Ophthalmic compositions

Abstract

A composition comprises: a base oil; an additive comprising a copolymer comprising hydrophobic and hydrophilic units; and a drug. The copolymer may for example have a comb structure in which the hydrophobic units and hydrophilic units are pendant chains on a backbone of the copolymer. The hydrophobic units and hydrophilic units may for example comprise polydimethylsiloxane moieties and ethylene glycol residues respectively. The composition may for example be used as a tamponade or as a component for a tamponade administered to the eye. The invention is useful for solubilising and/or releasing drugs.

Claims

1. A liquid ophthalmic composition comprising: i) a base oil comprising a silicone oil; ii) an additive comprising a copolymer comprising hydrophobic and hydrophilic units; and iii) a drug; wherein the liquid composition is adapted for use in or as an ophthalmic tamponade.

2. The composition of claim 1, wherein the base oil further comprises one or more of a further silicone oil, a fluorinated silicone oil, a perfluorocarbon oil, or mixtures thereof.

3. The composition of claim 1, wherein the base oil has a kinematic viscosity of from about 100 to about 10,000 cSt, or from about 1,000 to about 5,000 cSt, or from about 1,000 to about 2,000 cSt.

4. The composition of claim 1, wherein the copolymer is linear or branched.

5. The composition as claimed in claim 1, wherein the copolymer is a vinyl polymer.

6. The composition as claimed in claim 1, wherein the copolymer has a comb structure in which the hydrophobic and hydrophilic units are pendant chains on a backbone of the copolymer.

7. The composition of claim 1, wherein the hydrophilic unit is a polyethylene glycol unit.

8. The composition of claim 1, wherein the copolymer comprises a hydrophilic monomer unit, and the hydrophilic monomer unit is oligoethyleneglycol monomethyl ether methacrylate, or other oligoethyleneglycol methacrylate monomer.

9. The composition of claim 1, wherein the hydrophobic unit is selected from a methacrylate monomer, dimethacrylate monomer, or mixtures thereof.

10. The composition of claim 1, wherein the copolymer comprises a hydrophobic monomer unit, and the hydrophobic monomer unit is a polydimethylsiloxane methacrylate monomer, or polydimethylsiloxane dimethacrylate monomer, or other polydimethylsiloxane acrylate monomer.

11. The composition of claim 1, wherein the copolymer independently comprises from about 4 to about 100 hydrophilic and/or hydrophobic monomeric units, or from about 5 to about 90 hydrophilic and/or hydrophobic monomeric units, or from about 10 to about 80 hydrophilic and/or hydrophobic monomeric units, or from about 15 to about 70 hydrophilic and/or hydrophobic monomeric units, or from about 20 to about 60 hydrophilic and/or hydrophobic monomeric units.

12. The composition of claim 1, wherein the molar ratio of the monomeric units (hydrophobic:hydrophilic) is from about 80:20 to about 50:50; or from about 75:25 to about 50:50; or from about 70:30 to about 50:50; or from about 65:35 to about 50:50; or from about 60:40 to about 50:50; or from about 55:45 to about 50:50.

13. The composition of claim 1, wherein the copolymer has a weight average molecular weight of from about 30,000 to about 5,300,000 g/mol, or from about 35,000 to about 350,000 g/mol, or from about 40,000 to about 250,000 g/mol.

14. The composition of claim 1, wherein additive is present in an amount of from about 0.05% to about 20% v/v relative to the base oil, or from about 1% to about 15% v/v, or from about 2% to about 12% v/v, or from about 3% to about 10% v/v, or from about 4% to about 8% v/v, or from about 4% to about 7% v/v.

15. The composition of claim 1, wherein the copolymer comprises residues of a polydimethylsiloxane methacrylate monomer of the following formula: ##STR00007## wherein r is selected from about 1 to about 100, or from about 2 to about 90, or from about 4 to about 85, or from about 6 to about 80, or from about 8 to about 75, or from about 8 to about 70, or from about 9 to about 65, or from about 10 to about 60, or from about 12 to about 57, or from about 14 to about 55.

16. The composition of claim 1, wherein the copolymer comprises residues of a polydimethylsiloxane dimethacrylate monomer of the following formula: ##STR00008## wherein m is selected from about 1 to about 300, or from about 5 to about 290, or from about 10 to about 280, or from about 20 to about 250, or from about 20 to about 200, or from about 30 to about 180, or from about 40 to about 150, or from about 50 to about 100.

17. The composition of claim 1, wherein the drug is selected from an anti-inflammatory drug, an anti-proliferative, an anti-oxidant drug, an anti-neoplastic drug, an anti-growth factor, or mixtures thereof.

18. The composition of claim 1, wherein the drug is selected from all-trans retinoic acid and non-steroidal anti-inflammatories.

19. The composition of claim 1, wherein the drug is present in an amount of from about 1 to about 1000 μg per ml; or from about 5 to about 900 μg per ml; or from about 10 to about 800 μg per ml; or from about 15 to about 700 μg per ml.

20. The composition according to claim 1, wherein the copolymer is formed by controlled radical polymerisation, conventional free radical polymerisation, or other addition polymerisation.

21. The composition of claim 1, wherein the copolymer is a statistical copolymer.

Description

(1) The present invention will now be described by way of example with reference to the following examples and figures:

(2) FIG. 1 is GPC RI chromatogram overlay of A) p(OEGMA.sub.60) and B) p(PDMSMA.sub.(9)60) synthesised by ATRP (broad peak) and the RAFT synthesised equivalents (narrow peak);

(3) FIG. 2 GPC chromatogram overlays of A) RI chromatograms and B) RALS chromatograms for linear p(OEGMA.sub.60) (narrow peak) and branched p(OEGMA.sub.15-co-PDMSDMA.sub.(55)0.95) (broad peak);

(4) FIG. 3 GPC chromatogram overlays of A) RI and B) RALS signals for linear p(PDMSMA.sub.(9)47-co-OEGMA.sub.12) (black) and branched equivalents with either PDMSDMA.sub.(12)0.95 (red) or PDMSDMA.sub.(55)0.95 (blue) and p(PDMSMA.sub.(57)20-co-OEGMA.sub.5-co-PDMSDMA.sub.(55)9.95) (green). GPC analysis C) RI and D) RALS for linear p(PDMSMA.sub.(9)30-co-OEGMA.sub.30) (black) and branched equivalents with either PDMSDMA.sub.(12)0.95 (red) or PDMSDMA.sub.(55)0.95 (blue) and p(PDMSMA.sub.(57)20-co-OEGMA.sub.20-co-PDMSDMA.sub.(55)0.95) (green);

(5) FIG. 4 GPC chromatogram overlays of A) RI and B) RALS signals for linear p(PDMSMA.sub.(9)15-co-PDMSMA.sub.(57)15-co-OEGMA.sub.30) (black) and p(PDMSMA.sub.(9)24-co-PDMSMA.sub.(57)24-co-OEGMA.sub.12) (red) with their branched equivalents with PDMSDMA.sub.(55)0.95 (blue and green respectively);

(6) FIG. 5 .sup.1H NMR spectra (CDCl.sub.3, 400 MHz) of CPBD (top spectra) and p(PDMSMA.sub.(9)48-co-OEGMA.sub.12) after CTA removal (bottom spectra);

(7) FIG. 6 Resazurin assay with appropriate controls used to infer cytotoxicity of silicone oil (technical grade SiO.sub.1000 having a viscosity of from 900-1200 cSt at 25° C., and blends of p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) with silicone oil (technical grade SiO.sub.1000) at 10% (v/v). Pre- and post-confluent ARPE-19 cells (grown for 1 and 7 days respectively) were exposed to the oils for 1 and 7 d. A: Pre-confluent cells exposed for 1 day, B: Pre-confluent cells exposed for 7 days, C: Post-confluent cells exposed for 1 day and D: Post-confluent cells exposed for 7 days (mean, error bars represent ±1 standard deviation); n=3. *, Significance by ANOVA and Dunnett's T3 post-hoc evaluation (p≤0.05); and

(8) FIG. 7 ARPE-19 cells stained with phalloidin (green, F-actin cytoskeleton) and DAPI (blue, nuclei). Pre-confluent cells: Negative control (A), Exposed to silicone oil (technical grade SiO.sub.1000) (B), Exposed to a 10% (v/v) blend of p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) for 1 day (C) and post-confluent cells: Negative control (D), Exposed to silicone oil (technical grade SiO.sub.1000) (E), Exposed to a 10% (v/v) blend of p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) for 7 days (F) Scale bars represent 50 μm.

EXAMPLES

List of Abbreviations

(9) 5-FU 5-Fluorouracil

(10) ACN Acetonitrile

(11) AIBN 2,2′-Azobis(2-Methylpropionitrile)

(12) Ar Argon

(13) atRA All-trans Retinoic Acid

(14) ATRP Atom Transfer Radical Polymerisation

(15) bFGF Basic Fibroblast Growth Factor

(16) Bpy Bipyridyl

(17) CDCl.sub.3 Deuterated Chloroform

(18) CPBD 2-Cyano-2-Propyl Benzodithioate

(19) CTA Chain Transfer Agent

(20) D.sub.2O Deuterium Oxide

(21) DAPI Dianidine-2-Phenylindole

(22) DCM Dicholoromethane

(23) DMF Dimethylformamide

(24) DMSO Dimethyl Sulfoxide

(25) EtBrB Ethyl α-bromoisobutyrate

(26) EPGF Epidermal Growth Factor

(27) FGF Fibroblast Growth Factor

(28) FSiO Fluorinated Silicone Oil

(29) HCl Hydrochloric Acid

(30) Ibu Ibuprofen

(31) IF-γ Interferon-Gamma

(32) IL-Interleukin

(33) IPA Isopropanol

(34) MeOH Methanol

(35) Me-PEO Methyl Terminated Poly(ethyleneoxide)

(36) OEGMA Oligoethylene Glycol Methacrylate

(37) PDMS Poly(dimethylsiloxane)

(38) PDMSDMA Methacryloxypropyl Terminated Poly(dimethylsiloxane)

(39) PDMSMA Mono-Methacryloxypropyl Terminated Poly(dimethylsiloxane)

(40) PEO Poly(ethyleneoxide)

(41) PDGF Platelet-Derived Growth Factor

(42) PGA Polyglycolic Acid

(43) PLA Polylactic Acid

(44) PLGA Polylactic-stat-glycolic Acid

(45) PVA Polyvinyl Acetate

(46) RA Retinoic Acid

(47) RAFT Reversible Addition-Fragmentation Chain Transfer

(48) SFA Semi-Fluorinated Alkane

(49) SiO Silicone Oil

(50) TAA Triamcinolone Acetonide

(51) .sup.tBuOH Tertiary Butanol

(52) TGF-ß Tumour Growth Factor-Beta

(53) THF Tetrahydrofuran

(54) VEGF Vascular Endothelial Growth Factor

Preparation

RAFT Polymerisation of OEGMA and PDMSMA

(55) Due to the insolubility of the catalytic system within the ATRP solvent, an alternative polymerisation method was employed; the non-catalytic controlled radical polymerisation technique RAFT. Matching the exact conditions of an ATRP with RAFT polymerisation is not possible, as this method of polymerisation requires a chain transfer agent (CTA). 2-cyano-2-propyl benzodithioate (CPBD) was selected as the RAFT CTA due to the excellent compatibility with methacrylate monomers. 2,2′-azobis(2-methylpropionitrile) (AIBN) was also selected as the free radical initiator due to excellent solubility within .sup.tBuOH and copolymerisations were conducted at 30 wt. % monomer with respect to solvent and at 70° C., Scheme 1.

(56) ##STR00005##

Linear Amphiphilic Copolymers of OEGMA and PDMSMA Via RAFT

(57) Homopolymerisations of OEGMA and the two different chain length PDMSMA monomers, targeting DP.sub.n=60 monomer units, and copolymers of PDMSMA and OEGMA with identical targeted degrees of polymerisation at 80/20 and 50/50 ratios were synthesised. The resulting materials were analysed by .sup.1H NMR spectroscopy and GPC (THF eluent at 35° C.), Table 1.

(58) TABLE-US-00001 TABLE 1 .sup.1H NMR and GPC data of all polymers synthesised via RAFT in .sup.tBuOH GPC (THF).sup.a .sup.1H NMR Target Polymer M.sub.n M.sub.w Polymer Conversion (%) Composition Brancher (g/mol) (g/mol) Ð Composition PDMS OEGMA p(OEGMA.sub.60) — 24,800 30,400 1.23 p(OEGMA.sub.55) — 92 p(PDMSMA.sub.(9)60) — 51,200 59,400 1.16 p(PDMSMA.sub.(9)58) 96 — p(PDMSMA.sub.(57)60) — 376,000 439,900 1.17 p(PDMSMA.sub.(57)23) 38 — p(PDMSMA.sub.(9)48- — 47,650 52,700 1.11 p(PDMSMA.sub.(9)47- 9894 9498 stat-OEGMA.sub.12) stat-OEGMA.sub.11) p(PDMSMA.sub.(9)30- — 33,300 36,800 1.11 p(PDMSMA.sub.(9)24- 80 96 stat-OEGMA.sub.30) stat-OEGMA.sub.29) p(PDMSMA.sub.(57)48- — 350,600 380,200 1.08 p(PDMSMA.sub.(57)22- 45 77 stat-OEGMA.sub.12) stat-OEGMA.sub.9) p(PDMSMA.sub.(57)30- — 220,250 266,850 1.21 p(PDMSMA.sub.(57)14- 47 65 stat-OEGMA.sub.30) stat-OEGMA.sub.20) p(PDMSMA.sub.(9)48- PDMS- 56,200 74,200 1.32 — 94 98 stat-OEGMA.sub.12) DMA.sub.(12)0.95 p(PDMSMA.sub.(9)30- PDMS- 50,700 75,600 1.49 — 91 97 stat-OEGMA.sub.30) DMA.sub.(12)0.95 p(PDMSMA.sub.(9)48- PDMS- 423,900 3,590,000 8.47 — 91 98 stat-OEGMA.sub.12) DMA.sub.(55)0.95 p(PDMSMA.sub.(9)30- PDMS- 142,100 5,220,000 36.73 — 90 97 stat-OEGMA.sub.30) DMA.sub.(55)0.95 .sup.aDetermined by GPC (THF eluent at 35° C.) .sup.bDetermined by .sup.1H NMR in CDCl.sub.3

(59) All materials synthesised have low dispersities (<1.25) indicating control within the polymerisation, and the difference between ATRP and RAFT was readily obvious when comparing the RI traces of p(OEGMA) and p(PDMSMA) (M.sub.n 985 gmol.sup.−1, r=9) homopolymers, prepared by each technique (FIG. 1). High conversions could not be reached with the longer chain PDMSMA monomers (45-47% after 7 days), probably due to steric hindrance as RAFT requires a close proximity of two separate propagating polymer chain ends to undergo CTA exchange; however, these materials were still suitable for study after purification.

Branched Amphiphilic Terpolymers of OEGMA, PDMSMA and PDMSDMA Via RAFT

(60) Due to the successful synthesis of the linear homopolymers and copolymers via RAFT, the introduction of the divinyl PDMS branchers (M.sub.n=1,275 and 4,460 gmol.sup.−1; m=12 and 55 respectively) to form branched terpolymers was undertaken. The brancher to CTA ratio was kept at 0.95:1 for each attempt to ensure gelation did not occur.

(61) p(OEGMA.sub.15) was successfully branched with PDMSDMA.sub.(55), which was not possible when previously attempted by ATRP. The formation of a branched architecture is clearly evidenced by the GPC (THF eluent at 35° C.) analysis of the copolymer (see FIG. 2). This successful copolymerisation led to the synthesis of branched p(PDMSMA-stat-OEGMA-stat-PDMSDMA) terpolymer architectures with varying compositions, Table 2.

(62) TABLE-US-00002 TABLE 2 .sup.1H NMR and GPC data of all branched polymers synthesised via RAFT GPC (THF).sup.a .sup.1H NMR Target Polymer M.sub.n M.sub.w Polymer Conversion (%) Composition Brancher (g/mol) (g/mol) Ð Composition PDMS OEGMA p(OEGMA.sub.15) PDMSDMA.sub.(55) 44,000 470,300 10.69 — — 81 p(PDMSMA.sub.(9)48- PDMSDMA.sub.(12) 56,200 74,200 1.32 — 94 98 stat-OEGMA.sub.12) p(PDMSMA.sub.(9)30- PDMSDMA.sub.(12) 50,700 75,600 1.49 — 91 97 stat-OEGMA.sub.30) p(PDMSMA.sub.(9)48- PDMSDMA.sub.(55) 423,900 3,590,000 8.47 — 91 98 stat-OEGMA.sub.12) p(PDMSMA.sub.(9)30- PDMSDMA.sub.(55) 142,100 5,220,000 36.73 — 90 97 stat-OEGMA.sub.30) p(PDMSMA.sub.(57)20- PDMSDMA.sub.(55) 95,400 108,600 1.14 — 71 91 stat-OEGMA.sub.5) p(PDMSMA.sub.(57)20- PDMSDMA.sub.(55) 149,700 199,300 1.33 — 82 97 stat-OEGMA.sub.20) p(PDMSMA.sub.(9)15- — 95,400 108,600 1.14 p(PDMSMA.sub.(9.57)23- 77 95 stat- stat-OEGMA.sub.29) PDMSMA.sub.(57)15- stat-OEGMA.sub.30) p(PDMSMA.sub.(9)24- — 149,700 199,300 1.33 p(PDMSMA.sub.(9.57)37- 78 92 stat- stat-OEGMA.sub.11) PDMSMA.sub.(57)24- stat-OEGMA.sub.12) p(PDMSMA.sub.(9)15- PDMSDMA.sub.(55) 235,600 323,200 1.37 — 79 91 stat- PDMSMA.sub.(57)15- stat-OEGMA.sub.30) p(PDMSMA.sub.(9)24- PDMSDMA.sub.(55) 174,100 389,900 2.24 — 92 89 stat- PDMSMA.sub.(57)24- stat-OEGMA.sub.12) .sup.aDetermined by GPC (THF eluent at 35° C.) .sup.bDetermined by .sup.1H NMR in CDCl.sub.3

(63) When branched terpolymer architectures were targeted using PDMSDMA.sub.(12) (only PDMSMA.sub.(9) was used in this case due to obvious steric constraints), low dispersities were achieved, and no high molecular weight materials were obtained, suggesting that the branching had been unsuccessful. The use of the PDMS monofunctional monomer and bifunctional brancher of similar sizes (r=9 and m=12) may be expected to lead to steric constraints that prevent branching. When using the longer brancher, PDMSDMA.sub.(55), high molecular weight branched materials were obtained when the smaller monomer (PDMSMA.sub.(9)) was used, which appears to confirm this assumption. Similar problems were evident when branched architectures of p(PDMSMA.sub.(57)-co-OEGMA) were targeted, even with the longer PDMSDMA.sub.(55) branchers. Again, having PDMS monomer and brancher of similar sizes (r=57 and m=55) appears to lead to steric hindrance.

(64) The GPC chromatograms of the different architectures synthesised using RAFT are presented in FIG. 3. It can clearly be seen that only the combination of a small PDMS monomer (r=9) and a long PDMS brancher (m=55) leads to successful branching. The only architectures which display branching consist of p(PDMSMA.sub.(9)48-co-OEGMA.sub.12-co-PDMSDMA.sub.(55)0.95) and p(PDMSMA.sub.(9)30-co-OEGMA.sub.30-co-PDMSDMA.sub.(55)0.95) as evidenced by the very high molecular weights and Ð values obtained.

(65) In an attempt to obtain branched polymers using the long chain (PDMSMA.sub.(57)) monomer, the longer (PDMSDMA.sub.(55)) brancher was used Unfortunately, after leaving this reaction for 30 days, a high enough conversion was not reached in order to obtain highly branched materials (see FIG. 4).

Solubility of Amphiphilic Copolymers and Terpolymers in SiO

(66) As the aim of these studies was to investigate the ability to solubilise a hydrophilic polymer within SiO and evaluate drug release from the resulting mixture, it was essential to investigate the solubility of the amphiphilic co- and terpolymers, both linear and branched, in SiO. Table 3 summarises the materials tested for solubility in SiO and the results obtained.

(67) TABLE-US-00003 TABLE 3 Solubility of the synthesised polymers in SiO (% v/v) Target composition (mol % of total Miscibility monomer) in Silicone Target Polymer Composition Brancher EG DMS Oil (% v/v) p(OEGMA.sub.60) — 100 0  <1 p(PDMSMA.sub.(9)60) — 0 100 Miscible p(PDMSMA.sub.(57)60) — 0 100 Miscible p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) — 10 90 <30 p(PDMSMA.sub.(9)30-stat-OEGMA.sub.30) — 30.8 69.2  <5 p(PDMSMA.sub.(57)48-stat-OEGMA.sub.12) — 1.7 98.3 Miscible p(PDMSMA.sub.(57)30-stat-OEGMA.sub.30) — 6.5 93.5 Miscible p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) PDMSDMA.sub.(55) 9 91 <40 p(PDMSMA.sub.(9)30-stat-OEGMA.sub.30) PDMSDMA.sub.(55) 27 73 <40 p(PDMSMA.sub.(9)24-stat- — 2.9 97.1 Miscible PDMSMA.sub.(57)24-stat-OEGMA.sub.12) p(PDMSMA.sub.(9)15-stat- — 10.8 89.2 Miscible PDMSMA.sub.(57)15-stat-OEGMA.sub.30)

(68) The branched p(OEGMA.sub.15-co-PDMSDMA.sub.(55)0.95) had minimal SiO solubility (<1% v/v) and, as this contained the largest hydrophobic component possible through branching alone, co- and terpolymers of OEGMA with PDMSMA were clearly needed to increase solubility.

(69) All other polymers were soluble at 40-50% (v/v) apart from when a 50/50 ratio of OEGMA:PDMSMA.sub.(9) was copolymerised which represents the smallest content of hydrophobic monomer. This success represents a major increase in solubility compared to less than 1% (v/v) that had been seen without the incorporation of PDMSMA.

Removal of CTA from Amphiphilic Co- and Terpolymers Synthesised Via RAFT

(70) As the amphiphilic copolymers were soluble in SiO at high levels, the removal of the RAFT chain end was required as the CTA-end group is known to be highly coloured and the solutions were a bright pink colour (Scheme 2). In some cases, gelation was also observed when the amphiphilic copolymers and linear homopolymers of PDMSMA were stored, suggesting the presence of difunctionalised monomer impurities within the commercial PDMSMA. The removal and recovery of CPBD following RAFT polymerisations of PMMA has been reported. Excess AIBN was added to a solution of the purified polymer and subsequent AIBN thermal decomposition yields cyanoisopropyl radicals which react with the C═S bond of the thiocarbonyl-thio group. This results in an intermediate radical which can either fragment back to the previous state or free the thiocarbonyl-thio moiety from the polymer chain-end. The excess cyano-isopropyl radicals drive the equilibrium towards a radical chain end which is capped by additional cyano-isopropyl groups. It has been reported that too little AIBN under these conditions can lead to disproportionation and subsequent reactions between polymer chains. The temperature and length of the reaction are important for the complete removal of the thiocarbonyl-thio end group and should follow the half-life time of the radical initiator used (in this case AIBN).

(71) Similar conditions to those previously reported were employed for the removal of CPBD from p(PDMSMA.sub.(9)48-co-OEGMA.sub.12). The purified polymer was dissolved in toluene and a 20 molar excess of AIBN added to the solution which was degassed and heated to 80° C. for 150 minutes (the half-life time of AIBN in these conditions is 80 minutes). The polymer was isolated by precipitation into cold MeOH which afforded a white solution (Scheme 2).

(72) ##STR00006##

(73) The colour change indicated removal of the CPBD and the recovered polymer was analysed by .sup.1H NMR spectroscopy (FIG. 5) which confirmed the CTA removal as the peaks at 7.30, 7.38, and 7.93 ppm, characteristic of the dithiobenzoate moiety, were no longer present.

RAFT

(74) All RAFT polymerisations were conducted at a constant ratio of chain transfer agent to initiator [CPBD]:[AIBN]=1:0.2.

Linear Polymerisation: p(OEGMA)

(75) In a typical synthesis, targeting DP.sub.n=60 monomer units, AIBN (2.7 mg, 0.016 mmol), CPBD (18.4 mg, 0.083 mmol) and OEGMA (1.5 g, 5 mmol) were added to a 25 mL Schlenk tube equipped with a magnetic stirrer bar. .sup.tBuOH (4.5 mL, 30 wt % wrt. monomer, deoxygenated via N.sub.2 purge) was added and the resulting solution degassed by five cycles of freeze/pump/thaw. After the final thaw cycle, the flask was backfilled with N.sub.2. The reaction flask was placed into a pre-heated oil bath (70° C.) and stirred for 8 hours, after which the reaction medium was observed to be slightly turbid. The polymerization was stopped by cooling the flask to ambient temperature, exposing its contents to air and diluting the reaction medium with .sup.tBuOH. The solution was concentrated by rotary evaporation and precipitated into cold petroleum-ether (40-60) to give a pink solid. The sample was dried under vacuum at 40° C. for 24 hours and analysed by .sup.1H NMR in D.sub.2O and GPC with a mobile phase of DMF.

Linear Polymerisation p(OEGMA-stat-PDMSMA)

(76) In a typical synthesis, targeting DP.sub.n=60 monomer units (OEGMA/PDMSMA 50/50), AIBN (2.7 mg, 0.016 mmol), CPBD (18.4 mg, 0.083 mmol), OEGMA (0.148 g, 0.492 mmol) and PDMSMA (M.sub.n 985 gmol.sup.−1, 1.5 g, 1.524 mmol) were added to a 25 mL Schlenk tube equipped with a magnetic stirrer bar. .sup.tBuOH (4.96 mL, 30 wt % wrt. monomer, deoxygenated via N.sub.2 purge) was added and the resulting solution degassed by five cycles of freeze/pump/thaw. After the final thaw cycle, the flask was backfilled with N.sub.2. The reaction flask was placed into a pre-heated oil bath (70° C.) and stirred for 24 hours, after which the reaction medium was observed to be slightly turbid. The polymerisation was stopped by cooling the flask to ambient temperature, exposing its contents to air and diluting the reaction medium with .sup.tBuOH. The solution was concentrated by rotary evaporation and precipitated into cold MeOH to give a pink liquid. The sample was dried under vacuum at 40° C. for 24 hours and analysed by .sup.1H NMR in CDCl.sub.3 and GPC with a mobile phase of THF.

Branched Polymerisation: p(OEGMA-stat-PDMSMA-stat-PDMSDMA)

(77) In a typical synthesis, targeting DP.sub.n=60 monomer units (OEGMA/PDMSMA 50/50), AIBN (5.6 mg, 0.034 mmol), CPBD (37.5 mg, 0.169 mmol), OEGMA (1.524 g, 5 mmol), PDMSMA (M.sub.n 985 gmol.sup.−1, 5 g, 5 mmol) and PDMSDMA (M.sub.n 1,275 gmol.sup.−1, 0.205 g, 0.158 mmol) were added to a 100 mL Schlenk tube equipped with a magnetic stirrer bar. .sup.tBuOH (20.3 mL, 30 wt % wrt. monomer, deoxygenated via N.sub.2 purge) was added and the resulting solution degassed by five cycles of freeze/pump/thaw. After the final thaw cycle, the flask was backfilled with N.sub.2. The reaction flask was placed into a pre-heated oil bath (70° C.) and stirred for 24 hours, after which the reaction medium was observed to be slightly turbid. The polymerization was stopped by cooling the flask to ambient temperature, exposing its contents to air and diluting the reaction medium with .sup.tBuOH. The solution was concentrated by rotary evaporation and precipitated into MeOH) to give a pink liquid. The sample was dried under vacuum at 40° C. for 24 hours and analysed by .sup.1H NMR spectroscopy in CDCl.sub.3 and GPC with a mobile phase of THF.

CTA Removal from p(PDMS.SUB.(9)48.-stat-OEGMA.SUB.12.)

(78) A ratio of polymer:AIBN=1:20 was used. p(PDMS.sub.(9)48-co-OEGMA.sub.12) (5.3811 g, 0.112 mmol) was dissolved in toluene (73 mL, deoxygenated via Ar purge) in a 100 mL schlenk flask equipped with a stirrer bar. AIBN (369 mg, 2.24 mmol) was added to the reaction flask and purged with Ar. The temperature was raised to 80° C. for 2.5 hours. After the reaction with AIBN, the polymer was precipitated in cold MeOH and a white liquid was isolated by decanting the MeOH. The product was dried in vacuo then analysed by .sup.1H NMR spectroscopy in CDCl.sub.3.

Process

(79) All-trans retinoic acid (atRA) was purchased from Xian Bosheng Biological Technology Co., Ltd. and used as received. Ibuprofen (Ibu) was purchased from Tokyo Chemical Industry UK Ltd. and used as received. Silicone oil (technical grade: 1000; Viscosity=900-1200 cSt at 25° C.; 5000; Viscosity=4800-5500 cSt at 25° C.) was donated by Fluron GmbH and used as received. All deuterated solvents were purchased from Sigma-Aldrich and used as received apart from CDCl.sub.3 where 0.1% TMS was added. All solvents used were analytical grade and purchased from Fisher. Resazurin was purchased from Sigma and used as received. Alexa Fluor® 488 Phalloidin (Phalloidin) and 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) was purchased from Invitrogen and used as received; phalloidin was dissolved in methanol before use, according to the manufacturer's instructions. ARPE-19 cells were bought from American Type Culture Collection, Manassas, Va., USA, catalogue number CRL 2302 and frozen stocks were stored in-house. Dulbecco's Modified Eagle Medium/Ham's Nutrient Mixture F-12 Formulation (DMEM/F12, catalogue number D8437), Penicillin Streptomycin 10 mg/mL streptomycin in 0.9% NaCl (Pen-Strep), Amphotericin B solution 250 μg/mL in deionized water, Dulbecco's calcium and magnesium free phosphate buffered saline (PBS), Trypsin-EDTA containing 5 g porcine trypsin and 2 g ethylenediaminetetraacetic acid (Trypsin) and neutral buffered formalin (NBF) were purchased from Sigma-Aldrich and used as received. Foetal bovine serum (FCS) was purchased from BioSera and used as received. All tissue culture plates were purchased from Greiner, except black 96 well plates which were purchased from Costar. Poly (ethylene glycol) methyl ether methacrylate (M.sub.n=300 gmol.sup.−1) (98%) (OEGMA), 2-cyano-2-propyl benzodithioate (97%) (CPBD) and 2,2′-azobis(2-methylpropionitrile) (98%) (AIBN) were purchased from Sigma-Aldrich and used as received. Mono methacryloxypropyl polydimethylsiloxane methacrylate (molecular weight 985 and 4,600 gmol.sup.−1 PDMSMA.sub.9 and PDMSMA.sub.57 respectively) and methacryloxypropyl polydimethylsiloxane dimethacrylate (molecular weight 1,275 and 4,460 gmol.sup.−1; PDMSDMA.sub.12 and PDMSDMA.sub.55 respectively) were purchased from Gelest and used as received.

Characterization

(80) NMR spectra were recorded using a Bruker DPX-400 spectrometer operating at 400 MHz for .sup.1H NMR and 100 MHz for .sup.13C NMR in CDCl.sub.3. UV-Vis spectra were collected using a Thermo Fisher NanonDrop 2000c spectrophotometer, either with a quartz cuvette or directly with the nanodrop functionality of the equipment depending on the solvent used. Data was analyzed using the NanoDrop2000 software. ProSafe+ scintillation cocktail (Meridian Biotechnologies Ltd.) was used as received. All radiation measurements were carried out using a liquid scintillation counter (Packard Tri-carb 3100TR; Isotech). Triple detection GPC was performed to measure molecular weights and molecular weight distributions using Malvern Viscotek instruments. One instrument was equipped with a GPCmax VE2001 auto-sampler, two Viscotek D6000 columns (and a guard column) and a triple detector array TDA305 (refractive index, light scattering and viscometer) with a mobile phase of DMF containing 0.01 M lithium bromide at 60° C. and a flow-rate of 1 mL min.sup.−1. The second instrument was equipped with a GPCmax VE2001 autosampler, two Viscotek T6000 columns (and a guard column), a refractive index (RI) detector VE3580 and a 270 Dual Detector (light scattering and viscometer) with a mobile phase of THF containing 2 v/v % of trimethylamine at 35° C. with a flow rate of 1 mL min.sup.−1. A Nikon Eclipse Ti-E inverted microscope system was used to collect cell images.

Synthesis of Polymers

(81) All RAFT polymerizations were conducted at a constant ratio of chain transfer agent to initiator of [CPBD]:[AIBN]=1:0.2.

(82) For the synthesis of p(OEGMA), targeting DP.sub.n=60 monomer units, AIBN (2.7 mg, 0.016 mmol), CPBD (18.4 mg, 0.083 mmol) and OEGMA (1.5 g, 5 mmol) were added to a 25 mL Schlenk tube equipped with a magnetic stirrer bar. .sup.t-BuOH (4.5 mL, 30 wt % wrt. monomer, deoxygenated via N.sub.2 purge) was added and the resulting solution degassed by five cycles of freeze/pump/thaw. After the final thaw cycle, the flask was backfilled with N.sub.2. The reaction flask was placed into a pre-heated oil bath (70° C.) and stirred for 8 hours, after which the reaction medium was observed to be slightly turbid. The polymerization was stopped by cooling the flask to ambient temperature, exposing its contents to air and diluting the reaction medium with .sup.t-BuOH. The solution was concentrated by rotary evaporation and precipitated into cold petroleum-ether (40-60) to give a pink liquid. The sample was dried under vacuum at 40° C. for 24 hours and analysed by .sup.1H NMR in D.sub.2O and GPC with a mobile phase of DMF.

(83) In a typical synthesis of p(OEGMA-stat-PDMSMA.sub.9), targeting DP.sub.n=60 monomer units (OEGMA/PDMSMA.sub.9 50/50), AIBN (2.7 mg, 0.016 mmol), CPBD (18.4 mg, 0.083 mmol), OEGMA (0.148 g, 0.492 mmol) and PDMSMA.sub.9 (M.sub.n=985 gmol.sup.−1, 1.5 g, 1.524 mmol) were added to a 25 mL Schlenk tube equipped with a magnetic stirrer bar. .sup.t-BuOH (4.96 mL, 30 wt % wrt. monomer, deoxygenated via N.sub.2 purge) was added and the resulting solution degassed by five cycles of freeze/pump/thaw. After the final thaw cycle, the flask was backfilled with N.sub.2. The reaction flask was placed into a pre-heated oil bath (70° C.) and stirred for 24 hours, after which the reaction medium was observed to be slightly turbid. The polymerization was stopped by cooling the flask to ambient temperature, exposing its contents to air and diluting the reaction medium with .sup.t-BuOH. The solution was concentrated by rotary evaporation and precipitated into cold MeOH to give a pink liquid. The sample was dried under vacuum at 40° C. for 24 hours and analyzed by .sup.1H NMR in CDCl.sub.3 and GPC with a mobile phase of THF.

(84) In a typical branched polymerization synthesis of p(OEGMA-stat-PDMSMA.sub.9-stat-PDMSDMA.sub.12), targeting DP.sub.n=60 monomer units (OEGMA/PDMSMA.sub.9 50/50), AIBN (5.6 mg, 0.034 mmol), CPBD (37.5 mg, 0.169 mmol), OEGMA (1.524 g, 5 mmol), PDMSMA.sub.9 (M.sub.n=985 gmol.sup.−1, 5 g, 5 mmol) and PDMSDMA.sub.12 (M.sub.n=1,275 gmol.sup.−1, 0.205 g, 0.158 mmol) were added to a 100 mL Schlenk tube equipped with a magnetic stirrer bar. .sup.t-BuOH (20.3 mL, 30 wt % wrt. monomer, deoxygenated via N.sub.2 purge) was added and the resulting solution degassed by five cycles of freeze/pump/thaw. After the final thaw cycle, the flask was backfilled with N.sub.2. The reaction flask was placed into a pre-heated oil bath (70° C.) and stirred for 24 hours, after which the reaction medium was observed to be slightly turbid. The polymerization was stopped by cooling the flask to ambient temperature, exposing its contents to air and diluting the reaction medium with .sup.t-BuOH. The solution was concentrated by rotary evaporation and precipitated into MeOH) to give a pink liquid. The sample was dried under vacuum at 40° C. for 24 hours and analyzed by .sup.1H NMR spectroscopy in CDCl.sub.3 and GPC with a mobile phase of THF.

(85) CTA removal involved a ratio of polymer:AIBN=1:20. In a typical CTA removal p(PDMS.sub.(9)48-stat-OEGMA.sub.12) (5.3811 g, 0.112 mmol) was dissolved in toluene (73 mL, deoxygenated via Ar purge) in a 100 mL schlenk flask equipped with a stirrer bar. AIBN (369 mg, 2.24 mmol) was added to the reaction flask and purged with Ar. The temperature was raised to 80° C. for 2.5 hours. After the reaction with AIBN, the polymer was precipitated in cold MeOH and a white liquid was isolated by decanting the MeOH. The product was dried in vacuo then analyzed by .sup.1H NMR spectroscopy in CDCl.sub.3.

(86) In a typical solubilization experiment, polymer (1 mL) and SiO.sub.1000 (1 mL) were syringed into a glass vial to create a 50 v/v % mixture and placed on a roller for 3 days. The solutions were diluted systematically by adding SiO.sub.1000 to decrease the amount of polymer by 10 v/v %, rolled for 3 days each time, until a soluble concentration was reached (i.e. 40, 30, 20, 10 also 5 and 1 v/v % were tested).

Radiometric Studies and Analysis of Drug Solubility in Silicone Oils

(87) To determine solubility of drugs in silicone oil saturated solutions of atRA and Ibu in silicone oil were prepared by mixing atRA (11.6 mg) or Ibu (32 mg) with tritiated versions of the drug (10 μCi) in EtOH (2 mL); after evaporation of the solvent at ambient temperature, SiO.sub.1000 (5 mL) was added to the residual solid and the solution was stirred for 2 weeks. The sample was filtered using a syringe pump (4 mL/h) and 0.45 μm PTFE filters. Samples of the filtered oils (20 μL) were then solubilized in diethyl ether (8 mL) before scintillation cocktail (10 mL) was added. Radiation was then measured on a scintillation counter and saturation concentrations were determined.

(88) Amounts of drug added to the samples were altered depending on targeted final concentrations. Solutions of 200 μg/mL: atRA (1 mg) and tritiated atRA (6 μCi) were mixed in EtOH (2 mL) and the same protocol was followed. Solutions of 20 μg/mL: atRA (0.1 mg) and tritiated atRA (2 μCi) were mixed in EtOH (2 mL) and the same procedure was carried out. Solutions of 1 mg/mL: Ibu (5 mg) and tritiated Ibu (7.5 μCi) were mixed in EtOH (2 mL) and the same protocol was followed.

(89) For the preparation of both atRA and Ibu solutions in p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) blends, 5 and 10% vol content of polymer blends were prepared by mixing for 3 days and loaded with drug by following the same protocols as described above.

Cytotoxicity Assays of Drug Compounds and Polymers

(90) Cells were cultured at 37° C. in a dark, humid 5% CO.sub.2 incubator; media containing 1% Pen-Strep, 1% Amphotericin B and supplemented with 10% FCS was used. For these studies, cells were used between passages 22 and 25. Multiple assays were carried out on ARPE-19 cells to study cytotoxicity and the effects of different drug concentrations and blends of p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) in SiO (silicon oil). 18,000 cells/well were seeded in a 48 well tissue culture plate and left for 1 or 7 days to adhere to the plate. The 7 day samples were fed once within the week by replacing 450 μL old medium with 500 μL fresh culture medium. After the predetermined time period, the media was aspirated from all wells and replaced with 0.6 mL media containing drug, fresh media with p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) blends (0.2 mL) over the top or conditions required for controls. Controls included: media, SiO (0.2 mL) and a negative control (20% DMSO). Cells were then incubated for 1 to 7 days before the following assays could be performed.

(91) Sterile resazurin solution was added to wells at a concentration of 10 μg/mL Plates were incubated in the dark at 37° C. for 4 hours. Media was removed and put in black 96-well plastic plates; resorufin fluorescence was read using a Fluostar Optima spectrofluorometer (λ.sub.Excitation=530 nm; λ.sub.Emission=590 nm). All values were normalised to negative control wells on each plate.

(92) Following removal of resazurin solution, cells were washed with PBS (500 μL) then fixed for 10 minutes in 10% neutral buffered formalin (NBF; 10% formalin, approximately 4% formaldehyde). NBF was discarded and cells stained with phalloidin (6.667 μg/mL) for 30 minutes at 4° C. Cells were washed with PBS then counterstained with DAPI (0.01 μg/mL) for 10 minutes. Cells were imaged using a Zeiss Axiovert 400 microscope.

(93) Statistical analyses were carried out on SPSS Statistics V22 software; one way test of homogeneity of variances and ANOVA as well as Dunnett's T3 post-hoc evaluation were conducted, p<0.05 was regarded to be statistically significant.

Radiometric Studies of Drug Release

(94) 1 mL of silicone oil or p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) blend (5 or 10%) with determined concentration of drug was placed in a 24 well plate over 0.5 mL media. Samples of media (0.5 mL) were taken and replaced at determined time intervals; daily for the first critical week then every 2-3 days for the remainder of the study, using a 1 mL syringe and 25 gauge needle for up to 71 days. Sampling and withdrawing of the media was done very carefully in order to avoid any emulsification of the oil. Media (250 μL) was mixed with scintillation cocktail (10 mL) and analyzed by liquid scintillation counting.

Cytotoxicity of p(PDMSMA(9)48-stat-OEGMA12) with ARPE-19 Cells

(95) The metabolic activity and morphology of ARPE-19 cells was studied for pre- and post-confluent cells (1 day and 7 day growth respectively) which were exposed to silicone oil (SiO.sub.1000) and blends of p(PDMSMA.sub.(9)48-stat-OEGMA.sub.12) with silicone oil (SiO.sub.1000) at 10% (v/v) for 1 and 7 days. A resazurin assay was carried out, followed by phalloidin staining of the cells from the assay.

(96) The resazurin assay contained negative (healthy cells) and positive controls (cells exposed to 20% DMSO) as well as media with no resazurin present to determine background signals. As seen in FIG. 7, these controls confirmed the validity of the assay and statistical analyses were carried out on SPSS Statistics V22 software; one way test of homogeneity of variances and ANOVA as well as Dunnett's T3 post-hoc evaluation were conducted, p<0.05 was regarded to be statistically significant. There was a significant difference between the negative control and the positive control, however, no significant difference was observed between the positive control and cells exposed to silicone oil (SiO.sub.1000), p(PDMSMA(9)48-stat-OEGMA12) blends at 10% v/v in silicone oil (SiO.sub.1000), indicating that the oil and blends have no cytotoxic effect on the ARPE-19 cells.

(97) Phalloidin staining confirmed no cytotoxic effects, as evidenced by the presence of healthy cytoskeletons when cells were exposed to the oil and blend. Images of phalloidin stained cells exposed to silicone oil (SiO.sub.1000) and the 10% blends, are presented in FIG. 8, alongside negative control (healthy cells) for the two extreme time points examined. All images represent healthy cytoskeletons.