Ipragliflozin

Ipragliflozin

ASP-1941 ,   1(S)-[3-(1-benzothien-2-ylmethyl)-4-fluorophenyl]-1-deoxy-beta-D-glucopyranose L-proline cocrystal

Kotobuki (Originator)

Ipragliflozin (formerly ASP1941) has been filed in Japan on the back of phase III trials which showed that it could provide significant reductions in glycated haemoglobin levels (HbA1c) levels – a marker of glucose control over time – compared to placebo

According to Astellas’ latest R&D pipeline update in February 2013, Astellas is developing ipragliflozin only in Japan. The same document in August 2012 indicated it was also carrying out phase II studies with the drug in the US and Japan.

Astellas Pharma Inc.: Submits Application for Marketing Approval of
Ipragliflozin (ASP1941), SGLT2 Inhibitor for Treatment of
Type 2 Diabetes, in Japan
TOKYO, March 13, 2013 – Astellas Pharma Inc. (“Astellas”; Tokyo:4503; President and CEO:
Yoshihiko Hatanaka) announced today that it has submitted a market authorization application for aSGLT2 inhibitoripragliflozin (generic name; development code: ASP1941) to the Ministry of Health, Labour and Welfare in Japan seeking an approval forthe indication of type 2 diabetes.
Ipragliflozin is a selective SGLT2 (sodium-glucose co-transporter 2)inhibitor discovered through research collaboration with Kotobuki Pharmaceutical Co., Ltd. SGLTs are membrane proteins that
exist on the cell surface and transfer glucose into cells. SGLT2 is a subtype of the sodium-glucose co-transporters and plays a key role in the reuptake of glucose in the proximal tubule of the kidneys.
Ipragliflozin reduces blood glucose levels by inhibiting the reuptake of glucose.
In the Phase III pivotal study in monotherapy for type 2 diabetesin Japan, ipragliflozin
demonstrated significant decreases of HbA1c, an index of glycemic control, in change from baseline compared to placebo. Based on the safety resultsin this study, ipragliflozin was safe and well tolerated. Patients with type 2 diabetes generally need combination therapy, so it is important
for a novel oral hypoglycemic agent to be safe to use with existing diabetes therapies. In this regard, Astellas has conducted six Phase III studies to investigate the safety and efficacy of ipragliflozin
used in combination with other hypoglycemic agentsfor a long term period. In these Phase IIIstudies, effectiveness and favorable safety of ipragliflozin was confirmed even in combination with
other hypoglycemic agents.
Astellas expects to provide an additional therapeutic option and further contribute to the treatment of type 2 diabetes by introducing ipragliflozin, an oral hypoglycemic agent with a novel mechanism
of action, into the Japanese market.
About Type 2 Diabetes
Diabetes (medically known as diabetes mellitus) is a disorder in which the body has difficulty regulating its blood glucose (sugar) level. There are two major types of diabetes: type 1 and type 2.
Type 2 diabetes (formerly called non-insulin-dependent diabetes mellitus or adult-onset diabetes) is a disorder that is characterized by high blood glucose in the context of insulin resistance and relative insulin deficiency. Patients are instructed to increase exercise and diet restrictions, but most
require treatment with an anti-diabetic agent to control blood glucose.

structure:

The compound and methods of its synthesis are described in WO 2004/080990, WO 2005/012326 and WO 2007/114475 for example.

The gluconolactone method: In 1988 and 1989 a general method was reported to prepare C-arylglucosides from tetra-6>-benzyl protected gluconolactone, which is an oxidized derivative of glucose (see J. Org. Chem. 1988, 53, 752-753 and J. Org. Chem. 1989, 54, 610- 612). The method comprises: 1) addition of an aryllithium derivative to the hydroxy-protected gluconolactone to form a hemiketal (a.k.ci., a lactol), and 2) reduction of the resultant hemiketal with triethylsilane in the presence of boron trifluoride etherate. Disadvantages of this classical, but very commonly applied method for β-C-arylglucoside synthesis include:

1) poor “redox economy” (see J. Am. Chem. Soc. 2008, 130, 17938-17954 and Anderson, N. G. Practical Process Research & Development, 1st Ed.; Academic Press, 2000 (ISBN- 10: 0120594757); pg 38)— that is, the oxidation state of the carbon atom at CI, with respect to glucose, is oxidized in the gluconolactone and then following the arylation step is reduced to provide the requisite oxidation state of the final product. 2) due to a lack of stereospecificity, the desired β-C-arylglucoside is formed along with the undesired a-C-arylglucoside stereoisomer (this has been partially addressed by the use of hindered trialkylsilane reducing agents (see Tetrahedron: Asymmetry 2003, 14, 3243-3247) or by conversion of the hemiketal to a methyl ketal prior to reduction (see J. Org. Chem. 2007, 72, 9746-9749 and U.S. Patent 7,375,213)).

Oxidation Reduction

Glucose Gluconoloctone Hemiketal a-anomer β-anomer

R = protecting group

The metalated glucal method: U.S. Patent 7,847,074 discloses preparation of SGLT2 inhibitors that involves the coupling of a hydroxy-protected glucal that is metalated at CI with an aryl halide in the presence of a transition metal catalyst. Following the coupling step, the requisite formal addition of water to the C-arylglucal double bond to provide the desired C-aryl glucoside is effected using i) hydroboration and oxidation, or ii) epoxidation and reduction, or iii) dihydroxylation and reduction. In each case, the metalated glucal method represents poor redox economy because oxidation and reduction reactions must be conducted to establish the requisite oxidation states of the individual CI and C2 carbon atoms.

U.S. Pat. Appl. 2005/0233988 discloses the utilization of a Suzuki reaction between a CI -boronic acid or boronic ester substituted hydroxy-protected glucal and an aryl halide in the presence of a palladium catalyst. The resulting 1- C-arylglucal is then formally hydrated to provide the desired 1- C-aryl glucoside skeleton by use of a reduction step followed by an oxidation step. The synthesis of the boronic acid and its subsequent Suzuki reaction, reduction and oxidation, together, comprise a relatively long synthetic approach to C-arylglucosides and exhibits poor redox economy. Moreover, the coupling catalyst comprises palladium which is toxic and therefore should be controlled to very low levels in the drug substance.

R = protecting group; R’ = H or alkyl

The glucal epoxide method: U.S. Patent 7,847,074 discloses a method that utilizes an organometallic (derived from the requisite aglycone moiety) addition to an electrophilic epoxide located at C1-C2 of a hydroxy-protected glucose ring to furnish intermediates useful for SGLT2 inhibitor synthesis. The epoxide intermediate is prepared by the oxidation of a hydroxy- protected glucal and is not particularly stable. In Tetrahedron 2002, 58, 1997-2009 it was taught that organometallic additions to a tri-6>-benzyl protected glucal-derived epoxide can provide either the a-C-arylglucoside, mixtures of the a- and β-C-arylglucoside or the β-C-arylglucoside by selection of the appropriate counterion of the carbanionic aryl nucleophile (i.e., the

organometallic reagent). For example, carbanionic aryl groups countered with copper (i.e., cuprate reagents) or zinc (i.e., organozinc reagents) ions provide the β-C-arylglucoside, magnesium ions provide the a- and β-C-arylglucosides, and aluminum (i.e., organoaluminum reagents) ions provide the a-C-arylglucoside.

or Zn

The glycosyl leaving group substitution method: U.S. Patent 7,847,074, also disclosed a method comprising the substitution of a leaving group located at CI of a hydroxy-protected glucosyl species, such as a glycosyl halide, with a metalated aryl compound to prepare SGLT2 inhibitors. U.S. Pat. Appl. 2011/0087017 disclosed a similar method to prepare the SGLT2 inhibitor canagliflozin and preferably diarylzinc complexes are used as nucleophiles along with tetra- >-pivaloyl protected glucosylbromide.

Glucose Glucosyl bromide β-anomer

Methodology for alkynylation of 1,6-anhydroglycosides reported in Helv. Chim. Acta. 1995, 78, 242-264 describes the preparation of l,4-dideoxy-l,4-diethynyl^-D-glucopyranoses (a. La., glucopyranosyl acetylenes), that are useful for preparing but-l,3-diyne-l,4-diyl linked polysaccharides, by the ethynylating opening (alkynylation) of partially protected 4-deoxy-4-C- ethynyl-l,6-anhydroglucopyranoses. The synthesis of β-C-arylglucosides, such as could be useful as precursors for SLGT2 inhibitors, was not disclosed. The ethynylation reaction was reported to proceed with retention of configuration at the anomeric center and was rationalized (see Helv. Chim. Acta 2002, 85, 2235-2257) by the C3-hydroxyl of the 1,6- anhydroglucopyranose being deprotonated to form a C3-0-aluminium species, that coordinated with the C6-oxygen allowing delivery of the ethyne group to the β-face of the an oxycarbenium cation derivative of the glucopyranose. Three molar equivalents of the ethynylaluminium reagent was used per 1 molar equivalent of the 1,6-anhydroglucopyranose. The

ethynylaluminium reagent was prepared by the reaction of equimolar (i.e., 1:1) amounts of aluminum chloride and an ethynyllithium reagent that itself was formed by the reaction of an acetylene compound with butyllithium. This retentive ethynylating opening method was also applied (see Helv. Chim. Acta. 1998, 81, 2157-2189) to 2,4-di-<9-triethylsilyl- 1,6- anhydroglucopyranose to provide l-deoxy-l-C-ethynyl- -D-glucopyranose. In this example, 4 molar equivalents of the ethynylaluminium reagent was used per 1 molar equivalent of the 1,6- anhydroglucopyranose. The ethynylaluminium regent was prepared by the reaction of equimolar (i.e., 1: 1) amounts of aluminum chloride and an ethynyl lithium reagent that itself was formed by reaction of an acetylene compound with butyllithium.

It can be seen from the peer-reviewed and patent literature that the conventional methods that can be used to provide C-arylglucosides possess several disadvantages. These include (1) a lack of stereoselectivity during formation of the desired anomer of the C- arylglucoside, (2) poor redox economy due to oxidation and reduction reaction steps being required to change the oxidation state of CI or of CI and C2 of the carbohydrate moiety, (3) some relatively long synthetic routes, (4) the use of toxic metals such as palladium, and/or (5) atom uneconomic protection of four free hydroxyl groups. With regard to the issue of redox economy, superfluous oxidation and reduction reactions that are inherently required to allow introduction of the aryl group into the carbohydrate moiety of the previously mention synthetic methods and the subsequent synthetic steps to establish the required oxidation state, besides adding synthetic steps to the process, are particular undesirable for manufacturing processes because reductants can be difficult and dangerous to operate on large scales due to their flammability or ability to produce flammable hydrogen gas during the reaction or during workup, and because oxidants are often corrosive and require specialized handling operations (see Anderson, N. G. Practical Process Research & Development, 1st Ed.; Academic Press, 2000 (ISBN-10: 0120594757); pg 38 for discussions on this issue).

• The C-glycoside derivative represented by the formula (1) and its salt [hereinafter, they are referred to as "compound (1)" or "compound of formula (1)" in some cases] is known to be useful for treatment and prevention of diabetes such as insulin-dependent diabetes (type 1 diabetes), non-insulin-dependent diabetes (type 2 diabetes) and the like and various diabetes-related diseases including insulin-resistant diseases and obesity (Patent Literature 1).
• The method for producing the C-glycoside derivative represented by the formula (1), described in the Patent Literature 1 is understood to be represented by the below-shown reaction formula (I), by referring to the Examples and Reference Examples, described in the Patent Literature 1. Roughly explaining, it is a method which comprises reacting [1-benzothien-2-yl(5-bromo-2-fluorophenyl)methoxy]tert-butyl)dimethylsilane (synthesized in accordance with Reference Example 37 of the Literature) in a manner shown in Example 65 of the Literature, to obtain (1S)-1,5-anhydro-1-[3-(1-benzothien-2-ylmethyl)-4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucitol and then reacting the obtained compound in accordance with Example 100 of the Literature to synthesize intended (1S)-1,5-anhydro-1-C-[3-(1-benzothiophene-2-ylmethyl)-4-fluorophenyl]-D-glucitol.
• However, the method for producing the C-glycoside derivative of the formula (1), disclosed in the Patent Literature 1 is not industrially satisfactory in yield and cost, as is seen in later-shown Reference Example 1 of the present Description.
• For example, as described later, the method includes a step of low product yield (for example, a step of about 50% or lower yield) and the overall yield of the C-glycoside derivative (final product) represented by the formula (1) from the compound (8) (starting raw material) is below 7%; therefore, the method has problems in yield and cost from the standpoint of medicine production and has not been satisfactory industrially. In addition, the method includes an operation of purification by column chromatography which uses chloroform as part of purification solvents; use of such a solvent poses a problem in environmental protection and there are various restrictions in industrial application of such an operation; thus, the method has problems in providing an effective medicine.
• Also, an improved method of conducting an addition reaction with trimethylsilyl carbohydrate instead of benzyl carbohydrate and then conducting deprotection for acetylation, is known for a compound which has a structure different from that of the compound of the formula (1) but has a structure common to that of the compound of the formula (1) (Patent Literature 2). It is described in the Patent Literature 2 that the improved method enhances the overall yield to 6.2% from 1.4%. Even in the improved method, however, the yield is low at 6.2% which is far from satisfaction in industrial production.

  • Patent Literature 1: WO 2004/080990 Pamphlet
  • Patent Literature 2: WO 2006/006496 Pamphlet

http://www.google.com/patents/EP2105442A1

First step: synthesis of 1-benzothien-2-yl(5-bromo-2-fluorophenyl)methanol
• Into a tetrahydrofuran (20 ml) solution of benzo[b]thiophene (5.0 g) was dropwise added a n-hexane solution (25 ml) of n-butyl lithium (1.58 M) at -78°C in an argon atmosphere, followed by stirring at -78°C for 10 minutes. Into this solution was dropwise added a tetrahydrofuran (80 ml) solution of 5-bromo-2-fluorobenzaldehyde (8.0 g), followed by stirring at -78°C for 2.5 hours. The temperature of the reaction mixture was elevated to room temperature. Water was added thereto, followed by extraction with ethyl acetate. The organic layer was washed with a saturated aqueous sodium chloride solution, dried over anhydrous magnesium sulfate, filtered, and concentrated. The residue was purified by silica gel column chromatography (n-hexane/ethyl acetate) to obtain 1-benzothien-2-yl(5-bromo-2-fluorophenyl)methanol (10.5 g, yield: 83.6%).
  ¹H-NMR (CDCl₃): δ
  2.74 (1H, d), 6.35 (1H, d), 6.93 (1H, dd), 7.14 (1H, s), 7.27-7.38 (2H, m), 7.39 (1H, m), 7.68 (1H, dd), 7.74 (2H, m)

Second step: synthesis of [1-benzothien-2-yl(5-bromo-2-fluorophenyl)methoxy](tert-butyl)dimethylsilane

• To a dimethylformamide (20 ml) solution of 1-benzothien-2-yl(5-bromo-2-fluorophenyl)methanol (5.0 g) were added imidazole (1.3 g), a catalytic amount of 4-(dimethylamino)pyridine and tert-butyldimethylchlorosilane (2.7 g), followed by stirring at room temperature for 7 hours. To the reaction mixture was added a saturated aqueous ammonium chloride solution, followed by extraction with ethyl acetate. The organic layer was washed with a saturated aqueous ammonium chloride solution and a saturated aqueous sodium chloride solution, dried over anhydrous magnesium sulfate, filtered and concentrated. The residue was purified by silica gel column chromatography (n-hexane/ethyl acetate) to obtain [1-benzothien-2-yl(5-bromo-2-fluorophenyl)methoxy](tert-butyl)dimethylsilane (5.22 g, yield: 78.0%).
  MS: 451 (M⁺)
  ¹H-NMR (CDCl₃): δ
  0.05 (3H, s), 0.11 (3H, s), 0.95 (9H, s), 6.34 (1H, s), 6.91 (1H, t), 7.08 (1H, d), 7.23-7.38 (2H, m), 7.64-7.68 (1H, m), 7.75-7.28 (2H, m)

Third step: Synthesis of 1-C-[3-(1-benzothien-2-yl{[tert-butyl-(dimethyl)silyloxy}methyl)4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucopyranose

• Into a tetrahydrofuran (15 ml) solution of [1-benzothien-2-yl(5-bromo-2-fluorophenyl)methoxy](tert-butyl)dimethylsilane (1.5 g) was dropwise added a n-hexane solution (2.2 ml) of n-butyl lithium (1.58 M) in an argon atmosphere at -78°C, followed by stirring at -78°C for 30 minutes. Into the solution was dropwise added a tetrahydrofuran (20 ml) solution of 2,3,4,6-tetra-O-benzyl-D-glucono-1,5-lactone (1.9 g), followed by stirring at -78°C for 15 minutes and then at 0°C for 1.5 hours. To the reaction mixture was added a saturated aqueous ammonium chloride solution, followed by extraction with ethyl acetate. The organic layer was washed with a saturated aqueous ammonium chloride solution and a saturated aqueous sodium chloride solution, dried over anhydrous magnesium sulfate, filtered and concentrated. The residue was purified by silica gel column chromatography (n-hexane/chloroform/acetone) to obtain 1-C-[3-(1-benzothien-2-yl{[tert-butyl-(dimethyl)silyloxy}methyl)-4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucopyranose (1.52 g, yield: 50.2%). MS: 933 (M+Na)

Fourth step: Synthesis of 1-C-{3-[1-benzothien-2-yl(hydroxy)methyl]-4-fluorophenyl}-2,3,4,6-tetra-O-benzyl-D-glucopyranose

• To a tetrahydrofuran (15 ml) solution of 1-C-[3-(1-benzothien-2-yl{[tert-butyl-(dimethyl)silyloxy}methyl)-4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucopyranose (1.52 g) was added a tetrahydrofuran solution (2.0 ml) of tetrabutylammonium fluoride (1.0 M), followed by stirring at room temperature for 1 hour. The reaction mixture was concentrated per se. The residue was purified by silica gel column chromatography (n-hexane/ethyl acetate) to obtain 1-C-{3-[1-benzothien-2-yl(hydroxy)methyl]-4-fluorophenyl}-2,3,4,6-tetra-O-benzyl-D-glucopyranose (0.99 g, yield: 74.7%). MS: 819 (M+Na), 779 (M+H-H₂O)

Fifth step: Synthesis of (1S)-1,5-anhydro-1-[3-(1-benzothien-2-ylmethyl)-4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucitol

• To an acetonitrile (5.0 ml) solution of 1-C-{3-[1-benzothien-2-yl(hydroxy)methyl]-4-fluorophenyl}-2,3,4,6-tetra-O-benzyl-D-glucopyranose (500 mg) were added triethylsilane (175 mg) and boron trifluoride-diethyl ether complex (196 mg) in an argon atmosphere at -20°C, followed by stirring at -20°C for 5 hours. To the reaction mixture was added a saturated aqueous sodium bicarbonate solution, followed by extraction with chloroform. The organic layer was washed with a saturated aqueous sodium bicarbonate solution and a saturated aqueous sodium chloride solution, dried over anhydrous magnesium sulfate, filtered and concentrated. The residue was purified by silica gel column chromatography (n-hexane/ethyl acetate) to obtain (1S)-1,5-anhydro-1-[3-(1-benzothien-2-ylmethyl)-4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucitol (150 mg, yield: 30.2%) MS: 787 (M+Na)
  ¹H-NMR (CDCl₃): δ
  3.42-3.48 (1H, m), 3.55-3.58 (1H, m), 3.72-3.78 (4H, m), 3.83 (1H, d), 4.14-4.30 (3H, m), 4.39 (1H, d), 4.51-4.67 (4H, m), 4.83-4.94 (2H, m), 6.86-6.90 (1H, m), 6.98 (1H, brs), 7.06-7.37 (24H, m), 7.57-7.60 (1H, m), 7.66-7.69 (1H, m)

Sixth step: Synthesis of (1S)-1,5-anhydro-1-C-[3-(1-benzothiophene-2-ylmethyl)-4-fluorophenyl]-D-glucitol

• To a dichloromethane (10 ml) solution of (1S)-1,5-anhydro-1-[3-(1-benzothien-2-ylmethyl)-4-fluorophenyl]-2,3,4,6-tetra-O-benzyl-D-glucitol (137 mg) were added pentamethylbenzene (382 mg) and a n-heptane solution (0.75 ml) of boron trichloride (1.0 M) in an argon atmosphere at -78°C, followed by stirring at -78°C for 3 hours. Methanol was added to the reaction mixture, the temperature of the resulting mixture was elevated to room temperature, and the mixture was concentrated per se. The residue was purified by silica gel column chromatography (chloroform/methanol) to obtain (1S)-1,5-anhydro-1-C-[3-(1-benzothiophene-2-ylmethyl)-4-fluorohenyl]-D-glucitol  OR IPRAGLIFLOZIN (63 mg, yield: 87.8%).
  ¹H-NMR (CD₃OD): δ
  3.29-3.48 (4H, m), 3.68 (1H, dd), 3.87 (1H, dd), 4.11 (1H, d), 4.20-4.29 (2H, m), 7.03 (1H, s), 7.08 (1H, dd), 7.19-7.29 (2H, m), 7.35 (1H, m), 7.42 (1H, dd), 7.64 (1H, d), 7.72 (1H, d)

(1S)-1,5-anhydro-1-C-[3-(1-benzothiophene-2-ylmethyl)-4-fluorohenyl]-D-glucitol OR IPRAGLIFLOZIN

http://www.google.com/patents/EP2105442A1