Injectable composition of nonsteroidal Anti-inflammatory drugs, methods and uses thereof

Abstract

The present disclosure relates to injectable compositions, specifically viscosupplement compositions. More particularly, the invention relates to a novel formulation involving the combined use of a pharmaceutical drug, preferably a non-steroidal anti- inflammatory drug, and a low transition temperature mixture as a carrier. This composition is an injectable composition and significantly enhance bioavailability, optimize the release profile, and provide added lubricant properties, making it a suitable for the use in the treatment of joint-related conditions, such as osteoarthritis.

Description

D E S C R I P T I O NI NJECTABLE COM POSITION, M ETHODS AND USES TH EREOFTECH NICAL FIELD

[0001] The present disclosure relates to a viscosupplement composition, in particularto a combined formulation of non-steroidal anti-inflammatory drug (NSAID) and hyaluronic acid (HA), using LTTM (Low Transition Temperature Mixture) as carrier, for improved bioavailability, release profile, and additional lubricant properties.

[0002] The composition of the present disclosure is an injectable composition that significantly enhance bioavailability, optimize the release profile, and provide added lubricant properties, making it suitable for the use in the treatment of joint-related conditions, such as osteoarthritis.BACKGROU ND

[0003] Osteoarthritis (OA) is an inflammatory disease, characterized by cartilage loss between junctions, pain and loss of joint function. It affects 4% of the population worldwide, with a prevalence of 7% in Europe. Currently OA is managed either by the treatment of pain (through the prescription of pain and pain-related medications), viscosupplementation (for patients for who have failed analgesic therapy and from whom oral administration of non-steroidal anti-inflammatory drugs (NSAIDs) and CoOX- 2 inhibitors are contraindicated) or joint replacement surgery (knee arthroplasty is expected to increase to an even greater extent, 673% by 2030, resulting in nearly 3.5 million such operations).

[0004] Viscosupplements are hyaluronic acid (HA) solutions, administered through intra-articular (IA) injection as a supplement of the synovial fluid, a highly viscous liquid that lubricates, nourishes, and protects the joints from shocks and bone friction. This is a mildly invasive procedure, that can be applied 2 to 4-times per year but is sometimes associated with joint swelling and / or inflammation post-injection. In turn, drugs' efficiency, especially for NSAID is impaired by their low bioavailability, which also causes side effects.

[0005] NSAIDs are anti-inflammatory, antipyretic and analgesic medicines commonly used worldwide for the symptomatic relief of headaches, strains, sprains, viral infections as the flu or corona, arthritis and other cases of pain, fever and / or inflammation. Despite being regarded as generally safe, most NSAIDs require the intake of higher amounts than the therapeutic dose due to their general low solubility in the physiological media. Further, these drugs are associated to gastrointestinal, cardiovascular and other adverse effects potentiated by high doses and prolonged consumption. In the case of chronic diseases like osteoarthritis, this is specially concerning as this is the most administered treatment for the persistent symptoms of pain and inflammation. In this context, biocompatible formulations that improve the efficiency and safety of marketed drugs, while avoiding the long-term development of new drugs associated to the regulatory entities are of great importance.

[0006] Within NSAID, celecoxib (CEX) has shown differential therapeutic properties, such as chondroprotective effect. Intra-articular (IA) injections of CEX have shown positive therapeutic outcomes in preclinical research but is not yet used clinically. Low solubility and hence bioavailability of CEX is one of the major challenges associated.

[0007] Recently, the platform of LTTMs, including deep eutectic systems (DES) and therapeutic DES (THEDES) has been presented as a promising alternative strategy for drug delivery by increasing their solubility and / or their biological activity. As expressed by the name, LTTMs present low temperature transitions, which are commonly liquid at room temperature. Their liquefaction is solely promoted by interactions between their components, not involving any chemical reaction and rendering 100% atom economy, which complies with several green metrics. Moreover, their constituents can be chosen from non-toxic, biodegradable and biocompatible molecules, as required for pharmaceutical products. Envisaging therapeutic applications, drugs can be incorporated in LTTMs, either as a starting material or by solubilization. This type of therapeutic formulations has been described as propitious to improve drugs bioavailability by facilitating their permeation, dissolution and absorption. Formulations with increased drug solubility post-administration are highly promising as drug dosage can be minimized up to the effective amount, reaching therapeutic efficiency with lower drug intakes and reduced adverse effects. Regarding NSAIDs, some studies alreadyreported their improved solubility in LTTMs comparatively to aqueous media and other conventional solvents. Lu and co-authors reported 17 salt-based LTTM formulations in which the solubility of several NSAIDs was increased 5 to 17-fold in respect to their solubility in aqueous media [8], In a different study, Hyun et al. studied eutectic mixtures of celecoxib, a NSAID, with adipic acid or saccharin. In comparison to aqueous media or the respective physical mixtures, the eutectics significantly improved the dissolution rate of celecoxib and its wettability [9], Furthermore, Mokhtarpour and colleagues published a study involving 4 NSAIDs and 2 other drugs, with a solubility enhancement of 127 to 136444-fold when dissolved in 5 choline chloride-based LTTMs instead of water. Their experimental results were in accordance to the empiric solubility predictions performed using Hansen solubility parameters (HSP)

[0010] , An additional work, by Palmelund and the co-workers, studied the solubility of 11 drugs, including 5 NSAIDs, in 6 LTTMs, 4 of which choline chloride-based and 2 of betaine:glycerol:water and lactic acid:glucose:water. The drugs solubility in these systems was compared with 4 conventional solvents, water, ethanol, glycerol and polyethylene glycol 300 (PEG300). Despite the positive outputs of LTTMs improving NSAIDs solubility, few therapeutic combinations were studied, most of which based on choline chloride, that may be toxic

[0012] as well as the betaine:glycerol

[0013] ,

[0008] Moreover, the combination of LTTM and polymers to improve drugs bioavailability and control its release has been recently explored. Tuntarawongsa and Phaechamud were the first authors to describe the use of polymeric eutectic systems for controlled delivery of active ingredients [14,15], They used a mixture of two therapeutic compounds, menthol and camphor, to form a therapeutic DES and designed two different delivery systems. The addition of eudragit® polymers to the eutectic liquid up to 40% w / w, increased its viscosity, allowing to form a gel-product as an injectable formulation. The same DES system incorporating 30% w / w of eudragit® was used for the solubilization and delivery of an additional active ingredient, ibuprofen. The hydrophobic properties of the DES prolonged the time of drug delivery in comparison to a pharmaceutical solvent commonly used for drug solubilization. Additionally, the viscosity enhancement conferred by polymer addition contributed to an even higher delivery retardation

[0015] ,

[0009] LTTM-based hydrogels impregnated with silica nanoparticles were also described by Li et al. for the topical delivery of encapsulated methotrexate, an immune- suppressor for rheumatoid arthritis. The strong affinity of the citric acid:arginine:H2O 1:3:16 (molar) LTTM to the skin, allowed a high and sustained penetration of the nanoparticle-matrix into the skin

[0018] ,

[0010] The document CA2121454A1 describes the use of low molecular weight HA and NSAID to inhibit, control and / or regress angiogenesis. This document is however silent on a composition comprising high molecular weight (HMW) HA, needed for higher residence time and increased efficiency, an LTTM-based media for improved bioavailability, NSAID incorporation both dissolved and dispersed for an immediate and prolonged effect, and is also silent on the targeting of specific inflammatory diseases, such as arthritis.

[0011] Document EP3049091A1 discloses a composition combining HA and the anaesthetic mepivacaine. This document is silent on the combination of HA and NSAID, therefore, does not comprise anti-inflammatory action.

[0012] These facts are disclosed in order to illustrate the technical problem addressed by the present disclosure.GENERAL DESCRIPTION

[0013] Viscosupplements and drug delivery systems frequently encounter difficulties related to the solubility and bioavailability of active pharmaceutical ingredients. Traditional formulations often necessitate high doses to achieve therapeutic effects, which can lead to increased side effects. Consequently, there is a significant need for innovative compositions that enhance drug solubility, bioavailability, and therapeutic efficiency while minimizing side effects. The injectable composition of the present disclosure addresses these challenges by providing a solution that improves the solubility and bioavailability of pharmaceutical drugs, namely nonsteroidal antiinflammatory drug (NSAID), thereby enhancing therapeutic outcomes and reducing the risk of adverse effects.

[0014] In an embodiment, it is disclosed a composition comprising HA and an NSAID (in particular, celecoxib), combined in a formulation based on a low transition temperature mixture (LTTM), in particular of glyceroksorbitol (GS), a highly viscous liquid that confers additional lubrication properties, while increasing NSAID's solubility. Surprisingly, the resulting composition showed activity as a viscosupplement with enhanced therapeutic efficiency.

[0015] The present disclosure includes as delivery vehicle an LTTM based media. LTTM are mixtures of two or more components, whose interaction causes their liquefaction. They are normally composed by organic compounds, sugars, aminoacids, terpenes, polyols, and others with similar interaction potential. Thus, they represent a new and wide solvent platform, with more than 106possible combinations. In the present disclosure, LTTM were applied to increase the solubility and / or bioavailability of drugs, especially, NSAIDs. An example is the LTTM of glyceroksorbitol (GS). By increasing drugs bioavailability, LTTM allow higher efficiency and less side effects. Moreover, the polyol composition confers additional lubricant properties, which are beneficial for applications such as joint lubrication, cosmetics, and others. Physiological solutions, such as phosphate buffered solution (PBS), saline solutions or water, may be also used to adjust LTTM properties, such as viscosity, pH, solvation power, etc.

[0016] In an embodiment, the gel matrix of the present disclosure is accomplished by the inclusion of polymers such as hyaluronic acid, a natural lubricant, highly used in pharmaceutical and cosmetic products, such as ocular, cosmetics, viscosupplements, etc. In the pharmaceutic field, HA is used, for example, as main component and active principle, for more than 8 commercial viscosupplements, with proven therapeutic efficiency. Several amounts (most commonly up to l%(w / w)) and molecular weights of HA can be used, according to the desired application. For instance, high molecular weight (HMW) HA, is preferably used for intra-articular injection has it provides higher retention time. HMW HA is also advantageous to achieve complex matrixes for controlled drug delivery.

[0017] In a further embodiment, other natural or synthetic polymers, such as glycosaminoglycans, can be incorporated in the disclosed composition.

[0018] In an embodiment, the mixture of HA and LTTM form a delivery vehicle for the administration of active principles, including but not restricted to NSAIDs, for example, celecoxib. The selection and amount of drug incorporated might be such that it provides therapeutic efficiency. In the case of celecoxib, it provides anti-inflammatory action, which might be useful to treat and prevent inflammation associated to diseases (e.g.: arthritis) or medical procedures (e.g.: intra-articular injections). The inclusion of the selected drug in the dissolved and / or dispersed form, is tuned to the desired application: dissolved for quick release and fast action, and dispersed for higher retention time and delayed or prolonged action.

[0019] In a further embodiment, the disclosed composition also allows the incorporation of other beneficial compounds, such as, extracts, supplements, genetic material, proteins, cellular concentrates or mediators, immunogenic mediators, and others.

[0020] The composition of the present disclosure might be useful for several areas such as pharma, being particularly interesting and especially tuned for intra-articular intended applications.

[0021] In the state of the art, the rheological properties may be measured by many methods. In an embodiment, the rheological properties (storage (G') and loss (G") moduli) were measured using a Modular Compact Rheometer MCR102 (Anton Parr, Austria) coupled with a 2° angle cone-plate geometry (Cone CP25-2, 25 mm diameter, Anton Parr). Frequency sweeps were obtained using a predefined shear strain of 0.1. In a further embodiment, the composition of the present disclosure has an enhanced viscosity. In an embodiment, viscosity was measured, using a Modular Compact Rheometer MCR102 (Anton Parr, Austria), as a function of the temperature (5 to 45°C), with a heating rate of 2°C / min and a constant shear rate of 10 s’1, and as a function of the shear rate (0.01 to 1000 s1) at 37°C. Torque ranged from 0.00004 to 1.02 mN.m.

[0022] The present disclosure relates to an injectable composition comprising: 0.5 to 2% (w / w) of a natural polymer; 0.05 to 0.5% (w / w) of a pharmaceutical drug; and 50 to 90% (w / w) of a low transition temperature mixture. This injectable composition is a delivery vehicle, that surprisingly enhance the solubility and bioavailability of the pharmaceutical drug while conferring additional lubrication properties and stability. Theresulting composition shows improved activity as a viscosupplement with enhanced therapeutic efficiency.

[0023] In an embodiment for better results, the injectable composition comprises: 0.5 to 2% (w / w) of the natural polymer, preferably 0.8 to 1.5 % (\N / \N) 0.05 to 0.5% (w / w) of the pharmaceutical drug; and 50 to 90% (w / w) of the low transition temperature mixture; preferably 60 to 80% (w / w).

[0024] In an embodiment, the low transition temperature mixture includes the nomenclature of deep eutectic system. In a further embodiment, the low transition temperature mixture is a deep eutectic system.

[0025] In an embodiment for better results, the pharmaceutical drug is a nonsteroidal anti-inflammatory drug (NSAID). The composition of the present disclosure enhances the solubility and bioavailability of the NSAID, provides additional lubrication properties, and shows improved therapeutic efficiency as a viscosupplement. The composition of the present disclosure can be used in several applications, particularly for intra-articular injections.

[0026] In an embodiment for better results, the low transition temperature mixture comprises at least two components which are selected from a list consisting of: glycerol, sorbitol, ethyleneglycol, propyleneglycol, mannitol, glucose, fructose, galactose, inositol, xylitol, erythritol, arginine, or mixtures thereof; preferably glycerol, sorbitol, or a mixture of glycerol and sorbitol.

[0027] In an embodiment for better results, the molar ratio between the two components of the low transition temperature mixture ranges from 1:1 to 8:1.

[0028] In an embodiment for better results, the low transition temperature mixture comprises glycerol and sorbitol.

[0029] In an embodiment for better results, the molar ratio between glycerol and sorbitol ranges from 1:1 to 3:1, preferably from 1.5:1 to 2:1.

[0030] In an embodiment, a media based on the low transition temperature mixture further comprises water or a physiological solution, for example phosphate buffered solution (PBS), saline solutions, or mixtures thereof.

[0031] In an embodiment for better results, the natural polymer is selected from a list comprising hyaluronic acid, silk, ulvan, collagen, chondroitin sulphate, gelatin, gellan gum, chitosan, carrageenan, pectin, arabic gum, or mixtures thereof.

[0032] In an embodiment for better results, the natural polymer is hyaluronic acid.

[0033] In an embodiment for better results, the molecular weight of the hyaluronic acid is at least 0.5 MDa. In a further embodiment, the molecular weight of the hyaluronic acid ranges from 1 MDa to 3.6 MDa.

[0034] In an embodiment for better results, the natural polymer is crosslinked.

[0035] In an embodiment, the composition further comprises a synthetic polymer; preferably a glycosaminoglycan.

[0036] In an embodiment, the nonsteroidal anti-inflammatory drug is selected from a list comprising: celecoxib, naproxen, ketoprofen, flurbiprofen, ibuprofen, or mixtures thereof.

[0037] In an embodiment for better results, the injectable composition further comprises chondroitin sulfate, glucosamine sulphate, collagen, or mixtures thereof, preferably 0.5 to 2% (w / w) of chondroitin sulfate, glucosamine sulphate, collagen, or mixtures thereof.

[0038] In an embodiment for better results, the zero-shear-rate viscosity of the composition ranges from 0.3 to 300 Pa.s at 37°C, preferably 10 to 200 Pa.s.

[0039] In an embodiment for better results, the storage modulus at 0.5 Hz and 37°C ranges from 40 to 125 Pa; and / or the loss modulus at 0.5 Hz and 37°C ranges from 35 to 65 Pa.

[0040] In an embodiment for better results, the storage modulus at 2.5 Hz and 37°C ranges from 75 to 180 Pa; and / or the loss modulus at 2.5 Hz and 37°C ranges from 45 to 75 Pa.

[0041] In an embodiment for better results, the ratio between the viscosity at a shear rate of 0.1 s1and a shear rate of 240 s’1, at 37°C ranges from 50 to 300 (Pa.s / Pa.s).

[0042] In an embodiment for better results, the injectable composition further comprises natural extracts, supplements, genetic material, proteins, cells, immunogenic mediators, or mixtures thereof.

[0043] In an embodiment for better results, the injectable composition is for use in medicine.

[0044] In a further embodiment, the injectable composition is for use in the treatment of osteoarthritis, arthritis, or rheumatoid arthritis, preferably in situ treatment.

[0045] In a yet further embodiment, the injectable composition is for use in the treatment of joint disorders or diseases.

[0046] The present disclosure also relates to a viscosupplement composition comprising the disclosed injectable composition.

[0047] An aspect of the present disclosure relates to the use of the disclosed injectable composition for the manufacture of a medicament for the treatment of osteoarthritis, arthritis, or rheumatoid arthritis.

[0048] Another aspect of the present application relates to a method for treating or preventing osteoarthritis, arthritis, or rheumatoid arthritis in a subject, the method comprising administering the disclosed injectable composition.BRI EF DESCRIPTION OF TH E DRAWINGS

[0049] The following figures provide preferred embodiments for illustrating the disclosure and should not be seen as limiting the scope of invention.

[0050] Figure 1: Embodiment of results of frequency sweep profiles of a) GHA and b) GGIuS.

[0051] Figure 2: Embodiment of results of storage (G') and loss (G") moduli of GHA and GGIuS gels, in comparison to the ones reported for commercial viscosupplementations [19-22],

[0052] Figure 3: Embodiment of results of viscosity in function of temperature for a) GHA and b) GGIuS.

[0053] Figure 4: Viscosity as a function of shear rate measured for three different conditions: diluted GHA (dGHA), GHA with pre-shear; GHA without (w / ) pre-shear.

[0054] Figure 5: Embodiment of cell viability percentage determined in ATDC5 cells for the developed gels: GHA, GGIuS, GHA+CEX, GGIuS+CEX.

[0055] Figure 6: Representation of the mean values for the cumulative release of celecoxib (mg / L) from GHA+CEX (circles), GGIuS + CEX (squares), GS (triangles) and PBS (inverted triangles). The lines are a guide to the eye. a) High clearance rate scenario. Statistical analysis:nsGS equivalent to PBS (p-value > 0.9999); GHA equivalent to GGIuS (p-value > 0.18); *GHA significantly different from PBS (p-value = 0.015); **GHA significantly different from GS (p-value = 0.093). b) low clearance scenario. Statistical analysis:nsGHA equivalent to GGIuS (p-value > 0.117); GS equivalent to PBS (p-value > 0.9999); **GHA significantly different from GS and PBS (p-value < 0.005); **** GGIuS significantly different from GS and PBS (p-value < 0.0001).

[0056] Figure 7: Embodiment of POM images of GHA with celecoxib at 0.04 mg / mLGS:PBs (top) and GHA with excess celecoxib (bottom). Left: Non-polarized images; right: polarized images. Scale bar: 100 pm.

[0057] Figure 8: Embodiment of results of1H NMR spectra overtime. Bottom: GGIuS gel at a) day 0, freshly prepared, b) 1 week, c) 2 weeks, d) 1 month, e) 2 months; Top: GHA gels at a) day 0, b) 2weeks, c) 1 month, d) 2 months, e) 1 year. Similar spectra obtained for gels with and without celecoxib.

[0058] Figure 9: Embodiment of results of ipsilateral knee diameter upon OA induction (days -25 to -1) and after IA injection (days 1 to 55), in respect to baseline (healthy, prior OA induction). No knee edema was detected after OA induction.

[0059] Figure 10: Embodiment of dynamic weight bearing (DWB) results analysed by loading in the contralateral limb as response to OA-induced pain. Weight bearing by contralateral limb as a percentage of the total weight distributed by both hind limbs **p < 0.01; *p < 0.05; #p < 0.1.

[0060] Figure 11: Embodiment of pCT results (5 rats per group). A- Trabecular bone (TB) volume of the lateral tibial epiphysis. B- Trabecular (TB) total volume of the lateral tibial epiphysis. C-Trabecular bone (TB) volume fraction (BV / TV) of the lateral tibial epiphysis.D-Trabecular bone (TB) thickness of the medial tibia. Baseline refers to the day priorthe IA injection. ***p < 0.001; *p < 0.05.

[0061] Figure 12: Embodiment of histological evaluation of the injectable effects in OA synovitis and joint degeneration: A - Synovial inflammation as represented by the total Krenn score (0-9). B - Joint degeneration as represented by the total Mankin score (0- 14). C - Degeneration of joint structure denoted by the structure Mankin subscore (0-6), and represented in D - Safranin-O / Fast Green-stained medial joint sections. ***p < 0.001; **p < 0.01; *p < 0.05; #p < 0.1.

[0062] Figure 13: Embodiment of micro-computed tomography (pCT) results of the ipsilateral tibia plateau, analysed by differences between groups (data represented as endpoint / pre-injection) and raw data (changes at day 56 compared to day -1). A - Bone volume fraction (BV / TV) of the medial subchondral bone plate (SBP), corrected by preinjection. B - Thickness of the medial SPB, corrected by pre-injection. C - Raw data for the BV / TV of the medial SPB. D - Raw data for the thickness of the medial SPB. E - pCT frame for ipsilateral knee injected with PBS. F - pCT frame for ipsilateral knee injected with GHA+CEX. G - Bone volume fraction (BV / TV) of the medial trabecular bone (TB), corrected by pre-injection. H - Medial TB thickness, corrected by pre-injection. I - Raw data for the BV / TV of the medial TB. J - Raw data for the medial TB thickness. **p < 0.01; *p < 0.05; #p < 0.1.DETAILED DESCRIPTION

[0063] The present disclosure relates to injectable compositions, specifically viscosupplement compositions. More particularly, the invention relates to a novel formulation involving the combined use of a pharmaceutical drug, preferably a nonsteroidal anti-inflammatory drug; and a low transition temperature mixture as a carrier. Surprisingly, the disclosed composition is injectable and significantly enhances bioavailability, optimize the release profile, and provide added lubricant properties, making it a suitable for use in the treatment of joint-related conditions, such as osteoarthritis.

[0064] In an embodiment, an LTTM of glyceroksorbitol (GS) at 2:1 molar ratio, was mixed with phosphate buffered saline (PBS) as physiologic media at a mass ratio of 2:1. Then, 1.82 mg / mL of celecoxib and 1% HA (w / w) were dissolved in the GS:PBS media.

[0065] In an embodiment, the LTTM should be prepared by heating and stirring. The temperature should be increased in such a way that it does not overpasses the decomposition temperature of its constituents, and the stirring method should be adjusted to production scale. In an embodiment, the GS can be easily prepared at a temperature of 80°C.

[0066] In an embodiment, LTTM can then be mixed and homogenised with the physiologic solution (PBS), always using weight for precise ratios, due to the high viscosity of LTTM and its variability with temperature. The prepared media, for example GS:PBS 2:1 (w / w), is then used to solubilize drug(s). In an embodiment of the application of celecoxib, 1.82 mg / mL is then dispersed by stirring at 37°C for at least 24h, to ensure partial solubilization (forfast release) and homogeneous dispersion of the non-dissolved part (for prolonged / delayed release). The natural polymer, such as HA, is then added, at a 1% mass proportion, and stirred at 37°C, until a homogeneous gel is obtained.

[0067] In an embodiment envisioning intra-articular injections, the disclosed composition comprises both NSAID and viscosupplement agents in a unique formulation, while presenting rheological properties enhanced in respect to osteoarthritic synovial fluids, and in the range of commercial viscosupplements and healthy synovial fluids. Injectability was ensured up to needle gauges of 30G.

[0068] In an embodiment, the formulated gels were tested in terms of their rheological properties for viscosupplementation, their biocompatibility was evaluated in a chondrogenic cell line and the in vitro release profile of CEX investigated.

[0069] One of the most important features commonly characterized for injectable compositions, such as intra-articular (IA) injections, is their rheological behaviour, to understand how they respond to applied forces. This is highly relevant to assess the feasibility of manipulation and / or injectability, and their response to load forces, such as weight-bearing forces, upon injection. In the case of joint viscosupplementations, they must mimic healthy synovial fluids, in terms of its ability to lubricate the joint andabsorb mechanical shock, protecting cartilage. These properties are normally compromised in impaired synovial fluids, such as osteoarthritic ones. Therefore, IA viscosupplements should improve them and restore or enhance their normal function. Thus, they must have viscoelastic properties that allow its structure to quickly adjust when submitted to specific stimuli. These can be assessed though viscosity profiles as a function of several variables of interest, such as temperature and shear rate. For instance, the zero-shear-rate viscosity (rjo) is a parameter that characterizes the viscosity of a material at rest. Moreover, the loss (G") and storage (G') modulus are also widely studied, as they refer to the viscous and elastic component of the formulation, respectively. For weight-bearing joints, such as the knees, it is especially important to characterize the G' and G" under frequencies corresponding to walking (0.5 Hz) or running (2.5 Hz) activities.Example

[0070] In an embodiment, glycerol (99.5%, ref. GL0026) was purchased from Scharlau (Spain) and sorbitol (> 98%, ref. S1876) from Sigma Aldrich (France). Phosphate buffer saline tablets (PBS, ref. P4417) were acquired from Sigma Aldrich (USA) and reconstituted as described by the manufacturer (pH 7.4, at 25 °C). Hyaluronic acid sodium salt (HA, high molecularweight: > 1 MDa, ref. J66993) and glucosamine sulphate (98%, ref. J66271) were bought from AlfaAesar (China).

[0071] In an embodiment, the chondrogenic cell line ATDC5, derived from mouse teratocarcinoma AT805, was obtained from the European Collection of Authenticated Cell Cultures (ECACC 99072806, UK). DMEM / F-12 (Dulbecco's Modified Eagle Medium / Nutrient Mixture F-12) with (ref. 11320074) and without phenol red (ref. 21041) were purchased from Gibco (UK). The solutions of 0.25% trypsin (ref. 25-050-CI), Fetal Bovine Serum (FBS, ref. 35-079-CV) and the antibiotic-antimycotic solution (10000 lU / mL penicillin, 10 mg / mL streptomycin and 25 pg / mL amphotericin, ref. 30-004-CI) were acquired from Corning (USA). The MTS solution (3-(4,5-dimethylthiazol-2-yl)-5-(3- carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, <2%) was obtained from Promega (USA).

[0072] In an embodiment, for the high-performance liquid chromatography (HPLC), acetonitrile (>99.9%, ref 34967) from Honeywell, Riedel-de Haen (Germany) andpotassium dihydrogen phosphate (98-102%, ref 11591) from AlfaAesar (Germany) were used. The potassium dihydrogen phosphate 50 mM 50:50 (v / v) buffer was prepared in deionized water and the pH adjusted to 4.2, using phosphoric acid (85-90%, ref 79606) from Honeywell, Fluka (Germany).

[0073] In an embodiment, the normality of the results was assessed by the Shapiro- Wilk test. Those presenting a normal distribution were analysed by One-way ANOVA (viability profiles and drug release results) or Unpaired t-test with Welch's correction (frequency sweeps, GHA vs GGIuS); whereas the ones following a non-normal distribution were analysed by Mann-Whitney test (viscosity vs temperature, GHA vs GGIuS). The results were considered statistically different if p-value < 0.05.

[0074] In an embodiment of gel preparation, low Transition Temperature Mixtures (LTTM) of glyceroksorbitol 2:1 molar ratio (GS) were prepared by heating and stirring, as reported elsewhere

[0023] , The liquid GS was mixed with PBS at a mass ratio of 2:1, and used as media for gel preparation. Four types of gels were then produced. All gels are composed of 1% HA (w / w). Another group of gels was also supplemented with glucosamine sulphate (GluS), adding 10 mg of GluS per gram of gel. For the gels with CEX, this was incorporated by dispersing 2 to 4 mg / ml (above saturation) of CEX in the GS:PBS mixture and stirred for 24h before the addition of HA and / or glucosamine. The composition and nomenclature of the gels is discriminated in Table 1.Table 1 - Composition and name given to the gel samples formulated.Composition Gel nameGS:PBS+HA GHAGS:PBS+HA+GluS GGIuSGS:PBS+HA+CEX GHA+CEXGS:PBS+HA+GluS+CEX GGIuS+CEX

[0075] In an embodiment, the rheological properties of the gels were studied in a Modular Compact Rheometer MCR102 (Anton Parr, Austria) using a 2° angle cone-plate geometry (Cone CP25-2, 25 mm diameter, Anton Parr). All measurements were performed in the linear viscoelastic region (LVR) of the materials, determined by amplitude sweep at a frequency of 10 Hz, 37°C and within a shear strain ranging from 0.001 to 20 (LVR from 0.001 to 1). The real range was adjusted by the equipment,according to its ability to reach a stable reading. Frequency sweep analysis was performed at a shear strain of 0.1 and 37°C. The gels were equilibrated at 37°C for 5 min and the storage (G') and loss moduli (G") were then determined as function of the frequency (100 to 0.01 Hz). The viscosity of gels was also measured as a function of the temperature (5 to 45°C). The gels were stabilized at 5°C for 5 min and submitted to a heating ramp with a rate of 2°C / min and a constant shear rate of 10 s ~1. For the shear rate, the gels were pre-equilibrated at 37°C for 5 min, and viscosity obtained as a function of the shear rate, from 0.01 to 1000 s-1. The effect of pre-shear was also studied by applying a second shear rate cycle on the samples. All analyses were performed for three individual replicates of each sample.

[0076] The rheological properties of the formulated gels were characterized aiming to assess their ability to be used as viscosupplements for OA. Figure 1 represents the frequency sweep profiles of GHA and GGIuS, which were statistically equivalent. The storage modulus (G') represents the elastic component of the system, while the loss modulus (G") refers to the viscous component, thus being also referred to as elastic and viscous moduli, respectively. As it can be observed, above the crossover frequency (Fc), G' is higher than G", indicating the transition from viscous to elastic-fluid behaviour. In the synovial joint, this elasticity provides cartilage protection by absorbing mechanical energy

[0024] , Additionally, the G' (elastic component) is slightly dependent on frequency and the G” values (viscous component) have a close magnitude to G', mainly for lower frequencies. These supports some fluidity of the systems rather than a compact-solid- like structure. From these plots, it was possible to estimate the values of Fc, relaxation time (tp;) and the G' and G” at the frequency representative of walking (0.5 Hz) or running activities (2.5Hz). These results are summarized in Table 2.Table 2 - Results of crossover frequency (Fc), relaxation time (tR) and the storage modulus (G') and loss modulus (G") at the frequency representative of walking (0.5 Hz) or running activities (2.5Hz). G' and G" near 0.5 and 2.5 Hz are represented as mean ± SD. Fc and tRwere estimated from the crossover of G' and G' lines, plotted from their mean values (Figure 2).Walking RunningG' ~ 0.5Hz G" ~ 0.5Hz G' ~ 2.5Hz G" ~ 2.5HzSample Fc (Hz) tR(s)(Pa)GHA 0.02 50 78.7 ± 34.6 44.6 ± 9.1 119.4 ± 41.2 54.3 ± 8.6GGIuS 0.04 25 81.3 ± 42.3 45.0 ± 16.2 121.7 ± 55.2 54.5 ± 16.9

[0077] It is reported that healthy synovial fluids have Fc within 0.13 ± 0.02 to 0.41 ± 0.12

[0025] and commercial viscosupplementations present Fc from under 0.01 to over 10

[0019] , The Fc of the GHA replicates ranged from values bellow 0.01 to 0.14 Hz and for the GGIuS from 0.02 to 0.14 Hz. The corresponding mean Fc values obtained from the crossover between G' and G” (Figure 1), were 0.02 Hz for GHA and 0.04 Hz for GGIuS, with respective relaxation times near 50 and 25 s (Table 2).

[0078] In Figure 2, it is also possible to compare the G' and G” values at the walking (0.5 Hz) and running (2.5 Hz) frequencies, with the ones reported for commercial viscosupplementations [19-22], Considering both G' and G”, the values obtained forthe studied systems were more consistent with the commercial products Monovisc® and Orthovisc®, especially with Orthovisc®. In sum, the properties herein studied are in the range of commercial formulations and healthy synovial fluids.

[0079] In another embodiment, the gels were characterized in terms of their viscosity at different temperatures (Figure 3). As it can be observed for the GHA, there is a slight decrease of the viscosity profile when temperature increases. Within room temperature and body temperature (20-37°C), the gel has a viscosity between 10-12 Pa.s with no significant differences and therefore no substantial effect on the gel injectability is expected. The system GGIuS showed no significant differences from GHA.

[0080] Since the rheological properties of GHA and GGIuS previously analysed did not show significant differences, the remaining analysis were performed for GHA, as representative of both.

[0081] Knee joints have the highest volume of synovial fluid. Given that a normal knee joint contains about 6 mL of synovial fluid, this can be totally replaced with the maximum viscosupplementation volume (6 mL) or supplemented with lower volumes, most commonly, 2 mL. In this context, upon injection, the viscosupplementation might be in the concentrated form (GHA) or diluted 3-fold (dGHA) in the synovial fluid. In OA condition, the synovial fluid is commonly diluted due to effusion. Therefore, the 3-folddilution (dGHA) was performed with PBS to mimic a fictious worst -case scenario where the OA synovial fluid would have a viscosity near the physiological media.

[0082] In an embodiment envisioning an intra-articular injection, the gels are expected to be constrained and submitted to compression (or load) forces within the joint, especially in weight-bearing joints. High compression causes the gel to be more restrained (compacted), thus corresponding to low shear rates. Lower loading-forces allow higher gel flow, which corresponds to high shear rates

[0019] , Therefore, the shear rate profile is an important analysis for intra-articular-intended formulations, to assess the effect of the shear rate in the material properties. Herein, the shear rate analysis was performed for both concentrated (GHA) and diluted (dGHA) formulations, at 37°C. In Figure 4 it can be observed that the gels behave as non-Newtonian fluids, presenting shear thinning behaviour (viscosity decrease as shear rate increases), a characteristic of healthy synovial fluids and commercial viscosupplementations, that allows an adjustable while reversible response to loading forces. Moreover, the effect of pre-shear treatment was analysed to study the capacity of the gel to maintain or recover its properties upon exposure to shear rates (from low to high). The results support that the pre-shear treatment influences the viscosity profile of GHA for shear rates < 1 s ~1. Without preshear, the gels presented an almost linear decrease of viscosity with increasing shear rates. In turn, for the pre-sheared gels, lower but constant viscosity values were obtained for shear rates below 1 s ~1. Above 1 s’1, the viscosity profile of the pre-sheared and non-pre-sheared gels converged. These occurrences upon pre-shearing are consistent with thixotropic behaviour, reinforcing the reversible re-organization of the gels structure. In comparison to the non-pre-sheared GHA, the diluted gel (dGHA) had an equivalent viscosity profile for shear rates < 1 s’1; and above those, a deeper viscosity decrease was observed.

[0083] The phenomena of shear thinning and thixotropy are also good indicators of injectability, since the injection force increases the shear stress and shear rate, which lowers the gel viscosity, facilitating its administration.

[0084] In a further embodiment, the zero-shear-rate viscosity (qo) is an additional parameter, usually used to characterize the viscosity of a material at rest. Commonly, qo is defined as the viscosity measured at minimum shear rate tested. Here, the qo wasconsidered as qo.i, to be comparable with the marketed viscosupplementations characterized by Nicholls et al.

[0019] . The results obtained for the tested formulations, at 37°C, are summarized in Table 3. All of these formulations have a qo higher than the maximum value reported for OA synovial fluids (0.01-11 Pa.s), as well as within the range reported for healthy synovial fluids (1-175 Pa.s)

[0024] and commercial viscosupplementations (0.3-193 Pa.s)

[0019] , These attest the suitability of the disclosed composition for the proposed application as viscosupplement.Table 3 - Shear rate properties obtained for the GHA and dGHA formulations.Sample qo.i (Pa.s) q24o (Pa.s) no.i / r]24oGHA with pre-shear 39 ± 10 0.69 ± 0.06 56GHA w / pre-shear 185 ± 94 0.63 ± 0.07 290 dGHA 97 ± 71 0.05 ± 0.02 1925

[0085] Furthermore, the ratio between low shear rates (represented by rjo.i) and high shear rates (represented by r|24o) is commonly used as a quantitative measure of the shear thinning behaviour - shear thinning ratio (r|o.i / r|24o). According to Rebenda et al., the shear thinning ratio of healthy synovial fluids ranges from 70 to 250 while osteoarthritic synovial fluids have values between 5 and 40

[0024] , Moreover, the values reported by Nicholls et al. for commercial viscosupplementations vary from 2 to 651

[0019] , As observed for the experimental values of Table 3, the GHA based systems presented, once more, rheological properties superior to the ones of OA and within the range reported for normal synovial fluids. Moreover, in comparison to the shear thinning ratios reported for the commercial viscosupplementations, the range values of GHA (56-290) comprehend the magnitude of values stated for Monovisc®(52), Orthovisc® (170), Euflexxa® (237) and Gel-one® (243).

[0086] In the work of Bhuanantanondh and co-workers

[0020] , they reported the shear rate curves for mixtures of OA-synovial fluid (OA-SF) with three commercial viscosupplementations in a ratio of 1:1 (v / v): Suplasyn® (low molecular weight HA), Orthovisc® (high molecular weight HA) and Synvisc® (cross-linked HA). In comparison, the viscosity profile of GHA was surprisingly superior to the ones of OA-SF+suplasyn and OA-SF+orthovisc and equivalent to the profile of OA-SF+synvisc. Therefore, although GHA was produced from high molecular weight HA, like Orthovisc®, the combinationOA-SF+orthovisc presented viscosities clearly inferior up to shear rates of 1 s ~1, after which they converged. Thus, a full replacement of the synovial fluid by the GHA formulation, would be expected to render similar shear-rate properties to OA- SF+Synvisc, a viscosupplement where the HA is crosslinked.

[0087] Overall, the produced gels share rheological characteristics with some of the commercial viscosupplements, being therefore adequate for viscosupplementation. Moreover, these properties can be further tuned by polymer cross-linking or by changing the molecular weight.

[0088] In an embodiment, the in vitro biocompatibility of the developed gels was evaluated in the chondrogenic cell line ATDC5, representative of joint cells, by determining the cell viability through MTS colorimetric assay, in accordance to ISO 10993-5. The gel leachables were prepared according to ISO 10993-12, for gel concentrations from 0.6 to 0.075 g / mL, successively diluted in DMEM / F-12 with 0.5% (v / v) FBS. Cells were cultured in DMEM / F-12 with phenol red, supplemented with 10% (v / v) FBS and 1% (v / v) of penicillin-streptomycin solution (PS, see preparation of culture media). Media was changed every 2-3 days, until confluency (70-80%) was achieved. For the cell viability, cell suspensions of 1.5xl05cells / mL were prepared in culture media and seeded in a 96-well plate (100 pL per well), for 24h. The media was then removed, and the cells submitted to 100 pL of each gel concentration, for 24h. After this period, the cells were washed with PBS and 100 pL of MTS (diluted 62.5-fold, in DMEM / F- 12+0.5% (v / v) FBS) was added and incubated for 3h. The absorbance at 490 nm was then measured using a VICTOR Nivo™ microplate reader (PerkinElmer, Waltham, USA). Each sample was analysed in triplicate, in three independent cell assays. The cell viability was determined as previously described

[0023] , All cell incubation periods were performed in a humidified incubator, with 5% CO2, at 37 °C.

[0089] As it is possible to observe from Figure 5, all the gels present a similar viability profile and no statistically significant differences were observed between the four formulations tested (p-value > 0.98). Therefore, the results for GHA and GGIuS were considered as representative. According to the international standards of ISO 10993-5 and 10993-12, a polymeric leachable in a concentration of 0.2 g / mL (± 10%) with aviability > 70%, is considered non-cytotoxic. These criteria were met since their mean viability at 0.2 g / mL was 73.5% (64±21% for GHA and 83±14% for GGIuS).

[0090] For gel concentrations > 0.3 g / mL, the viability was below 2%. At this higher concentration range, the limiting component(s) might be the glyceroksorbitol content which had an IC50 near 0.09 g / mL. As reported in a previous work

[0023] , the glyceroksorbitol mixture may cause some osmotic stress in vitro, mainly induced by sorbitol. However, given the dynamic tendency of cells to achieve homeostasis, this is not expected to occur in vivo. In fact, in vivo, glycerol can be a potential adjuvant in joint lubrication while its combination with sorbitol results in a more viscous lubricant, which is advantageous for the intended application. Moreover, glycerol and sorbitol are FDA- approved excipients for several human products, and sorbitol is even reported for administration via intra-articular routes up to a maximum unit dose of 45% (w / v)

[0029] , a concentration higher than the formulations herein tested contain.

[0091] For gel concentrations < 0.175 g / mL, the cell viability increased up to 26% compared to the control (100%). This supports that, at certain concentration, the gels are biocompatible and can even stimulate cell metabolic activity, most probably due to the extracellular matrix (ECM) biopolymers included in the gel.

[0092] In an embodiment, the drug release study was performed mimicking two different scenarios, one representing an extreme gel clearance rate and the other a low gel clearance rate. Briefly, about 0.67 g of the gels (GHA+CEX and GGIuS+CEX) or the controls (GS and PBS) were placed in 1.5 mLglass vials closed with a cellulose membrane (0.45 pm, Sartorius, Germany) in one end. These were placed in 15 mL conical tubes, previously filled with 2 mL of PBS, as release media, with the membrane facing down. The conical tubes were then placed in a thermostatic water bath (Grant, LSB 12 Aqua Pro, UK) at 37°C, with a longitudinal agitation of 60 rpm. At pre -determined time points, aliquots of the released media were collected and replaced by the same volume of fresh PBS. For the extreme gel clearance scenario, all the released media was removed and replaced by 2 mL of fresh PBS and for the setting mimicking a low gel clearance, 200 pL were collected and replaced by fresh PBS. The amount of celecoxib in the collected aliquots was quantified by high performance liquid chromatography (HPLC), as previously described elsewhere

[0023] ). Briefly, 20 pL were injected in a Smartline HPLCSeries (Knauer, Germany), eluted in acetonitrile:potassium dihydrogen phosphate 50 mM 50:50 (v / v) at pH 4.2, using a Keystone Kromasil C18 column (250 x 4.6 mm 5 pm, Thermo Scientific, USA) with an isocratic flow of 0.6 mL / min, and detected at 220 nm in a Smartline 2500 UV-detector (Knauer, Germany).

[0093] Mimicking the drug release of the intra-articular environment in vitro is not straightforward as this is a very dynamic environment. Osteoarthritic synovial fluids tend to have HA clearance rates elevated, in comparison to healthy ones. The lower the HA clearance rate, the higher its retention time and effectiveness, both in terms of action and durability. Ideally, upon viscosupplementation, the normal properties of synovial fluids are restored, thus restoring a normal, low clearance rate. Herein, both scenarios were studied for a more comprehensive understanding of the developed formulations: Figure 6a represents the release profile at an extreme gel clearance scenario, where the released media is totally removed at each time point; and Figure 6b represents a low clearance of the gel, where only 10% of the media is removed at each time point. The gel formulations GHA+CEX and GGIuS+CEX were compared with the same matrix without the natural polymer (named GS) and with PBS, as reference, to understand how the gel constituents modulate the release of CEX in vitro. From the results, in both scenarios there is an increased release of CEX from the gels in comparison to GS or PBS. However, there is a clear difference between their maximum CEX release and their release profiles.

[0094] In an embodiment of results, for the high clearance method (Figure 6a), the maximum release was achieved at 8 h, with a CEX concentration up to 6-fold higherthan for GS or PBS. After these first hours, the release reaches a plateau. In turn, for the low clearance path (Figure 6b), a release profile with supersaturation points (e.g.: 6 h and 48 h for GHA) is observed for all the conditions tested, except for PBS. These are related to the solubility and recrystallization of CEX, which depend on its solvation in each of the formulations and on the gel structure, when present

[0031] , In this low clearance set-up, the maximum CEX release was achieved forthe gel formulations at 48 h, with an increase up to more than 17-fold or 73-fold in comparison to GS or PBS, respectively. Surprisingly, at the 48 h release from the gels, the CEX amount retrieved in low clearance was more than 10-times higher than the amount obtained at high clearance. This differencereflects the importance of the gel structure in the increment of celecoxib's concentration. In the case of a high clearance scenario, the gel disintegration is faster which translates into a lower and less prolonged CEX release. In a low clearance situation, a fluid gel structure is maintained, confining higher CEX amounts in the polymeric network. Therefore, the longer the gel structure is maintained upon injection, the more extended is the release of CEX at higher concentration. Still, even if celecoxib is not retained for more than 48 h, it can contribute to the immediate reduction of local inflammation and prevent possible inflammation and subsequent sweeling associated with post-intra-articular injection procedures. Reduced inflammation allows to restore or maintain the viscoelasticity of the synovial fluids' or injected viscosupplements', thus reducing the clearance rate of HA and prolonging their protective and therapeutic function.

[0095] In an embodiment to quantitatively determine the maximum CEX solubility in the gel, a GHA+CEX gel was prepared as previously described, with a celecoxib concentration of 0.04 mg or near 2 mg (excess CEX) per mL of GS:PBS. The obtained mixtures were placed on a microscope glass slide and observed by Polarized Optical Microscopy (POM), using a BX-53 polarized optical microscope (Olympus, Japan) in transmission mode. Images were obtained from an equipped camera (Olympus SC50), through the software Olympus Stream Start 2.4.2 (Olympus, Japan). Celecoxib was considered soluble when no crystals were observed by POM.

[0096] Figure 7 depicts an embodiment of results of the POM observation. At least 0.04 mg / mL of CEX appear to be soluble in GHA+CEX, since no CEX crystals are observed at this concentration (Figure 7, top), in contrast to the saturated formulation (Figure 7, bottom). This amount is near the maximum CEX concentration obtained from the gels release (Figure 6) and it represents improvements above 5-fold and 25-fold than the maximum solubility previously reported for GS-rich media and PBS, respectively

[0023] ,

[0097] In an embodiment of evaluation of long-term stability of the gels, proton nuclear magnetic resonance (1H NMR) spectra were recorded on a Bruker Avance III 400 spectrometer (Bruka, Billerica, USA) at 400 MHz. About 300 pL of each sample was mixed with 500 pL of dimethyl sulfoxide-d6 (DMSO-d6) for analysis. For stability studies,reagents and prepared samples were kept at room temperature. MestReNova software (11.0.4-18998) was used for data analysis and the chemical shifts expressed in ppm.

[0098] In an embodiment of results, freshly prepared GHA and GGIuS presented equivalent results for most of the analysis performed. However, the study of gels' stability over time demonstrated that GGIuS gels were not stable for long periods. As it can be observed in the NMR spectra obtained for GGIuS at different time points (Figure 8, bottom), after 1 month, the signals between 6 4 and 5 ppm converge. This can be justified by the occurrence of Maillard reaction promoted by the presence of glucosamine, therefore modifying the composition and properties of the gel. Thus, for the GGIuS gels to be suitable for viscosupplementation or drug delivery system, strategies for the stabilization of glucosamine would have to be studied. In turn, the NMR spectra obtained for the GHA gels (Figure 8, right) did not present alterations up to 1 year, which support chemical stability under this period, at room temperature. No differences between the gels with and without celecoxib were noticed.

[0099] In an embodiment of the in vivo studies, all the reagents used were pharma grade. Glyceroksorbitol 2:1 (GS, molar ratio) was prepared by heating and stirring. Four injectables were prepared: Phosphate buffer saline (PBS) for the control group, HA solution in PBS (HA group), a solution of HA and celecoxib in PBS (HA+CEX) and a solution of HA and celecoxib in GS:PBS 2:1 (w / w) (GHA+ CEX). HA was added at lmg / g solution and celecoxib at 1.82 mg / mL solution.

[0100] In an embodiment, the in vivo experiments were conducted in agreement with the Dutch Law of Animal Experimentation and approved by the Animal Ethics Committee in Utrecht, the Netherlands (project number AVD11500202114837) and in compliance with the ARRIVE guidelines. Knee osteoarthritis was induced unilaterally by the combination of anterior cruciate ligament transection (ACLt) and partial medial meniscectomy (pMMx) in the right knee of 24 rats (Sprague-Dawley female, 12-weeks old, Crl:CD(SD) strain code 001, Charles River), at day -28. The animals were priorly acclimatized, monitored, fed with standard chow diet and weighted weekly through all study. Prior surgery, the rats were sedated by inhalation with isoflurane (induction 4%, maintenance 1.5-3%). Buprenorfine (0.03 mg / kg) and Carprofen (4mg / kg) were subcutaneously administered at least 30 min prior-surgery, and buprenorphinereinforced post-surgery, 6 hours after the first administration. The 2 following days (day -27 and -26), the animals also received a daily dose of 4 mg / g of carprofen, for pain management.

[0101] Four weeks after OA induction (day 0), the animals were randomized in 4 groups of 6 rats each, (sample size calculation from previously reported data by Rudnik- Jansen et al.

[0032] , using G-Power v3.1.9.7). Each rat received a 25 pL unilateral intraarticular injection (30G needle), through the patellar tendon, in the OA-induced knees. The injectable formulations were administered to the respective groups: HA, HA+celecoxib (HA+CEX), HA+GS+celecoxib (GHA+CEX), and control (PBS). The following assays and data analysis were performed in random order by blinded observers.

[0102] Euthanasia was performed at day 56 by aortic excision under anaesthesia. 1 rat of PBS and HA+CEX groups were sacrificed at days 1 and 48, respectively, due to lesions external to the experiment. Upon sacrifice, knees were harvested for further analysis.

[0103] In an embodiment to assess knee sweeling, a digital calliper (Kunzer 7EMS01, Germany) was used to measure rat knees' diameter, at predetermined time points, along the study, before and after OA induction and intra-articular injection. A mean value was calculated from three consecutive manual measures for each timepoint.

[0104] In an embodiment, pain associated behaviour was assessed through dynamic weight bearing (DWB) analysis. Rats were individually placed in a 22 x 22 x 30 cm plexiglass chamber with floor sensors containing pressure transducers (44 x 44 captors, 10.89 mm2per captor), and allowed to move freely during all procedure. 5 min videos were recorded using a high-resolution camera (640x480 pixel), coupled to DWB software (Bioseb, module vl.4.2.98, France) that measured, in grams, the average weight exerted by each limb. The zone detection parameters were set as follows: low weight threshold > lg, weight threshold > 1.5 g, surface threshold > 2. The weightbearing limbs were validated by a blinded observer, for at least 1.5 min video frames, through association between rat position and sensor activation. 3 day-measures before OA induction were considered as baseline. Data is expressed in Figure 10 as the weight on the contralateral limb as a percentage of the total weight distributed by both hind limbs (mean ± SD).

[0105] In a further embodiment, the systemic CEX levels were taken as indirect measure of intra-articular concentration. The GHA+CEX reduced CEX systemic exposure, due to a prolonged CEX retention in the joints, and / or slower CEX release in the joints, but at such low levels that these result in undetectable systemic levels. Surprisingly, the increased CEX bioavailability conferred by GS, and the longer therapeutic action of GHA+CEX, provides the analgesic effect for GHA+CEX but not for HA+CEX, compared to PBS. A CEX improved action also results in a higher therapeutic potential of GHA+CEX in comparison to HA+CEX in terms of the ability to inhibit bone changes and cartilage degeneration.

[0106] In an embodiment, micro-computed tomography scans (pCT) were acquired in a Quantum FX CT scanner (PerkinElmer, Massachusetts), one day prior the intra-articular injection (day -1) and post-mortem (endpoint). The in vivo measurements were carried under isoflurane anaesthesia (3.5 - 4% (v / v) induction, 1.5 - 2.5% (v / v) maintenance). The hind legs were extended and scanned for 3 min at an isotropic voxel size of 42 pm, 90 kV voltage, 180 pA current and a field of view of 21 mm. 3D reconstructed images were obtained, and 2D images were reconstructed using software Analyze 11.0 (PerkinElmer; RRID:SCR_009120). 512 serial slides (2D) were then analysed using ImageJ software (ImageJ, 1.50i, NIH, USA). Images were converted to 8-bit, the brightness / contrast parameter set from 100 to 255 and, an autolocal threshold algorithm (vl.16.12, Bernsen method, 5 radius) was applied. Semi-manual segmentation for the tibial subchondral bone plate and trabecular bone was done and segmented (polygon selection) regions of interest (ROI) registered. ROI were generated each 5 slides, starting from the union of the medial and lateral side of the tibial epiphysis (back to front joint direction). Interpolation was applied and corrected when necessary.

[0107] In an embodiment, the harvested knees were surgically cleaned and fixated using 4 to 10% (v / v) neutral buffered formalin (NBF), preferably 10% (v / v) NBF, for 5 to 8 days, preferably 5 days. Thereafter, joints were decalcified in 0.5M ethylenediamine tetraacetic acid (EDTA) solution for a total of 7 weeks, with a refixation step overnight in 10% NBF every 2 week. The decalcified knees were then dehydrated in series of alcohols using a tissue processor (Leica biosystems, Netherlands), cleared in xylene, and further infiltrated and embedded in paraffin wax blocks. 5 pm transversalsections were obtained using a Leica micrometre (RM2245, China) and stained (Leica autostainer XL, Germany) with Safranin-O, Fast Green and haematoxylin (Saf-O), as well as with Hematoxylin and Eosin (HE). HE stained sections were used to assess synovitis, by the Krenn score

[0034] , Briefly, the parameters of synovial lining cell layer enlargement, resident cells density and inflammatory infiltrate were scored from 0 to 3, reaching a total of 0-9, where 0-1 means no synovitis; 2-4: low-grade synovitis and 5-9: high-grade synovitis. In turn, joint degeneration was evaluated in Saf-0 stained slides through the Mankin score

[0033] , In short, 4 categories were scored as follows: cartilage degradation (0-5), cellularity (0-3), Saf-0 staining (0-4), tidemark integrity (0-1). Their sum provides total scores ranging from 0 (healthy) to 14 (severe joint degeneration). Both score systems were randomly and individually attributed by two blind evaluators. Averaged scores were used for data representation, plotted as mean ± SD.

[0108] For the scope and interpretation of the present disclosure it is defined that "room temperature" should be regarded as a temperature between 15-30 °C, preferably between 18-25 °C, more preferably between 20-22 °C.

[0109] In an embodiment, an in vivo experiment was conducted using HA, HA+CEX or GHA+CEX as experimental conditions and PBS as control, using for each condition 5 rats treated with the respective composition. Micro-computed tomography (pCT) results, and histology evaluation through Mankin and Krenn score are shown. Statistical differences were explored by Brown-Forsythe and Weich ANOVA. P-values < 0.05 were considered as significant. pCT analysis showed a significant increase from baseline of bone volume (BV) (Figure 11 A) and total volume (TV) (Figure 11, B) in untreated animals, which was not observed for any of the treatments. No significant differences between treatments or compared to the baseline were found for bone volumetric fraction (BV / TV, Figure 11, C). Surprisingly, there was a trend towards inhibition of the OA-induced increase from baseline in thickness of the medial trabecular bone (TB) (Figure 11, D) by treatment with GHA+CEX but not the other HA-based injections.

[0110] In an embodiment, statistical analysis for in vivo results (max. 6 rats per group) was performed with GraphPad Prism 8.0.1. Normality was assessed by Shapiro- Wilk tests and Q-Q plots. Knee diameter measurements were assessed by mixed modelanalysis for group and timepoints comparison. pCT analysis was performed using Brown- Forsythe and Welch ANOVA tests for normal data, while non-normal data was assessed by Kruskal-Wallis tests. The histological scoring was analysed by two-way-ANOVA, to correct for inter-observer variability. P-values < 0.05 were considered significant. To decrease the chance of discarding false negatives, p-values < 0.1 were also reported.

[0111] In an embodiment for the in vivo results (max. 6 rats per group), the knee diameter measures prior to OA induction were subtracted to the diameter of OA knees at each time-point, to assess for knee edema derived from OA progression. No OA- induced knee edema and no difference between groups were found regarding knee diameter (Figure 9). Regarding weight bearing analysis, differences between the nonaffected limb (contralateral) and OA limb (ipsilateral) were used as pain indicator. Painbehaviour was noted by weight load transfer from the affected to the non-affected limb, as measured by dynamic weight bearing. All HA treatments showed analgesic action (Figure 10).

[0112] Histological analysis depicted reduced synovitis in all HA treatment groups (Figure 12, A). Surprisingly, cartilage degeneration was only inhibited by HA in combination with CEX and GS (GHA+CEX), but not with HA+CEX or HA alone (Figure 12, B-D), highlighting the added value of the GS in preventing OA progression. In terms of the pCT results (Figure 13), changes in the tibial subchondral bone plate and trabecular bone were assessed by comparing data of the endpoint (day 56) to the one obtained pre-injection (day -1). Post hoc comparison of group effects was performed upon data normalization by dividing the endpoint by the pre-injection data. Results were expressed as mean ± SD. When evaluating healthy subchondral bone plate (SBP), bone volume fractions (BV / TV) near 1 are expected as the volume of mineralized bone should match the total volume. OA-progression as measured by the decreased BV / TV of the medial subchondral bone plate was observed for all treatments, except for GHA+CEX, which, unexpectedly, prevented decreases in BV / TV caused by deterioration, therefore protecting the SBP integrity (Figure 13, A, C). Additional trabecular bone changes were inhibited by GHA+CEX but not the other treatments. Surprisingly, intra-articular CEX formulation was able to inhibit subchondral bone deterioration, at a single dose, 180x up to 5000x lower than the oral doses required to inhibit bone loss in arthritis ratmodels. In sum, HA+GS+CEX presented equivalent therapeutic potential to HA viscosupplementation in terms of pain and synovitis, while conferring additional therapeutic outcomes: inhibition of bone-changes and chondroprotective action. Thus, HA+GS+CEX not only provided chondroprotective action, but it also prevented bone lesions, emphasising the synergistic effect between HA, GS and CEX in potentiating those effects, retarding OA-progression.

[0113] The term "comprising" whenever used in this document is intended to indicate the presence of stated features, integers, steps, components, but not to preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

[0114] The disclosure should not be seen in any way restricted to the embodiments described and a person with ordinary skill in the art will foresee many possibilities to modifications thereof. The above-described embodiments are combinable.

[0115] The following dependent claims further set out particular embodiments of the disclosure.References:1. Ghlichloo, I.; Gerriets, V. Nonsteroidal Anti-inflammatory Drugs (NSAIDs); StatPearls Publishing, 2020;2. Markowicz-Piasecka, M.; Mikiciuk-Olasik, E. 2.3.1 Non-steroidal antiinflammatory drugs (NSAIDs). In Nanobiomaterials in Drug Delivery. Applications of Nanobiomaterials Volume 9; Grumezescu, A.M., Ed.; William Andrew, 2016; p. 44.3. Aroso, I.M.; Silva, J.C.; Mano, F.; Ferreira, A.S.D.; Dionisio, M.; Sa-Nogueira, I.; Barreiros, S.; Reis, R.L.; Paiva, A.; Duarte, A.R.C. Dissolution enhancement of active pharmaceutical ingredients by therapeutic deep eutectic systems. Eur. J. Pharm. Biopharm. 2016, 98, 57-66.4. Faggian, M.; Sut, S.; Perissutti, B.; Baldan, V.; Grabnar, I.; Dall'Acqua, S. Natural Deep Eutectic Solvents (NADES) as a tool for bioavailability improvement: Pharmacokinetics of rutin dissolved in proline / glycine after oral administration in rats: Possible application in nutraceuticals. Molecules 2016, 21, 1531.5. Silva, E.; Oliveira, F.; Silva, J.M.; Matias, A.; Reis, R.L.; Duarte, A.R.C. Optimal design of THEDES based on perillyl alcohol and ibuprofen. Pharmaceutics 2020, 12, 1121.6. Francisco, M.; Van Den Bruinhorst, A.; Kroon, M.C. Low-transition-temperaturemixtures (LTTMs): A new generation of designer solvents. Angew. Chemie - Int. Ed. 2013, 52, 3074-3085. Nguyen, C.-H.; Augis, L.; Fourmentin, Sophie Barratt, G.; Legrand, F.-X. Deep Eutectic Solvents for Innovative Pharmaceutical Formulations. In Deep Eutectic Solvents for Medicine, Gas Solubilization and Extraction of Natural Substances; Fourmentin, S., Gomes, M.C., Lichtfouse, E., Eds.; Springer Nature, 2020; pp. 41- 102. Lu, C.; Cao, J.; Wang, N.; Su, E. Significantly improving the solubility of nonsteroidal anti-inflammatory drugs in deep eutectic solvents for potential nonaqueous liquid administration. Medchemcomm 2016, 7, 955-959. Hyun, S.-M.; Lee, B.J.; Abuzar, S.M.; Lee, S.; Joo, Y.; Hong, S.-H.; Kang, H.; Kwon, K.-A.; Velaga, S.; Hwang, S.-J. Preparation , characterization , and evaluation of celecoxib eutectic mixtures with adipic acid / saccharin for improvement of wettability and dissolution rate. Int. J. Pharm. 2019, 554, 61-71. Mokhtarpour, M.; Shekaari, H.; Zafarani-Moattar, M.T.; Golgoun, S. Solubility and solvation behavior of some drugs in choline based deep eutectic solvents at different temperatures. J. Mol. Liq. 2020, 297, 1-11. Wen, Q.; Chen, J.-X.; Tang, Y.-L.; Wang, J.; Yang, Z. Assessing the toxicity and biodegradability of deep eutectic solvents. Chemosphere 2015, 132, 63-69. Ahmadi, R.; Hemmateenejad, B.; Safavi, A.; Shojaeifard, Z.; Mohabbati, M.; Firuzi, O. Assessment of cytotoxicity of choline chloride-based natural deep eutectic solvents against human HEK-293 cells: A QSAR analysis. Chemosphere 2018, 209, 831-838. Benlebna, M.; Ruesgas-Ramon, M.; Bonafos, B.; Fouret, G.; Coudray, C.; Durand, E.; Figueroa-Espinoza, M.C.; Feillet-Coudray, C. Toxicity of Natural Deep Eutectic Solvent Betaine:Glycerol in Rats. J. Agric. Food Chem. 2018, 66, 6205-6212. Tuntarawongsa, S.; Phaechamud, T. Polymeric Eutectic System. Adv. Mater. Res. 2012, 528, 180-183. Tuntarawongsa, S.; Phaechamud, T. Polymeric Eutectic Drug Delivery System. J. Met. Mater. Miner. 2012, 22, 27-32. Liu, Y.; Zhang, Y.; Chen, S.N.; Friesen, J.B.; Nikolic, D.; Choules, M.P.; McAlpine, J.B.; Lankin, D.C.; Gemeinhart, R.A.; Pauli, G.F. The influence of natural deep eutectic solvents on bioactive natural products: studying interactions between a hydrogel model and Schisandra chinensis metabolites. Fitoterapia 2018, 127, 212-219. Qin, H.; Owyeung, R.E.; Sonkusale, S.R.; Panzer, MJ. Highly stretchable and nonvolatile gelatin-supported deep eutectic solvent gel electrolyte-based ionic skins for strain and pressure sensing. J. Mater. Chem. C 2019, 7, 601-608. Li, M.; Cui, H.; Cao, Y.; Lin, Y.; Yang, Y.; Gao, M.; Zhang, W.; Wang, C. Deep eutectic solvents— Hydrogels for the topical management of rheumatoid arthritis. J. Control. Release 2023, 354, 664-679. Nicholls, M.; Manjoo, A.; Shaw, P.; Niazi, F.; Rosen, J. Rheological properties of commercially available hyaluronic acid products in the United States for the treatment of osteoarthritis knee pain. Clin. Med. Insights Arthritis Musculoskelet. Disord. 2018, 11, 1-5. Bhuanantanondh, P.; Grecov, D.; Kwok, E.; Guy, P. Rheology of osteoarthritic synovial fluid mixed with viscosupplements: A pilot study. Biomed. Eng. Lett.2011, 1, 213-219. Bhuanantanondh, P.; Grecov, D.; Kwok, E. Rheological study of viscosupplements and synovial fluid in patients with osteoarthritis. J. Med. Biol. Eng. 2012, 32, 12. Mazzucco, D.; McKinley, G.; Scott, R.D.; Spector, M. Rheology of joint fluid in total knee arthroplasty patients. J. Orthop. Res. 2002, 20, 1157-1163. Roda, A.; Paiva, A.; Duarte, A.R.C. Therapeutic Liquid Formulations Based on Low Transition Temperature Mixtures for the Incorporation of Anti-Inflammatory Drugs. Pharmaceutics 2021, 13, 1620. Rebenda, D.; Vrbka, M.; Cipek, P.; Toropitsyn, E.; Necas, D.; Pravda, M.; Hartl, M. On the Dependence of Rheology of Hyaluronic Acid Solutions and Frictional Behavior of Articular Cartilage. Materials (Basel). 2020, 13, 2659. Balazs, E.A. The physical properties of synovial fluid and the special role of hyaluronic acid. In Disorders of the Knee; Helfet, A.-, Ed.; T. B. Lippincott Company: Philadelphia, 1974; pp. 63-75 ISBN 0397503229. Heilmann, H.-H.; Lindenhayn, K.; Walther, H.-U. Das Synovia-Volumen gesunder und arthrotischer menschlicher Kniegelenke. Z. Orthop. Unfall. 2008, 134, 144- 148. Schurz, J.; Ribitsch, V. Rheology of synovial fluid. Biorheology 1987, 24, 385-399. Chen, M.H.; Wang, L.L.; Chung, J.J.; Kim, Y.-H.; Atluri, P.; Burdick, J. A. Methods to Assess Shear-Thinning Hydrogels for Application As Injectable Biomaterials. ACS Biomater. Sci. Eng. 2017, 3, 3146-3160. FDA Inactive Ingredient Search for Approved Drug Products Available online: https: / / www.accessdata.fda.gov / scripts / cder / iig / index.cfm (accessed on Dec 6, 2021). Strecanska, M.; Danisovic, L.; Ziaran, S.; Cehakova, M. The Role of Extracellular Matrix and Hydrogels in Mesenchymal Stem Cell Chondrogenesis and Cartilage Regeneration. Life 2022, 12, 2066. Patel, D.D.; Anderson, B.D. Maintenance of Supersaturation II: Indomethacin Crystal Growth Kinetics Versus Degree of Supersaturation. J. Pharm. Sci. 2013, 102, 1544-1553. Rudnik-Jansen, L; Tellegen, A.R.; Pouran, B.; Schrijver, K.; Meij, B.P.; Emans, P.J.; de Gendt, E.; Thomas, R.E.; Kik, M.J.L.; de Visser, H.M.; et al. Local controlled release of corticosteroids extends surgically induced joint instability by inhibiting tissue healing. Br. J. Pharmacol. 2019, 176, 4050-4064. Mankin, H.J.; Dorfman, H.; Lippiello, L.; Zarins, A. Biochemical and metabolic abnormalities in articular cartilage from osteo-arthritic human hips. II. Correlation of morphology with biochemical and metabolic data. J. Bone Joint Surg. Am. 1971, 53, 523-37. Krenn, V.; Morawietz, L.; Burmester, G.-R.; Kinne, R.W.; Mueller-Ladner, U.; Muller, B.; Haupl, T. Synovitis score: discrimination between chronic low-grade and high-grade synovitis. Histopathology 2006, 49, 358-364.

Claims

C L A I M S1. Injectable composition comprising:0.5 to 2% (w / w) of a natural polymer;0.05 to 0.5% (w / w) a pharmaceutical drug; and50 to 90% (w / w) of a low transition temperature mixture.

2. Injectable composition according to the previous claim wherein the pharmaceutical drug is a nonsteroidal anti-inflammatory drug.

3. Injectable composition according any of the previous claims comprising:0.8 to 1.5 % (w / w) of a natural polymer; and60 to 80% (w / w) of a low transition temperature mixture.

4. Injectable composition according any of the previous claims wherein the low transition temperature mixture is a deep eutectic system.

5. Injectable composition according to any of the previous claims wherein the low transition temperature mixture comprises at least two components which are selected from a list consisting of: glycerol, sorbitol, ethyleneglycol, propyleneglycol, mannitol, glucose, fructose, galactose, inositol, xylitol, erythritol, arginine, or mixtures thereof.

6. Injectable composition according to the previous claim wherein the molar ratio between the two components of the low transition temperature mixture ranges froml:l to 8:1.

7. Injectable composition according to any of the previous claims wherein the low transition temperature mixture comprises glycerol and sorbitol.

8. Injectable composition according to the previous claim wherein the molar ratio between glycerol and sorbitol ranges from 1:1 to 3:1, preferably is 1.5:1 to 2:1.

9. Injectable composition according to any of the previous claims wherein the low transition temperature mixture further comprises water or a physiological solution, preferably phosphate buffered solution (PBS), saline solutions, or mixture thereof.

10. Injectable composition according to any of the previous claims wherein the natural polymer is selected from a list comprising hyaluronic acid, silk, ulvan, collagen, chondroitin sulphate, gelatin, gellan gum, chitosan, carrageenan, pectin, arabic gum, or mixtures thereof.

11. Injectable composition according to any of the previous claims wherein the natural polymer is hyaluronic acid.

12. Injectable composition according to the previous claim wherein the molecular weight of the hyaluronic acid is at least 0.5 MDa.

13. Injectable composition according to the previous claim, wherein the molecular weight of the hyaluronic acid ranges from 1 MDa to 3.6 MDa.

14. Injectable composition according to any of the previous claims wherein the natural polymer is crosslinked.

15. Injectable composition according to any of the previous claims wherein the composition further comprises a synthetic polymer; preferably a glycosaminoglycan.

16. Injectable composition according to any of the previous claims 2-15 wherein the nonsteroidal anti-inflammatory drug is selected from a list comprising: celecoxib, naproxen, ketoprofen, flurbiprofen, ibuprofen, or mixtures thereof.

17. Injectable composition according to any of the previous claims further comprising chondroitin sulfate, glucosamine sulphate, collagen, or mixtures thereof, preferably 0.5 to 2% (w / w) of chondroitin sulfate, glucosamine sulphate, collagen, or mixtures thereof.

18. Injectable composition according to any of the previous claims wherein the zero- shear-rate viscosity of the composition ranges from 0.3 to 300 Pa.s at 37°C; preferably 10 to 200 Pa.s.

19. Injectable composition according to any of the previous claims wherein the storage modulus at 0.5 Hz and 37°C ranges from 40 to 125 Pa; and / or the loss modulus at 0.5 Hz and 37°C ranges from 35 to 65 Pa.

20. Injectable composition according to any of the previous claims wherein the storage modulus at 2.5 Hz and 37°C ranges from 75 to 180 Pa; and / or the loss modulus at 2.5 Hz and 37°C ranges from 45 to 75 Pa.

21. Injectable composition according to any of the previous claims wherein the ratio between the viscosity at a shear rate of 0.1 s1and a shear rate of 240 s ~1, at 37°C ranges from 50 to 300 (Pa.s / Pa.s).

22. Injectable composition according to any of the previous claims further comprising natural extracts, supplements, genetic material, proteins, cells, immunogenic mediators, or mixtures thereof.

23. Injectable composition according to any of the previous claims for use in medicine.

24. Injectable composition according to the previous claim for use in the treatment of joint disorders or diseases.

25. Injectable composition according to the previous claim for use in the treatment of osteoarthritis, arthritis, or rheumatoid arthritis.

26. Viscosupplement composition comprising the injectable composition described in any of the previous claims.

27. The use of an injectable composition as described in any of the previous claims 1- 25 for the manufacture of a medicament for the treatment of osteoarthritis, arthritis, or rheumatoid arthritis.

28. A method for treating or preventing osteoarthritis, arthritis, or rheumatoid arthritis in a subject, the method comprising administering the injectable composition of any of the previous claims 1-25 to the subject.

Images