NZ729479B2 - Crystallization process of aripiprazole derivatives in extended release formulations for treatment of schizophrenia - Google Patents

Abstract

Processes for providing depot injections of recrystallized aripiprazole lauroxil in which particles of the aripiprazole lauroxil have a surface area of about 0.50 to about 3.3 m 2/g; and crystals of aripiprazole lauroxil produced by such processes. The present invention provides a process for making a compound of Formula (A) in crystal form wherein Ra is CH20C(O)R and wherein R1 is a substituted or unsubstituted aliphatic moiety. a compound of Formula (A) in crystal form wherein Ra is CH20C(O)R and wherein R1 is a substituted or unsubstituted aliphatic moiety.

Description

LLIZATION PROCESS OF ARIPIPRAZOLE DERIVATIVES IN EXTENDED RELEASE FORMULATIONS FOR TREATMENT OF SCHIZOPHRENIA RELATED APPLICATIONS This application claims priority to US. Provisional ation Serial No. 62/041 ,341, filed on August 25, 2014, the content of which is incorporated herein by reference in its entirety.

FIELD OF THEINVENTION The present invention is directed to the preparation of crystalline forms of aripiprazole derivatives including aripiprazole lauroxil and aripiprazole cavoxil. More particularly, the present invention is directed to lling the recrystallization of aripiprazole lauroxil and aripiprazole cavoxil to produce particles useful in extended e injectable formulations for the treatment of schizophrenia and other psychiatric conditions.

RELATEDART Aripiprazole is an atypical antipsychotic drug used in the treatment of phrenia and other psychiatric conditions, such as r disorder and major depressive disorder. razole, which is a ne D2 and serotonin 5-HT1A receptor t, and an antagonist of the serotonin 5-HT2A receptor, has been formulated as a tablet and as a solution, both for oral administration. However, concerns with patient compliance with oral antipsychotics have been reported, and other methods of delivering ychotics, such as intramuscular or subcutaneous injection, have been developed.

ABILIFY®, which is a drug containing aripiprazole as the active agent, is available from Otsuka as an oral tablet (aripiprazole dosage of 2 mg, 5 mg, 10 mg, 15 mg, 20 mg, or 30 mg), an orally disintegrating tablet (dosage of 10 mg or 15 mg), an oral solution (dosage of 1 mg/mL), and as an injection for intramuscular use (9.75 mg/1.3 mL in a single-dose vial). Y® is indicated for schizophrenia, bipolar I disorder, adjunctive treatment of major depressive disorder, irritability associated with autistic disorder, and agitation associated with schizophrenia or bipolar mania. Abilify Maintena® is an extended release inj ectable suspension of aripiprazole available from Otsuka, and which is indicated for schizophrenia.

There is a need in the art for formulations containing an aripiprazole prodrug that when administered to a patient can provide for improved therapeutic amounts of aripiprazole.

There is also a need in the art for methods of preparing an razole prodrug that can be formulated into a long-acting or extended-release formulation that when administered to a patient can provide for improved eutic amounts of aripiprazole over an ed period of time.

SUMMARYOF THEINVENTION The present invention provides a process for making a compound of Formula (A) in crystal form CI Nd Formula (A) wherein Ra is O)R1 and wherein R1 is a tuted or unsubstituted aliphatic moiety, comprising the steps of: (a) obtaining a drug solution by combining the compound of Formula (A) or a salt or solvate thereof with a first solvent; (b) optionally combining the drug solution with a second solvent to form a mixture; (c) cooling the e; and ((1) when the temperature of the e is within the range of about 0-5 °C above a target temperature, homogenizing the mixture to form crystallized les of the compound of Formula (A) having a surface area of about 0.50 m2/g to about 3.3 m2/g.

In another embodiment of the , the compound of Formula (A) is selected from the group consisting of: Formula (1) Formula (II).

In a particular embodiment of the method, the compound of Formula (A) has the structure of Formula (I). In another particular embodiment, the compound of Formula (A) has the structure of Formula (II).

The present invention provides a process for making a nd of Formula (I) in crystal form {IE [‘ifxi‘é’wx ' g; E\VL‘w hixJ \rr‘fikvr" Formula (1) comprising the steps of: (a) ing a drug solution by ing the compound of Formula (A) or a salt or solvate thereof with a first solvent; (b) optionally combining the drug solution with a second solvent to form a e; (c) cooling the mixture; and (d) when the temperature of the mixture is within the range of about 0-5 °C above a target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (A) having a surface area of about 0.50 m2/g to about 3.3 mZ/g.

The present ion also provides a s for making a compound of Formula (II) in crystal form Formula (11) comprising the steps of: (a) obtaining a drug solution by combining the compound of Formula (A) or a salt or e thereof with a first solvent; (b) optionally combining the drug solution with a second solvent to form a mixture; (c) cooling the mixture; and (d) when the temperature of the mixture is within the range of about 0-5 °C above a target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (A) having a surface area of about 0.50 m2/g to about 3.3 mZ/g.

The processes provided herein encompass a number of embodiments, including the ing: In one embodiment, the first solvent of step (a) is a single solvent. In another embodiment, the first solvent of step (a) is a mixture of two or more solvents. In a ular embodiment, the first solvent of step (a) is a mixture of two or more solvents and step (b) is absent. Suitable solvents are known to persons having skill in the art of crystallization. Examples of solvents are provided infra. In a particular embodiment, the first t of step (a) is isopropyl e. In another particular embodiment, the first solvent of step (a) is a mixture of isopropyl acetate and n-heptane.

In one embodiment, step (b) comprises combining the drug solution with a second solvent to form a mixture. In a particular embodiment, the second solvent of step (b) is n-heptane. The e of step (b) may be a homogeneous e. In certain embodiments, homogeneity of the mixture of step (b) is achieved or maintained by heating or preheating the first solvent of step (a) and/or the drug solution of step (a) and/or the second solvent of step (b). In a particular embodiment, the temperature of the mixture of step (b) is in the range of about 55 °C to about 65 °C. In another embodiment, step (b) is absent. When step (b) is , step (c) ses cooling the drug solution of step (a), and step (d) comprises homogenizing the drug solution to form crystallized particles of the compound of Formula (A) having a surface area of about 0.50 m2/g to about 3.3 m2/g when the temperature of the mixture is within the range of about 0-5 °C above a target temperature, In one embodiment, step (c) comprises cooling the mixture to the point of supersaturation. The temperature at which the mixture s aturated may be in the range of about 50 0C to about 55 0C. In another embodiment, step (c) comprises cooling to mixture to so that its temperature approaches a target temperature. In a particular embodiment, the target temperature is about 34 °C.

In one embodiment, the target temperature of step (d) is in the range of about 31 CC to about 35 °C. In a particular embodiment, the target temperature of step (d) is about 34 CC. In another embodiment of step (d), homogenizing begins when the temperature of the mixture is about 0 °C to about 4 °C above the target temperature (e.g., at about 31 0C to about 38 oC).

The ing s may further comprise the following steps: (e) stopping homogenization and re—dissolving the crystallized particles of the compound of Formula (A) (e.g., compounds having the structure of Formula (I) or Formula (II)) by heating the mixture; (f) cooling the mixture; and (g) when the temperature of the mixture is within the range of about 0-5 °C above the target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (A) (e.g., compounds having the structure of Formula (I) or Formula (11)) having a surface area of about 0.50 m2/g to about 3.3 m2/g.

Steps (c), (d) and (e) may be performed once, or two or more times, prior to proceeding to step (f).

Any one or more of steps (a), (b), (c), (d), (e), and (I) may be performed under agitation The foregoing methods may fitrther se the following steps: filtering the crystallized particles; rinsing the crystallized particles; and drying the crystallized particles.

The crystallized particles produced in accordance with the processes described herein may have a e area of about 0.80 to about 1.1 m2/g. In one ment, the crystallized les have a surface area of about 1.00 m2/g. In another embodiment, the Dv[50] of the crystallized les is about 10 to about 30 microns. In still another embodiment, the Dv[50] of the crystallized particles is about 10 to about 20 microns. In yet another embodiment, the crystallized les are le for use in a depot injection.

The invention provides crystallized particles of the compound of Formula (I) and the compound of Formula (11) ed by the foregoing process. Preferably, the crystallized particles may have a e area of about 0.80 to about 1.1 m2/g, and more ably, about 1.00 mZ/g. The Dv[50] of the crystallized particles may be about 10 to about 30 microns, preferably about 10 to about 20 microns.

The present invention provides a process for providing a depot ion comprising the compound of Formula (I) in crystal form CE [gmN .M‘xvxe‘fi'xmg '~T§.r"?§<\s ,3 ] wry -"‘-‘J’\\’«ffifNV-xfl\\l’fi§:’ “‘2’. N \T/x o 6;} L”, iv, Formula (I), the process comprising the steps of: (a) obtaining a drug solution by combining the compound of a (I) or a salt or solvate f with a first solvent; (b) combining the drug solution with a second solvent to form a mixture with reduced solubility relative to the solubility of the drug solution; (c) g the mixture so that it becomes supersaturated; (d) cooling the mixture so that its temperature approaches a target temperature; and (e) when the temperature of the mixture is within the range of about 5 °C above the target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (I) having a surface area of about 0.50 to about 3.3 m2/g. Any one or more of steps (a) through (d) of the foregoing process may be performed under agitation. The foregoing process may fithher comprise the steps of (f) filtering the crystallized particles, (g) rinsing the crystallized particles, and (h) drying the crystallized particles.

In the foregoing process, the first solvent may be ethyl acetate, propyl acetate, isopropyl acetate, butyl acetate, isobutyl acetate, tert—butyl acetate, acetone, and the like, with isopropyl acetate being preferred; and the second solvent may be pentane, cyclopentane, hexane, cyclohexane, methyl cyclohexane, heptanes, , nonane, decane, undecane, ne, ethanol, ol, and the like, with n-heptane being preferred.

Preferably, in step (b), the temperature of the e is in the range of about 55 °C to about 65 CC. In step (c), the ature at which the mixture becomes supersaturated may be in the range of about 50 CC to about 55 CC. The target temperature d in step (d) may be in the range of about 31 °C to about 35 °C, such as about 34 oC.

Preferably, in step (e), the homogenizing begins when the temperature of the mixture is about 0 °C to about 4 oC above the target temperature.

The invention also provides for crystallized les of the compound of a (I) produced by the foregoing process. ably, the crystallized particles may have a surface area of about 0.80 to about 1.1 mZ/g, and more preferably, about 1.00 m2/g. The Dv[50] ofthe crystallized particles may be about 10 to about 30 s, preferably about 10 to about 20 microns.

Further, the invention provides a process for providing a depot injection comprising the compound of Formula (I) in crystal form CE if TN: ctr. a-5L. 3.14;, {J “IN-‘1‘ K»,-A? ca»\\._f"J... 3 \Nf’ NT._ (x 5, Lx r “(v-“N,“'v/- C} Formula (I), the process sing the steps of: (a) obtaining a drug solution by combining the compound of Formula (I) or a salt or solvate thereof with heated pyl acetate; (b) combining the drug solution with n-heptane to form a mixture with reduced solubility relative to the solubility of the drug solution; (c) cooling the mixture so that it becomes supersaturated; (d) cooling the mixture so that its temperature approaches a target temperature of about 34 °C; and (e) when the temperature of the mixture is within the range of about 5 °C above the target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (I) having a surface area of about 0.50 to about 3.3 m2/g. Any one or more of steps (a) through (d) of the foregoing process may be performed under agitation. The foregoing process may further comprise the steps of (f) filtering the crystallized particles, (g) rinsing the crystallized particles, and (h) drying the llized particles.

Preferably, in step (b), the temperature of the mixture is in the range of about 55 °C to about 65 °C. In step (c), the temperature at which the mixture becomes supersaturated may be in the range of about 50 CC to about 55 CC. The target temperature reached in step (d) may be in the range of about 31 °C to about 35 °C, such as about 34 oC.

Preferably, in step (e), the homogenizing begins when the temperature of the mixture is about 0 °C to about 4 oC above the target temperature.

The invention also es for crystallized particles of the nd of a (I) produced by the foregoing process. ably, the llized particles may have a surface area of about 0.80 to about 1.1 mZ/g, and more preferably, about 1.00 mz/g. The Dv[50] ofthe crystallized particles may be about 10 to about 30 microns, preferably about 10 to about 20 microns.

Still further, the invention provides a process for providing a depot injection sing the compound of Formula (I) in l form Formula (I), the process comprising the steps of: (a) obtaining a drug solution by combining the compound of Formula (I) or a salt or solvate thereof with a first solvent; (b) combining the drug on with a second solvent to form a mixture with reduced solubility relative to the solubility of the drug solution; (c) cooling the mixture so that it s supersaturated; (d) cooling the mixture so that its temperature approaches a target temperature; (e) when the temperature of the mixture is within the range of about 5 °C above the target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (I); (f) stopping homogenization, and re-dissolVing the crystallized particles of the compound of Formula (I) by heating the mixture; (g) cooling the mixture so that its temperature approaches the target ature; and (h) when the temperature of the mixture is within the range of about 5 °C above the target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (I) having a surface area of about 0.50 to about 3.3 m2/g.

WO 32950 Steps (d), (e), and (f) of the ing process can be performed a second time (or three times, four times, etc.) prior to proceeding to step (g). For example, the process steps can be carried out in the order (a), (b), (c), (d), (e), (f), (d), (e), (f), (g), and (h); that is, steps (d) through (f) are performed twice in succession.

In the foregoing process, the first solvent may be ethyl acetate, propyl acetate, pyl acetate, butyl acetate, isobutyl acetate, utyl acetate, acetone, and the like, with isopropyl e being preferred; and the second solvent may be pentane, cyclopentane, , cyclohexane, methyl cyclohexane, heptanes, octane, nonane, decane, undecane, ne, ethanol, methanol, and the like, with n-heptane being preferred.

Any one or more of steps (a), (b), (c), (d), (e), (f), and (g) of the foregoing process may be performed under agitation. The foregoing process may further comprise the steps of (i) filtering the crystallized particles, (j ) rinsing the crystallized particles, and (k) drying the crystallized particles.

Preferably, in step (b), the temperature of the mixture is in the range of about 55 °C to about 65 °C. In step (c), the temperature at which the e becomes supersaturated may be in the range of about 50 °C to about 55 OC. The target temperature reached in steps (d) and (g) may be in the range of about 31 °C to about 35 0C, such as about 34 °C. ably, in steps (e) and (h), the homogenizing begins when the temperature of the mixture is about 0 CC to about 4 °C above the target temperature.

The ion also provides for llized particles of aripiprazole lauroxil produced by the foregoing process. Preferably, the crystallized particles may have a surface area of about 0.80 to about 1.1 m2/g, and more preferably, about 1.00 mz/g. The Dv[50] of the crystallized particles may be about 10 to about 30 microns, preferably about 10 to about 20 microns.

In alternative embodiments, the homogenizing of the processes described herein can be replaced with sonicating or the use of an ultrasound device.

The depot injection provided by any of the foregoing processes can provide for extended release of aripiprazole in viva. Such extended release can occur, for example, over from a period of about one month to a period of about three months. Preferably, such extended release can occur, for example, over from a period of about one month to a period of about two months. The depot injection provided by any of the foregoing WO 32950 processes can be administered, for example, as a once-monthly injection, a once-every- two-months injection, or a once-every-three months injection.

BRIEFDESCRIPTIONOF THEDRA WINGS depicts a typical cooling profile for the recrystallization process of the present invention. is a graph showing that homogenizer initiation d llization of aripiprazole lauroxil at a target temperature. depicts temperature profiles for recrystallization tests at the 4 kg scale. is a graph showing the relationship between surface area of particles of recrystallized razole il and exotherm onset ature. is a graph showing the relationship between particle size and e area.

FIGS. 6, 7, 8, 9, and 10 show models built from the results of an augmented multi- factor DOE study used to evaluate the combinatory effect of homogenizer initiation temperature, homogenizer speed, and heat er temperature gradient on cess crystal surface area, particle size, and exotherm onset temperature. shows the relationship between surface area of particles of recrystallized aripiprazole lauroxil and exotherm onset temperature as revealed by an augmented multi-factor DOE study. shows several cooling profiles for runs from a multi—factor DOE (central composite design) study that used no homogenization. shows plots of transformed temperature data for the cooling profiles from FIG. shows a plot of exotherm onset ature (Tmin) versus the calculated Exponential Primary Cooling Parameter (i.e., cooling rate). shows particle size distributions for several batches of recrystallized aripiprazole lauroxil made according to a 200 gram process. shows particle size distributions for several batches of recrystallized aripiprazole lauroxil made according to a modified 200 gram process. shows aripiprazole pK profiles resulting from uscular administration of a single dose of recrystallized aripiprazole lauroxil (20 mg aripiprazole equivalents) suspended in either NaCMC or SML vehicle, to male rats to assess the effect of injection vehicle on the in vivo profile. shows particle size distribution (PSD) profiles for four lots of aripiprazole lauroxil recrystallized drug substance. shows aripiprazole pK profiles resulting from IM administration to male rats of recrystallized razole lauroxil from the same lots as in suspended in SML vehicle.

DETAILED DESCRIPTION Crystallization Crystallization is a process of forming crystals through precipitation of solids from a solution, which occurs by variation of the solubility conditions of the solute in the solvent. The process is governed by both thermodynamic and kinetic s, which can make it highly variable and difficult to l. These factors include component concentrations, impurity levels, mixing regime, vessel design, and cooling profile. All can have a major impact on the size, number, and shape of crystals produced. dynamically, llization is impossible below the theoretical solution solubility threshold (saturation). At values above this old, the solution is supersaturated (contains more solute than could be dissolved by the solvent under normal stances) and crystallization may proceed. Supersaturation is a fundamental factor in llization dynamics, where the level of supersaturation affects the crystallization rate and indicates that crystallization is under kinetic, rather than thermodynamic, control.

Crystallization consists of two major kinetic driven events: nucleation and crystal growth. Nucleation is the step where the solute molecules dispersed in the solvent start to gather into clusters (nuclei) that become stable under the t operating conditions. The crystal growth is the subsequent growth of the nuclei that d in producing stable crystals. Nucleation and grth continue to occur aneously while solution supersaturation exists. tion is the initiation of crystallization and is the sum effect oftwo categories, primary and secondary. Primary nucleation is the initial formation of nuclei where there are no other crystals present. This typically occurs through the influence/presence of other solids (i.e., walls of the crystallizer vessel and les of any foreign substance). Secondary nucleation is the formation of nuclei attributable to the influence of already—existing crystals in the solution. Typically, this is a on of fluid shear and collisions of crystals and results in the formation of new nuclei. Several factors used to influence tion rate are use of seed crystals, equipment surface imperfections, and high shear homogenization.

The combination of solution supersaturation level and factors such as homogenization governs the tion rate, which in turn influences the crystal particle size and surface area. In general, a fast nucleation rate leads to smaller crystals, while a slow nucleation results in larger crystals. This is best understood through the concept of population e. A fast nucleation rate creates a large number of small nuclei in a specified period, while a slow nucleation rate creates a lesser number in the same period. After the nuclei are ted, they then being to grow. Given a finite amount ofmass available for , and assuming equivalent growth rates, the larger number population of nuclei will achieve a final particle size that is smaller (conversely larger surface area) than the lesser number population.

Aripiprazole lauroxil Aripiprazole lauroxil, an N—lauroyloxymethyl g form of aripiprazole, has been developed for formulation into extended-release injectable formulations, such as for intramuscular ion. Aripiprazole lauroxil is a non-hygroscopic white crystalline solid with a melting point of 81.3 to 83.0 CC, and it exists as a stable form, which has not been observed in any rphic modifications to date. The compound is insoluble in water (<4 ng/mL at room temperature) and shows highest room temperature solubility in the following organic solvents: THF (~400 mg/mL), dichloromethane (~500 mg/mL), and toluene (~300 mg/mL). The IUPAC name for aripiprazole lauroxil is (7-(4-(4-(2,3 oropheny1)piperazin- l -yl)butoxy)oxo-3 ,4- dihydro-ZH—quinolin-l-yl)methyl dodecanoate, corresponding to the molecular formula C36H51C12N304 and a molecular weight of 660.7. razole lauroxil may also be referred to as N—lauroyloxymethyl aripiprazole. The chemical structure of aripiprazole lauroxil is as follows: {:E [fanN M‘KKVJWHKVJ,‘Q ‘Nfi’fta‘fig ii ’”’ w» {3 N __,,,l and is also referred to herein as Formula (I).

Pre-processed aripiprazole lauroxil suitable for the recrystallization process described herein may be ed, for example, by following the synthesis described in US.

Patent No. 8,431,576. This document is incorporated herein by reference in its entirety.

Salts and es of aripiprazole lauroxil, which are disclosed and described in US.

Patent No. 8,431,576, are also suitable for the recrystallization process described herein.

Aripiprazole lauroxil oes hydrolysis to lauric acid, formaldehyde, and aripiprazole, which is an important antipsychotic used in the treatment of schizophrenia and other psychiatric conditions, such as bipolar disorder and major depressive disorder.

Conversion of aripiprazole lauroxil to aripiprazole in vivo is governed by slow dissolution of the aripiprazole lauroxil drug crystals and subsequent enzyme—mediated ge to the N—hydroxymethyl aripiprazole intermediate, which spontaneously converts to aripiprazole.

The slow ution of the aripiprazole lauroxil drug crystals in viva results in systemic exposure of aripiprazole over several weeks. The rate of aripiprazole il release is a function of the amount of exposed surface area, ented by particle size distribution (PSD) and shape/morphology of the drug crystals.

An ed release IM injection offers the potential for an improved safety profile and ent compliance; therefore, it has the potential to provide more effective management of schizophrenia.

The present invention provides a process for providing a depot injection comprising aripiprazole lauroxil, i.e., the compound of Formula (I), in crystal form. g t ii-‘eg‘ N xxx] ““xjf ”a ~.,/’£‘\‘ w ‘\ x “‘x.”’fl‘PK-"‘-'3 “f Formula (I) The process comprises the steps of: (a) obtaining a drug solution by combining the compound of Formula (I) or a salt or solvate thereof with a first solvent; (b) combining the drug solution with a second solvent to form a mixture with reduced solubility ve to the solubility of the drug solution; (c) cooling the mixture so that it becomes supersaturated; (d) cooling the mixture so that its temperature approaches a target ature; and (e) when the temperature of the mixture is within the range of about 5 °C above the target temperature, homogenizing the mixture to form llized particles of the compound of Formula (I) having a surface area of about 0.50 to about 3.3 m2/g. Any one or more of steps (a) through (d) of the foregoing process may be performed under agitation. The foregoing process may r comprise the steps of (t) filtering the crystallized particles, (g) rinsing the llized particles, and (h) drying the crystallized les.

In the foregoing process, the first solvent may be ethyl acetate, propyl acetate, isopropyl acetate, butyl acetate, isobutyl acetate, tert-butyl acetate, acetone, and the like, with isopropyl acetate being preferred; and the second solvent may be pentane, cyclopentane, hexane, cyclohexane, methyl cyclohexane, heptanes, octane, nonane, decane, undecane, dodecane, ethanol, methanol, and the like, with n-heptane being preferred.

Preferably, in step (b), the temperature of the e is in the range of about 55 °C to about 65 °C. In step (c), the temperature at which the e becomes supersaturated may be in the range of about 50 CC to about 55 CC. The target temperature reached in step (d) may be in the range of about 31 °C to about 35 °C, such as about 34 °C. ably, in step (e), the homogenizing begins when the temperature of the mixture is about 0 °C to about 4 °C above the target temperature.

The homogenizing in step (e) initializes and promotes llization, and allows for control of particle size and surface area. A suitable homogenization speed is from about 4800 to about 9600 rpm. The drying in step (h) can be conducted under nitrogen purge and vacuum.

The invention also provides for crystallized particles of the compound of a (I) produced by the foregoing s. Preferably, the crystallized particles may have a surface area of about 0.80 to about 1.1 mZ/g, and more preferably, about 1.00 m2/g. The Dv[50] ofthe crystallized particles may be about 10 to about 30 microns, preferably about 10 to about 20 s. r, the invention provides a s for providing a depot injection comprising the compound of Formula (I) in crystal form CE r“ “*3; {3; N xxx-“J [ 1} «I "“xgtifii‘x.

RM, f,x ”waft.“ 5:: a, J33 if] L,» a.» o 5:} Formula (I), the process comprising the steps of: (a) obtaining a drug solution by combining the compound of a (I) or a salt or solvate thereof with heated isopropyl acetate; (b) combining the drug solution with n-heptane to form a mixture with reduced lity relative to the solubility of the drug solution; (c) cooling the mixture so that it becomes supersaturated; (d) cooling the mixture so that its temperature approaches a target temperature of about 34 °C; and (e) when the temperature of the mixture is within the range of about 5 °C above the target temperature, homogenizing the mixture to form crystallized particles of the compound of Formula (I) having a surface area of about 0.50 to about 3.3 m2/g. Any one or more of steps (a) through (d) of the foregoing s may be performed under agitation. The foregoing process may fiarther comprise the steps of (f) filtering the crystallized particles, (g) rinsing the crystallized particles, and (h) drying the crystallized particles.

Preferably, in step (b), the temperature of the mixture is in the range of about 55 °C to about 65 0C. In step (c), the temperature at which the mixture becomes supersaturated may be in the range of about 50 °C to about 55 CC. The target temperature reached in WO 32950 step (d) may be in the range of about 31 0C to about 35 0C, such as about 34 OC.

Preferably, in step (e), the homogenizing begins when the temperature of the mixture is about 0 0C to about 4 oC above the target ature.

The homogenizing in step (e) initializes and promotes crystallization, and allows for control of particle size and surface area. A suitable homogenization speed is from about 4800 to about 9600 rpm. The drying in step (h) can be conducted under nitrogen purge and vacuum.

The invention also provides for crystallized particles of the compound of Formula (I) produced by the foregoing process. ably, the crystallized particles may have a surface area of about 0.80 to about 1.1 mZ/g, and more preferably, about 1.00 m2/g. The Dv[50] of the crystallized particles may be about 10 to about 30 microns, preferably about 10 to about 20 microns.

Still further, the invention provides a process for providing a depot injection sing the compound of a (I) in crystal form Formula (I), the process comprising the steps of: (a) obtaining a drug solution by combining the compound of a (I) or a salt or solvate thereof with a first solvent; (b) combining the drug solution with a second solvent to form a mixture with reduced solubility relative to the solubility of the drug solution; (c) cooling the e so that it becomes supersaturated; (d) cooling the mixture so that its temperature approaches a target temperature; (e) when the temperature of the mixture is within the range of about 5 °C above the target ature, homogenizing the mixture to form crystallized particles of the compound of Formula (I); (f) stopping homogenization, and re-dissolving the crystallized particles of the nd of Formula (I) by heating the mixture; (g) cooling the mixture so that its temperature approaches the target ature; and (h) when the temperature of the mixture is within the range of about 5 OC above the target temperature, homogenizing the mixture to form crystallized les of the compound of a (I) having a surface area of about 0.50 to about 3.3 m2/g.

Steps ((1), (e), and (f) of the foregoing process can be performed a second time (or three times, four times, etc.) prior to proceeding to step (g). For example, the process steps can be carried out in the order (a), (b), (c), (d), (e), (f), (d), (e), (f), (g), and (h); that is, steps ((1) through (f) are performed twice in succession.

In the foregoing s, the first solvent may be ethyl acetate, propyl acetate, isopropyl e, butyl acetate, isobutyl acetate, tert-butyl acetate, acetone, and the like, with isopropyl acetate being preferred; and the second solvent may be e, cyclopentane, hexane, cyclohexane, methyl cyclohexane, heptanes, octane, nonane, decane, undecane, dodecane, ethanol, methanol, and the like, with n-heptane being preferred.

Any one or more of steps (a), (b), (c), (d), (e), (f), and (g) of the ing process may be performed under agitation. The foregoing process may further comprise the steps of (i) filtering the crystallized particles, (1') rinsing the crystallized particles, and (k) drying the llized particles.

Preferably, in step (b), the temperature of the mixture is in the range of about 55 °C to about 65 °C. In step (c), the temperature at which the mixture becomes supersaturated may be in the range of about 50 °C to about 55 CC. The target temperature reached in steps (d) and (g) may be in the range of about 31 °C to about 35 °C, such as about 34 oC. Preferably, in steps (e) and (h), the homogenizing begins when the temperature of the mixture is about 0 °C to about 4 °C above the target ature.

The homogenizing in steps (e) and (h) initializes and promotes crystallization, and allows for control of particle size and surface area. A suitable homogenization speed is from about 4800 to about 9600 rpm. The drying in step (k) can be conducted under nitrogen purge and vacuum.

The invention also provides for crystallized particles of the nd of Formula (I) produced by the foregoing process. Preferably, the crystallized les may have a surface area of about 0.80 to about 1.1 mZ/g, and more preferably, about 1.00 m2/g. The Dv[50] ofthe crystallized particles may be about 10 to about 30 microns, preferably about 10 to about 20 microns. 2015/046525 Each of the foregoing processes may use, instead of the compound of Formula (I), a salt or solvate thereof such as the salts or es of pre-processed aripiprazole lauroxil disclosed and described in US. Patent No. 8,431,576.

Aripiprazole cavoxil Aripiprazole cavoxil, an N—hexanoyloxymethyl prodrug form of aripiprazole, has been developed for formulation into extended-release injectable formulations, such as for intramuscular injection. The IUPAC name for aripiprazole cavoxil is (7-(4—(4-(2,3- dichlorophenyl)piperazin- l -yl)butoxy)oxo-3 ,4-dihydroquinolin- l (2H)-yl)methyl hexanoate, corresponding to the molecular formula C30H39C12N304 and a molecular weight of 576.56. The chemical structure of aripiprazole cavoxil is as follows: C! [JV/RN “FR, “fixv’fl‘v’xxfi mmi‘ixcxw«AE‘\ 5’3“} If] "a“, “K vx”‘"\§lx°fl‘\vw : \fr" 9 Q and is also ed to herein as Formula (II). Aripiprazole cavoxil suitable for the processes described herein may be obtained, for example, by the synthesis methods described in US. Patent No. 8,431,576. This document is incorporated herein by reference in its entirety. Salts and solvates of aripiprazole cavoxil, which are disclosed and described in US. Patent No. 8,431,576, are also suitable for the ses described herein.

Process Equipment The following s equipment was used to recrystallize aripiprazole lauroxil according to the present invention. Other suitable process equipment may be used, as would be well understood by one skilled in the art in light of present disclosure.

Drug dissolution vessel: To produce a 4.0 kg batch of tallized aripiprazole lauroxil, a single closed, jacketed, ed 20 liter vessel was used to dissolve and er the drug through a sterilizing filter to the ized recrystallization vessel in a single step. A r pilot process (producing 1.75 kg of recrystallized aripiprazole lauroxil) used two small stock pots and hot plates to dissolve the drug, and multiple transfer steps to a 4 liter pressure vessel to sterile filter the solution into the recrystallization vessel. At both 4.0 kg and 1.75 kg scales, warm isopropyl acetate was used to dissolve the pre-processed drug crystals of aripiprazole lauroxil. As would be well understood by one of skill in the art, a “jacket” refers to heat transfer fluid and to the d space around the vessel containing the heat er fluid that acts as a heat ger to cool or heat the inside of the vessel, and a “glycol jacket” is a jacket where the heat transfer fluid is glycol or a mixture of water and glycol. The glycol jacket temperature s primary cooling, which transitions the system into a meta—stable zone where crystallization of aripiprazole lauroxil can be initiated with homogenization.

Drug solution filter: The drug solution filter used for both scales was Milliport Aervent al Cartridge, PTFE Hydrophobic, LAGR04TP6 (112-00783).

Filter heat tape: The filter heat tape used for both scales was Fiberglass Cloth Heating Tape with Glas-Col Poerrol Controller (10 amps/120 volts).

Recrystallization vessel: The recrystallization vessel used for both scales was a DCI er Cone Shaped (16” upper ID / 23° angle) Stainless Steel Jacketed Vessel (DCI Serial #: J82884) with a 3.75” Radial Lower Impeller, 3.75” Axial Upper Impeller on an angled (non-vertical) agitator.

Homogenizer: The homogenizer used for both scales was a Kinematica Polytron PT—D 50-6 F/G (installed in the recrystallization vessel) with 50 mm Stator Diameter and 45 mm Rotor Diameter.

Sonicator: An exemplary tor le for use with the process of the present invention is Transsonic T310 from Lab-Line Instruments Inc.

Dryer: The dryer used in the 4.0 kg scale process was a closed, agitated, 15” self- discharging vacuum filter dryer (Powder Systems Limited; PSL). The 1.75 kg scale process used two 8” static vacuum filter dryers that required manual aseptic stirring of the recrystallized drug crystals prior to drying and discharge. The mode of drying was the same at both scales, namely, the recrystallized drug crystals were dried under vacuum at room temperature with a dry gas purge to facilitate removal of processing solvents to acceptably low levels.

Filtrate vessel: The te vessel used in both scales was a DCI 10—gallon Stainless Steel Jacketed Vessel (DCI Serial #: JS2060).

Recrystallization process The recrystallization process of the present invention can produce crystallized particles of aripiprazole lauroxil having a surface area of about 0.50 to about 3.3 m2/g, preferably about 0.80 to about 1.1 mZ/g, more preferably about 1.00 m2/g. The Dv[50] of the crystallized particles may be about 10 to about 30 microns, preferably about 10 to about s.

Recrystallized aripiprazole lauroxil can be produced h the following procedure: Drug Dissolution: Dissolve pre-processed aripiprazole lauroxil or a salt or solvate thereof in a first solvent such as isopropyl acetate or another suitable first solvent as described herein, and sterile filter the result into a recrystallization vessel.

Crystallization: Mix the drug solution (aripiprazole lauroxil dissolved in, e.g., isopropyl e) and a second solvent such as heptane or another suitable second solvent as bed herein and then cool at a controlled rate; initiate homogenization at a target ature to induce crystallization.

Collection: Transfer contents of recrystallization vessel and filter crystals from the solvent in a dryer.

Rinsing: Use a fresh portion of the second solvent to recover any crystals remaining in the tallization vessel and remove gross residual first solvent from the l surface. Rinsing can also remove residual amounts of acetonitrile, which may have been present in the ocessed aripiprazole lauroxil.

Drying: Use vacuum drying to reduce levels of both the first and second ts.

The first t may be ethyl acetate, propyl acetate, isopropyl acetate, butyl acetate, isobutyl acetate, tert-butyl acetate, acetone, or another suitable solvent as would be well understood by one skilled in the art in light of the present disclosure, or mixtures of the foregoing ts. Isopropyl acetate is a preferred first t.

The second solvent may be pentane, cyclopentane, hexane, exane, methyl cyclohexane, es, octane, nonane, decane, undecane, dodecane, ethanol, methanol, or another suitable solvent as would be well understood by one skilled in the art in light of the present disclosure, or mixtures of the foregoing solvents. N-heptane is a preferred second solvent.

A preferred selection and ratio of first solvent and second solvent is isopropyl acetate and heptanes, in a ratio of 1:2 (V/V).

When the drug solution of aripiprazole lauroxil in a first solvent (such as isopropyl acetate) is combined with the second solvent (such as e), this forms a e with reduced solubility relative to the lity of the drug solution. The drug solution and the second solvent are preferably combined at a ature of from about 55 °C to about 65 °C, and then the mixture is cooled at a specified rate, such as 1.5 °C per minute, so that the mixture becomes supersaturated. The temperature at which the mixture becomes aturated may be in the range of about 50 0C to about 55 °C.

Then the mixture is cooled so that its temperature approaches a target temperature.

This target temperature may be in the range of about 31 °C to about 35 CC, such as about 34 °C. When the temperature of the mixture is within the range of about 0 °C to about 4 °C above the target ature, homogenization of the mixture is initiated. A suitable speed for homogenization is from about 4800 to about 9600 rpm.

Drug dissolution, combining the drug solution and the second solvent, cooling the mixture of the drug solution and the second t, cooling the mixture once it has become supersaturated, homogenizing the mixture, and re-dissolving the crystallized particles of razole lauroxil by heating the mixture can each be performed under agitation. The agitation may be carried out with an agitator such as an overhead stirrer.

The agitator helps to maintain a uniform l suspension and control over the temperature.

Suitable salts and solvates of pre-processed aripiprazole lauroxil that can be obtained, synthesized, and used with the present invention include those disclosed in US. Patent No. 8,431,576.

Formulations The recrystallized aripiprazole lauroxil prepared according to the methods disclosed herein can be ded in injection es to e injectable compositions suitable, for example, for 1M administration. Such vehicles include a phosphatebuffered saline injection vehicle comprising sorbitan monolaurate in an amount of imately 0.37% by weight relative to the weight of the inj ectable composition; polysorbate 20 in an amount of approximately 0.15% by weight relative to the weight of the inj ectable composition; and an s carrier. The recrystallized aripiprazole lauroxil prepared ing to the methods disclosed herein can also be incorporated into other vehicles and formulations, such as those disclosed in US. Patent Application ation No. 23 8552.

The following is the formulation of an exemplary depot injection ition comprising recrystallized aripiprazole lauroxil prepared according to the s disclosed herein: Recrystallized Aripiprazole 26.6 Lauroxil Drug Substance P—hosphateSodium Dibasic Anhydrous Sodium D1hydrogen Phosphate Monobasic Dihydrate Water for Injection QS to 100 Relation of particle size and release rate Particle size of the aripiprazole lauroxil produced from the recrystallization process of the present invention was shown to relate to release rate in animal studies and thus required controlling to within an acceptable range. The particle size distribution (PSD) ofrecrystallized aripiprazole il produced according to the process disclosed herein can be measured, for e, using a light scattering particle size analyzer such as those available from HORIBA or Beckman-Coulter, or by other suitable instruments and methods as would be well understood by one skilled in the art in light of the present disclosure.

As used herein, “Dv[50]” refers to the 50th percentile of the particle size distribution, which is hangeable with median diameter or the average particle diameter by volume. As used herein, “Dv[10]” refers to the 10th percentile of the particle size distribution, “Dv[90]” refers to the 90th percentile of the particle size distribution, and “Dv[X]” refers to the Xth percentile of the particle size distribution.

An acceptable Dv[50] range of particles of aripiprazole il produced from the recrystallization process of the present invention is 10-30 microns, with a DV[50] of 10- microns being red.

The relation of particle size and release rate is further explained in the studies and examples further below.

Relation of surface area and release rate Drug release was found to be proportional to the e area of aripiprazole lauroxil ed from the recrystallization process of the present invention.

The e area of recrystallized aripiprazole lauroxil particles can be measured, for example, using an rated surface area and porosimetry analyzer, or by other suitable instruments and methods as would be well tood by one skilled in the art in light of the present disclosure.

An acceptable surface area range for particles of aripiprazole lauroxil produced from the recrystallization process of the present invention is from about 0.50 to about 3.3 m2/g. A surface area range of 0.80 to about 1.1 mZ/g is preferred, and a surface area of about 1.0 m2/g is more red.

The relation of surface area and release rate is further explained in the studies and examples further below.

Cooling profile depicts a typical cooling profile for the recrystallization process of the present invention. Cooling of a mixture ning razole lauroxil, first solvent, and second solvent causes the temperature of the mixture to decrease, and the mixture becomes supersaturated. Aripiprazole lauroxil precipitates, causing an increase in temperature of the system. This is ed by further cooling of the system. As used herein, the term “exotherm” refers to the increase in temperature of the system due to precipitation of the drug. The “precipitation zone,” in which the exotherm occurs, begins when the temperature starts to increase and covers the entire period during which aripiprazole lauroxil is precipitating or crystallizing. The “arrest temperature” or target temperature is the ature at which no further decrease in temperature of the system is observed and the start (or onset) of crystallization occurs. Homogenization is preferably initiated when the temperature of the supersaturated mixture of aripiprazole lauroxil is a few degrees above the arrest or target temperature. nization promotes crystallization and allows for control of particle size and surface area.

“Tmin” indicates the initial temperature increase due to exothermic heating from the major crystallization event. Tmin, which defines both the “onset of crystallization” and the “onset of exotherm,” is ly correlated with particle size and surface area of the recrystallized particles of aripiprazole lauroxil. “Tmax,” or the exotherm maximum temperature, denotes the completion of cant exothermic heating from the major crystallization event. ing the major crystallization event, the slurry is fiarther cooled in the “final cooling” stage (growth zone). “Tming” is the temperature equal to Tmin that occurs upon g of the system following the exotherm associated with crystallization.

STUDIES AND EXAMPLES Homogenization As illustrated in homogenizer initiation induces llization at a target ature. The results of five crystallization tests at the 1.75 kg scale are reflected in All of the llization tests had the same cooling rate, but each used a different homogenizer initiation temperatures and resulted in different exotherm onset temperatures and, as a result, different crystal sizes. In four of the tests, the homogenizer was turned on at the respective temperature specified in the plot, and crystallization was induced soon after the initiation of homogenization. The Batch 5 test shows the point when spontaneous crystallization ed, when no homogenization was used at the given cooling rate.

Impact of homogenizer initiation temperature on crystal size (single factor screening) The objective of this study was to screen the impact of homogenization initiation temperature (“Homogenizer ON”) on the surface area and particle size of recrystallized aripiprazole lauroxil, as well as the llization induction time and exotherm onset temperature. The study evaluated homogenizer initiation temperature at three values (35, 36, and 37 °C) while the following parameters were held constant: Homogenizer speed — 75% (120 Hz / 7200 rpm); Vessel agitation speed — 375 rpm; Vessel jacket glycol temperature set-point — 30 0C (this parameter dictates the solution y g rate).

Table 1 summarizes l tests at the 4 kg scale used to evaluate the effect of homogenizer initiation temperature enizer ON) on in—process crystal particle size. By “exotherm onset temperature” is meant the temperature at which dissolved aripiprazole lauroxil begins to recrystallize.

Table 1: Effect of Homogenizer 0N Temperature (4 kg Scale) shows the temperature profiles for the tests at the 4 kg scale. The plots showed that crystallization (observed by the exotherm) occurred shortly after nizer initiation, as was previously demonstrated at the 1.75 kg scale (.

Impact of crystallization variables on particle size and surface area (multi-factor screening) The objective of this study was to characterize the impact of homogenizer initiation temperature and homogenizer speed on e area and particle size as well as the llization ion time and exotherm onset temperature. The study evaluated the two factors of homogenizer initiation temperature and homogenizer speed, both factors being d at three levels. The study used a fiill factorial experimental design with a center-point resulting in ten (10) crystallizations. The jacket glycol temperature set- point was adjusted as a function of the homogenizer initiation temperature. The intent was to maintain a heat transfer temperature gradient (at crystallization) between 5-7 0C.

This value is defined as the difference n the jacket glycol temperature set-point and homogenizer initiation temperature.

Table 2 summarizes tests from the multi-factor screening study used to evaluate the combinatory effect of nizer initiation temperature (Homogenizer ON) and homogenizer speed on in-process crystal surface area and particle size. The jacket glycol temperature set-point was varied to maintain the gradient in a range of 5-7 0C in order to minimize the under-cooling temperature delta. and present plots of surface area versus rm onset temperature and particle size versus surface area. The plots demonstrate that the strong relationship observed between these attributes at the 1.75 kg scale ued to be present at the 4 kg scale.

WO 32950 Screening) (Multi-factor n—m-nflII—Elna Parameters Ch. O.

Crystallization Ln l\ m m m m Effect C) 2: H Table m m m m N m U U Impact of crystallization variables on particle size and surface area (multi-factor DOE) The objective of this study was to characterize the impact of homogenizer tion temperature and homogenizer speed on surface area and particle size as well as the crystallization induction time and exotherrn onset temperature. This study was equivalent to the study Impact of crystallization variables on particle size and surface area (multi- factor screening) above, but executed in a different process train. This study evaluated the two factors of homogenizer initiation temperature and homogenizer speed, each at three levels. The study used a full factorial experimental design resulting in nine (9) crystallizations. The jacket glycol temperature set-point was adjusted as a function of the homogenizer initiation temperature. The intent was to in a heat transfer temperature gradient (at crystallization) of 6 0C. This value is defined as the ence n the jacket glycol temperature set-point and homogenizer initiation temperature.

Table 3 summarizes tests fiom the multi—factor DOE (design of ments) study used to evaluate the combinatory effect of homogenizer initiation ature (Homogenizer ON) and homogenizer speed on in-process crystal surface area and le size. The jacket glycol temperature set—point was varied to maintain the gradient at a specified 6 °C in order to ze the under-cooling temperature delta. The study showed that the impact of Homogenizer ON temperature and the impact of homogenizer speed parameters were both statistically significant.

WO 32950 PCT/U52015/046525 nunnun DOE) (Multi-factor Parameters ———— Crystallization '\. 7‘. “3. \—c m H m m m Effect nN <r N m m m Table o m o £0 00 U) N N N N m \—: N m D Q Impact of crystallization variables on particle size and surface area (multi-factor Augmented DOE) The objective of this study was to characterize the impact of homogenizer initiation temperature, homogenizer speed, and heat transfer temperature gradient (at llization) on surface area and particle size as well as the crystallization induction time and rm onset temperature. This study augmented the foregoing multi—factor DOE study by orating the additional parameter of heat transfer temperature gradient. The study evaluated the three factors of nizer initiation temperature, homogenizer speed, and heat transfer temperature gradient (at crystallization), each at three levels. The study used a central composite experimental design with center-point replicates resulting in seventeen (l7) crystallizations. The jacket glycol temperature set-point was adjusted as a function of the homogenizer initiation temperature in order to set the heat transfer temperature gradient (at crystallization). This value is defined as the difference between the jacket glycol temperature int and homogenizer initiation temperature.

Table 4 summarizes tests from the augmented multi-factor DOE study used to evaluate the combinatory effect of homogenizer initiation temperature (Homogenizer ON), homogenizer speed, and heat transfer temperature gradient on in-process crystal surface area, particle size, and exotherm onset temperature. This DOE ted the immediately previous experimental design by including heat transfer temperature gradient as a factor.

Again, the jacket glycol temperature set-point was varied to maintain the gradient at specified values of 4, 6, and 8 0C.

FIGS. 6, 7, 8, 9, and 10 show several models built from the results. Table 5 izes the findings from these models. All models were statistically significant based on ANOVA ues < 0.05) and demonstrate no Lack of Fit. shows the relationship n e area and exotherm onset temperature.

Linear regression analysis trated statistical significance per ANOVA (p-values < 0.05). The excellent relationship of surface area to exotherm onset temperature was ageous because it provided an in-process measurement of crystallization progression or performance. Advantageously, in the event that the exotherm onset temperature fell WO 32950 outside the predicted or target range during a crystallization, the run could be discontinued, the material re-heated, and the crystallization repeated.

WO 32950 PCT/U52015/046525 DOE) nunnunnun—nn Augmented (Multi-factor '\.

Parameters m_ Crystallization 4 4 Effect N Table Ln |\ nun—n an34 [In L|_| Eannual.“fl “man“nunnunnu 26 flflfl WO 32950 PCT/U82015/046525 nunl-nnl-nn Lnm\—| Human-n nun—n m_o00NVonLG|\ E15 LO PCT/U52015/046525 Table 5: Summary of Multi-factor Augmented DOE Models e area Model (Multi-factor Augmented DOE) Exotherm Onset Temperature Model (Multi- factor Augmented DOE) Particle Size (Dv[10]) Model (Multi-factor 0.599 0.0003 No Augmented DOE) le Size (Dv[50]) Model (Multi-factor 0.574 0.0004 Yes Augmented DOE) : Particle Size (Dv[90]) Model (Multi-factor 0.340 0.0140 No Augmented DOE) Characterization of the process operational zone (Multi-factor DOE) The objective of this study was to characterize the impact of solution cooling rate (as a fimction ofj acket glycol temperature), homogenization speed, and crystallization type aneous versus induced) on the crystallization induction time and exotherm onset temperature. The study used a designed experiment al composite design) consisting of 10 runs that incorporated two factors at three levels: Vessel Jacket Temperature (3 0C, 16.5 0C, 30 0C); and nizer Speed (0%, 37.5%, 75%).

Table 6 summarizes tests from the factor DOE (central composite design) consisting of 10 runs that incorporated the two factors at three levels: Vessel Jacket Temperature (3 0C, 16.5 0C, 30 0C); and Homogenizer Speed (0%, 37.5%, 75%). Each run that used a 0% homogenizer speed (no homogenization) was cooled until it spontaneously crystallized. In the runs that used homogenizer speeds of 37.5% and 75%, the homogenizer was turned on when the process ature cooled to 53.6 °C. (This temperature represents the solubility limit of the s solution below which the solution enters a meta-stable state.) The solution then continued to cool with homogenization until it crystallized. shows several cooling profiles for runs that used no nization.

An exotherm onset temperature (Tmin) was recorded for each run and the cooling rate was calculated. The process cooling followed an exponential decay profile. Therefore, an Exponential Primary g Parameter (i.e., cooling rate) was calculated by plotting the process temperature as a function of time using the following data transformation method: y=m*x+b Where T—Ta 3]-_ Ln[T0—Ta] x [=] Time (Min.) T [=] Temperature (0C) To = 58°C Ta = 20°C A -regression fit of the data resulted in slope-m (l/°C) and intercept—b, where slope represents the Exponential Primary Cooling Parameter. displays plots of the transformed temperature data for the corresponding cooling profiles from .

Table 6: Process Operational Zone Results shows a plot of exotherm onset temperature (Tmin) versus the calculated Exponential y g Parameter (i.e., cooling rate). The following terms describe key features of the plot: Solubility Limit: The temperature at which the process solution becomes saturated. Below this temperature, the on is supersaturated. The supersaturated solution is kinetically persistent since it is ng relatively slowly but has not yet reached thermodynamic equilibrium, which results in crystallization.

Lower ing Boundary: The lowest process temperature the system can achieve at a specified cooling rate prior to spontaneous crystallization (i.e., crystallization in the absence of homogenization).

Upper Operating ry: The highest process temperature the system can achieve at a specified cooling rate prior to forced crystallization (i.e., crystallization that occurs in the presence of homogenization). In this case, homogenization was initiated immediately after the on cooled to the solubility limit (53.6 0C), thereby imparting the maximum homogenization time within the table zone.

Operational Zone: The temperature range available to the system for a specified cooling rate where crystallization of the supersaturated solution can be induced through homogenization at a target temperature. The operational zone provided ce for selection of optimum target cooling rate and exotherm onset temperature (Tmin) combinations, which ed in robust sing to target surface areas.

Additional synthesis examples 200 gram scale Aripiprazole lauroxil was recrystallized using the ing procedure. 246.4 g of isopropyl acetate was heated in a 1 liter Erlenmeyer flask to 70-75 0C. 383.0 g of heptane was heated in a 1 liter Erlenmeyer flask to 45—50 c’C. The hot isopropyl acetate was added to a 2 liter Erlenmeyer flask containing 200 g of aripiprazole lauroxil. The mixture was heated with swirling until all the white solids dissolved and a clear solution was obtained at 65-70 0C. Hot e was added to the clear solution in three portions with gentle heating and swirling to avoid crash-out.

The flask containing the clear solution was placed in a 12 inch sieve pan or equivalent. A homogenizer probe was placed into the solution and turned on to # 3 (13.5 l/min set on the machine). Ice was added up to the capacity of the pan. The homogenizer was stopped once the solution crystallized. The flask was kept in ice until the temperature was 15-20 c’C. The flask was removed from the ice bath.

A filtration set-up was assembled using a 2 liter filtering flask, Buchner funnel with a rubber connector, and filter paper. The filter paper was wetted with heptane (~ 5 ml). The recrystallized white solid was filtered and washed with heptane (~ 60 ml).

The filtered material was spread into a dish. The al was dried inside a vacuum oven at room temperature for 18-24 hrs with a nitrogen purge. The dried al was transferred into a 250um sieve. 5 PTFE sieve rings were added to the sieve, a cover with o-ring was placed on the sieve, and sieving took place using an Analysette 3 PRO at an amplitude set point of 2.7.

Four batches were made following this procedure and combined together to prepare a suspension of razole lauroxil. Table 7 lists the particle size distribution summary statistics for each batch as well as the ed t averaged) final batch. shows the particle size distributions for each batch as well as the combined (weight averaged) final batch.

Table 7: Particle size distribution summary statistics (post dry and sieving) Batch Dv[10], DV[50], Dv[90], % Total um um um MadiZzed 200 gram scale Using a graduated cylinder, 280 ml of isopropyl e were measured and transferred to a 2 liter Erlenmeyer flask. Using a ted cylinder, 560 ml of n-heptane were measured and mixed with the measured isopropyl acetate. 200 g of aripiprazole lauroxil were weighed into a 2 liter Erlenmeyer flask. The solvent mixture was heated to 70 °C and then added to the WO 32950 aripiprazole lauroxil containing flask. The slurry was heated back to 65 °C to obtain a clear solution.

The solution was then poured into a 1 liter jacketed glass vessel with a first ulator ulating water and an overhead high shear mixing probe. The probe was turned on immediately on setting #3 (13.5 l/min set on the machine). As soon as the internal temperature reached 2 OC above the target arrest temperature, the recirculating water was switched to a second recirculator in order to arrest the cooling. As soon as the temperature started to rise the time was noted and the recirculating water was switched back to the first recirculator. The probe was then stopped 90 seconds after the start of the temperature rise, after which it was replaced with an overhead mixer. The slurry was left to cool down to 18 0C.

When the slurry reached 18 °C the slurry was filtered using a Buchner funnel with Whatman filter paper 4. The solids were then washed with approximately 100 ml of n-heptane. The solids were spread into a crystallization dish and left to dry in a vacuum oven at room temperature, house , and a nitrogen purge for approximately 18 hours.

Table 8: Differences between 200 9 scale and modified 200 9 scale processes _200gram scale Modified 200 gram s Drug Aripiprazole lauroxil is dissolved in Aripiprazole lauroxil is dissolved in a dissolution 70-75°C isopropyl acetate and then 65 °C isopropyl acetate/heptane 45-50 °C heptane is added. mixture.

Crystallization 2 liter Erlenmeyer flask in an ice 1 liter jacketed glass vessel with two vessel bath. recirculating jacket fluid temperatures.

Manual swirling and homogenizer. Overhead mixer and homogenizer.

Crystallization Homogenizer is switched on and Homogenizer is switched on and solution is cooled with ice bath until solution cooled with recirculator 1. it reaches 15'20 OC- As soon as the internal ature Homogenizer is switched off once reached 2 °C above the target arrest the solution crystallizes. temperature, the recirculating water is switched to recirculator 2 in order to arrest the cooling.

As soon as the ature started to rise the time the recirculating water is switched back to recirculator 1 until the slurry reaches 15-20 °C. nizer is switched off 90 s after the start of the ature rise, after which it is replaced with an ad mixer.

The modified 200 gram process was used successfully in four aripiprazole lauroxil recrystallization batch campaigns (each consisting of five recrystallization batches). A particle size distribution similar to that from the (unmodified) 200 gram process was ucibly ed as seen in . Summary statistics are listed in Table 9. A comparison between the le size distribution tics of recrystallized aripiprazole lauroxil from the modified process versus that from the unmodified process revealed a tighter spread in DV[10], DV[SO], and DV[90], clearly showing an improvement in process robustness and reproducibility.

WO 32950 Table 9: Particle size distribution summary statistics of four batch campaigns mDv[10], pm Dv[50], pm Dv[90], pm M1 5 21 36 1.75 kg scale Aripiprazole lauroxil was recrystallized using the following procedure. 2156.0 g of isopropyl acetate was added to a recrystallization vessel containing 1750.0 g of aripiprazole lauroxil.

The mixture was heated under agitation to 55-65 0C. When the drug was visibly dissolved in solution, 3351.0 g of heptane heated to 55-65 °C was added to the recrystallization vessel. The resulting mixture was heated to 60-65 °C, at which point cold glycol was introduced into the jacket of the recrystallization vessel in order to cool the mixture. When the temperature of the mixture of isopropyl acetate, e, and razole lauroxil reached 34 oC, homogenization was initialized. The temperature was continuously monitored for the onset of the exotherm (start of precipitation or llization) and the exotherm maximum. When the e temperature reached the value Tmin after the exotherm (Tming), homogenation was stopped.

More cold glycol was introduced into the vessel jacket in order to cool the mixture to 18 0C, at which point the mixture was held for 5 minutes.

Then, hot glycol was introduced into the vessel jacket to reheat the mixture toward 60-65 °C, at which point cold glycol was again introduced into the jacket of the tallization vessel in order to cool the mixture. When the temperature of the mixture of isopropyl acetate, heptane, and aripiprazole il reached 34 °C, homogenization was initialized. The ature was continuously monitored for the onset of the exotherm and the exotherm maximum. When the mixture temperature reached the value Tmin after the rm (Tminz), homogenation was stopped. More cold glycol was introduced into the vessel jacket in order to cool the e to 18 °C. The recrystallized aripiprazole lauroxil was filtered under vacuum in a dead end filter dryer and rinsed with 2187.0 g of heptanes at ambient temperature. The solids were dried under vacuum (80 torr) for 40 hours in the same vessel and collected. 4 kg scale Aripiprazole lauroxil was recrystallized using the ing procedure. 4940.0 g of isopropyl acetate was added to a recrystallization vessel containing 4000.0 g of aripiprazole lauroxil and mixture under agitation. When the drug was visibly ved in solution (2 55 0C), 7670.0 g of heptane heated to 55-65 °C was added to the recrystallization vessel. The resulting mixture was heated to 2 60 oC and then held for 5 minutes. Cold glycol (28 0C) was then introduced WO 32950 into the jacket of the recrystallization vessel in order to cool the mixture, and when the temperature of the mixture of isopropyl acetate, e, and aripiprazole lauroxil reached 33.8 c’C, homogenization was initialized. The temperature was continuously monitored for the onset of the exotherm (start of precipitation or crystallization) and the exotherm maximum. (If Tmin was less than 33 CC, then another cycle of recrystallization was performed.) When the e temperature d the value Tmin after the exotherm ), the homogenation was stopped. More cold glycol was introduced into the vessel jacket in order to cool the mixture to 18 °C.

Then, hot glycol was introduced into the vessel jacket to reheat the mixture toward 60-65 0C, at which point cold glycol was again uced into the jacket of the recrystallization vessel in order to cool the mixture. When the temperature of the e of isopropyl acetate, heptane, and aripiprazole lauroxil reached 33.8 °C, homogenization was initialized. The temperature was continuously monitored for the onset of the rm and the exotherm maximum. When the mixture temperature reached the value Tmin after the exotherm (Tmin2), homogenization was stopped. More cold glycol was introduced into the vessel jacket in order to cool the mixture to 18 °C. The recrystallized aripiprazole lauroxil was filtered under vacuum in a dead end filter dryer and rinsed with 9 kg of heptanes at ambient temperature. The solids were dried under vacuum (20 torr) for 20 hours in the same vessel and collected.

Examgle using sonication instead of homogenization Aripiprazole lauroxil (10 g) was dissolved in hot isobutyl acetate (14 mL). N—heptane (28 mL) was added to the hot solution and the mixture was heated fiirther to dissolve all solids. The hot solution was placed in a sonication bath and sonicated for 2 minutes. Ice was added to the sonication bath to cool down the mixture. White crystals were formed. The crystals were filtered using a Buchner funnel and washed with cold n-heptane (10 mL). The white solid was then dried under vacuum at room temperature overnight and resulted in 9.6 g of recrystallized razole il (96% yield).

Impact of razole lauroxil particle size distribution (PSD), surface area, and e on in vivo release profiles In order to explore the effect of injection vehicle on razole plasma exposure, a single— dose 1M muscular) rat dosing study was conducted. The closing amount in the study was 29 mg of recrystallized aripiprazole lauroxil prodrug, which is equivalent to 20 mg aripiprazole base. The following two formulations were prepared and dosed intramuscularly to male rats: (1) aripiprazole lauroxil bulk recrystallized drug substance suspended in a phosphate- ed saline injection vehicle with sodium carboxymethylcellulose (NaCMC) (2 wt%) and polysorbate 20 (0.2 wt%); and (2) aripiprazole lauroxil recrystallized drug substance suspended in a phosphate- ed saline injection vehicle using sorbitan monolaurate (SML) (0.5 wt%) and polysorbate (0.2 wt%). shows pharmacokinetic profiles of aripiprazole resulting from intramuscular administration of a single dose of recrystallized aripiprazole lauroxil (20 mg razole equivalents) suspended in either the NaCMC or SML vehicle, to male rats to assess the effect of injection vehicle on the in vivo profile. Two lots of recrystallized aripiprazole lauroxil drug nce were tested. Pharmacokinetic analysis showed that suspensions prepared from the same drug substance lot resulted in essentially overlapping in vivo pK profiles of aripiprazole, independent of injection vehicle. The pK parameters, summarized in Table 10, indicated that the injection vehicle did not significantly impact Cmax, Tmax, or AUC0_T1ast.

Table 10: pK parameters from a single IM administration of recrystallized aripiprazole lauroxil in male rats a b AUCo “a“_ drug substance lot, vehicle Cmax (ng/mL) Tmx (day) (dav*ng/mL) Lot A, suspended in NaCMC 34.2 1r 6.21 13.8 i 3.66 735 i 82.7 vehicle, Lot A, suspended in SML 43.1 1r 8.60 16.0 i 1.55 836 i 73.0 vehicle.

Lot B, suspended in NaCMC 22.7 i 1.17 17.2 i 5.74 638 i 53.7 vehicle_ Lot B, suspended in SML 23.4 1r 6.47 17.2 i 3.66 643 1r 146 vehicle. aCmaxz The maximum precipitated plasma concentration observed. bTmax: Time at which Cmax occurred.

CAUCMaSg Area under the precipitated plasma tration-time curve from Time 0 to the last measured precipitated plasma concentration.

The impact of the particle size bution (PSD) and surface area on aripiprazole pK was further investigated using four lots of recrystallized aripiprazole lauroxil in the rat IM pK model at the same dose of 29 mg aripiprazole lauroxil equivalent to 20 mg of razole.

PSD and surface area ements for the four lots are presented in Table 11, and the PSD profiles are illustrated in .

Table 11: PSD and surface area measurements for aripiprazole lauroxil bulk recrystallized drug nce lots Surface area a a a Lat # Dv[10] um Dv[50] um Dv[90] um (mZ/g) aFor a single preparation and measurement, PSD method error for the volume metrics are: Dv[10]=i0.9 pm, : i3.5 pm, and Dv[90]: $5.7 pm. For an average of three preparations and measurements, PSD method error would be Dv[10]=i0.5 pm, Dv[50]= $2.0 pm, and Dv[90]: i3.3 um.

To examine the influences of the PSD and surface area on aripiprazole pK, the characterized lots of recrystallized aripiprazole lauroxil listed in Table 11 were suspended in SML vehicle and administered by IM administration to male rats. The aripiprazole pK profiles are plotted in , and the pK parameters are presented in Table 12. These data show that material with a smaller PSD and sed surface area (i.e., lot X3) will result in faster release rates (, top—most curve). (Release rate is expressed as razole plasma exposure (dependent on both the rate of dissolution of recrystallized aripiprazole lauroxil and the rate of esterase mediated conversion of aripiprazole lauroxil to aripiprazole). Material with larger PSD and decreased surface area (i.e., lot X2) will result in slower e rates (, bottom-most curve). Importantly, ation of the calculated AUCinf indicates that systemic exposure was similar for all groups in this experiment.

Table 12: Aripiprazole pK parameters following a single IM administration of tallized aripiprazole lauroxil suspended in SML injection vehicle in male rats Surface AUCo-tlast area (ml/g) g/mL) aCmaxz The maximum precipitated plasma concentration observed. bTmax: Time at which Cmax occurred.

CAUCo_t|ast: Area under the precipitated plasma concentration—time curve from Time 0 to the last measured precipitated plasma tration. dAUCinf: Area under the precipitated plasma concentration-time curve from Time 0 to infinity.

The conclusions from these experiments are that both PSD and surface area measurements were important aspects of the physical ty characterization of the aripiprazole lauroxil drug ls. Consistent with a release mechanism that is ted by crystal dissolution, the data obtained from these pK studies highlight the rank order with regard to surface areas and pK profiles. This ordering is consistent with the character of the insoluble prodrug crystals, namely, that the particle size distribution and surface area of aripiprazole lauroxil are the key attributes influencing in vivo performance.

In conclusion, as demonstrated in the pK studies described above, the mance of recrystallized aripiprazole lauroxil drug product was dominated by the physical properties of the product crystals. Dissolution of aripiprazole lauroxil following injection was d by slow dissolution of the drug crystals, and was a function of the amount of exposed surface area of the aripiprazole il material. The particle size distribution and surface area of aripiprazole lauroxil were the key attributes influencing in viva mance.

Two-pass recrystallization A two-pass recrystallization process was developed to fithher improve reproducibility and particle size control.

As bed earlier, crystallization of aripiprazole lauroxil occurs after cooling the mixture of the drug, the first solvent (such as isopropyl acetate), and the second t (such as n- heptane) to a supersaturated condition. Control of the solution temperature to target a specific onset temperature (exotherm mum” or “Tmin”) for crystallization is important to control the final particle size distribution and surface area of the aripiprazole lauroxil crystals.

Nucleation and crystallization can be induced by initiating high-shear mixing as the supersaturated mixture approaches a target temperature. The recrystallization process of the present invention reproducibly produces crystals of aripiprazole lauroxil with desirable particle size bution and surface area parameters.

Subsequent to the studies described above ing one-pass recrystallization or a single pass of tallization, additional studies were performed in ultra-clean equipment.

In sterile pharmaceutical manufacturing, processing is ted in ultra-clean equipment to ensure quality and reduce ination. Equipment is cleaned and steam sterilized in place before use. Surface finish is controlled to be very smooth to aid in cleaning. After cleaning and during use, equipment must be kept totally closed to the environment to prevent contamination. A first pass of recrystallization may not always behave as predictably as desired. For instance, variation in llization onset time or crystallization temperature may be unacceptably large, resulting in tallized particles that have sub-optimal particle size distribution and surface area parameters. Moreover, since the process is conducted in totally closed equipment, conventional means of adding solid crystals to facilitate crystallization can be highly impractical.

Accordingly, a process to fiarther facilitate reproducible recrystallization of razole lauroxil was developed. ls of aripiprazole lauroxil were first formed by cooling and itating at supersaturated conditions with or without homogenization. The solution was then re-warmed and the ls were solved. When the solution was re-cooled, crystals were precipitated in a reliable , with the aid of high—shear mixing as the target temperature was ched, much as in the one-pass crystallization process described earlier.

In replicate experiments, the first pass recrystallization resulted in large variation in time from homogenizer onset to crystal formation. Time to crystallization ranged from about 1 minute to over 37 minutes. A larger variation in crystallization Tmin was noted. Results of first pass recrystallization are shown in Table 13.

TABLE 13 Induction time (Crystallization Homogenize Crystallization onset time — r ON Crystallization Homogenizer onset (Tmin) homogenizer pass ON time time on time) temeperatur After seeding, the subsequent second pass of recrystallization occurred quickly and reproducibly around 1 minute or less afler homogenization onset. A much smaller variation in Tmin was achieved. Results of the second pass of recrystallization are shown in Table 14. 2015/046525 TABLE 14 Induction time (Crystallization nizer Crystallization onsettime— ON Sample Crystallization Homogenizer Tmin) homogenizer temperature, pass ON time time on time) Tmcin, While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood in light of the present disclosure by those skilled in the art that various changes in form and details may be made therein Without departing from the scope of the invention encompassed by the appended claims. 572570 ALT-024PC

Claims (19)

What is claimed is:

1. A process for making the compound of Formula (I) in crystal form Formula (I), 5 the process comprising the steps of: (a) obtaining a drug solution by combining the compound of Formula (I) or a salt or solvate thereof with a first solvent; (b) optionally ing the drug solution with a second solvent to form a mixture; (c) g the mixture; and 10 (d) when the temperature of the mixture is within the range of 31 °C to 43 °C, homogenizing the mixture to form crystallized particles of the compound of Formula (I) having a surface area of 0.50 to 3.3 m2/g; wherein the first solvent is ed from the group consisting of ethyl acetate, propyl acetate, isopropyl acetate, butyl acetate, isobutyl acetate, tert-butyl e, 15 acetone, and a mixture of isopropyl acetate and n-heptane; and the second solvent is selected from the group consisting of pentane, entane, hexane, cyclohexane, methyl cyclohexane, heptane, octane, nonane, decane, undecane, dodecane, ethanol, and methanol.

2. The process of claim 1, r comprising the steps of: 20 (e) stopping homogenization, and re-dissolving the crystallized particles of the compound of Formula (I) by heating the mixture; (f) cooling the mixture; and (g) when the temperature of the e is within the range of 31 °C to 43 °C, homogenizing the mixture to form llized particles of the compound of 25 Formula (I) having a surface area of 0.50 to 3.3 m2/g. 572570 ALT-024PC

3. The process of claim 1 or claim 2, wherein the crystallized les have a surface area of 0.80 to 1.1 m2/g.

4. The process of claim 1 or claim 2, wherein the crystallized particles have a surface area of 1.00 m2/g.

5 5. The process of claim 1 or claim 2, wherein the Dv[50] of the crystallized particles is 10 to 30 microns.

6. The process of claim 5, wherein the Dv[50] of the llized particles is 10 to 20 microns.

7. The process of any one of claims 1 to 6, wherein the first solvent is isopropyl 10 acetate.

8. The process of any one of claims 1 to 6, wherein the first solvent is a e of isopropyl acetate and n-heptane.

9. The process of any one of claims 1 to 8, wherein the second solvent is n-heptane.

10. The process of any one of claims 1 to 9, n in step (b), the temperature of the 15 mixture is in the range of 55 °C to 65 °C.

11. The process of any one of claims 1 to 10, wherein the temperature in step (d) is in the range of 31 °C to 35 °C.

12. The process of any one of claims 1 to 11, wherein the temperature in step (d) is 34 20

13. The process of claim 12, wherein in step (d), the homogenizing begins when the temperature of the mixture is 0 °C to 4 °C above 34 °C.

14. The process of any one of claims 1 to 13, wherein one or more of steps (a) through (c) is performed under agitation.

15. The process of any one of claims 2 to 13, wherein one or more of steps (a), (b), (c), 25 (d), (e), (f), and (g) is performed under agitation. 572570 ALT-024PC

16. The process of any one of claims 1 to 15, further sing the step of filtering the crystallized particles.

17. The process of claim 16, further comprising the step of rinsing the crystallized particles. 5

18. The s of claim 17, further comprising the step of drying the crystallized particles.

19. The process of any one of claims 1 to 10 and 14 to 18, wherein the temperature in step (d) is in the range of 31 °C to 38 °C.