % of the instilled NLC-ADA, by 0.4 pg/uL NLC-ADA. When 0.5 pL of 107 pg/uL NLC-ADA was intravitreally administered the amount of ADA reaching the retina may even be too high to have a more neuroprotective effect and the dose would need to be lowered. Therefore, studies with lower doses of ADA are carrying out.

Claims

1. A lipid nanoparticle comprising: a) a lipid matrix comprising a mixture of lipids and/or molecules of lipophilic nature comprising a.1) at least one lipid whose melting point is equal or greater than 25°C, a.2) at least one lipid whose melting point is less than 25°C, b) a TN Fa-antibody, entrapped in the lipid matrix, c) at least one surfactant.

2. The lipid nanoparticle according to claim 1 , wherein the TNFa antibody is adalimumab or a biosimilar thereof which is an antibody that binds to the same epitope that adalimumab and has been approved for its commercialization for the same medicinal purposes.

3. The lipid nanoparticle according to claim 1 or 2, wherein the TNFa antibody is a biosimilar of adalimumab selected from the group of adalimumab-adaz, adalimumab- bwwd, adalimumab-afzb, adalimumab-fkjp.

4. The lipid nanoparticle according to claim 1 or 2, wherein the TNFa antibody is adalimumab.

5. The lipid nanoparticle according to any one of claims 1 to 4, wherein the lipid nanoparticle lacks phospholipids and/or any conjugated peptide and/or any nucleic acid.

6. The lipid nanoparticle according to any one of claims 1 to 5, wherein the ratio of at least one or more lipids whose melting point is equal or greater than 25°C and the at least one or more lipids whose melting point is less than 25°C is from 70 : 30 to 99.5 : 0.5 w/w., respectively.

7. The lipid nanoparticle according to claim 6, wherein the ratio of at least one or more lipids whose melting point is equal or greater than 25°C and the at least one or more lipids whose melting point is less than 25°C is 10:1 w/w.

8. The lipid nanoparticle according to any of the preceding claims, wherein there is only one lipid whose melting point is less than 25°C which is glyceryl palmitostearate and only one lipid whose melting point is equal or greater than 25°C, which is glycerol octanoate decanoate.

9. The lipid nanoparticle according to any of the preceding claims, wherein the monoclonal antibody adalimumab or a biosimilar thereof is present in the nanoparticle in a weightweight relation range with regard to the total nanoparticle weight of 1 :15 to 1 :35.

10. The lipid nanoparticle according to claim 9, wherein the monoclonal antibody adalimumab or the biosimilar thereof is present in the nanoparticle in a weightweight relation range with regard to the total nanoparticle weight of 1 : 25±5.

11. The lipid nanoparticle according to any of the preceding claims, wherein the nanoparticle comprises at least a nonionic surfactant, and at least one surfactant being a triblock copolymer composed of a central hydrophobic chain flanked by two hydrophilic chains is present.

12. The lipid nanoparticle according to claim 11 , wherein the nanoparticle comprises polysorbate 80 and poloxamer 188.

13. The lipid nanoparticle according to any one of claims 10 to 12, wherein the weightweight relation of the total surfactant amount with regard to the total weight of the complete nanoparticle is 3:14 w/w.

14. A pharmaceutical composition comprising at least one lipid nanoparticle according to any one of claims 1 to 13, which is in the form of a powder or as an aqueous dispersion.

15. The pharmaceutical composition according to claim 14, wherein the TNFa antibody is adalimumab.

16. A procedure for the preparation of a lipid nanoparticle of any of claim 1 to 13, which comprises the following steps:

(i) preparing a mixture of the lipids and freeze-dried antibody adalimumab by heating to a temperature slightly above the melting point of the lipid whose melting point is equal or greater than 25°C;

(ii) preparing an aqueous solution comprising one or more surfactants,

(iii) heating the aqueous solution obtained in (ii) up to the same temperature than the lipophilic solution with adalimumab prepared in (i),

(v) cooling down the emulsion obtained in step (iv) at 15°C-25°C for at least 1 hour under magnetic stirring and, subsequently, at refrigeration temperature to allow lipid recrystallization and nanoparticle formation,

(vii) lyophilizing the nanoparticles with a cryoprotectant.

17. A nanoparticle of any one of claims 1 to 13 or a pharmaceutical composition of claim 14 or 15 for use in the treatment of a retinal degenerative disease by the ocular topical route.

18. A nanoparticle of any one of claims 1 to 13 or a pharmaceutical composition of claim 14 or 15 for use according to claim 17, wherein the retinal degenerative disease is retinitis pigmentosa.

19. A nanoparticle of any one of claims 1 to 13 or a pharmaceutical composition of claim 14 or 15 for use according to claim 17 or 18, for use in the treatment of ocular inflammation, increased gene expression of pro-inflammatory molecules, reduced vision, reduced light perception, reduced retinal electrical response to light stimuli and/or photoreceptor degeneration.

20. A nanoparticle of any one of claims 1 to 13 or a pharmaceutical composition of claim 14 or 15 for use according to any one of claims 17 to 19, wherein the TN Fa antibody is adalimumab.