PEROXIDE REMOVAL FROM DRUG DELIVERY VEHICLE

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. provisional application serial number

60/702,546, filed July 26, 2005, and U.S. application serial number , filed July 24,

2006, each of which is incorporated herein in its entirety.

FIELD OF THE INVENTION

[0002] The present invention relates to methods for reducing peroxide levels in non- polymeric preparations and to compositions used in and prepared by such methods.

BACKGROUND OF THE INVENTION

[0003] Sucrose acetate isobutyrate ("SAIB") is a hydrophobic liquid with limited water solubility. It is soluble in a large number of biocompatible solvents. SAIB has an unusual property - it undergoes a dramatic change in viscosity with small additions of heat or with the

addition of solvents. It is a very viscous liquid, having a viscosity of approximately 3200 poise at 37° C. SAIB is produced by the controlled esterification of natural sugar (sucrose) with acetic and isobutyiϊc anhydrides. SAEB metabolizes to sucrose, acetic acid and isobutyric acid.

[0004] SAIB is orally non-toxic and is currently used to stabilize emulsions in the food industry. In one example, SAIB is commonly found in the beverage industry, where it is used as a weighting agent to help stabilize the final beverage formula. Also, SAD3 has been reported to be useful as a gelling system-type drug excipient that allows for sustained or controlled release of drugs. When in solution or in an emulsion, SAIB can be applied via injection or an aerosol spray. SAEB is compatible with cellulose esters and other polymers that can affect the rate of delivery of the substance, hi one example, SAIB is the main ingredient for the SABER drug delivery system, which also consists of a pharmaceutically acceptable solvent.

[0005] Drug delivery systems, including SAEB delivery systems, are still confronted by various issues of drug instability, as such systems are considered for longer and longer drug delivery durations. Drug instability can occur via a number of factors, including denaturation, precipitation, oxidation, aggregation, and others, hi particular, a number of excipients used to facilitate delivery and release of drugs have peroxides or are susceptible to the formation of peroxides, which may lead to oxidation of active ingredient in the formulation, hi the example of SAEB, the presence of peroxides is deleterious to a drug incorporated in an SAEB drug formulation as the drug is likely to undergo oxidative degradation. Thus, in order to formulate any drug formulation based on SAIB that provides enough of a stable environment to facilitate the delivery of a drug, the peroxide levels must be reduced.

[0006] There is no known process for removal of peroxides from SAIB at present, despite availability of processes for the removal of peroxides from other materials such as

polymers. Therefore, there still remains a need for a drug formulation of SAIB having improved properties to reduce the degradation of the drug therein.

SUMMARY OF THE INVENTION

[0007] An aspect of the present invention comprises methods of treating sucrose acetate isobutyrate (SAIB) formulations to be used as drug delivery vehicles comprising adding to the formulations an amount of bisulfite salt effective to substantially remove peroxides, the bisulfite salt comprising sodium metabisulfite, potassium metabisulfite, sodium bisulfite, or potassium bisulfite, or a mixture thereof.

[0008] In another aspect of the present invention, provided are drug delivery vehicles adapted to provide prolonged stability of a drug that is to be delivered in vivo comprising sucrose acetate isobutyrate having substantially reduced levels of peroxide, the drug delivery vehicle being treated with an amount of bisulfite salt effective to substantially reduce levels of peroxide in said drug delivery vehicle, the bisulfite salt comprising sodium metabisulfite, potassium metabisulfite, sodium bisulfite, or potassium bisulfite, or a combination thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The invention is illustrated by way of example and is not intended to be limited by the accompanying figures.

[0010] Figure 1 illustrates a bar graph of the results of Study I - Stability of omega- interferon in untreated SAIB.

[0011] Figure 2 illustrates a bar graph of the results of Study Ha - Stability of omega- interferon in alumina treated SAIB.

[0012] Figure 3 illustrates a bar graph of the results of Study lib - Stability of omega- interferon in alumina treated SAIB.

[0013] Figure 4 illustrates a bar graph of the results of Study III - Stability of omega- interferon in untreated SAIB.

[0014] Figure 5 illustrates a bar graph of the results of Study VIb - Stability of omega- iuterferon in untreated SAIB.

[0015] Figure 6 illustrates a bar graph of the results of Study Via - Stability of omega- interferon in sodium metabisulfite treated SAIB.

[0016] Figure 7 illustrates a bar graph that provides comparisons of oxidation of omega-IFN in sodium metabisulfite treated and untreated SAIB.

[0017] Figure 8 illustrates an osmotically pump-driven implantable device, Duros® being an example, that facilitates in vivo delivery of an active agent in an SAIB vehicle.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0018] In an aspect of the present invention, provided are methods of treating sucrose acetate isobutyrate formulations (SABB) that are to be used as drug delivery vehicles comprising adding an amount of a bisulfite salt effective for substantially removing peroxide from the formulations, the bisulfite salt comprising sodium metabisulfite, potassium metabisulfite, sodium bisulfite, or potassium bisulfite, or a combination thereof. Preferably, the bisulfite salt is sodium metabisulfite. A ratio ranging from about 1:1 to about 1:4 (weight: volume) SAIB:aqueous solution of bisulfite salt ("aqueous bisulfite salt") can be used. Preferably, the bisulfite salt is a metabisulfite salt, hi some embodiments, the bisulfite salt is preferably sodium metabisulfite. Preferably, the ratio of the aqueous bisulfite salt to SAIB is 1:1. In one example, to purify 1 kg of SAJB, a volume of sodium metabisulfite solution can be made up to 1 liter, and an approximate proportion of 1 : 1 of S AIB :aqueous sodium metabisulfite was used. The aqueous bisulfite salt in SAIB can be from about 0.1 % weight to volume of water (w/v) to about 50 %

w/v; preferably, from about 0.5 % w/v to about 30 % w/v. In some embodiments, the aqueous bisulfite salt is preferably from about 1 % w/v to about 15 % w/v. In some embodiments, the aqueous bisulfite salt is about 5 % w/v solution in water.

[0019] The method removes peroxide to a level that is at least less than 50 % of the levels before the method, or starting levels, and, preferably, less than 20 % of the starting levels. In some embodiments, peroxide is removed to less than 10 % of the starting levels. While in some embodiments, the method removes peroxide to a level that is less than 5 % of the starting levels. Furthermore, the method can remove peroxide so that the resulting SAIB formulation contains peroxide in amounts less than 20 ppm, and, preferably, less than 10 ppm. In some embodiments, the method removes peroxide to result in an SAIB formulation containing less than 5 ppm. In some embodiments, the resulting SAIB formulation from this method can serve as a drug delivery vehicle for use with a medical delivery device, including a drag eluting stent, a catheter, or other drug delivery implants. In one example, the SAIB formulation can be loaded into an osmotically pump-driven implantable device of the type disclosed in U.S. Patent No. 6,395,292, for example. Preferably, the osmotically pump-driven implantable device is a Duros® device (Alza Corporation, Mountain View, California), hi other embodiments, the SAIB formulation can serve as a drug depot for drug delivery.

[0020] hi some embodiments, the step of adding the bisulfite salt comprises mixing a solution of the bisulfite salt with the sucrose acetate isobutyrate formulation. The SATB formulation can be further comprised of a cosolvent, which can be selected from a number of solvents including pharmaceutically acceptable solvents, e.g., hexane, ethyl acetate, ethanol, benzyl benzoate, N-methyl pyrrolidone, and iso-propyl alcohol, among others. Preferably, the cosolvent is hexane or ethyl acetate, hi some embodiments, the methods further comprise vacuum treating the formulation to remove the cosolvent. Also, some embodiments comprise

the additional step of removing bisulfite salt from the formulation. This removal step comprises washing the formulation with water to remove the bisulfite salt, hi the embodiments that incorporate the washing step, a further step of drying the formulation over magnesium sulfate can be utilized to remove the water. Alternatively, calcium chloride anhydrous, calcium sulfate anhydrous, activated silica gel, phosphorous pentoxide, or drying under vacuum, or a combination thereof can be used to also remove the water, hi alternative embodiments, glycerin can be used to wash the bisulfite-added formulation to remove the bisulfite salt. Afterwards, residual glycerin can be removed by washing with water and then drying to remove water.

[0021] hi some aspects of the present invention, the methods of substantially removing peroxide from a sucrose acetate isobutyrate formulation (SAIB) comprising the steps of adding the aqueous bisulfite salt, washing the formulation, and drying the formulation are repeated at least once. The steps can be repeated to further reduce the levels of peroxide in the SAEB formulation.

[0022] hi another aspect, the present invention includes a drug delivery vehicle comprising SAD3 that provides for prolonged stability of a drug that is to be delivered by maintaining substantially reduced levels of peroxide, the drug delivery vehicle being treated with sodium metabisulfite. The prolonged stability comprises reduced oxidation, deamidation, or aggregation, e.g., dinierization, of the drug over extended periods of time in which drug is within environment of delivery vehicle. Preferably the prolonged stability is reduced oxidation. The extended periods of time can be periods from about one week to a few months, and up to about a year. Preferably, the prolonged stability is evidenced by significant improvements in oxidation, deamidation, or aggregation levels of the drug when the delivery vehicle has been treated with a bisulfite salt versus untreated delivery vehicle, hi some preferred embodiments, the prolonged stability is characterized as about 50 % less oxidation, about 33 % less deamidation, or about 75

% less dimerization as compared to untreated delivery vehicles. The drug can be selected from any known and desired biomolecular material that can act as therapeutics and other therapeutic active agents that are susceptible to oxidative degradation. As it is used herein, the term "biomolecular material" refers to peptides, polypeptides, proteins, nucleic acids, viruses, antibodies, small molecules susceptible to oxidation, and any other naturally derived, synthetically produced, or recombinantly produced active agent that includes nucleic or amino acid. In some embodiments, for example, drags can be selected from among the following: a steroid, NSAIDS, peptides, proteins such as growth factors or hormones, anti-tumor agents, antibiotics, analgesics, local anesthetics, antiviral agents, antipsychotics, anticoagulants, oligonucleotides for gene therapy, active small molecules, and others.

[0023] As used herein, the term "removing" and all variations thereof, refer to decreasing by any measurable degree the level of peroxide present in a drag formulation. The term "substantially removing" is used herein to describe a dramatic decrease in the level of peroxide present in a drag formulation, such as SAIB formulation. The dramatic decrease is at least 50 % of original levels (levels before treatment) and in some instances is 10 % of original levels. In preferred aspects of the present invention, the "substantial removal" describes a decrease to less than 5 % of original levels.

[0024] As used herein, the term "drag delivery vehicle" or "delivery vehicle" refers to a formulation that is biocompatible and used to carry a drag without reacting with the same drug. Also, the vehicle does not alter or minimally alters the activity of the drag. Furthermore, the vehicle allows for the transport of the drug in vivo and eventual delivery of the drug to a biological site for therapeutic effect.

[0025] As used herein, the term "prolonged stability" is used to refer to the stabilizing effect of the drug delivery vehicles of the present invention on the carried drug. Prolonged

stability can be evidenced by significant improvements in oxidation, deamidation, or aggregation of the drug over extended periods of time.

Examples

[0026] Different approaches were investigated for removal of peroxides from SATR, as indicated in Table 1.

Preparation of Suspension

[0027] Each of the experiments involved protein particles consisting of omega- interferon, which were suspended in SATB at a particle loading of either 4 % or 10 % by weight. The suspensions were prepared in a dry box under nitrogen at 45°'C. The suspension was mixed for 15 minutes while maintaining the temperature. Suspension mixing was performed by hand. Aliquots from the prepared suspensions were transferred to clear crimp-top glass vials and sealed under nitrogen. Each aliquot contained at least six milligrams of protein to allow for stability testing in triplicate. These samples were stored in an oven at 40° C. Samples were withdrawn at regular intervals (as indicated in Table 1) and analyzed for omega-interferon content and purity was assessed using reverse phase HPLC and size exclusion chromatography.

Size Exclusion Chromatography

[0028] Size exclusion chromatography (SEC) was used to monitor the omega- interferon content and purity in the formulations. The percentages of monomer and dimer in the formulation were quantified using SEC. The stability of omega-interferon was judged by using a stability indicating chromatographic technique based on reverse phase HPLC (rp-HPLC). This technique was used to monitor the oxidation, deamidation and formation of an unknown species of omega-interferon in the formulations. The peroxide content of the vehicle was determined

using EP 2002, 2.5.5 (Method A with auto titration). See Extra Pharmacopoeia. 2002 Ed. Content and purity assay of omega-interferon by size exclusion chromatography (SEC).

Reverse Phase High Performance Liquid Chromatography

[0029] Purity assay and identity of omega-interferon recombinant in suspension systems by reverse phase high performance liquid chromatography (rp-HPLC).

[0030] The stability of omega-interferon was monitored in two different lots of untreated SAEB (as received) and in treated SAD3 (removal of peroxides), when treatment was applied.

[0031] The studies are outlined below: . Study I: Stability in untreated SAIB (lot #TD1030507) for 2 weeks

. Study Ha: Treatment of SAB3 (lot #TD1030507) with neutral alumina by heating and stability in this treated SAIB for 4 weeks • Study lib: Treatment of SAIB (lot #TD1030507) with neutral alumina in presence of ethanol and stability in this treated SAIB for 4 weeks

. Study in: Stability in untreated SAIB (lot #TD2032663) for 2 weeks . Study IV: Treatment of SAIB (lot #TD2032663) with basic alumina by heating . Study V: Treatment of SAIB (lot #TD2032663) with 10 % aqueous methionine solution by heating . Study Via: Treatment of SAIB (lot #TD2032663) with 5 % aqueous solution of sodium metabisulfite and stability in treated S AIB for 8 weeks . Study VIb: Stability in untreated SAIB (lot #TD2032663) for 8 weeks

Table 1. Details about stability studies of omega-interferon in SAIB

Study SAIB Treatment Particle Time Tests

# (Lot #) loading points

I TD1030507 Untreated 4% 0, 4, 7, SEC, RP-HPLC 14 days

Ha TD1030507 Treated with neutral 10% 0 2 SEC, RP-HPLC alumina by heating 4 weeks

lib TD1030507 Treated with neutral 10% 0, 2, SEC, RP-HPLC alumina using ethanol 4 weeks

III TD2032663 Untreated 10% 0, 1, SEC, RP-HPLC 2 weeks

IV TD2032663 Treated with basic NA NA NA alumina by heating

V TD2032663 Treated with 10 % NA NA NA aqueous solution of methionine

Via TD2032663 Treated with hexane 10% SEC, RP-HPLC and 0, 1, 2, 4, sodium metabisulfite 8 weeks

VIb TD2032663 Untreated 10% 0, 1, 2, 4, SEC, RP-HPLC 8 weeks

Materials and Equipment

[0032] The following tables, Table 2 and Table 3, provide a list of materials and equipment that can be utilized to perform the experiments described, below.

Table 2. List of materials

Materials

Spray dried omega-interferon particles

SAIB, Eastman Chemical Company

Aluminum oxide (Powder)

Ethanol, absolute, 200 proof, AAPER

Aluminum oxide, basic, standard activity I, 50 -200 μm, Sorbent Technologies

Aluminum oxide, basic, Super I, 50 -200 μm, Sorbent Technologies

Methionine, USP, Ph Eur, JP

Table 3. List of equipment

Example 1

Study I: Stability in untreated SAIB (lot #TD1030507) for 2 weeks

Table 4. Stabilit of ome a-interferon in untreated SAB lot #: 1030507 - Stud I

"sampled by scraping container walls, so values might not be representative of the bulk

ND = Not detected,

** standard deviation in parenthesis; *** protein particles - 1 =0 for suspension

[0033] The preliminary stability study of omega interferon in untreated SAIB ( lot # TD1030507, peroxide value - 71.4 ppm) was over 2 weeks. The results indicated that up to 8.31% of omega-interferon was oxidized in two weeks, which corresponds to an increase of 5.51% with respect to particles (2.8% oxidation at t=0). See Table 4, Figure 1. Furthermore, a small increase occurred in the percentage of deamidated form (+ 0.83%) of omega-interferon and the dimer (+ 0.58%). The high level of oxidation can be attributed to the high peroxide content of SAEB.

Example 2

Study Ha and lib: Stability of SAIB (lot #TD1030507) treated with neutral alumina with heating or neutral alumina in presence of ethanol for 4 weeks

Treatment of SAIB with neutral alumina with heating

[0034] SAIB was heated to 75° C. Alumina (15 % w/w) was added to the heated SAIB. The mixture was stirred for 40 minutes and filtered though a 5.0 μm filter at 75° C. The treated SAEB was then collected, sampled for peroxide testing, and used for preparation of suspension for stability testing.

Treatment of SAEB with neutral alumina in presence of ethanol

[0035] SAEB was mixed with 15% absolute ethanol to reduce the viscosity. Basic alumina (15 % w/w) was added to the SAEB containing ethanol. The resulting mixture was stirred for 1 hour and filtered though a 0.2 μm filter. The filtered SAEB was placed overnight under vacuum at 60° C to remove the ethanol. This treated SAEB was then collected, sampled for peroxide testing, and used for preparation of suspension for stability testing.

Table 5. Stability of omega-interferon in alumina treated SAIB ((lot #: 1030507) - Studies Ha and lib

SAIB treated with neutral alumina b heatin - Study Ha

SAIB treated with neutral alumina using ethanol - Study lib

**standard deviation in parenthesis

[0036] The stability of omega-interferon in alumina treated SAEB was tested. After one month in the neutral alumina treated SABB (Study Ha and lib), oxidation of omega-interferon increased by about 5.7% for both Ha and lib. This indicates that alumina treatment of SAD3 did not improve the stability of omega-interferon in SAJB. See Table 5. In addition, this analysis is also reflected in the high peroxide content of alumina treated SACB (66.3 and 62.9 ppm, respectively). Treatment with neutral alumina was not effective in decreasing peroxide content.

Example 3

Study III: Stability in untreated SAIB (lot #TD2032663) for 2 weeks

Table 6. Stabilit of omega-interferon in untreated SAIB ((lot #: 2032663) - Study III

ND = Not detected

*n = 6

** standard deviation in parenthesis

[0037] Stability of omega-interferon in untreated SAIB was again tested. The results of a two week stability study (Study III) of omega-interferon in SAIB (lot # TD 2032663) are comparable to studies I and II. See Table 6, Figure 4. The amount of oxidation was found to have increased by 5.58%, while deamidation increased by 0.39% and dimerization increased by 0.42%.

Example 4

Study IV and V: Treatment of SAIB (lot #TD2032663) with basic alumina with heating or with 10 % aqueous methionine solution

Treatment of SAIB with basic alumina with heating

[0038] SAIB was heated to 90° C. Basic alumina (15 % w/w) was added to the heated SAIB. Two different grades of alumina were used - Basic Super I and Basic Standard Activity I. The resulting mixture was stirred for 40 minutes. The mixture was then centrifuged at 4000 rpm while temperature was maintained at 75° C. After centrifugation, the supernatant was collected and sampled for peroxide analysis.

Treatment of SAIB with 10% aqueous solution of methionine

[0039] One part of SAIB was vigorously agitated with 4 parts of 10 % aqueous solution of methionine at 80° C for 45 minutes using a magnetic stirrer. (Evaporated water was replenished as necessary) Afterwards, the methionine solution was decanted. SAEB was then washed with 4 parts of water by agitating for 15 minutes at 70° - 80° C. This washing step was carried out three times. SAIB was placed overnight in vacuum oven at 70° C to remove residual water, and, afterwards, was sampled for peroxide analysis.

[0040] The peroxide content of S AD3 treated with basic alumina or with aqueous methionine solution was determined to be 109.3 and 95.7 respectively (Study IV and V), indicating that these approaches were not successful in the removal of peroxides. See Figure 7.

Example 5

Study Via and VIb: Stability of SAIB (lot #TD2032663) treated with 5 % aqueous solution of sodium metabisulfite or untreated for 8 weeks

Treatment of S AIB with 5 % aqueous solution of sodium metabisulfite in presence of hexane

[0041] SAIB was dissolved in two parts of hexane. The resulting solution was treated with a 5 % aqueous solution of sodium metabisulfite by vigorous shaking. The aqueous layer was removed and the SAIB layer was washed with water. The SAIB layer was dried with MgSO ₄ . Hexane was removed from SAD3 by evaporation under vacuum at 50° C. The treated SAD3 was sampled for peroxide analysis and used for preparation of suspension for stability testing.

Table 7. Stability of omega-interferon in untreated SATS and treated SAIB - Study Via and VIb Stabilit of ome a-IFN in Untreated SAIB Lot: TD 2032663)

% omega-IFN 88.91 (0.39) 87.29 (0.25) 83.10 (0.08) 81.62 80.17 79.35 %Oxidized 1.90 (0.39) 3.38 (0.19) 7.86 (0.14) 8.54 (0.07) 8.94 (0.08) 8.86 (0.06)

% Deamidated 2.02 (0.01) 2.15 (0.03) 2.24 (0.09) 2.59 (0.15) 3.33 (0.04) 4.46 (0.07) % Unlαiown 7.17 (0.44) 7.18 (0.47) 6.80 (0.02) 7.25 (0.36) 7.55 (0.05) 7.33 (0.47)

Note: The omega content in the suspension was 1.00 % and not 1.66% because the particles contained 11.45% omega and the loading of particles in suspension was at 10%

Stability of omega-IFN in Treated SAIB (Lot: TD 2032663)

Analysis by SEC (n=3)**

Initial (t=0) Protein particles Initial (t=0) 1 week 2 weeks 4 weeks 8 weeks

AR 48219 AR 48445 AR48441 AR 48440 AR 50132 AR 50161

% Monomer 99.67 (0.01) 99.59 (0.02) 99.34 (0.02) 99.41 99.42 99.00 % Dimer 0.25 (0.01) 0.35 (0.02) 0.53 (0.02) 0.54 (0.02) 0.29 (0.01) 0.94 (0.06 Unknown 0.08 (0.00) 0.05 (0.00) 0.13 (0.01) 0.05 (0.01) 0.29 (0.01) 0.06 (0.01

Note: The omega content in the suspension was 1.17 % and not 1.66% because the particles contained 11.45% omega and the loading of particles in suspension was at 10%. ** standard deviation in parenthesis

[0042] The stability study (Study Via and VIb, Table 7, Figures 5 - 7) conducted in treated (5 % aqueous solution of sodium metabisulfite) and untreated SAIB shows that oxidation levels are reduced at 8 weeks, along with the reduction of peroxide levels - 4.16 % in treated SAIB versus 8.86 % in untreated SAIB equivalent to a change of 2.26 % and 6.96 %, respectively, from t=0 values of the protein particles. (For all relative changes reported herein,

the changes are based on differences between the percentage values, e.g., percent oxidation, at t _n and t =0 of the particles as opposed to relative percent change from value at t=0). Deamidation increased by 2.44 % and 1.62 % in untreated and treated SAIB, respectively. Dimerization increased by 2.28 % and 0.59 % in untreated and treated SAIB%, respectively. The quantities of unknown did not change significantly over time, which indicates that the extent of oxidation, deamidation and dimerization in treated SAIB (low peroxide value of 2.6 ppm) was lower than in untreated material. This treatment decreased the peroxide content substantially.

Table 8. Peroxide content of SAEB

Study # SAIB Treatment Peroxide AR (Lot #) value (ppm)* numbers

I TD1030507 Untreated 71.4 48557

Ha TD1030507 Treated with neutral 66.3 48568 alumina by heating

lib TD1030507 Treated with neutral 62.9 48568 alumina using ethanol

III TD2032663 Untreated 115.9 48581 rv TD2032663 Treated with basic 109.3 48581 alumina by heating

V TD2032663 Treated with 10% aqueous 95.7 48446 solution of methionine

Via TD2032663 Treated with hexane and 2.6 49648 sodium inetabisulfite

VIb TD2032663 Untreated 115.9** 48581

*oxidative activity equivalent to hydrogen peroxide (n =1) **peroxide content determined during Study III

[0043] As shown in Figure 7, along with data provided in Table 8, treatment with an aqueous solution of sodium metabisulfite was effective in significantly reducing peroxide levels from 115.9 ppm to 2.6 ppm - almost 45 times, or 45 fold decrease. In comparison, treatment with neutral alumina, either with heat or with ethanol, resulting in only a nominal change in peroxide levels - a 7 % or l2 % decrease, respectively. In addition, treatment with basic alumina with heat or 10% aqueous methionine only resulted in nominal change in peroxide levels - a 6 % or 18 % decrease, respectively.

[0044] Figure 8 illustrates an osmotically pump-driven implantable device for delivering an SAIB formulation acting as a drug delivery vehicle, active agent within. Depicted in Fig. 8 is an osmotically pump-driven implantable device 10 shown comprising an impermeable reservoir 12. The reservoir 12 is divided into two chambers by a piston 16. The first chamber 18 is adapted to contain an SAIB formulation 19 containing an active agent 20 and the second chamber 21 is adapted to contain a fluid-imbibing agent. A back-diffusion regulating outlet 22 is inserted into the open end of the first chamber 18 and a semipermeable membrane 24 encloses the open end of the second chamber 21. The piston 16 is driven towards the open end of the first chamber 18 by the osmotic pressure generated by the fluid-imbibing agent in the second chamber 21. The pressure created by the piston 16 can force the contents of the first chamber 18 out the opening, i.e., the SAIB formulation 19 comprising active agents 20. The release rate of the active agent can be governed by the osmotic pumping rate.

[0045] It is to be appreciated that certain features of the invention which are, for clarity, described above in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention that are, for brevity, described in the context of a single embodiment, may also be provided separately or in any

subcombination. Further, reference to values stated in ranges includes each and every value within that range, unless clearly expressed otherwise.

[0046] The entire disclosure of each patent, patent application, and publication cited or described in this document is incorporated herein by reference.