MICROSPHERE-BASED INJECTIBLE CELECOXIB FORMULATION

This application claims the benefit of U.S. Provisional Application No. 62/884,526, filed August 8, 2019, the contents of which are incorporated herein by reference.

Throughout this application, various publications are cited. The disclosure of these publications is hereby incorporated by reference into this application to describe more fully the state of the art to which this invention pertains.

Field of the Invention

The present invention relates to methods for treating joint-related disorders via localized injection of celecoxib-containing biodegradable microspheres.

Background of the Invention

NSAIDs and Celecoxib

Non-steroidal anti-inflammatory drugs (NSAIDs) treat inflammation and pain. For treating joint diseases, oral NSAIDs have been widely used to decrease pain and improve physical function. However, the vast majority of NSAIDs cause severe side effects due to the inhibition of COX-1 and COX-2 targets. Indeed, each such side effect-causing drug carries a black-box warning on its product label, as required by the FDA. These labels also state that patients should use the lowest effective dosage for the shortest duration. In addition to the cardiovascular and gastrointestinal adverse effects listed in the black-box warnings, the labels also warn of other systemic toxicities such as hepatotoxicity and renal toxicity.

There are two types of NSAIDs. Non-selective COX-1 /2 inhibitors, such as ibuprofen and naproxen, inhibit joint cartilage production due to the COX-1 mechanism. Although these NSAIDs are used to treat arthritic pain, they may actually promote the progression of the underlying cartilage degeneration in osteoarthritis. Selective COX-2 inhibitors, such as celecoxib (sold as Celebrex ^® (Pfizer)), do not inhibit cartilage production and cause fewer gastrointestinal adverse effects than non- selective COX-1 /2 inhibitors, but their cardiovascular side effects are still significant.

Celecoxib Formulations

G. Gaudriault, et al. (International Publication No. WO/2017/085561, “A method for morselizing and/or targeting pharmaceutically active principles to synovial tissue”) describes a hydrogel formulation of celecoxib that maintained a therapeutic level in the sheep knee for 14 days. The hydrogel is based on poly D,L-lactide and polyethylene glycol (PEG).

M. Homar, et al. (“Influence of polymers on the bioavailability of microencapsulated celecoxib”, J. of Microencapsulation, Nov. 2007; 24(7): 621-633) describes a list of formulations using PLGA, poly-caprolactone, ethylcellulose, and others. Celecoxib release was observed for seven days. The PLGA microspheres had mean diameters between 10-20 pm. Particles with diameters under 10 pm are subject to phagocytosis and removal by macrophages. Furthermore, the phagocytosed particles cause inflammation (T. Green, et al. (“Polyethylene particles of a ‘critical size’ are necessary for the induction of cytokines by macrophages in vitro”, Biomaterials, Dec. 1998; 19(24): 2297-2302; H. Chikaura, et al. (“Effect of particle size on biological response by human monocyte-derived macrophages”, Biosurface and Biotribology, March 2016. 2(1):18-25); J. Matthews, et al. (“Comparison of the response of primary human peripheral blood mononuclear phagocytes from different donors to challenge with model polyethylene particles of known size and dose”, Biomaterials. Oct. 2000. 21(20):2033-2044)). With mean diameters of from 10 pm to 20 pm, a significant portion of the described PLGA microspheres would be subject to macrophage phagocytosis and removal from the injected site, leading to inflammation, and thus rendering the formulations unsuitable as pharmaceutical depots. Furthermore, some formulations stopped releasing drug after an initial phase of burst release.

H.Y. Yang, et al. (“Applicability of a newly developed bioassay for determining bioactivity of anti-inflammatory compounds in release studies - celecoxib and triamcinolone acetonide released from novel PLGA-based microspheres”, Pharm. Res. (2015) 32:680-690) describes PLGA-polythioester formulations that achieve continuous release of celecoxib for over 90 days. However, polythioester has never been used in any FDA-approved pharmaceuticals.

H. Thakkar, et al. (“Enhanced retention of celecoxib-loaded solid lipid nanoparticles after intra-articular administration”, Drugs R D 2007; 8(5):275-285) describes solid lipid nanoparticle formulations that released celecoxib for up to 180 hours.

A. Petit, et al. (“Release behavior and intra-articular biocompatibility of celecoxib- loaded acetyl-capped PCLA-PEG-PCLA thermogels”, Biomaterials 35 (2014) 7919- 7928); and A. Petit, et al. (“Sustained intra-articular release of celecoxib from in situ forming gels made of acetyl-capped PCLA-PEG-PCLA triblock copolymers in horses”, Biomaterials 35 (2015): 426-436), describe hydrogel formulations that released therapeutic concentrations of celecoxib (100 ng/ml) for two weeks after subcutaneous injection in rats. Also described are hydrogel formulations that released therapeutic concentrations of celecoxib for four weeks after intra-articular injection in horses. However, PCLA (poly-caprolactone-co-lactide) has never been used in any FDA-approved pharmaceuticals.

M. Janssen, et al. (“Celecoxib-loaded PEA microspheres as an auto regulatory drug- delivery system after intra-articular injection”, J. Control Release, 2016 Dec. 28:244 (Pt. A): 30-40) describes polyester amide (PEA) formulations that released celecoxib for 80 days and beyond. However, polyester amide has never been used in any FDA-approved pharmaceuticals.

H. Thakkar, et al. (“Celecoxib incorporated chitosan microspheres: in vitro and in vivo evaluation”, J. of Drug Targeting, October- December 2004, Vol. 12 (9-10), pp. 549-557) describes chitosan formulations that released celecoxib for 100 hours. Chitosan has never been used in any FDA-approved pharmaceuticals. PLGA Microspheres

Biodegradable microspheres made of polylactic co-glycolic acid copolymer (PLGA) are known. PLGA is made of polylactic acid (PLA), polyglycolic acid (PGA), and typically both. PLGA is an FDA-approved biodegradable polymer. It has been extensively investigated in many medical and pharmaceutical fields due to its good biodegradability and biocompatibility. PLGA-containing microspheres have shown sustained release characteristics due to degradation and diffusion mechanisms. The drug release profile of a PLGA microsphere preparation is dependent on certain factors, such as the specific properties of the drug, the ratio of PLA to PGA, the type of end cap of the polymer (i.e. , ester or acid), the molecular weight/inherent viscosity of the polymer, the loading ratio of drug to the polymer, and the size of the microspheres.

In 2017, Zilretta ^® , a PLGA-triamcinolone-acetonide microsphere formulation for knee injection, was approved by the FDA. Zilretta ^® exhibits pain reduction in knee osteoarthritis patients for 12 weeks. Meanwhile, U.S. Patent No. 9,555,048, assigned to Flexion and relating to Zilretta ^® , describes tests of various combinations of PLGA, drug, and other co-polymers, and their effects on the release of triamcinolone acetonide. The longest drug release duration of any formulation described in this patent is 40-50 days.

An Unmet Need

There is an unmet need for a superior way to treat joint disorders using celecoxib, while minimizing the drug’s side effects observed with systemic delivery.

Summary of the Invention

This invention provides a biodegradable microsphere, wherein the microsphere (i) has a diameter of from 1 pm to 500 pm; (ii) comprises a polylactic-co-glycolic acid copolymer (PLGA) matrix; (iii) carries pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, releases celecoxib for at least one month.

This invention also provides a plurality of biodegradable microspheres, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a polylactic-co- glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month.

This invention further provides an injectable formulation comprising (a) a pharmaceutically acceptable carrier and (b) a plurality of biodegradable microspheres wherein the microspheres (i) have a d9o value from 1 pm to 500 pm;

(ii) comprise a polylactic-co-glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month.

This invention still further provides a method for treating a joint-related disorder in a subject comprising introducing biodegradable microspheres into suitable tissue in or around one or more of the subject’s joints, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a polylactic-co-glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month.

Finally, this invention provides a kit comprising, in separate compartments, (a) a diluent, and (b) plurality of biodegradable microspheres, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a polylactic-co-glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month. Brief Description of the Fiqures

Figure 1

This figure shows the release of diclofenac acid in a range of PLGA and PDLA microspheres.

Figure 2

This figure shows the release of diclofenac acid in PLGA50:50, 0.2 dl/g, acid- terminated microspheres.

Figure 3

This figure shows the diclofenac acid release from a range of PLGA and PDLA formulations.

Figure 4

This figure shows that both formulations involving PLGA85:15, 0.6 dl/g, ester- terminated, resulted in complete diclofenac acid release within 10 days.

Figure 5

This figure shows complete diclofenac acid release in 10 days for a range of PLGA formulations.

Figure 6

This figure shows the microsphere formation of lornoxicam-PLGA microspheres (PLGA50:50, 0.2 dl/g, acid-terminated).

Figure 7

This figure shows cumulative celecoxib release for a range of PLGA formulations. Figure 8

This figure shows that the celecoxib-loading amount from 1 mg to 2.5 mg, mixed with 2.5 mg PLGA, did not change drug release significantly. Figure 9

This figure shows that increasing celecoxib loading to 3-3.5 mg, mixed with 2.5 mg PLGA50:50, 0.5 dl/g, ester-terminated, resulted in continuous celecoxib release, which is different from the little drug release seen in day 9 to day 15 in formulations with lower (1 mg) celecoxib loading.

Figure 10

In this figure, 2.5 mg of 50% of PLGA50:50, 0.5 dl/g, acid-terminated, and 50% of PDLA, 0.4 dl/g, ester-terminated, with 1 mg celecoxib loading, resulted in continuous celecoxib release over 45 days.

Figure 11

In this figure, 2.5 mg of PLGA with all ratios of PLGA50:50, 0.5 dl/g, acid-terminated, mixed with PLGA50:50, 0.4 dl/g, ester-terminated, and PLGA50:50, 0.5 dl/g, ester- terminated, resulted in continuous drug release at celecoxib loading of 1 mg.

Figure 12

In this figure, 2.5 mg of PLGA with 2.5 mg celecoxib loading was tested. 25% PDLA, 0.4 dl/g, ester-terminated, with 75% PLGA50:50, 0.5 dl/g, acid-terminated, resulted in continuous celecoxib release over 25 days.

Figure 13

In this figure, 2.5 mg of PLGA with 3.5 mg celecoxib loading was tested. 50-75% of PLGA50:50, 0.5 dl/g, acid-terminated, mixed with PLGA75:25, 0.6 dl/g, ester- terminated, resulted in continuous celecoxib release over 25 days.

Figure 14

This figure shows that PEG1450 blending with PLGA slightly increased celecoxib release. With 0 mg of PEG1450, celecoxib release was minimal from day 11 to day 15, whereas all three formulations with PEG1450 blending showed continuous celecoxib release until day 30. Figure 15

This figure shows that celecoxib loading of 1-2.5 mg with 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 100 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous drug release over 30 days.

Figure 16

This figure shows that celecoxib loading of 1-3.5 mg with 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane (DCM) and stirred at 1 ,000 rpm, resulted in continuous drug release over 30 days.

Figure 17

This figure shows that celecoxib loading of 3.5 mg and 5 mg with 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous drug release over 60 days.

Figure 18

This figure shows that celecoxib loading of 5 mg and 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm, resulted in continuous drug release over 60 days.

Figure 19

This figure shows that celecoxib loading of 3.5 mg and 5 mg, and 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1,200 rpm, resulted in continuous drug release over 60 days.

Figure 20

This figure shows that celecoxib loading at 4 mg, 5 mg, and 6 mg, mixed with PLGA50:50, 0.5 dl/g, ester-terminated, and PLGA50:50, 0.6 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 30 days. Figure 21

This figure shows that celecoxib loading at 6 mg and 7.5 mg, mixed with PLGA50:50, 0.5 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 20 days.

Figure 22

This figure shows that celecoxib loading at 4 mg, 5 mg, and 6 mg, mixed with PLGA50:50, 0.7 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 70 days. Unexpectedly, increased loading ratio did not result in higher release rate.

Figure 23

This figure shows that celecoxib loading at 4 mg, 5 mg, and 6 mg, mixed with 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 111 days. Higher celecoxib loading was associated with faster drug release.

Figure 24

This figure shows that celecoxib loading at 5 mg, 6 mg, and 7.5 mg, mixed with 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 92 days. Higher celecoxib loading was associated with faster drug release.

Figure 25

This figure shows that celecoxib loading at 5 mg and 6 mg, mixed with 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, dissolved in 300 mI or 400 mI and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 90 days.

Figure 26

This figure shows celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 90 days. Figure 27

This figure shows that celecoxib loading of 5 mg, 6 mg, and 7.5 mg, mixed with PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI or 400 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 80 days.

Figure 28

This figure shows that celecoxib loading of 5 mg, 6 mg, and 7.5 mg, mixed with PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI or 400 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 80 days.

Figure 29

This figure shows that celecoxib loading of 5 mg and 7.5 mg, mixed with PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 100 days and 50 days, respectively.

Figure 30

This figure shows that celecoxib loading of 5 mg and 6 mg, mixed with PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 70 days.

Figure 31

This figure shows that celecoxib loading of 5 mg and 6 mg, mixed with PLGA75:25, 1.2 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 70 days.

Figure 32

This figure shows that 7.5 mg of celecoxib, mixed with 2.5 mg PLGA85:15, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 40 days. Figure 33

This figure shows that 6 mg of celecoxib, mixed with 2.5 mg PLGA85:15, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over approximately 30 days.

Figure 34

This figure shows that celecoxib loading of 7.5 mg, mixed with 2.5 mg PLGA85:15, 1.5 dl/g, ester-terminated, dissolved in 300 mI and 400 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 35 days.

Figure 35

This figure shows that celecoxib loading of 7.5 mg, mixed with 2.5 mg PLGA85:15, 1.5 dl/g, ester-terminated, dissolved in 500 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 50 days.

Figure 36

This figure shows that celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PDLA, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 40 days.

Figure 37

This figure shows that celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PDLA, 0.6 dl/g, ester-terminated, dissolved in 300 mI and 400 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 35 days.

Figure 38

This figure shows that celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PDLA, 0.4 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm, resulted in little celecoxib release from day 20 to day 30 and from day 50 to day 60, which is unsuitable for continuous celecoxib release over multiple months. Figure 39

This figure shows that mixing PLGA50:50, 0.7 dl/g, ester-terminated, with PLGA75:25, 0.6 dl/g, ester-terminated, slowed down celecoxib release and prolonged duration of continuous drug release.

Figure 40

This figure shows that mixing PLGA50:50, 0.7 dl/g, ester-terminated, with PLGA75:25, 0.6 dl/g, ester-terminated, slowed down celecoxib release. All formulations resulted in continuous drug release for at least 60 days.

Figure 41

This figure shows that mixing PLGA75:25, 0.4 dl/g, acid-terminated, with PLGA75:25, 0.6 dl/g, ester-terminated, slowed down celecoxib release. All formulations resulted in continuous drug release for at least 60 days.

Figures 42A-42D

The first two figures show that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib (Figure 42A) or 6 mg celecoxib (Figure 42B), dissolved in 200 pi dichloromethane and stirred at 1 ,200 rpm in PVA (with optical density 230nm (OD230) at 0.160), resulted in continuous drug release over 180 days. Figures 42C (top) and 42D (bottom) show microscopic images of microspheres formed by 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated and 5 mg (Figure 42C) or 6 mg (Figure 42D) celecoxib, dissolved in 200 mI dichloromethane and stirred at 1,200 rpm in 50 ml PVA (OD230 = 0.160) for 2 minutes, followed by stirring at 500 rpm for 2 hours, harvesting, washing with water, and lyophilization.

Figure 43

This figure shows that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 6 mg celecoxib, dissolved in 300 mI dichloromethane and stirred at 1,000 rpm in PVA (OD230 = 0.160), resulted in continuous drug release over 175 days. Figure 44

This figure shows that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 7.5 mg celecoxib, dissolved in 200 or 300 mI dichloromethane and stirred at 1 ,000 or 1 ,200 rpm in PVA (OD230 = 0.160), resulted in continuous drug release over 180 days.

Figure 45

This figure shows that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm or 1 ,200 rpm in PVA (OD230 = 0.160), resulted in continuous drug release over 180 days.

Figures 46A and 46B These figures show 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib (Figure 46A) or 6 mg celecoxib (Figure 46B), dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm in PVA (OD230 = 0.05), resulted in continuous drug release over 180 days.

Figures 47A and 47B Figure 47A shows 2.5mg PLGA75:25, 0.6 dl/g or 0.9 dl/g, ester-terminated, and 5 mg celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm in PVA (OD230 = 0.160), resulted in continuous drug release over at least 100 days. Figure 47B shows a microscopic image of microspheres formed by 2.5 mg PLGA75:25, 0.9 dl/g, ester-terminated and 5 mg celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm for 2 minutes in 50 ml PVA (OD230 = 0.160), followed by stirring at 500 rpm for 2 hours, harvesting, washing with water, and lyophilization.

Figure 48

This figure shows plasma celecoxib concentration after injection of 5 mg of the present formulation into bilateral knees of Lewis rats.

Figures 49A-49C

These figures show results from using the present formulation in a rat monoiodoacetate (MIA) osteoarthritis pain model. Specifically, the data shown include results from a Von Frey filament test (Figure 49A), a thermal hyperalgesia test (Figure 49B), and a test for differential weight bearing on hind limbs (Figure 49C).

Figures 50A-50D These figures show celecoxib concentration in joint tissues (i.e. , synovial fluid (Figure 50A), synovial membrane (Figure 50B), femur cartilage (Figure 50C), and tibia cartilage (Figure 50D)) after injection of 100 mg PLGA-encapsulated celecoxib formulation into bilateral knees of Beagle dogs. Figure 51

This figure shows celecoxib plasma concentration after injection of 100mg PLGA- encapsulated celecoxib formulation into bilateral knees of Beagle dogs.

Figure 52 This figure compares celecoxib concentrations in plasma and joint tissues after oral daily 5 mg/kg dosing of celecoxib in Beagle dogs.

Figures 53A and 53B

Figure 53A shows the concentration ratios of celecoxib within joint tissues (i.e., (i) synovial membrane to synovial fluid, and (ii) femur cartilage to synovial fluid) after oral daily 5 mg/kg dosing of celecoxib in Beagle dogs. Figure 53B shows the concentration ratios of celecoxib within joint tissues (i.e., (i) synovial membrane to synovial fluid, (ii) femur cartilage to synovial fluid, and (iii) tibia cartilage to synovial fluid) after knee injection of 100 mg PLGA-celecoxib oral in Beagle dogs.

Detailed Description of the Invention

This invention provides celecoxib-containing biodegradable microspheres and methods for using them to treat joint-related disorders.

Definitions

In this application, certain terms are used which shall have the meanings set forth as follows.

As used herein, a “biodegradable microsphere” comprises a polylactic-co-glycolic acid copolymer (PLGA) matrix, which matrix can include solely polylactic acid (PLA), solely polyglycolic acid (PGA), or a polymeric combination of lactic acid and glycolic acid units. For certain lactic acid to glycolic acid ratios (e.g., 50:50 to 100:0), the higher a microsphere’s lactic acid content, the slower it degrades and, thus, the more stable it is. Conversely, for such ratios, the higher a microsphere’s glycolic acid content, the faster it degrades and the less stable it is. In one embodiment, the biodegradable microsphere contains a combination of lactic acid and glycolic acid units wherein the molar ratio of lactic acid to glycolic acid units (i.e. , the “lactic acid to glycolic acid ratio”, or“L:G ratio”) is 0:100, 5:95, 10:90, 15:85, 20:80, 25:75, 30:70, 35:65, 40:60, 45:55, 50:50, 55:45, 60:40, 65:35, 70:30, 75:25, 80:20, 85:15, 90:10, 95:5, or 100:0. In another embodiment, the biodegradable microsphere contains a combination of lactic acid and glycolic acid units wherein the molar ratio of lactic acid to glycolic acid is from 5:95 to 20:80, from 20:80 to 40:60, from 40:60 to 50:50, from 50:50 to 60:40, from 60:40 to 80:20, from 80:20 to 100:0, from 50:50 to 100:0, from 60:40 to 90:10, from 70:30 to 80:20, from 50:50 to 80:20, from 50:50 to 90:10, from 60:40 to 70:30, from 80:20 to 90:10, or from 90:10 to 100:00. The population of biodegradable microspheres used in this invention can be homogeneous or heterogeneous with respect to the microspheres’ molar ratio of lactic acid to glycolic acid. In one embodiment, the population of biodegradable microspheres is homogeneous with respect to the microspheres’ molar ratio of lactic acid to glycolic acid (e.g., the population includes only microspheres wherein the molar ratio of lactic acid to glycolic acid is 75:25). In another embodiment, the population of biodegradable microspheres is heterogeneous (e.g., the population includes both (i) microspheres wherein the molar ratio of lactic acid to glycolic acid is 70:30, and (ii) microspheres wherein the molar ratio of lactic acid to glycolic acid is 80:20). In a preferred embodiment, the instant microspheres contain PLGA having an inherent viscosity of 0.1 to 2.4 dl/g (e.g., 0.16 to 1.7 dl/g), and a molecular weight from 1,000 to 600,000 (e.g., from 7,000 to 240,000).

The subject biodegradable microsphere (i) has a diameter from 1 pm to 500 pm, (ii) can non-covalently carry a therapeutic agent (e.g., celecoxib), and (iii) depending on its polymeric composition, degrades over a period lasting, for example, from one month to over six months when placed in suitable joint-related tissue. Microsphere diameters (and d9o values) include, for example, the following: (i) from 1 pm to 20 pm, from 20 pm to 40 pm, from 40 pm to 60 pm, from 60 pm to 80 pm, from 80 pm to 100 pm, from 100 pm to 120 pm, from 120 pm to 140 pm, from 140 pm to 160 pm, from 160 pm to 180 pm, from 180 pm to 200 pm, from 200 pm to 250 pm, from 250 pm to 300 pm, from 300 pm to 350 pm, from 350 pm to 400 pm, from 400 pm to 450 pm, or from 450 pm to 500 pm; (ii) 20 pm, 40 pm, 60 pm, 80 pm, 100 pm, 120 pm, 140 pm, 160 pm, 180 pm, 200 pm, 250 pm, 300 pm, 350 pm, 400 pm, 450 pm, or 500 pm; and (iii) from 20 pm to 100 pm, from 20 pm to 150 pm, from 50 pm to 100 pm, or from 50 pm to 150 pm. The subject biodegradable microspheres can further comprise polyethylene glycol (PEG). Biodegradable PLGA microspheres (including homogeneous and heterogeneous populations thereof having defined molar ratios of lactic acid to glycolic acid units) are commercially available from, among other sources, Millipore-Sigma in the form of Degradex ^® products (Burlington, MA) and Evonik Industries in the form of Resomer ^® products (Essen, Germany).

As used herein, the term “carry”, with respect to pharmaceutical celecoxib and a biodegradable microsphere, means that the pharmaceutical celecoxib is non- covalently bound to, or otherwise contained in or on, the biodegradable microsphere in a manner permitting release from the microsphere during its biodegradation.

As used herein, the term “celecoxib” shall mean 4-[5-(4-methylphenyl)-3- (trifluoromethyl)-l H-pyrazol-1 -yl] benzenesulfonamide, with CAS number 169590 42 5 Celecoxib is a nonsteroidal anti-inflammatory drug. It is commercially available, and is sold by Pfizer under the trade name Celebrex ^® . As used herein, the term “diluent” includes, without limitation, carboxymethylcellulose sodium, polysorbate 80, mannitol (which can optionally be incorporated on and/or into the microspheres to improve suspendability), and water.

As used herein, the term “d9o value”, with respect to a population of biodegradable microspheres having a specified size range, means that 90% of the biodegradable microspheres have a diameter in the specified range.

As used herein, “introducing”, with respect to biodegradable microspheres, means delivering to a specified part of the body, such as joint fluid. Methods of introducing biodegradable microspheres to joint fluid are known and include, for example, intra- articular injection. See, e.g., the Zilretta ^® label. Methods for injecting biodegradable microspheres into discs are known and can be performed based on known animal studies. For example, polyester amide microspheres were injected to the intervertebral discs in a canine model of disc degeneration, and were shown to have good cytocompatability and biocompatibility. In a rat discitis model, intra-discal vancomycin-loaded PLGA microspheres were shown to control and reduce infective discitis, with superior efficacy to intravenous vancomycin. See, e.g., Williems, et al. , and Wang, et al.

As used herein, a “joint-related disorder” includes, without limitation, osteoarthritis, synovitis, hemophilic arthropathy, rheumatoid arthritis, psoriatic arthritis, juvenile rheumatoid arthritis, gouty arthritis, septic or infectious arthritis, reactive arthritis, reflex sympathetic dystrophy, scleroderma, ankylosing spondylitis, algodystrophy, achondroplasia, Paget’s disease, Tietze syndrome or costal chondritis, fibromyalgia, neurogenic or neuropathic arthritis, arthropathy, sarcoidosis, amylosis, cartilage injury, hydarthrosis, periodical disease, rheumatoid spondylitis, osteochondritis dissecans, hypertropic arthritis, Yersinia arthritis, pyrophosphate arthritis, an endemic form of arthritis, fibromyalgia, systemic lupus erythematosus, scleroderma, ankylosing spondylitis, degenerative disc disease, chronic lower back pain, chronic neck pain, hip dysplasia, osteochondrosis, elbow dysplasia, joint injury caused by trauma, acute and subacute bursitis, acute and subacute nonspecific tenosynovitis and epicondylitis, acute rheumatic carditis and ankylosing spondylitis, tenosynovitis, epicondylitis, synovitis, sciatica, and other forms of radicular pain. In one embodiment, a joint-related disorder includes discomfort, inflammation, or other indication associated with recovery from a surgical joint procedure such as a total or partial knee replacement, a total or partial hip replacement, a total or partial ankle replacement, an arthroscopic or open joint surgery, a microfracture, an autologous chondrocyte implantation, mosaicplasty, debridement and lavage, a ligament repair, a tendon repair, a rotator cuff repair, meniscus surgery, or synovectomy.

As used herein, the term “pharmaceutical celecoxib” includes, without limitation, celecoxib and pharmaceutical salts (e.g., celecoxib sodium) and esters thereof.

“Pharmaceutically acceptable carriers” are well known and include, without limitation, the diluents described herein.

As used herein, a biodegradable microsphere “releases” celecoxib when some or all of the celecoxib contained by the microsphere is freed into the microsphere’s surrounding milieu. Preferably, the release is continuous. For example, in a plurality of celecoxib-carrying biodegradable microspheres having an average release per day of X mg, the celecoxib released per day is, e.g., from 0.1X mg to 10X mg, from 0.2X mg to 5X mg, or from 0.5X mg to 2X mg. In another example, in a plurality of celecoxib-carrying biodegradable microspheres having an average release per week of X mg, the celecoxib released per week is, e.g., from 0.2X mg to 10X mg, or from 0.5X mg to 2X mg. Celecoxib release into the joint-related tissue can precede, and is distinct from, its efficacy in that tissue. For example, biodegradable microspheres that release a therapeutically effective amount of pharmaceutical celecoxib into a joint’s synovial fluid for two months might yield a therapeutic effect for three months.

As used herein, the term “subject” includes, without limitation, a mammal such as a human, a non-human primate, a dog, a cat, a horse, a sheep, a goat, a cow, a rabbit, a pig, a rat and a mouse. Preferably, the subject is human. In another preferred embodiment, the subject is a cat, a dog, or a horse.

As used herein, the phrase “suitable joint-related tissue” includes any portion of a joint, the joint’s surrounding tissue, an intervertebral disc, or the intervertebral disc’s surrounding tissue that is capable of holding the instant celecoxib-containing biodegradable microspheres such that the celecoxib released therefrom can act on the joint. Suitable joint-related tissue includes, without limitation, (i) articular and periarticular spaces; (ii) the bursa, synovial cavity, joint capsule with synovial lining, and the fluids contained therein; and (iii) connective and contractile tissue (e.g., articular cartilage, ligaments, tendons and muscles).

As used herein, a “suitable matrix” for a biodegradable microsphere includes, without limitation, a polylactic-co-glycolic acid copolymer (PLGA) matrix; a hydrogel matrix (e.g., one based on poly D,L-lactide and polyethylene glycol (PEG)); a poly- caprolactone matrix; an ethylcellulose matrix; a PLGA-polythioester matrix; a solid lipid nanoparticle matrix; an acetyl-capped PCLA-PEG-PCLA thermogel matrix; a polyester amide (PEA) matrix; and a chitosan matrix. In the preferred embodiment, the suitable matrix is a polylactic-co-glycolic acid copolymer (PLGA) matrix.

As used herein, the term “therapeutically effective amount”, with respect to pharmaceutical celecoxib carried in biodegradable microspheres, refers to the amount of pharmaceutical celecoxib collectively carried by the total dose of biodegradable microspheres introduced in or around one of a subject’s joints. In one embodiment, the effective amount is 1 pg, 5 pg, 10 pg, 15 pg, 20 pg, 25 pg, 30 pg,

40 pg, 50 pg, 60 pg, 70 pg, 80 pg, 90 pg, 100 pg, 150 pg, 200 pg, 250 pg, 300 pg, 350 pg, 400 pg, 450 pg, 500 pg, 550 pg, 600 pg, 650 pg, 700 pg, 750 pg, 800 pg,

850 pg, 900 pg, 950 pg, 1 mg, 5 mg, 10 mg, 15 mg, 20 mg, 25 mg, 30 mg, 40 mg, 50 mg, 60 mg, 70 mg, 80 mg, 90 mg, 100 mg, 150 mg, 200 mg, 250 mg, 300 mg, 350 mg, 400 mg, 450 mg, 500 mg, 550 mg, 600 mg, 650 mg, 700 mg, 750 mg, 800 mg, 850 mg, 900 mg, 950 mg, 1 ,000 mg, 1 ,050 mg, 1 ,100 mg, 1 ,150 mg, 1 ,200 mg,

1 ,250 mg, 1 ,300 mg, 1 ,350 mg, 1 ,400 mg, 1 ,450 mg, 1 ,500 mg, 1 ,550 mg, 1 ,600 mg, 1 ,650 mg, 1 ,700 mg, 1 ,750 mg, 1 ,800 mg, 1 ,850 mg, 1 ,900 mg, 1 ,950 mg, or 2,000 mg. In another embodiment, the effective amount is from 1 pg to 10 pg, from 10 pg to 50 pg, from 50 pg to 100 pg, from 100 pg to 150 pg, from 150 pg to 200 pg, from 200 pg to 250 pg, from 250 pg to 300 pg, from 300 pg to 350 pg, from 350 pg to 400 pg, from 400 pg to 450 pg, from 450 pg to 500 pg, from 500 pg to 550 pg, from 550 pg to 600 pg, from 600 pg to 650 pg, from 650 pg to 700 pg, from 700 pg to 750 pg, from 750 pg to 800 pg, from 800 pg to 850 pg, from 850 pg to 900 pg, from 900 pg to 950 pg, from 950 pg to 1 mg, 1 mg to 10 mg, from 10 mg to 50 mg, from 50 mg to 100 mg, from 100 mg to 150 mg, from 150 mg to 200 mg, from 200 mg to 250 mg, from 250 mg to 300 mg, from 300 mg to 350 mg, from 350 mg to 400 mg, from 400 mg to 450 mg, from 450 mg to 500 mg, from 500 mg to 550 mg, from 550 mg to 600 mg, from 600 mg to 650 mg, from 650 mg to 700 mg, from 700 mg to 750 mg, from 750 mg to 800 mg, from 800 mg to 850 mg, from 850 mg to 900 mg, from 900 mg to 950 mg, from 950 mg to 1 ,000 mg, from 1 ,050 mg to 1 ,100 mg, from 1 ,100 mg to 1 , 150 mg, from 1 , 150 mg to 1 ,200 mg, from 1 ,200 mg to 1 ,250 mg, from 1 ,250 mg to

1 ,300 mg, from 1 ,300 mg to 1 ,350 mg, from 1 ,350 mg to 1 ,400 mg, from 1 ,400 mg to

1 ,450 mg, from 1 ,450 mg to 1 ,500 mg, from 1 ,500 mg to 1 ,550 mg, from 1 ,550 mg to

1 ,600 mg, from 1 ,600 mg to 1 ,650 mg, from 1 ,650 mg to 1 ,700 mg, from 1 ,700 mg to

1 ,750 mg, from 1 ,750 mg to 1 ,800 mg, from 1 ,800 mg to 1 ,850 mg, from 1 ,850 mg to

1 ,900 mg, from 1 ,900 mg to 1 ,950 mg, or from 1 ,950 mg to 2,000 mg. In a further embodiment, the effective amount is from 1 pg to 250 pg, from 250 pg to 500 pg, from 500 pg to 750 pg, from 750 pg to 1 mg, 1 mg to 250 mg, from 250 mg to 500 mg, from 500 mg to 750 mg, from 750 mg to 1 ,000 mg, from 1 ,000 mg to 1 ,250 mg, from 1 ,250 mg to 1 ,500 mg, from 1 ,500 mg to 1 ,750 mg, or from 1 ,750 mg to 2,000 mg. In yet a further embodiment, the effective amount is from 1 pg to 500 pg, from 500 pg to 1 mg, 1 mg to 500 mg, from 10 mg to 500 mg, from 500 mg to 1 ,000 mg, from 1 ,000 mg to 1 ,500 mg, or from 1 ,500 mg to 2,000 mg. In a further embodiment, the therapeutically effective amount of celecoxib is an amount sufficient to achieve a celecoxib concentration in the synovial fluid of, for example, 0.1 ng/ml, 0.2 ng/ml, 0.3 ng/ml, 0.4 ng/ml, 0.5 ng/ml, 0.6 ng/ml, 0.7 ng/ml, 0.8 ng/ml, 0.9 ng/ml, 1 ng/ml, 2 ng/ml, 3 ng/ml, 4 ng/ml, 5 ng/ml, 6 ng/ml, 7 ng/ml, 8 ng/ml, 9 ng/ml, 10 ng/ml, 20 ng/ml, 40 ng/ml, 60 ng/ml, 80 ng/ml, 100 ng/ml, 120 ng/ml, 140 ng/ml, 160 ng/ml,

180 ng/ml, 200 ng/ml, 250 ng/ml, 300 ng/ml, 350 ng/ml, 400 ng/ml, 450 ng/ml, 500 ng/ml, 550 ng/ml, 600 ng/ml, 650 ng/ml, 700 ng/ml, 750 ng/ml, 800 ng/ml, 850 ng/ml, 900 ng/ml, 950 ng/ml, 1 ,000 ng/ml, 2,000 ng/ml, 3,000 ng/ml, 4,000 ng/ml, 5,000 ng/ml, 6,000 ng/ml, 7,000 ng/ml, 8,000 ng/ml, 9,000 ng/ml, 10,000 ng/ml, 20,000 ng/ml, from 0.1 ng/ml to 0.5 ng/ml, from 0.5 ng/ml to 1 ng/ml, from 1 ng/ml to 5 ng/ml, from 5 ng/ml to 10 ng/ml, from 10 ng/ml to 15 ng/ml, from 15 ng/ml to 20 ng/ml, from 20 ng/ml to 25 ng/ml, from 25 ng/ml to 50 ng/ml, from 50 ng/ml to 75 ng/ml, from 75 ng/ml to 100 ng/ml, from 100 ng/ml to 125 ng/ml, from 125 ng/ml to 150 ng/ml, from 150 ng/ml to 200 ng/ml, from 200 ng/ml to 300 ng/ml, from 300 ng/ml to 400 ng/ml, from 400 ng/ml to 500 ng/ml, from 500 ng/ml to 600 ng/ml, from 600 ng/ml to 700 ng/ml, from 700 ng/ml to 800 ng/ml, from 800 ng/ml to 900 ng/ml, from 900 ng/ml to 1 ,000 ng/ml, from 1 ,000 ng/ml to 2,000 ng/ml, from 2,000 ng/ml to 3,000 ng/ml, from 3,000 ng/ml to 4,000 ng/ml, from 4,000 ng/ml to 5,000 ng/ml, from 5,000 ng/ml to 6,000 ng/ml, from 6,000 ng/ml to 7,000 ng/ml, from 7,000 ng/ml to 8,000 ng/ml, from 8,000 ng/ml to 9,000 ng/ml, from 9,000 ng/ml to 10,000 ng/ml, or from 10,000 ng/ml to 20,000 ng/ml.

As used herein, “treating” a subject afflicted with a disorder shall include, without limitation, (i) slowing, stopping or reversing the disorder's progression, (ii) slowing, stopping or reversing the progression of the disorder’s symptoms (e.g., pain), (iii) reducing the likelihood of the disorder’s recurrence, and/or (iv) reducing the likelihood that the disorder’s symptoms will recur. In the preferred embodiment, treating a subject afflicted with a disorder means (i) reversing the disorder's progression, ideally to the point of eliminating the disorder, and/or (ii) reversing the progression of the disorder’s symptoms, ideally to the point of eliminating the symptoms.

Embodiments of the Invention

This invention solves an unmet need in the art by providing an unexpectedly superior way to treat joint-related disorders using celecoxib. The invention does this via celecoxib-carrying microspheres that release celecoxib overtime.

Specifically, this invention provides a biodegradable microsphere, wherein the microsphere (i) has a diameter of from 1 pm to 500 pm; (ii) comprises a polylactic- co-glycolic acid copolymer (PLGA) matrix; (iii) carries pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, releases celecoxib for at least one month.

In one embodiment of the instant biodegradable microsphere, it has a lactic acid to glycolic acid molar ratio of from 100:0 to 50:50. Preferably, the instant microsphere (i) has a diameter of from 20 pm to 200 pm; and (ii) has a lactic acid to glycolic acid molar ratio of 75:25. In another embodiment of the instant biodegradable microsphere, it further comprises polyethylene glycol (PEG). The PEG can be any type suitable for use in forming biodegradable microspheres (e.g., PEG1450 (Polysciences, Inc.,

Warrington, PA)). Moreover, the ratio of PEG to PLGA can be any ratio suitable for use in forming biodegradable microspheres (e.g., 25:100, 50:100, 75:100 or 100:100).

In yet another embodiment of the instant biodegradable microsphere, when it is present in a suitable joint-related tissue, it releases celecoxib for longer than one month. Preferably, the microsphere, when present in a suitable joint-related tissue, releases celecoxib for at least two months, at least three months, at least four months, at least five months, at least six months, at least seven months, at least eight months, at least nine months, at least ten months, at least eleven months, or at least twelve months.

The following celecoxib-containing biodegradable microspheres are preferred embodiments of this invention. The microsphere population comprises PLGA75:25 (i.e. , wherein the L:G ratio is 75:25), 0.5-1.0 dl/g, ester-terminated, with celecoxib loading of 50%-75% (calculated as celecoxib/(celecoxib+PLGA)). For example, for each 2.5 mg of PLGA75:25, 0.6 dl/g, ester-terminated, 5 mg celecoxib is encapsulated to form 7.5 mg microspheres. In another example, for each 2.5 mg of PLGA75:25, 0.9 dl/g, ester-terminated, 5 mg celecoxib is encapsulated to form 7.5 mg microspheres. For each 2.5 mg PLGA and 5 mg celecoxib, the microspheres are prepared by dissolving the PLGA and celecoxib in 200 pi of dichloromethane, injected into the center of 50 ml of 0.5-1.5% poly-vinyl-alcohol (PVA, w/v, molecular weight 31-50K, 98-99% hydrolyzed), stirred with a magnetic stir bar (2 cm in length, 0.7 cm in diameter) at 1 ,200 rpm for 2 minutes in a 50 ml beaker and then at about 500 rpm for 2 hours. The PVA concentration is adjusted to have an absorption of 0.160 at 230 nm wavelength.

This invention also provides a plurality of biodegradable microspheres, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a polylactic-co- glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month.

In an embodiment of the instant plurality of biodegradable microspheres, the microspheres further comprise polyethylene glycol (PEG).

In another embodiment of the instant plurality of biodegradable microspheres, the microspheres, when present in a suitable joint-related tissue, release celecoxib for longer than one month. Preferably, the microspheres, when present in a suitable joint-related tissue, release celecoxib for at least two months, at least three months, at least four months, at least five months, at least six months, at least seven months, at least eight months, at least nine months, at least ten months, at least eleven months, or at least twelve months.

In a further embodiment of the instant plurality of biodegradable microspheres, the microspheres (i) have a d9o value from 20 pm to 200 pm; (ii) have a lactic acid to glycolic acid molar ratio of from 100:0 to 50:50; and (iii) carry from 1 pg to 500 mg of pharmaceutical celecoxib.

This invention further provides an injectable formulation comprising (a) a pharmaceutically acceptable carrier and (b) a plurality of biodegradable microspheres wherein the microspheres (i) have a d9o value from 1 pm to 500 pm;

(ii) comprise a polylactic-co-glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month.

In an embodiment of the instant formulation, the microspheres further comprise polyethylene glycol (PEG).

In another embodiment of the instant formulation, the microspheres, when present in a suitable joint-related tissue, release celecoxib for longer than one month. Preferably, the microspheres, when present in a suitable joint-related tissue, release celecoxib for at least two months, at least three months, at least four months, at least five months, at least six months, at least seven months, at least eight months, at least nine months, at least ten months, at least eleven months, or at least twelve months.

This invention still further provides a method for treating a joint-related disorder in a subject comprising introducing biodegradable microspheres into suitable tissue in or around one or more of the subject’s joints, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a polylactic-co-glycolic acid copolymer (PLGA) matrix; (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month.

In an embodiment of the instant method, the microspheres further comprise polyethylene glycol (PEG).

In a preferred embodiment of the instant method, the subject is human. In another preferred embodiment of the instant method, the subject is a cat, a dog, or a horse.

In another preferred embodiment of the instant method, the disorder is arthritis (e.g., osteoarthritis or rheumatoid arthritis).

In this invention, the biodegradable microspheres can be introduced into the suitable tissue in or around one or more of the subject’s joints using any known method appropriate for the tissue in question. For example, in a preferred embodiment of the instant method where the tissue is synovial fluid in the knee joint, the method comprises intra-articularly injecting the biodegradable microspheres into the synovial fluid of one or both of the subject’s knees. In another embodiment, the instant method is performed a plurality of times (e.g., two times, three times, four times, five times, or more). In that embodiment, each subsequent time the method is performed, it is performed after a suitable period has lapsed since the preceding time the method was performed. This suitable time can be, for example, one month, two months, three months, four months, five months, six months, one year, or longer. Microsphere-based drug products and methods of delivering them are known, at least generally (e.g., Zilretta ^® (triamcinolone acetonide extended-release injectable suspension for intra-articular use (Flexion)); and Sandostatin LAR ^® Depot (octreotide acetate for injectable suspension) (Novartis)). In a further preferred embodiment of the instant method, the microspheres (i) have a d9o value from 20 pm to 200 pm (e.g., from 20 pm to 150 pm); (ii) have a lactic acid to glycolic acid molar ratio of from 100:0 to 50:50 (e.g., 75:25); and (iii) carry from 1 pg to 500 mg of pharmaceutical celecoxib.

In yet another embodiment of the instant method, the microspheres, when present in a suitable joint-related tissue, release celecoxib for longer than one month. Preferably, the microspheres, when present in a suitable joint-related tissue, release celecoxib for at least two months, at least three months, at least four months, at least five months, at least six months, at least seven months, at least eight months, at least nine months, at least ten months, at least eleven months, or at least twelve months.

This invention further provides an article of manufacture (kit) comprising, in separate compartments, (a) one of, and ideally both of, (i) a diluent and (ii) a label instructing the user to introduce the biodegradable microspheres (described below) into suitable tissue in or around one or more of a subject’s joints, and (b) a plurality of biodegradable microspheres, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm (e.g., from 20 pm to 200 pm, or from 20 pm to 150 pm); (ii) comprise a polylactic-co-glycolic acid copolymer (PLGA) matrix (preferably having a lactic acid to glycolic acid molar ratio of from 100:0 to 50:50 (e.g., 75:25)); (iii) carry a therapeutically effective amount of pharmaceutical celecoxib (e.g., from 1 pg to 500 mg of pharmaceutical celecoxib); and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month (and optionally release celecoxib for at least two months, at least three months, at least four months, at least five months, at least six months, at least seven months, at least eight months, at least nine months, at least ten months, at least eleven months, or at least twelve months). In one embodiment of the instant kit, the microspheres further comprise polyethylene glycol (PEG). Where applicable, the embodiments described above for the instant methods are also envisioned for this article of manufacture.

In a preferred embodiment, the instant kit is supplied as a single-dose kit, and contains (i) a single dose vial of celecoxib-carrying biodegradable microspheres, and (ii) a single dose vial of diluent (e.g., sterile, clear liquid solution of 0.9% w/w sodium chloride, 0.5% - 1% w/w sodium carboxymethylcellulose, and 0.1% w/w polysorbate- 80).

This invention provides a biodegradable microsphere, wherein the microsphere (i) has a diameter of from 1 pm to 500 pm; (ii) comprises a suitable matrix (e.g., a polylactic-co-glycolic acid copolymer (PLGA) matrix); (iii) carries pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, releases celecoxib for at least one month, wherein the celecoxib so released is characterized (e.g., when measured at 14 days post-injection) by (i) a synovial membrane to synovial fluid celecoxib concentration ratio (i.e. , the ratio of synovial membrane celecoxib concentration to synovial fluid celecoxib concentration) of at least 20, and/or (ii) a joint cartilage to synovial fluid celecoxib concentration ratio (i.e., the ratio of joint cartilage (e.g., femur cartilage or tibia cartilage) celecoxib concentration to synovial fluid celecoxib concentration) of at least 10. The biodegradable microsphere components, dimensions, features, and celecoxib release duration described above are envisioned, mutatis mutandis, for this biodegradable microsphere. In a preferred embodiment, the synovial membrane to synovial fluid celecoxib concentration ratio is at least 100, at least 500, at least 1 ,000, at least 1 ,500, at least 2,000, at least 2,500, at least 3,000, at least 3,500, at least 4,000, or at least 5,000. In another preferred embodiment, the joint cartilage (i.e., femur cartilage) to synovial fluid celecoxib concentration ratio is at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 60, at least 70, at least 80, at least 90, or at least 100. In a further preferred embodiment, the joint cartilage (i.e., tibia cartilage) to synovial fluid celecoxib concentration ratio is at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 60, at least 70, at least 80, at least 90, or at least 100. Methods for measuring celecoxib concentrations in synovial membrane, synovial fluid, and joint cartilage are known, and are exemplified by a number of references. For example, methods for measuring celecoxib concentration in plasma and synovial fluid are exemplified by Hunter et al. (2004). Methods for measuring celecoxib concentration in synovial membrane are exemplified by G. Gaudriault, et al.

This invention also provides a plurality of biodegradable microspheres, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a suitable matrix (e.g., a polylactic-co-glycolic acid copolymer (PLGA) matrix); (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month, wherein the celecoxib so released is characterized (e.g., when measured at 14 days post injection) by (i) a synovial membrane to synovial fluid celecoxib concentration ratio of at least 20, and/or (ii) a joint cartilage to synovial fluid celecoxib concentration ratio of at least 10. The biodegradable microsphere components, dimensions, features, celecoxib dosing, and celecoxib release duration described above are envisioned, mutatis mutandis, for this plurality of biodegradable microspheres. In a preferred embodiment, the synovial membrane to synovial fluid celecoxib concentration ratios and the joint cartilage to synovial fluid celecoxib concentration ratios are as described above.

This invention further provides an injectable formulation comprising (a) a pharmaceutically acceptable carrier and (b) a plurality of biodegradable microspheres wherein the microspheres (i) have a d9o value from 1 pm to 500 pm;

(ii) comprise a suitable matrix (e.g., a polylactic-co-glycolic acid copolymer (PLGA) matrix); (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month, wherein the celecoxib so released is characterized (e.g., when measured at 14 days post-injection) by (i) a synovial membrane to synovial fluid celecoxib concentration ratio of at least 20, and/or (ii) a joint cartilage to synovial fluid celecoxib concentration ratio of at least 10. The biodegradable microsphere components, dimensions, features, celecoxib dosing, and celecoxib release duration described above are envisioned, mutatis mutandis, for this injectable formulation. In a preferred embodiment, the synovial membrane to synovial fluid celecoxib concentration ratios and the joint cartilage to synovial fluid celecoxib concentration ratios are as described above.

This invention still further provides a method for treating a joint-related disorder in a subject comprising introducing biodegradable microspheres into suitable tissue in or around one or more of the subject’s joints, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a suitable matrix (e.g., a polylactic-co- glycolic acid copolymer (PLGA) matrix); (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month, wherein the celecoxib so released is characterized (e.g., when measured at 14 days post-injection) by (i) a synovial membrane to synovial fluid celecoxib concentration ratio of at least 20, and/or (ii) a joint cartilage to synovial fluid celecoxib concentration ratio of at least 10. The (i) biodegradable microsphere components, dimensions, features, celecoxib dosing, and celecoxib release duration, and (ii) therapeutic indications, subjects, and modes of administration described above are envisioned, mutatis mutandis, for this method. In a preferred embodiment, the synovial membrane to synovial fluid celecoxib concentration ratios and the joint cartilage to synovial fluid celecoxib concentration ratios are as described above.

Finally, this invention provides a kit comprising, in separate compartments, (a) a diluent, and (b) plurality of biodegradable microspheres, wherein the microspheres (i) have a d9o value from 1 pm to 500 pm; (ii) comprise a suitable matrix (e.g., a polylactic-co-glycolic acid copolymer (PLGA) matrix); (iii) carry a therapeutically effective amount of pharmaceutical celecoxib; and (iv) when present in a suitable joint-related tissue, release celecoxib for at least one month, wherein the celecoxib so released is characterized (e.g., when measured at 14 days post-injection) by (i) a synovial membrane to synovial fluid celecoxib concentration ratio of at least 20, and/or (ii) a joint cartilage to synovial fluid celecoxib concentration ratio of at least 10. The diluents, biodegradable microsphere components, dimensions, features, celecoxib dosing, and celecoxib release duration described above are envisioned, mutatis mutandis, for this kit. In a preferred embodiment, the synovial membrane to synovial fluid celecoxib concentration ratios and the joint cartilage to synovial fluid celecoxib concentration ratios are as described above.

The subject biodegradable microspheres, formulations, methods, and kits are envisioned in this invention for (a) administration to (i) epidural and perineural spaces, (ii) the foramenal space at or near the site of a patient’s pain or inflammation, and (iii) sites of injury or surgery, as well as (b) treatment of non-joint- related indications such as post-surgical pain relief (e.g., relief of pain due to hernia repair, abdominoplasty, or bunionectomy), mutatis mutandis, as they are for suitable joint-related tissue. This invention will be better understood by reference to the examples which follow, but those skilled in the art will readily appreciate that the specific examples detailed are only illustrative of the invention as described more fully in the claims which follow thereafter.

Examples

Example 1 . Experimental Materials and Methods Generally

PLGA refers to poly-lactic-co-glycolic acid; PDLA refers to poly-D-lactic acid, which is one kind of poly-lactic acid (PLA); DMSO refers to dimethyl sulfoxide; PVA refers to poly-vinyl-alcohol, molecular weight: 31-50K, 98-99% hydrolyzed; PBS refers to phosphate buffered saline, pH 7.4; and MW refers to molecular weight. The magnetic stir bar used in all of the following examples is 2 cm in length and 0.7 cm in diameter.

(i) PDLA, 0.4 dl/g, ester-terminated: Intrinsic viscosity = 0.35-0.45 dl/g. MW:20,000- 30,000; (ii) PDLA, 0.6 dl/g, ester-terminated: Intrinsic viscosity = 0.55-0.65 dl/g; (iii) PDLA, 2 dl/g, ester-terminated: Intrinsic viscosity = 1.6-2.4 dl/g. MW:300,000- 600,000; (iv) PLGA50:50, 0.2 dl/g, acid-terminated: Intrinsic viscosity = 0.16-0.24 dl/g. MW:7, 000-17,000; (v) PLGA50:50, 0.2 dl/g, ester-terminated: Intrinsic viscosity = 0.16-0.24 dl/g. MW:7, 000-17,000; (vi) PLGA50:50, 0.4 dl/g, acid-terminated: Intrinsic viscosity = 0.32-0.44 dl/g. MW:24, 000-38, 000; (vii) PLGA50:50, 0.4 dl/g, ester- terminated: Intrinsic viscosity = 0.32-0.44 dl/g. MW:24, 000-38, 000; (viii) PLGA50:50, 0.5 dl/g, acid-terminated: Intrinsic viscosity = 0.45-0.6 dl/g. MW:38, 000-54, 000; (ix) PLGA50:50, 0.5 dl/g, ester-terminated: Intrinsic viscosity = 0.45-0.6 dl/g. MW:38, GOO- 54, 000; (x) PLGA50:50, 0.6 dl/g, ester-terminated: Intrinsic viscosity = 0.50-0.65 dl/g; (xi) PLGA50:50, 0.7 dl/g, ester-terminated: Intrinsic viscosity = 0.61-0.74 dl/g. MW: 54, 000-69, 000; (xii) PLGA65:35, 0.4 dl/g, acid-terminated: Intrinsic viscosity = 0.32-0.44 dl/g. MW:24, 000-38, 000; (xiii) PLGA75:25, 0.2 dl/g, acid-terminated:

Intrinsic viscosity = 0.15-0.25 dl/g. MW: 10,000-18,000; (xiv) PLGA75:25, 0.2 dl/g, ester-terminated: Intrinsic viscosity = 0.16-0.24 dl/g; (xv) PLGA75:25, 0.4 dl/g, acid- terminated: Intrinsic viscosity = 0.32-0.44 dl/g; (xvi) PLGA75:25, 0.6 dl/g, ester- terminated: Intrinsic viscosity = 0.5-0.7 dl/g; (xvii) PLGA75:25, 0.65 dl/g, ester- terminated: Intrinsic viscosity = 0.55-0.75 dl/g. MW: about 97,000; (xviii) PLGA75:25, 0.9 dl/g, ester-terminated: Intrinsic viscosity = 0.71-1.0dl/g. MW:76, 000-115,000; (xix) PLGA75:25, 1.2 dl/g, ester-terminated: Intrinsic viscosity = 0.9-1 .3 dl/g; (xx) PLGA85:15, 0.6 dl/g, ester-terminated: Intrinsic viscosity = 0.55-0.75 dl/g. MW: 190,000-240,000; and (xxi) PLGA85:15, 1.5 dl/g, ester-terminated: Intrinsic viscosity = 1.3-1.7 dl/g.

In experiments described herein, one type of PVA (i.e. , of one molecular weight and degree of hydrolysis) at one concentration (i.e., 1%) is used to produce microspheres. However, in this invention, other types of PVA (for example, EMRPOVE PVA 4-88, Millipore Sigma) and other PVA concentrations are also envisioned to yield the same microspheres. For example, where 1% PVA, 200 pi dichloromethane, and 1,000 rpm stirring, is used to produce a certain population of microspheres, 0.5% PVA, 300 mI dichloromethane, and 1,400 rpm stirring, can also be used to produce essentially the same population of microspheres. Surfactants other than PVA can also be used in this invention to produce microspheres. These other surfactants include, for example, the commonly known surfactants vitamin E, Tween-20, Tween-80, poloxamers, poloxamines, pluronic polymers (such as F68 and F127), and sodium cholate. Similarly, in experiments described herein, dichloromethane is used to produce microspheres. However, in this invention, other types of organic solvents (e.g., ethyl acetate, chloroform, acetone, propylene carbonate, and tetrahydrofuran) may be used instead of dichloromethane to yield essentially the same microspheres, assuming other experimental parameters are adjusted accordingly. Moreover, in this invention, any of a plurality of physical methods for preparing PLGA microspheres (e.g., spinning disk, spray drying, and microfluidics) may be used to yield the subject microspheres.

Example 2. Microsphere Formulations of Diclofenac Acid: Effects of Polymer

Composition on Diclofenac Acid Release (I)

A pharmaceutical depot was prepared comprised of the NSAID, diclofenac acid (2- [2-(2,6-dichloroanilino)phenyl]acetic acid), incorporated into microspheres. Formulations were prepared by dissolving 50 mg of PLGA or PDLA and 20 mg of diclofenac acid in 440 mI of dichloromethane and 60 mI of DMSO. The 500 mI of solution was injected into the center of 200 ml of 1 % PVA (w/v, pH adjusted to 2) in a 250 ml beaker, stirred at 2,000 rpm with a magnetic stir bar to form an emulsion.

The emulsion was stirred at 2,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with water for three times.

For testing diclofenac acid release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. Diclofenac acid concentration was determined by UV absorption at a wavelength of 276 nm.

Figure 1 shows the release of diclofenac acid in a range of PLGA and PDLA microspheres. The PLGA50:50, 0.6 dl/g, ester-terminated formulation showed complete diclofenac acid release within 30 days. The PLGA75:25, 0.65 dl/g, ester- terminated formulation showed little diclofenac acid release from day 6 to day 16, which is unsuitable for continuous drug delivery. The PLGA85:15, 0.6 dl/g, ester- terminated formulation showed little diclofenac acid release from day 10 to day 16, which is unsuitable for continuous drug delivery. The PDLA, 2 dl/g, ester-terminated formulation showed little diclofenac acid release from day 6 to day 16, which is unsuitable for continuous drug delivery.

Figure 2 shows the release of diclofenac acid in PLGA50:50, 0.2 dl/g, acid- terminated microspheres. Release was complete by day 14, which is unsuitable for continuous drug delivery over multiple months. In some experiments, cumulative drug release may not reach 100%, due to minor variations in instrumental readings.

Example 3. Microsphere Formulations of Diclofenac Acid: Effects of Polymer

Composition on Diclofenac Acid Release (II)

In another set of experiments, formulations were prepared by dissolving 5 mg of PLGA or PDLA and 2 mg of diclofenac acid in 180 pi of dichloromethane and 20 mI DMSO. The 200 mI of solution was injected into the center of 50 ml of 1 % PVA (w/v, pH adjusted to 2) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing diclofenac acid release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. Diclofenac acid concentration was determined by UV absorption at a wavelength of 276 nm.

Figure 3 shows the diclofenac acid release from a range of PLGA and PDLA formulations. The PLGA50:50, 0.2 dl/g, acid-terminated formulation resulted in complete drug release in about 10 days. All other formulations resulted in little drug release from day 7 to day 10. None of these formulations is suitable for continuous drug delivery over multiple months.

Example 4. Microsphere Formulations of Diclofenac Acid: Effects of Mixed PLGA or

PDLA on Diclofenac Acid Release

Since PLGA50:50, 0.2 dl/g, acid-terminated formulations released diclofenac acid very quickly within 15 days, and some other formulations (such as PLGA75:25, 0.65 dl/g, ester-terminated, PLGA85:15, 0.6 dl/g, ester-terminated, PDLA, 0.4 dl/g, ester- terminated) released diclofenac acid too slowly, or stopped drug release after day 7, mixed PLGA or PDLA formulations were tested to obtain formulations that may achieve continuous drug release over 30 days.

Formulations were prepared by dissolving 5 mg of PLGA or PDLA and 2 mg of diclofenac acid in 180 pi of dichloromethane and 20 mI DMSO. The 5 mg of PLGA or PDLA was comprised of various ratios of PLGA50:50, 0.2 dl/g, acid-terminated, PLGA85:15, 0.6 dl/g, ester-terminated, PLGA75:25, 0.65 dl/g, ester-terminated, and PDLA, 0.4 dl/g, ester-terminated. The 200 mI of solution was injected into the center of 50 ml of 1 % PVA (w/v, pH adjusted to 2) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing diclofenac acid release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. Diclofenac acid concentration was determined by UV absorption at a wavelength of 276 nm.

Figure 4 shows that both formulations involving PLGA85:15, 0.6 dl/g, ester- terminated resulted in complete diclofenac acid release within 10 days. Both formulation involving PDLA, 0.4 dl/g, ester-terminated released little diclofenac acid from day 4 to day 9. Figure 5 shows complete diclofenac acid release in 10 days. None of these formulations is suitable for continuous drug release over multiple months.

Example 5. Microsphere Formulations of Lornoxicam

Microsphere formulations of another NSAID, lornoxicam (6-ch!oro-4-hydroxy-2- methy!-1 ,1-dioxo-A/-pyridin-2-yi-thieno[2,3-e]thiazine-3-carboxamide ), were prepared with various PLGA and PDLA polymers, including PLGA50:50, 0.2 dl/g, acid- terminated, PLGA75:25, 0.65 dl/g, ester-terminated, PLGA50:50, 0.6 dl/g, ester- terminated, PLGA85:15, 0.6 dl/g, ester-terminated, PDLA, 0.4 dl/g, ester-terminated, and PDLA, 2 dl/g, ester-terminated. Formulations were prepared by dissolving 20 mg of PLGA or PDLA and 2 mg lornoxicam in 900 pi dichloromethane and 100 mI methanol. The 1 ml of solution was injected into the center of 50 ml of 1 % PVA (w/v, pH adjusted to 2) in a 50 ml beaker, stirred at 800 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 800 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane.

Figure 6 shows the microsphere formation of lornoxicam-PLGA microspheres (PLGA50:50, 0.2 dl/g, acid-terminated). Crystal rods and spikes formed during evaporation of dichloromethane and precipitation of lornoxicam. The rods and spikes of lornoxicam crystals prevented the formation of microspheres. These formulations are unsuitable for continuous drug delivery of lorxicam over multiple months.

Example 6. Microsphere Formulations of Celecoxib: Effects of Polymer Composition on Celecoxib Release

Formulations were prepared by dissolving 2.5 mg PLGA and 1 mg celecoxib in 100 pi dichloromethane. The 100 mI of solution was injected into the center of 50 ml of 1% PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 7 shows cumulative celecoxib release for a range of PLGA formulations. PLGA50:50, 0.2 dl/g, acid-terminated, PLGA50:50, 0.4 dl/g, acid-terminated, PLGA50:50, 0.5 dl/g, acid-terminated, and PLGA65:35, 0.4 dl/g, acid-terminated resulted in continuous celecoxib release over 7 days, 20 days, 24 days, and 25 days, respectively, which are unsuitable for continuous drug release over multiple months.

PLGA50:50, 0.4 dl/g, ester-terminated resulted in continuous celecoxib release over 30 days, and PLGA50:50, 0.6 dl/g, ester-terminated resulted in continuous celecoxib release over 37 days. Low molecular weight and correspondingly low inherent viscosity PLGA is known to allow for more complete and faster release of pharmaceutical agents incorporated into microparticles than their higher molecular weight and higher inherent viscosity counterparts. (Anderson, et al. “Biodegradation and biocompatibility of PLA and PLGA microspheres.” Advanced Drug Delivery 28(1997): 5-24; Bouissou et al., “Poly(lactic-co-glycolic-acid)

Microspheres.” Polymer in Drug Delivery (2006): Chapter 7). Unexpectedly, PLGA50:50, 0.5 dl/g, ester-terminated resulted in slower celecoxib release than the PLGA50:50, 0.6 dl/g, ester-terminated formulation. Also, the PLGA50:50, 0.5 dl/g, ester-terminated formulation resulted in little celecoxib release from day 9 to day 15, which is unsuitable for continuous drug release over months.

PLGA75:25, 0.4 dl/g, acid-terminated resulted in continuous celecoxib release over 35 days. Unexpectedly, compared to the release curve of PLGA50:50, 0.6 dl/g, ester-terminated formulation, the celecoxib release rate was more uniform throughout the entire period of drug release for PLGA75:25, 0.4 dl/g, acid-terminated. In other words, the release rate between day 1 and day 23 was higher for PLGA75:25, 0.4 dl/g, acid-terminated, and the release rate between day 24 to day 35 was lower for PLGA75:25, 0.4 dl/g, acid-terminated. A more uniform drug release is desirable for a pharmaceutical depot formulation because it would require a lower total dosage of drug administration in order to maintain the same efficacious drug concentration throughout the entire release period in vivo.

Example 7. Microsphere Formulations of Celecoxib: Effects of Celecoxib Loading on

Drug Release

Formulations were prepared by dissolving 2.5 mg PLGA and different amounts of celecoxib in 200 pi dichloromethane. The 200 mI of solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 8 shows that the celecoxib-loading amount from 1 mg to 2.5 mg, mixed with 2.5 mg PLGA, did not change drug release significantly. Figure 9 shows that increasing celecoxib loading to 3-3.5 mg, mixed with 2.5 mg PLGA50:50, 0.5 dl/g, ester-terminated, resulted in continuous celecoxib release, which is different from the little drug release seen in day 9 to day 15 in formulations with lower (1 mg) celecoxib loading. Furthermore, the two PLGA50:50, 0.7 dl/g, ester-terminated formulations showed continuous celecoxib release till at least day 50. In comparison, both PLGA75:25, 0.6 dl/g, ester-terminated formulations had little celecoxib release from day 30 to day 45, which is unsuitable for continuous celecoxib release over multiple months.

Example 8. Microsphere Formulations of Celecoxib: Effects of Mixed PLGA

Polymers on Celecoxib Release

Since a PLGA50:50, 0.5 dl/g, acid-terminated formulation resulted in continuous celecoxib release over 24 days (Figure 7), mixing with other PLGA polymers may accelerate their respective celecoxib release. Formulations were prepared by dissolving 2.5 mg PLGA and different amounts (1 mg, 2.5 mg, or 3.5 mg) of celecoxib in 200 pi dichloromethane. 2.5 mg PLGA was comprised of various ratios of PLGA50:50, 0.5 dl/g, acid-terminated, and other PLGA/PDLA polymers. The 200 mI of solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times. For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 pi was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

In Figure 10, 2.5 mg of 50% of PLGA50:50, 0.5 dl/g, acid-terminated, and 50% of PDLA, 0.4 dl/g, ester-terminated, with 1 mg celecoxib loading, resulted in continuous celecoxib release over 45 days.

In Figure 11 , 2.5 mg of PLGA with all ratios of PLGA50:50, 0.5 dl/g, acid-terminated, mixed with PLGA50:50, 0.4 dl/g, ester-terminated, and PLGA50:50, 0.5 dl/g, ester- terminated, resulted in continuous drug release at celecoxib loading of 1 mg.

In Figure 12, 2.5 mg of PLGA with 2.5 mg celecoxib loading was tested. 25% PDLA, 0.4 dl/g, ester-terminated, with 75% PLGA50:50, 0.5 dl/g, acid-terminated, resulted in continuous celecoxib release over 25 days.

In Figure 13, 2.5 mg of PLGA with 3.5 mg celecoxib loading was tested. 50-75% of PLGA50:50, 0.5 dl/g, acid-terminated, mixed with PLGA75:25, 0.6 dl/g, ester- terminated, resulted in continuous celecoxib release over 25 days.

Example 9. Microsphere Formulations of Celecoxib: Effects of Poly-Ethylene Glycol

(PEG) on Celecoxib Release

PEG/PLGA blends are known to allow for more complete and faster release of pharmaceutical agents incorporated into microparticles than PLGA alone (Cleek et al. , “Microparticles of poly(DL-lactic-coglycolic acid)/poly(ethylene glycol) blends for controlled drug delivery.” J Control Release 48(1997):259-268; Morlock, et al., “Erythropoietin loaded microspheres prepared from biodegradable LPLG-PEO-LPLG triblock copolymers: protein stabilization and in vitro release properties.” J Control Release, 56(1-3) (1998): 105-15; Yeh, “The stability of insulin in biodegradable microparticles based on blends of lactide polymers and polyethylene glycol.” J Microencapsul, 17(6) (2000):743-56.).

Formulations were prepared by dissolving 2.5 mg PLGA50:50, 0.5 dl/g, ester- terminated, 2.5 mg of celecoxib, and 0.625 mg, 1 .25 mg, or 2.5 mg PEG1450 in 200 mI dichloromethane. The 200 mI of solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 14 shows that PEG1450 blending with PLGA slightly increased celecoxib release. With 0 mg of PEG1450, celecoxib release was minimal from day 11 to day 15, whereas all three formulation with PEG1450 blending showed continuous celecoxib release until day 30.

Example 10. Microsphere Formulations of Celecoxib: Microsphere Formulations

Based on PLGA75:25, 0.4 dl/q, Acid-Terminated

Formulations were prepared by dissolving 2.5 mg PLGA and different amounts (1 mg-5 mg) of celecoxib in 100 pi or 200 mI dichloromethane. The solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm or 1 ,200 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm or 1 ,200 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 pi was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 15 shows that celecoxib loading of 1-2.5 mg with 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 100 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous drug release over 30 days.

Figure 16 shows that celecoxib loading of 1-3.5 mg with 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous drug release over 30 days.

Figure 17 shows that celecoxib loading of 3.5 mg and 5 mg with 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous drug release over 60 days.

Figure 18 shows that celecoxib loading of 5 mg and 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm, resulted in continuous drug release over 60 days.

Figure 19 shows that celecoxib loading of 3.5 mg and 5 mg, and 2.5 mg PLGA75:25, 0.4 dl/g, acid-terminated, dissolved in 200 mI dichloromethane and stirred at 1,200 rpm, resulted in continuous drug release over 60 days. Example 11. Microsphere Formulations of Celecoxib (I)

Formulations were prepared by dissolving 2.5 mg PLGA and different amounts (4 mg - 7.5 mg) of celecoxib in 200 pi dichloromethane. The solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 20 shows that celecoxib loading at 4 mg, 5 mg, and 6 mg, mixed with PLGA50:50, 0.5 dl/g, ester-terminated, and PLGA50:50, 0.6 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 30 days.

Figure 21 shows that celecoxib loading at 6 mg and 7.5 mg, mixed with PLGA50:50, 0.5 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 20 days.

Figure 22 shows that celecoxib loading at 4 mg, 5 mg, and 6 mg, mixed with PLGA50:50, 0.7 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 70 days. Unexpectedly, increased loading ratio did not result in higher release rate. Example 12. Microsphere Formulations of Celecoxib (II)

Formulations were prepared by dissolving 2.5 mg PLGA and different amounts (4 mg - 7.5 mg) of celecoxib in 200 pi - 400 mI dichloromethane. The solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 23 shows that celecoxib loading at 4 mg, 5 mg, and 6 mg, mixed with 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 111 days. Higher celecoxib loading was associated with faster drug release.

Figure 24 shows that celecoxib loading at 5 mg, 6 mg, and 7.5 mg, mixed with 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 92 days. Higher celecoxib loading was associated with faster drug release.

Figure 25 shows that celecoxib loading at 5 mg and 6 mg, mixed with 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, dissolved in 300 mI or 400 mI and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 90 days. Example 13. Microsphere Formulations of Celecoxib (III)

Formulations were prepared by dissolving 2.5 mg PLGA and different amounts (5 mg-7.5 mg) of celecoxib in 300 pi or 400 mI dichloromethane. The solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 26 shows celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 90 days.

Figures 27, 28, 29, and 30 show that celecoxib loading of 5 mg, 6 mg, and 7.5 mg, mixed with PLGA75:25, 0.9 dl/g, ester-terminated, dissolved in 300 mI or 400mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 80 days.

Figure 31 shows that celecoxib loading of 5 mg and 6 mg, mixed with PLGA75:25,

1.2 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 70 days. Example 14. Microsphere Formulations of Celecoxib (IV)

Formulations were prepared by dissolving 2.5 mg PLGA or PDLA and different amounts (5 mg - 7.5 mg) of celecoxib in 200 pi, 300 mI or 400 mI dichloromethane. The solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm with a magnetic stir bar for 2 minutes to form an emulsion. The emulsion was stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figure 32 shows that 7.5 mg of celecoxib, mixed with 2.5 mg PLGA85:15, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 40 days.

Figure 33 shows that 6 mg of celecoxib, mixed with 2.5 mg PLGA85:15, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over about 30 days.

Figure 34 shows that celecoxib loading of 7.5 mg, mixed with 2.5 mg of PLGA85:15,

1 .5 dl/g, ester-terminated, dissolved in 300 mI or 400 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 35 days.

Figure 35 shows that celecoxib loading of 5 mg and 7.5 mg, mixed with 2.5 mg of PLGA85:15, 1.5 dl/g, ester-terminated, dissolved in 500 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 50 days. Figure 36 shows that celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PDLA, 0.6 dl/g, ester-terminated, dissolved in 300 mI dichloromethane and stirred at 1,000 rpm, resulted in continuous celecoxib release over 40 days.

Figure 37 shows that celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PDLA, 0.6 dl/g, ester-terminated, dissolved in 300 mI or 400 mI dichloromethane and stirred at 1 ,000 rpm, resulted in continuous celecoxib release over 35 days.

Figure 38 shows that celecoxib loading of 5 mg and 6 mg, mixed with 2.5 mg PDLA, 0.4 dl/g, ester-terminated, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm, resulted in little celecoxib release from day 20 to day 30 and from day 50 to day 60, which is unsuitable for continuous celecoxib release over multiple months.

Example 15. Microsphere Formulations of Celecoxib: Effects of Mixed PLGA

Compositions on Celecoxib Release

Formulations were prepared by dissolving 2.5 mg PLGA (comprising of different proportions of mixed polymers) and different amounts (3.5 mg - 5 mg) of celecoxib in 200 pi or 300 mI dichloromethane. The solution was injected into the center of 50 ml of 1 % PVA (w/v) in a 50 ml beaker, stirred at 1 ,000 rpm or 1 ,200 rpm with a magnetic stir bar to form an emulsion. The emulsion was stirred at 1 ,000 rpm or 1 ,200 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times.

For testing celecoxib release from the microspheres, all prepared microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm. Figure 39 shows that mixing PLGA50:50, 0.7 dl/g, ester-terminated, with PLGA75:25, 0.6 dl/g, ester-terminated, slowed down celecoxib release and prolonged duration of continuous drug release. With increasing proportion of PLGA75:25, 0.6 dl/g, ester- terminated, the duration of continuous drug release was longer, until the proportion reached 75%, which resulted in continuous drug release for at least 87 days.

Figure 40 shows that mixing PLGA50:50, 0.7 dl/g, ester-terminated, with PLGA75:25, 0.6 dl/g, ester-terminated, slowed down celecoxib release. 25% PLGA50:50, 0.7 dl/g, ester-terminated +75% PLGA75:25, 0.6 dl/g, ester-terminated, had continuous celecoxib release over 130 days.

Figure 41 shows that mixing PLGA75:25, 0.4 dl/g, acid-terminated, with PLGA75:25, 0.6 dl/g, ester-terminated, slowed down celecoxib release. 25% PLGA75:25, 0.4 dl/g, acid-terminated + 75% PLGA75:25, 0.6 dl/g, ester-terminated, had continuous celecoxib release over 120 days.

Example 16. Microsphere Formulations of Celecoxib with Lyophilization (I)

Formulations were prepared by dissolving 2.5 mg PLGA75:25, 0.6 dl/g, ester- terminated, and 5 mg, 6 mg, or 7.5 mg celecoxib in 200 pi or 300 mI dichloromethane The solution was injected into the center of 50 ml of PVA (concentration adjusted to have absorption at 230 nm of 0.160) in a 50 ml beaker, stirred at 1 ,000 rpm or 1 ,200 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times, and lyophilized overnight to form a dry powder.

To determine the encapsulation efficiency of celecoxib into the microspheres (encapsulated celecoxib in the microspheres/total celecoxib input), the absorption of the PVA at 230 nm wavelength was measured after microsphere harvesting and normalized against the background absorption of 0.160. This differential was due to the “leaked” celecoxib into the PVA. By subtracting the amount of leaked celecoxib into the PVA from total celecoxib, the amount of encapsulated celecoxib was determined. To determine the loading ratio of celecoxib in the microspheres (celecoxib/(celecoxib+PLGA)), 10 mg of dry microspheres were added to 40 ml ethanol and shaken overnight to allow complete dissolution of celecoxib in ethanol. Celecoxib concentration in ethanol was analyzed by UV absorption at 230 nm.

For testing celecoxib release from the microspheres, 8 mg of dry microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 pi was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figures 42A and 42B show that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg (Figure 42A) or 6 mg (Figure 42B) celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm resulted in continuous celecoxib release over at least 180 days. The encapsulation efficiencies for the 5 mg and 6 mg group were 94.4% and 93.8%, respectively. This indicates near complete encapsulation of celecoxib into the microspheres. The loading ratios for the 5 mg and 6 mg group were 62.7% and 62.8%, respectively. Figures 42C and 42D show microscopic images of microspheres formed by 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated and 5 mg (Figure 42C) or 6 mg (Figure 42D) celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm for 2 minutes, followed by stirring at 500 rpm for 2 hours, harvesting, washing with water, and lyophilization.

Figure 43 shows that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 6 mg celecoxib, dissolved in 300 mI dichloromethane and stirred at 1 ,000 rpm resulted in continuous celecoxib release over 175 days. Figure 44 shows that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 7.5 mg celecoxib, dissolved in 200 mI or 300 mI dichloromethane and stirred at 1 ,200 rpm or 1 ,000 rpm resulted in continuous celecoxib release over 180 days. Figure 45 shows that that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,000 rpm or 1 ,200 rpm resulted in continuous celecoxib release over 180 days.

Example 17. Microsphere Formulations of Celecoxib with Lyophilization (II)

Formulations were prepared by dissolving 2.5 mg PLGA75:25, 0.6 dl/g, ester- terminated, and 5 mg or 6 mg celecoxib in 200 pi dichloromethane. The solution was injected into the center of 50 ml of PVA (concentration adjusted to have absorption at 230 nm of 0.05) in a 50 ml beaker, stirred at 1 ,000 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times, and lyophilized overnight to form a dry powder.

For testing celecoxib release from the microspheres, 8 mg of dry microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 mI was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm.

Figures 46A and 46B show that 2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg (Figure 46A) or 6 mg (Figure 46B) celecoxib, dissolved in 200 mI dichloromethane and stirred at 1,000 rpm resulted in continuous celecoxib release over 180 days.

Example 18. Microsphere Formulations of Celecoxib with Lvophilization (III)

Formulations were prepared by dissolving 2.5 mg PLGA75:25, 0.6 dl/g or 0.9 dl/g, ester-terminated, and 5 mg celecoxib in 200 mI dichloromethane. The solution was injected into the center of 50 ml of PVA (concentration adjusted to have absorption at 230nm of 0.160) in a 50 ml beaker, stirred at 1 ,200 rpm for 2 minutes and stirred at about 500 rpm for 2 hours to evaporate the dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, washed with 50 ml water for three times, and lyophilized overnight to form a dry powder.

For testing celecoxib release from the microspheres, 8 mg of dry microspheres were added to 250 ml of 10 mM PBS and shaken at 80 rpm at 37 °C. At each time point in the release study, microspheres were allowed to settle, and an aliquot of 650 pi was taken for measuring UV absorption and added back to the original solution. If the released celecoxib in PBS reached about one third of the fully saturated concentration, 200 ml out of the 250 ml of PBS (without microspheres) was taken out and replaced with fresh 200 ml of PBS. Celecoxib concentration was determined by UV absorption at a wavelength of 230 nm. Three samples were analyzed per formulation.

Figure 47A shows that both PLGA75:25, 0.6 dl/g and 0.9 dl/g, ester-terminated formulations resulted in continuous celecoxib release over least 100 days. The release rate of 0.9 dl/g PLGA was slightly lower than 0.6 dl/g PLGA. Figure 47B shows a microscopic image of microspheres formed by 2.5 mg PLGA75:25, 0.9 dl/g, ester-terminated and 5 mg celecoxib, dissolved in 200 mI dichloromethane and stirred at 1 ,200 rpm for 2 minutes, followed by stirring at 500 rpm for 2 hours, harvesting, washing with water, and lyophilization.

Example 19. Pharmacokinetics of the Present Formulation in Lewis Rats

2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib were dissolved in 200 mI dichloromethane, which was injected into the center of 50 ml PVA solution (concentration adjusted to have absorption at 230 nm of 0.160) and stirred at 1 ,200 rpm for 2 minutes, followed by stirring at about 500 rpm for 2 hours to allow evaporation of dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, treated for two hours with 100 units/ml penicillin, 100 pg/ml streptomycin, and 0.25 pg/ml amphotericin B, washed with 50 ml sterile water for three times, and lyophilized overnight to form a dry powder (“present formulation”). The dry powder from multiple batches was pooled for injection to rats. A diluent was made as 0.9% NaCI, 0.1% polysorbate-80, and 1% carboxymethylcellulose. Prior to injection to rats, the dry powder was re-suspended in 65 pi of diluent. Celecoxib concentration was analyzed by LC-MS (AB Sciex TRIPLE QUAD 6500+).

Nine male Lewis rats (250-275 grams) were injected in bilateral knees with 5 mg of the present formulation. Three rats were sacrificed at day 7, 14, and 21 with blood collection. Figure 48 shows the plasma level of celecoxib, which is much lower than the plasma level observed with human use of oral celecoxib 200 mg QD (705 ng/ml in peak plasma concentration and about 250 ng/ml mean plasma concentration). This suggests a reduced incidence of systemic side effects.

Example 20. Rat Monoiodoacetate (MIA) Osteoarthritis Pain Model

2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib were dissolved in 200 mI dichloromethane, which was injected into the center of 50 ml PVA solution (concentration adjusted to have absorption at 230 nm of 0.160) and stirred at 1 ,200 rpm for 2 minutes, followed by stirring at about 500 rpm for 2 hours to allow evaporation of dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, treated for two hours with 100 units/ml penicillin, 100 pg/ml streptomycin, and 0.25 pg/ml amphotericin B, washed with 50 ml sterile water three times, and lyophilized overnight to form a dry powder (“present formulation”). The dry powder from multiple batches was pooled for injection to the left knees of rats. A diluent was made containing 0.9% NaCI, 0.1% polysorbate-80, and 0.5% methylcellulose. Prior to injection, 5 mg of the present formulation was re suspended in 65 pi of diluent.

Thirty female Sprague-Dawley rats (about 200 grams) were equally divided into three groups: diluent only injection as a negative control; diluent injection with daily oral celecoxib as a positive control; and injection of the present formulation. On day 0, all rats received injection of diluent or the present formulation in the left knee. The positive control group received QD 12.5 mg/kg oral celecoxib from day 0 to day 7, and QD 40 mg/kg oral celecoxib from day 8 to day 10. On day 2, 50 pi of 60 mg/ml monoiodoacetate was injected to the left knee of all rats to induce arthritis. Von Frey filament tests, thermal hyperalgesia tests, and tests for differential weight bearing on hind limbs were performed at five time points: before dosing on day 0; before oral dosing on day 7 and day 10; and 3 hours post-oral dosing on day 7 and day 10. Knee injection of the present formulation resulted in superior pain relief compared to vehicle (diluent only) injection and oral celecoxib. Results are presented in Figures 49A-49C.

Von Frey filament test: Mechanical allodynia of the left hind paw was measured by determining withdrawal thresholds to an electronic Von Frey filament (Bioseb, France). The filament was applied perpendicularly to the plantar surface of the paw with increasing force. The force required to induce a reflex paw withdrawal was calculated.

Differential weight bearing test: With a weight balance changing instrument (YLS- 11 A, Jinan Yi Yan Technology Development Co, China), the rats were tested to register the weight load exerted by the hind paws by means of a force plate inserted in the floor. The mean weight bearing in grams between the MIA injected paw and contra-lateral paw were determined in 10 seconds.

Thermal hyperalgesia test: Thermal hyperalgesia of the left hind paw was measured by using a planter heat test, which measures the paw withdrawal latency period (in seconds) in response to radiant heat. txample 21. Pharmacokinetics ot PLGA-tncapsulated Celecoxib Microspheres in

Beaqle Dogs

2.5 mg PLGA75:25, 0.6 dl/g, ester-terminated, and 5 mg celecoxib were dissolved in 200 mI dichloromethane, which was injected into the center of 50 ml PVA solution (concentration adjusted to have absorption at 230 nm of 0.160) and stirred at 1 ,200 rpm for 2 minutes, followed by stirring at about 500 rpm for 2 hours to allow evaporation of dichloromethane. The microspheres were harvested by centrifugation at 4,000 g for 5 minutes, treated for two hours with 100 units/ml penicillin, 100 pg/ml streptomycin, and 0.25 pg/ml amphotericin B, washed with 50 ml sterile water three times, and lyophilized overnight to form a dry powder (“present formulation”). The dry powder from multiple batches was pooled for injection to Beagle dogs. A diluent was made as 0.9% NaCI, 0.1% polysorbate-80, and 1% carboxymethylcellulose.

Prior to injection to dogs, the dry powder was re-suspended in 1 ml of diluent.

Nine male Beagle dogs (about 10 kg in weight) received injections of 100 mg of the present formulation in bilateral knee joints. At day 7, 14, and 21 , three dogs received blood draw immediately before sacrifice to collect synovial fluid, synovial membrane, femur cartilage, and tibia cartilage from both knees. Celecoxib concentration in these tissues was analyzed by liquid chromatography-mass spectrometry (LC-MS). Briefly, blood was collected from the cephalic vein into tubes with EDTA-K2 as anti coagulant, and centrifuged at 5,000 rpm for 10 minutes to get plasma. Synovial membrane and cartilage were immersed in acetonitrile with volume (in ml) equal to 5 times the wet weight of the tissue (in g), homogenized with mechanical force, and sonicated for 30 minutes to extract celecoxib. The extracted celecoxib was quantified with Acquity UPLC with MS/MS (API4000 with Analyst 1.6.3 AB Sciex).

Figures 50A-50D show celecoxib concentrations in joint tissues (i.e. , synovial fluid (Figure 50A), synovial membrane (Figure 50B), femur cartilage (Figure 50C), and tibia cartilage (Figure 50D)). From day 7 to 21 , the geometric mean of synovial fluid concentrations was above 100 ng/ml. The geometric mean of synovial membrane concentrations was above 100,000 ng/ml. The geometric mean of cartilage concentrations was above 1 ,000 ng/ml. Since the biochemical IC50 of celecoxib is 40 nM (about 15 ng/ml), the observed celecoxib concentrations in vivo greatly exceed the IC50, which indicates sufficient level to deliver therapeutic efficacy.

Figure 51 shows celecoxib concentrations in plasma, which are significantly lower than the peak (705 ng/ml) and mean (about 250 ng/ml) celecoxib concentration in plasma observed in humans at the FDA-approved oral 200 mg QD dose (References: FDA label of celecoxib; SK Paulson et al. , “Pharmacokinetics of celecoxib after oral administration in dogs and humans: effect of food and site of absorption” The Journal of Pharmacology and Experimental Therapeutics. (2001). 297(2):638-45). The high plasma concentration by oral administration gives rise to systemic side effects in gastrointestinal and cardiovascular systems, among others. The low plasma concentration observed with the present formulation should result in significantly reduced systemic side effects. In a separate study, three male Beagle dogs (about 10 kg in weight) received QD 5 mg/kg oral celecoxib under overnight fasting condition. After eight days of dosing, at three hours after the final dose (time of peak plasma concentration), the dogs received blood draw and were sacrificed to collect synovial fluid, synovial membrane, and femur cartilage. Celecoxib concentration in these tissues was analyzed by LC- MS. Briefly, blood was collected from the cephalic vein into tubes with EDTA-K2 as anti-coagulant, and centrifuged at 5,000 rpm for 10 minutes to get plasma. Synovial membrane and cartilage were immersed in acetonitrile with volume (in ml) equal to 5 times the wet weight of the tissue (in g), homogenized with mechanical force, and sonicated for 30 minutes to extract celecoxib. The extracted celecoxib was quantified with Acquity UPLC with MS/MS (API4000 with Analyst 1.6.3 AB Sciex).

Figure 52 shows celecoxib concentrations in plasma and joint tissues. The synovial fluid concentration was significantly lower than in plasma, and the levels in plasma, synovial membrane, and femur cartilage were similar.

Surprisingly, the concentration ratio of synovial membrane to synovial fluid, and the concentration ratio of cartilage to synovial fluid were significantly higher for the present formulation than for oral administration of celecoxib. Despite the different routes of administration (knee injection versus oral), the altered distribution of celecoxib among different joint compartments within the same joint was unexpected. In Figures 53A and 53B, the geometric mean of the concentration ratio of synovial membrane to synovial fluid is 5.4 for oral celecoxib (Figure 53A) versus 2,522 for the present formulation as a knee injection (Figure 53B). The geometric mean of the concentration ratio of femur cartilage to synovial fluid is 4.2 for oral celecoxib versus 49 for the present formulation as a knee injection, which is consistent with the concentration ratio (54) of tibia cartilage to synovial fluid for the present formulation as a knee injection. In other words, the present formulation as a knee injection enabled high enrichment of celecoxib in synovial membrane and cartilage relative to synovial fluid, which is unexpected.

The sites of therapeutic action for celecoxib are synovial membrane and cartilage. The unexpectedly high enrichment in synovial membrane and cartilage relative to synovial fluid is a favorable property, since it permits joint injection of PLGA- encapsulated celecoxib at a much lower dose to achieve therapeutic efficacy, which in turn permits further reducing celecoxib’s systemic side effects.

References

Anderson, et al. , Biodegradation and biocompatibility of PLA and PLGA microspheres, Advanced Drug Delivery 28(1997): 5-24.

Bodick, et al., U.S. Patent No. 9,555,048.

Bouissou, et al. , Poly(lactic-co-glycolic-acid) Microspheres. Polymer in Drug Delivery (2006): Chapter 7.

H. Chikaura, et al., Effect of particle size on biological response by human monocyte-derived macrophages, Biosurface and Biotribology, March 2016. 2(1 ): 18- 25.

Cleek et al., Microparticles of poly(DL-lactic-coglycolic acid)/poly(ethylene glycol) blends for controlled drug delivery. J Control Release 48(1997):259-268.

FDA Label for Celebrex ^® (celecoxib).

FDA Label for Zilretta ^® (a PLGA-triamcinolone-acetonide microsphere formulation).

G. Gaudriault, et al. , International Publication No. WO/2017/085561.

T. Green, et al., Polyethylene particles of a "critical size" are necessary for the induction of cytokines by macrophages in vitro, Biomaterials, Dec. 1998; 19(24): 2297-2302.

M. Homar, et al., Influence of polymers on the bioavailability of micro-encapsulated celecoxib, J. of Microencapsulation, Nov. 2007; 24(7): 621-633.

Hunter et al. (2004) Plasma pharmacokinetics and synovial fluid concentrations after oral administration of single and multiple doses of celecoxib in Greyhounds. AJVR. Vol 66, No. 8. M. Janssen, et al. , Celecoxib-loaded PEA microspheres as an auto regulatory drug- delivery system after intra-articular injection, J. Control Release, 2016 Dec. 28:244 (Pt. A): 30-40.

J. Matthews, et al., Comparison of the response of primary human peripheral blood mononuclear phagocytes from different donors to challenge with model polyethylene particles of known size and dose, Biomaterials. Oct. 2000. 21(20):2033-2044.

Morlock, et al. Erythropoietin loaded microspheres prepared from biodegradable LPLG-PEO-LPLG triblock copolymers: protein stabilization and in vitro release properties. J Control Release, 56(1-3) (1998):105-15.

S.K. Paulson, et al., Pharmacokinetics of celecoxib after oral administration in dogs and humans: effect of food and site of absorption, The Journal of Pharmacology and Experimental Therapeutics. (2001) 297(2):638-45.

A. Petit, et al., Release behavior and intra-articular biocompatibility of celecoxib- loaded acetyl-capped PCLA-PEG-PCLA thermogels, Biomaterials 35 (2014) 7919- 7928.

A. Petit, et al., Sustained intra-articular release of celecoxib from in situ forming gels made of acetyl-capped PCLA-PEG-PCLA triblock copolymers in horses,

Biomaterials 35 (2015): 426-436.

H. Thakkar, et al., Celecoxib incorporated chitosan microspheres: in vitro and in vivo evaluation, J. of Drug Targeting, October - December 2004, Vol. 12 (9-10), pp. 549-557.

H. Thakkar, et al., Enhanced retention of celecoxib-loaded solid lipid nano-particles after intra-articular administration, Drugs R D 2007; 8(5):275-285.

Wang, et al. (2011) Intra-discal vancomycin-loaded PLGA microsphere injection for MRSA discitis: an experimental study. Arch Orthop Trauma Surg. 131:111-119. Williems, et al. (2017) Safety of intradiscal injection and biocompatibility of polyester amide microspheres in a canine model predisposed to intervertebral disc degeneration. Journal of Biomedical Materials Research Part B. 105(4):707-714. H.Y. Yang, et al., Applicability of a newly developed bioassay for determining bioactivity of anti-inflammatory compounds in release studies - celecoxib and triamcinolone acetonide released from novel PLGA-based microspheres, Pharm. Res. (2015) 32:680-690. Yeh, The stability of insulin in biodegradable microparticles based on blends of lactide polymers and polyethylene glycol. J Microencapsul, 17(6) (2000):743-56.