`
`(19) World Intellectual Property
`Organization
`International Bureau
`
`(43) International Publication Date
`30 September 2004 (30.09.2004)
`
`
`
`PCT
`
`(10) International Publication Number
`WO 2004/082618 A2
`
`(51) International Patent Classification’:
`
`A61K
`
`(21) International Application Number:
`PCT/US2004/007894
`
`(22) International Filing Date:
`
`16 March 2004 (16.03.2004)
`
`(25) Filing Language:
`
`(26) Publication Language:
`
`English
`
`English
`
`(30) Priority Data:
`10/390,526
`
`17 March 2003 (17.03.2003)
`
`US
`
`(71) Applicant (for all designated States except US): ASH
`STEVENSINC [US/US]; 5861 John C Lodge Freeway,
`Detroit, MI 48202 (US).
`
`(72) Inventors; and
`IONESCU, Du-
`(75) Inventors/Applicants (for US only):
`mitru [RO/US]; 2828 Grant Drive, Ann Arbor, MI 48108
`(US). BLUMBERGS,Peter [US/US]; 4105 Springer,
`Royal Oak, MI 48073 (US).
`
`(81) Designated States (unless otherwise indicated, for every
`kind of national protection available): AE, AG, AL, AM,
`AT, AU, AZ, BA, BB, BG, BR, BW, BY, BZ, CA, CH, CN,
`CO, CR, CU, CZ, DE, DK, DM, DZ, EC, EE, EG, ES, FI,
`GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE,
`KG,KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MA, MD,
`MG, MK, MN, MW,MX, MZ, NA, NI, NO, NZ, OM, PG,
`PH, PL, PT, RO, RU, SC, SD, SE, SG, SK, SL, SY, TJ, TM,
`TN, TR, TT, TZ, UA, UG, US, UZ, VC, VN, YU, ZA, ZM,
`ZW.
`
`(84) Designated States (unless otherwise indicated, for every
`kind of regional protection available): ARIPO (BW, GH,
`GM, KE, LS, MW, MZ, SD, SL, SZ, TZ, UG, ZM, ZW),
`Eurasian (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), Euro-
`pean (AT, BE, BG, CH, CY, CZ, DE, DK, EE, ES, FI, FR,
`GB, GR, HU,IE, IT, LU, MC, NL, PL, PT, RO, SE, SI, SK,
`TR), OAPI (BF, BJ, CE, CG, CI, CM, GA, GN, GQ, GW,
`ML, MR, NE, SN, TD, TG).
`
`Published:
`
`without international search report and to be republished
`upon receipt of that report
`
`(74) Agents: SWANSON & BRATSCHUNLLCetal.; 1745
`Shea Center Drive, Suite 330, Highlands Ranch, CO 80129
`(US).
`
`For two-letter codes and other abbreviations, refer to the "Guid-
`ance Notes on Codes and Abbreviations" appearing atthe begin-
`ning of each regular issue of the PCT Gazette.
`
`(54) Title: SYNTHESIS OF 5-AZACYTIDINE
`
`N
`
`NA
`LA
` (1)
`
`(57) Abstract: The present invention provides a method for the preparation of
`5-azacytidine, wherein 5-azacytidine is represented by the structure: The method
`involvesthesilylation of 5-azacytosine, followed by the coupling of silylated 5-aza-
`cytosine to a protected B-D-ribofuranose derivative. The coupling reaction is cat-
`alyzed by trimethylsilyl] trifluoromethanesulfonate (TMS-Triflate).
`
`CELGENE 2044
`CELGENE 2044
`APOTEX v. CELGENE
`APOTEX v. CELGENE
`IPR2023-00512
`IPR2023-00512
`
`
`
`WO2004/082618A2IMMUNIMIINATITMIEMMAITAAAi
`
`
`
`
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`SYNTHESIS OF 5-AZACYTIDINE
`
`FIELD OF THE INVENTION
`
`The invention relates to the synthesis of 5-azacytidine (also known as azacitidine and 4-
`
`amino-1-B-D-ribofuranosy!-S-triazin-2(14)-one). 5-azacytidine may be used in the treatment
`
`10
`
`of disease, including the treatment of myelodysplastic syndromes (MDS).
`
`BACKGROUND OF THE INVENTION
`
`5-azacytidine (also knownas azacitidine and 4-amino-1-B-D-ribofuranosyl-S-triazin-
`
`2(1H)-one; Nation Service Center designation NSC-102816; CAS Registry Number 320-67-2)
`
`15
`
`has undergone NCI-sponsoredtrials for the treatment of myelodysplastic syndromes (MDS).
`
`See Kornblithet al., J. Clin. Oncol. 20(10): 2441-2452 (2002) and Silverman et al., J. Clin.
`
`Oncol. 20(10): 2429-2440 (2002). 5-azacytidine may be defined as having a formula of
`
`CsHi2N4QOs, a molecular weight of 244.20 and a structure of
`
`20
`
`29
`
`LA
`
`
`
`The s-triazine ring of 5-azacytidine has a particular sensitivity to water (see J. A. Beisler,
`
`J. Med. Chem., 21, 204 (1978)); this characteristic has made the synthesis of 5-azacytidine a
`
`30
`
`challenge, especially in manufacturing at commercial scale. A numberofprior art methods
`
`have been developed in order to avoid the use of water; however, these methodsall have
`
`additional problemsthat render them undesirable for the productionof large-scale batches of
`
`-1-
`
`
`
`WO2004/082618
`
`PCT/US2004/007894
`
`5
`
`5-azacytidine. For example, Piskala and Sorm teach the following synthesis schemein (see
`
`United States Patent No 3,350,388; A. Piskala and F. Sorm, Collect. Czech. Chem. Commun.,
`
`29, 2060 (1964); and A. Piskala and F. Sorm, Ger. 1922702 (1969), each of which is
`
`incorporated herein by reference in its entirety):
`
`Ac
`
`N=C=O
`
`O
`
`AcO
`
`OAc
`
`Ae
`
`_
`‘HoN-C=NH 85 %
`
`=
`
`7R
`N=C—NH>
`|
`o NHe=0
`HC(OEt)3
`75%
`
`AcO
`
`OAc
`
`SR
`
`NH
`
`A
`AN
`WAN
`AJ
`cowed
`O
`Oy
`O
`MeoH 4
`68 %
`
`—>
`
`H
`
`H
`
`10
`
`The overall yield of this scheme is 43.3%. This method involves a reactive starting
`
`material (isocyanate) with a controlled stereochemistry (1-B configuration). Such a compound
`
`cannot be regarded as a starting material. The drawbacks of this schemeinclude the presence
`
`of steps that are difficult to scale-up, the use of benzeneas solvent in one step, and the
`
`requirement for a deprotection step performed in a closing pressure vessel using dry ammonia.
`
`15
`
`Furthermore, the final 5-azacytidine product was isolated from the reaction mixture by
`
`filtration with no further purification; this is not acceptable for the synthesis of an Active
`
`Pharmaceutical Ingredient (API) for human use. The addition of further purification steps will
`
`further reduce the overall yield.
`
`Winkley and Robins teach an 5-azacytidine synthesis processthat relies on the coupling
`
`20
`
`ofa"bromosugar" with a silyl derivative of 5-azacytosine (see M. W. Winkley and R. K.
`
`Robins, J. Org. Chem., 35, 491(1970), incorporated by reference in its entirety):
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`NH>
`
`NHTMS
`
`N~~n
`Aw HMDS
`Ay 3
`JS (NH,)2SO
`~*ruso
` N
`
`0”
`
`~N
`I
`
`0
`
`AC
`+
`
`AcO
`
`OAc
`
`a Ne
`
`NH»
`
`an NH
`ot MeOH
`
`————»
`
`O
`
`H
`
`AcO
`
`NH»
`
`f
`
`Oy
`es 3
`
`N
`
`O
`
`AcO
`
`OAc
`
`HO OH
`
`In this procedure, 5-azacytosine was treated with excess hexamethyldisilazane (HMDS)
`
`in the presence ofcatalytic amounts of ammonium sulfate at reflux until a complete solution
`
`was generated (TMS = (CH3)3Si). See E. Wittenburg, Z. Chem., 4, 303 (1964) for the general
`
`10
`
`procedure. The excess HMDS was removed by vacuum distillation and the residue was used
`
`directly (without further purification) in the coupling with 2,3,5-tri-O-acetyl-D-ribofuranosyl
`
`bromidein acetonitrile. The coupled product was deprotected with methanolic ammonia
`
`solution.
`
`There are many significant weaknesses in this procedure. First, the fact that the
`
`15
`
`bromosugar was a mixture of anomers, which meansthatthe final coupled product was also a
`
`mixture of anomers. Second, the work-up in the coupling step involved a great manysteps,
`
`specifically: concentration of the reaction mixture to dryness; treatment of the residue with
`
`sodium bicarbonate, water and methanol; removal of the water by co-evaporation with
`
`absolute ethanol; extraction of the residue with chloroform twice; and finally the concentration
`
`20
`
`to dryness of the combined chloroform extract. Third, ammonia in MeOH wasusedin the
`
`deprotection step, which requires the use of a pressure vessel. Fourth, the crude 5-azacytidine
`
`wasisolated in only a 35 % yield. This crude material was then dissolved in warm water and
`
`the solution was decolorized with charcoal. Evaporation then gave crystals of 5-azacytidine
`
`with a yield 11 %. This material was further recrystallized from aqueous ethanol(charcoal).
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`5
`
`The low recovery during purification can be correlated with the poor anomeric ratio and with
`
`the knownlowstability of 5-azacytidine in water.
`
`Piskala and Sormalso teach the following process for coupling involving the use of a
`
`"chlorosugar" (A. Piskala and F. Sorm, Nucl. Acid Chem. 1, 435 (1978), incorporated herein
`
`by reference in its entirety):
`
`10
`
`NH>
`
`NHTMS
`
`cl MeCN
`‘
`As J
`J ainse
`See
`SN
`(NH4)2S04
`
`O07 ~N~*57999, TMSO 65-70%
`I
`BzO
`OBz
`
`
`
`15
`
`NH»
`
`NH>
`
`Zz
`
`on
`nw MeONa
`okI MeOH ok
`
`BzO—— 0
`94-96 % HO—— O
`
`BzO OBz
`
`HO OH
`
`20
`
`2,3,5-Tri-O-Benzoyl-D-ribofuranosy] chloride was prepared by saturating a solution of
`
`1-O-acetyl-2,3,5-tri-O-benzoyl-B-D-ribose in CICH2CH2Cl-AcCl with gaseous HCl (with ice-
`
`cooling) and then keeping the mixture overnight at room temperature. This procedure is
`
`difficult to scale-up with plant equipment dueto the special handling requirements of gaseous
`
`HCl. Also, the typical a/B ratio in the chlorosugar is unknown,as is the impact of the a/B
`
`25
`
`ratio on the yield andfinal purity of 5-azacytidine.
`
`Piskala, Fiedler and Sormteach a procedure forthe ribosylation of silver salts of 5-
`
`azapyrimidine nucleobasesin A. Piskala, P. Fiedler and F. Sorm, Nucleic Acid Res., Spec.
`
`Publ. 1, 17 (1975), incorporated herein by referencein its entirety. Specifically, they teach
`
`that the ribosylation ofthesilver salt of 5-azacytosine with 2,4,5-tri-O-benzoyl-D-ribosyl
`
`30
`
`chloride gives 5-azacytidine. This is clearly not a procedure that is amenable to scale up for
`
`the large-scale production of 5-azacytidine.
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`Niedballa and Vorbriiggen teach the procedure that has been usedhistorically for the
`large-scale synthesis of 5-azacytidine for the above-mentioned NCI-sponsoredtrials for the
`
`treatment of myelodysplastic syndromes. See H. Vorbrtiggen and U. Niedballa, Ger.
`
`2,012,888 (1971) and U. Niedballa and H. Vorbriiggen, J. Org. Chem., 39, 3672 (1974), each
`
`of which is incorporated herein by reference in its entirety. The procedure involves the
`
`10
`
`following steps:
`
`NH
`
`—<<——>
`
`An
`on
`SnCl4
`O .
`EDCer BOL
`
`NHTMS
`Le
`
`TMSO~
`
`“N
`
`BzO
`
`0.
`
`OAc
`
`Bro
`
`OBz
`
`81%
`
`NHTMS
`|
`NN
`Be J
`Tuso~ SN
`
`
`
`SnCl,
`——* Ac
`MeCN
`50%
`
`beoO
`ry
`ZA
`o
`Oo
`
`N
`
`AcO OAc
`There are at least three major drawbacksto this procedure. First, and most importantly,
`
`after purification, variable amounts of tin from one batch to another were found in the API.
`
`The lack of control of the tin level means that the procedureis not suitable for producing an
`
`API for human use. Second, emulsions developed during the workup of the coupling mixture.
`
`15
`
`Indeed, H. Vorbriiggen and C. Ruh-Pohlenz in Organic Reactions, Vol. 55, , 2000 (L. A.
`
`Paquette Ed., John Wiley & Sons, New York), p 100, have previously noted that silylated
`
`heterocycles and protected 1-O-acyl or 1-O-alkyl sugars in the presence of Friedel-Crafts
`
`catalysts like SnCl, often form emulsions and colloids during work-up. The phase separation
`
`of the emulsion is slow, so the water-sensitive protected 5-azacytidine was exposed to water
`
`20
`
`for variable periods of time leading to variable amounts of decomposition. Third, a filtration
`
`step was performedin orderto isolate the insoluble tin salt. Typically, this filtration is very
`
`slow, andis likely the reason that variations in the final yield were noted. These problems
`
`mean that the process is not conveniently amenable to scale-up.
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`Vorbriiggen et al. in Chemische Berichte, 114: 1234-1255 (1981) teach the use of certain
`
`Lewis acids as Friedel-Crafts catalysts for the couplingofsilylated bases with 1-O-acyl
`
`sugars. In particular, they teach the coupling of silylated bases with 1-O-acyl sugars in the
`
`presence oftrimethylsilyl trifluoromethanesulfonate (TMS-Triflate) in 1,2-dichloroethane or
`
`acetonitrile. The reaction mixture was then diluted with dichloromethane and the organic
`
`10
`
`phase extracted with ice-cold saturated NaliCO3. The use of this procedure to synthesize 5-
`
`azacytidine is not taught or suggested.
`
`Vorbriiggen and Bennua in Chemische Berichte, 114: 1279-1286 (1981) also teach a
`
`simplified version of this nucleosidesynthesis method in which base silylation, generation of
`
`the Lewis acid Friedel-Crafts catalyst, and coupling of the silylated base to the 1-O-acyl sugar
`
`15
`
`takes place in a one step/one pot procedure employing a polar solvent such as acetonitrile.
`
`Following reaction, dichloromethane is added, and the mixture is extracted with aqeous
`
`NaHCO3. The use of this procedure to synthesize 5-azacytidine is not taught or suggested.
`
`Moreover,this one step/one pot reaction is not suitable for the synthesis of 5-azacytidine
`
`because the extraction is done is the presence of acetonitrile. Acetonitrile is a polar solvent,
`
`20
`
`and is therefore miscible with water. As a consequence, the protected 5-azacytidine in the
`
`acetonitrile is exposed during extraction to the aqueous phase for variable amounts of time,
`
`which in turn leads to variable arnounts of decomposition of the protected 5-azacytidine.
`
`Thus, there is an unmetneedin the field for the provision of a simple, controlled
`
`procedure for the synthesis of 5-azacytidine that provides an APIthatis suitable for use in
`
`25
`
`humans, minimizes the exposure of 5-azacytidine to water, and is amenable to scaling-up for
`
`the productionof large quantities of 5-azacytidine.
`
`SUMMARYOF THE INVENTION
`
`30
`
`The present invention providesfor the first time a method that synthesizes 5-azacytidine
`
`that is suitable for use in humansand is amenableto large scale synthesis.
`
`In oneseries of embodiments, 5-azacytidine is prepared by:
`
`a)
`
`reacting 5-azacytosine with a silylating reagent to yield a compoundofthestructure:
`
`
`
`WO2004/082618
`
`PCT/US2004/007894
`
`5
`
`HNSi(R1)3aN
`AZ
`
`(Ri)3SiO
`
`(A)
`
`10
`
`wherein each R;is an optionally substituted C-C29 alkyl group independently selected from
`the group consisting of straight chain alkyl groups, branched alkyl groups, and cyclic alkyl
`
`groups;
`
`b) coupling (A) with a compoundofthe structure:
`
`15
`
`R20
`
`(B)
`
`wherein each Ry is an optionally substituted C) - C29 acyl group independently selected from
`
`the group consisting ofstraight chain acyl groups, branched acyl groups, and benzoyl groups,
`
`20
`
`wherein the coupling of (A) and (B) is carried out in the presence of trimethylsilyl]
`
`trifluoromethanesulfonate (TMS-Triflate), and wherein the coupling yields a compoundofthe
`
`structure
`
`25
`
`30
`
`R20
`
`; and
`
`a)s
`aLJ
`
`
`Oo
`
`N
`
`"
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`5
`
`c) removing said Si(R,); and Rz groups from (C).
`
`In preferred embodiments,the silylating reaction takes place in the absence of a solvent
`
`using an excess ofsilylating reagent, and optionally in the presence of a catalyst. If a catalyst
`
`is used, a preferred catalyst is ammonium sulfate. Preferably the silylating reagent is a
`
`trimethylsilyl (TMS) reagent(i.c., Ri=CHs3), or a mixture of two or more TMSreagents in
`
`10
`
`excess over the 5-azacytosine. Preferred TMS reagents include hexamethyldisilizane (HMDS)
`
`and chlorotrimethylsilane (TMSCI). The silylated 5-azacytosine is preferably isolated prior to
`
`coupling by removingthe silylating reagents using vacuum distillation, or by filtration.
`
`Preferably, the compound (B) of coupling step b)is
`
`No QW
`
`A
`
`BzO.
`
`or
`
`wherein Bz=
`
`o
`
`2
`
`%,
`
`and the coupling reactionis carried out in a dry organic solvent, more preferably a dry organic
`
`non-polar solvent that is not miscible with water. Most preferably, the TMS-Triflate is
`
`quenched by extracting the reaction product of b) with, for example, an aqueous bicarbonate
`
`solution.
`
`30
`
`In another series of embodiments, a "one pot" synthesis of 5-azacytidine is provided
`
`comprising the stepsof:
`
`a)
`
`inadry organic solvent, reacting 5-azacytosine with one or moresilylating reagents to
`
`yield a compound having the structure;
`
`-8-
`
`
`
`WO2004/082618
`
`PCT/US2004/007894
`
`5
`
`(R1)3SiO
`
`HNSi(R1)3
`
`\—
`v7
`3
`
`N
`
`A)
`
`wherein each R;is an optionally substituted Cj-C29 alkyl group independently selected from
`
`the group consisting of straight chain alkyl groups, branched alkyl groups, and cyclic alkyl
`
`groups;
`
`10
`
`b) adding directly to the reaction mixture of a) TMS-Triflate and a compound having
`
`the structure
`
`(B)
`
`wherein each Ro is an optionally substituted C; - C29 acyl group independently selected
`
`15
`
`from the group consisting of straight chain acyl groups, branched acyl groups, and benzoyl
`
`group to yield a compound havingthe structure;
`
`HNSI(R4)s3
`
`A
`LI
`R,O
`
`o
`
`c)
`
`d)
`
`extracting the reaction mixture of b) with an aqueous quenchingsolution; and
`
`removing said Si(Ry)3 and Rz groups.
`
`20
`
`Preferably, the dry organic solvent of step a) is a polar solvent, most preferably
`
`acetonitrile. Preferably the polar solvent is removed between steps b) and c) and the reaction
`
`products of b) are dissolved in a dry organic non-polar solvent, most preferably
`
`dichloromethane of 1,2-dichlorethane, prior to step c).
`
`-9.
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`In some embodiments, the crude 5-azacytidine produced by the above-described
`
`processes is subjected to one or morerecrystallization procedures. For example, the crude 5-
`
`azacytidine may be dissolved in dimethylsulfoxide (DMSO), and then recrystallized by the
`
`addition of methanol.
`
`The methods provided by the instant invention are amenable to scale-up, and avoid the use
`
`10
`
`oftin catalysts and other metal ions, thereby providing 5-azacytidine that is suitable for use as
`
`an API. The methods also avoid the formation of emulsions during the work up
`
`(quenching/extraction) of the coupling reaction, thereby avoiding hydrolysis of the s-triazine
`
`ring.
`
`-10-
`
`
`
`WO 2004/082618
`
`~ PCT/US2004/007894
`
`5
`
`DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
`
`In the most basic embodimentof the invention, 5-azacytidine is synthesized according to
`
`the following process wherein each R,is an optionally substituted C;-C29 alkyl group
`
`independently selected from the group consisting of straight chain alkyl groups, branched
`
`alkyl groups, and cyclic alkyl groups, and wherein each Rois an optionally substituted C; - C29
`
`10
`
`acyl group independently selected from the group consisting of straight chain acyl groups,
`
`branched acyl groups, and benzoyl (Bz) groups.
`
`°
`
`N
`()
`
`HNSI(Ri)3
`
`R30
`
`. OK J
`
`(Ry)3SiO
`
`N
`
`\ 2 f
`Si
`Ss
`a“ \ Vi
`3
`oO
`TMS-Triflate
`
`OR,
`
`oO
`
`H
`
`bp,
`
`OR
`| H
`
`ase
`
`aoN
`
`N
`
`(4)
`
`NH
`
`HNSi(Ry)s
`at
`aeJ SilylatingReagent
`NA,
`~ J
`(R,)3Si0 Sy
`(2)
`
`N Ic
`
`(2)
`
`(3)
`
`HNSi(Ry)3
`
`“AJ
`
`R20
`
`0
`
`Deprotection
`
`According to this scheme, 5-azacytosine (1) is reacted with a silylating reagent to yield a
`
`silylated 5-azacytosine (2). Preferably, the silylating reagent is a trimethylsilyl (TMS) reagent
`
`-11-
`
`
`
`WO 2004/082618
`
`~ PCT/US2004/007894
`
`or a mixture of two or more TMSreagents. Preferred TMSreagents include
`
`hexamethyldisilizane (HMDS: (CH;)SiNHSi(CHs3)s) and chlorotrimethylsilane (TMSCL
`
`(CH;)3;SiCl). The silylated 5-azacytosine is then reacted with a protected §-D-ribofuranose
`
`derivative (3) in the presence of TMS-Triflate (trimethylsilyl trifluoromethanesulfonate).
`
`TMS-Triflate catalyzes the coupling reaction, resulting in the formation of a protected 5-
`
`10
`
`azacytidine (4). The protecting groups can be removed by any technique known intheart,
`
`including, but not limited to, treatment with methanol/sodium methoxide. The individual
`
`reactions of the scheme will now be discussed in detail.
`
`Preparation of Silylated 5-Azacytosine
`
`In one embodiment, the silylated 5-azacytosine is prepared by heating a suspension of 5-
`
`iS
`
`azacytosine (1), one or more TMS reagents (present in excess over the 5-azacytosine) and a
`
`catalyst, preferably ammonium sulfate, at reflux without a solvent until a clear solution results.
`
`Most preferably, the TMS reagent is HMDS, whichproducesa trimethylsilyl 5-azacytosine
`
`derivative (Ri= CH; in the scheme above). By cooling to ambient temperature, the silylated 5-
`
`azacytosine crystallizes from the reaction mixture. Thesilylated 5-azacytosine can then be
`
`isolated by any technique known in the art. For example, the silylated 5-azacytosine may be
`
`isolated by partially removing excess TMSreagent, followed byaddition ofa suitable solvent
`
`(for example, heptane) andfiltration under inert atmosphere. The silylated 5-azacytosine thus
`
`isolated is used with or without drying in the coupling step. Alternatively, silylated 5-
`
`azacytosine may be isolated by removing TMSreagent by vacuum distillation and then
`
`25
`
`dissolving the residue is in dichloromethane, acetonitrile, or 1,2-dichloroethane for use in the
`
`coupling step.
`
`In another embodiment, the silylated 5-azacytosine is prepared "in situ" from 5-
`
`azacytosine and an equivalent amount of one or moresilylating reagents (preferably a mixture
`
`of HMDS and TMSC1I)in a suitable solvent in the presence or absenceofa catalyst at reflux.
`
`30
`
`Preferably, the solvent is a dry organic solvent, more preferably a dry polar organic solvent,
`
`including but not limited to acetonitrile. The resulting silylated 5-azacytosine can be used
`
`directly in the coupling step without isolation as described below.
`
`Coupling of Silylated 5-Azacytosine to Sugar
`
`35
`
`In one embodimentof the invention, coupling of the silylated 5-azacytosine to the sugar
`
`is performed by first preparing a cooled mixture(preferably in the range of about 0°C to about
`
`-12-
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`5°C)ofsilylated 5-azacytosine and 1,2,3,5-tetra-O-acetyl-B-D-ribofuranose (or 1-O-acetyl-
`
`2,3,5-tri-O-benzoy1-B-D-ribofuranose) in dichloromethane, acetonitrile, or 1,2-dichloroethane.
`
`Preferably, the solvent for the coupling step is dichloromethane or 1,2-dichloroethane, most
`
`preferably dichloromethane. TMS-triflate is then added to the mixture, preferably at a rate that
`keeps the temperature below 25°C. After the addition is complete, the clear solutionis stirred
`
`10
`
`at ambient temperature for about 2 hours to about 3 hours.
`
`In embodiments whichthe silylated 5-azacytosine is generated "in situ," the coupling
`
`reaction mixture may instead be prepared by adding the sugar and TMS-Triflate directly to the
`
`silylation reagents (silylating agent and 5-azacytosine). The sugar and TMS-Triflate can be
`
`added concurrently with the silylating reagents, or they may be added at the conclusion of the
`
`15
`
`silylation reaction. Preferably, the TMS-Triflate and the sugar are in the same solvent as used
`
`in the silylation reaction, which solvent, as described above, is preferably a dry organic polar
`
`solvent including, but not limited to acetonitrile. Using "in situ" generated silylated 5-
`
`azacytosine in this manner thus allows one to perform "one pot" silylation and coupling. See
`
`Examples 5 and 6.
`
`20
`
`In embodiments where acetonitrile or other polar solvent is present during the coupling
`
`reaction (for example, in embodiments where "one pot" silylation and coupling are performed
`
`in a polar solvent), the acetonitrile or other polar solventis first removed, preferably in
`
`vacuum, and the residue is dissolved in dichloromethane or 1,2-dichloroethaneprior to
`
`quenching. Because polar solvents such are acetonitrile are miscible with water, removing
`
`2S
`
`such solvents from the coupling product and then dissolving the product in dry organic non-
`
`polar solvents such as dichloromethane or 1,2-dichloromethane minimizes the exposure of the
`
`water-sensitive 5-azacytidine to the aqueous phase during extraction/quenching.
`
`Quenching/extraction preferably is performed in a 1/1 w/w NaHCO; / NazCO3 solution
`
`at about 0°C to about 5°C. Using cooled quenching solution further minimizes the
`
`30
`
`decomposition of the protected 5-azacytidine product during quenching. The organic phase of
`
`the quenched reaction is then separated and the water phase extracted with dichloromethane or
`
`1,2-dichloroethane. The combined organic extract is washed with cooled (preferably in the
`range of about 0°C to about 5°C) NaHCO;solution (preferably 10%) and water, then dried
`
`over MgSOn, filtered, andthe filtrate concentrated in vacuum. The residueis a protected 5-
`
`35
`
`azacytidine (4). Methanolis then charged to the residue. When dichloromethane is used
`
`(either as the coupling solvent or following use of acetonitrile as the coupling solvent), the
`
`-13-
`
`
`
`WO 2004/082618
`
`~ PCT/US2004/007894
`
`dichloromethane may be partially removed in vacuum,followed by charging methanolto the
`mixture, and finally by continued vacuum distillation was continued until substantially all
`
`dichloromethane is removed.
`
`As described above, the exposure of protected 5-azacytidine to water can be minimized
`
`by using a non-polar dry organic solvent for the coupling step. Alternatively, if a dry organic
`
`10
`
`polar solvent is present at the coupling step, that solvent can be removed and replaced with a
`
`dry non-polar organic solvent prior to quenching. The duration of exposure ofthe protected
`
`azacitidine to water (during quenching) also depends on the size of the batch that is processed
`
`as small batches can be processed in a shorter time than large batches. Thus, in preferred
`
`embodiments of the invention, a single batch of coupling reaction product is split into smaller
`
`15
`
`sub-batches, and each sub-batch is separately subjected to quenching/extraction.
`
`In preferred embodiments, the protecting groups are removed from the protected 5-
`
`azacytidine (4) by diluting the methanolic solution of protected 5-azacytidine (4) with
`
`methanol, then adding sodium methoxide in methanol (preferably about 25% w/w)to the
`
`mixture with stirring at ambient temperature. During this procedure, a white solid separates.
`
`The mixture is preferably left stirring for about 8 hours to about 16 hours, following which the
`
`solid is filtered off and washed with methanol (until the filtrate is about pH 7). The solid is
`then dried, preferably in vacuum at about 55°C. to about 65°C until the weight of the solid
`
`remains constant. The solid is crude 5-azacytidine (5).
`
`The crude 5-azacytidine (5) may be purified by any technique knownin the art. In
`
`25
`
`preferred embodiments,purification is performed by dissolving the crude product in dimethyl
`
`sulfoxide (DMSO)at about 85°C to about 90°C understirring and in an inert atmosphere.
`
`Methanolis gradually addedto the resulting solution under slow heating, and the mixture is
`
`stirred at ambient temperature for about 8 hours to about 16 hours. The resulting recrystallized
`
`solid is filtered off, washed with methanol, and then dried, preferably under vacuum at about
`
`30
`
`85°C to about 95°C until the weight remains constant. The overall yield is about 30-40%.
`
`The 5-azacytidine synthesis methods provided by the invention provides a numberof
`
`clear advantages over the prior art methods. First, the methods allow the manufacturing of
`
`pilot plant scale uniform batches of 5-azacytidine. Second, the procedure assures an API
`
`without tin or other metallic ion contaminants. Third, there are no difficult to handle phase
`
`35
`
`separation (emulsion) problemsin the work-up of the coupling step. Fourth, by removing
`
`polar solvents from the coupling reaction prior to quenching/extraction and then dissolving the
`
`-14-
`
`
`
`WO 2004/082618
`
`~ PCT/US2004/007894
`
`reaction productis dichloromethane or 1,2-dichloroethane, the exposure ofthe water-senstive
`5-azacytidine to the aqueous phase is minimized. Finally, the decomposition ofthe water-
`sensitive 5-azacytidine is further minimized during the quenching/extraction step by using
`
`cooled quenching solutions.
`
`The following examples are provided for illustrative purposes only. They are not to be
`
`10
`
`interpreted as limiting the scope of the invention in any way.
`
`-15-
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`Example 1: Preparation of Silylated 5-Azacytosine
`
`EXAMPLES
`
`N
`
`N a N
`
`HNSi(CHs)s
`X
`antl(HMDS)
`pe
`A,J 2 J J
`
`NH
`
`(H3C)3SiO
`
`ammomium sulfate
`
`5
`
`10
`
`(1)
`
`(6)
`
`In a 22L, 3-necked flask, a mixture of 5-azacytosine (1) (2.0kg, 17.8mol, 1.07 molar
`
`15
`
`eq.), HMDS(9.162kg) and ammonium sulfate (40.0g) was heated at reflux for 2 hours. A
`
`fresh amount of ammonium sulfate (20.0g) was added, and the reflux was continued for 6
`
`hours longer. The initial slurry turned into a clear, pale-yellow, solution and no more gas
`
`evolved at the end of the reflux. The excess HMDS was evaporated off in vacuumto obtain
`
`an off-white residue, which is trimethylsilylated 5-azacytosine (6).
`
`-16-
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`5
`
`Example 2: Coupling of Silylated 5-Azacytosine to Sugar
`
`ye A
`
`HNSi(CHs)3
`
`e
`
`?
`
`n>
`
`A3 *\7 8
` « T (7 T
`
`(6)
`
`\A§Si
`dichloromethane “ \ 4
`
`O
`
`S
`
`F
`TMS-Triflate
`
`-
`
`HNSi(CHa)s
`
`AN erO
`
`H
`
`Ye ye
`
`O
`
`(8)
`Trimethylsilylated 5-azacytosine (6) prepared according to the method of Example 1
`
`was diluted with anhydrous dichloromethane (18.1kg) in a SOL, 3-necked,flask and solid,
`
`1,2,3,5-Tetra-O-acetyl-B-D-ribofuranose (5.330kg, 16.7mol) (7) was charged to the mixture.
`
`An anhydrous dichloromethane rinse (0.533kg) was used andthe slurry was cooled to 0-5 °C.
`
`10
`
`TMS-triflate (4.75kg, 1.2 molar eq.) was added to the mixture over 5-10 minutes. During the
`
`addition, the reaction temperature increased to 15 - 20°C,and the initial suspension turned into
`
`-17-
`
`
`
`WO2004/082618
`
`PCT/US2004/007894
`
`a clear, pale-yellow, solution. After 2 hours ofstirring, the solution was poured over a mixture
`
`of Na2CO3 (2.00kg), NaHCOs (2.00kg), water (29.9kg) and ice (20.0kg). The layers were
`
`separated. The water layer was extracted with dichloromethane (8.0 kg). The combined
`organics were washed with cold (0 -5°C) 10% NaHCO; (2x10L). The combined washings
`
`were extracted with dichloromethane (8.0kg). The combined organics were washed with cold
`
`10
`
`water (2x5kg), dried on MgSO, (2.0kg), and filtered. The filtrate and dichloromethane washes
`
`on the pad (2x1.32kg) were combined and reduced in volume using vacuum (~200mmHg,
`
`30°C). The distillation was continued until the majority of dichloromethane (app. 85-95%
`
`total) was removed. The residue was taken up in methanol (4.0kg) and the remaining
`
`dichloromethane was removedto give a protected 5-azacytidine (8) as an off-white to yellow
`
`15
`
`foam.
`
`Example 3: Deprotection of Protected 5-azacytidine
`
`er
`
`GC
`
`HNSI(CH3)3
`
`J
`
`r
`le
`~
`
`
`a
`[
`
`sodiummethoxide
`
`er
`
`methanol
`
`HO.
`
`NH
`
`NA
`a.J
`
`0
`
`
`3
`ny
`|
`OH
`
`H
`
`H
`
`t
`y
`
`i
`
`(5)
`(8)
`Protected 5-azacytidine (8) trom Example 2 was diluted with methanol (35.5kg), then
`
`25% NaOMein methanol (439g, 0.11 mol. eq.) was charged. The initial clear solution
`
`becameturbid and a solid started to precipitate. The slurry wasleft under nitrogen overnight.
`
`The solids were isolated and washed with methanol (7x2.4kg). The solids were dried (~28
`
`inHg and ~85°C) to a constant weight to give crude 5-azacytidine (1.835kg; 44.9%) (5).
`
`Example 4: Purification of Crude 5-azacytidine
`
`Crude 5-azacytidine was purified from DMSO/MeOHasfollows: Crude 5-
`
`azacytidine (1.829kg) was dissolved in preheated DMSO (5.016kg; 87-90°C) undernitrogen.
`
`25
`
`The solution was diluted with methanol in portions at approximate 10-minute intervals
`
`(9x1.4kg then 1x0.58kg) while slowly cooling. After the addition, 45-55°C was maintained
`for 1 hour and then the mixture wasleft to cool to ambient temperature overnight. The next
`day, the solids were isolated at ambient temperature, washed with MeOH (6x0.83kg), and
`
`- 18 -
`
`
`
`WO 2004/082618
`
`PCT/US2004/007894
`
`dried in vacuum (~30 inHg and ~85°C)to a constant weight to give 5-azacytidine (1.504kg;
`
`82.2% recovery).
`
`Example 5: One Pot Synthesis of 5-azacytidine
`
`A mixture of 5-azacytosine (5.0 g, 44.6 mol), HMDS(6.3 mL, 29.8 mol), and TMSCI(6
`
`mL, 47.3 mmol) in acetonitrile (78 mL) was heated to reflux for 20 hours under an inert
`
`10
`
`atmosphere. TMS-triflate (9 mL, 50 mmol) and 1,2,3,5-tetra-O-acetyl-$-D-ribofuranose (14.2
`
`g, 44.6 mmol were added directly to the silylated 5-azacytosine in acetonitrile. The addition
`
`was performed at ambient temperature and underan inert atmosphere. The reaction mixture
`
`was maintained understirring for 20 hours, then poured over a pre-cooled (0-5°C) sodium
`
`bicarbonate solution (10%, 500 mL). The resulting mixture was extracted with
`
`15
`
`dichloromethane (3x75 mL). The combined organic extract was washed with cooled (0-5°C)
`
`10% sodium bicarbonate (2x25 mL) and brine (2x25 mL), then dried over magnesium sulfate
`
`(10.0 g), filtered, and the filtrate concentrated in vacuum to dryness. The off-white foam
`
`dissolved in methanol (120 mL) wastreated with a solution of 25% sodium methoxide in
`
`methanol (1.0 g, 4.62 mmol). Soon a white solid started to separate. The suspension was
`
`20
`
`stirred at ambient temperature for 15 hours, then the solid wasfiltered off, washed with
`
`methanol (3x5 mL) and anhydrous ether (2x5 mL), then dried in vacuum. The crude 5-
`
`azacytidine (4.5 g, 41.3 %) was further purified from DMSO and methanol(fordetails see
`
`Example 4).
`
`25
`
`Example 6: One Pot Synthesis of 5-azacytidine
`
`A mixture of 5-azacytosine, HMDS, and TMSCI in acetonitrile is heated to reflux for 20
`
`hours under an inert atmosphere. TMS-triflate and 1,2,3,5-tetra-O-acetyl-B-D-ribofuranose
`
`are then added directly to the silylated 5-azacytosine in acetonitrile. The addition is performed
`
`at ambient temperature and under an inert atmosp