WO2017205622A1

WO2017205622A1 - Method of making benznidazole

Info

Publication number: WO2017205622A1
Application number: PCT/US2017/034469
Authority: WO
Inventors: Lynsey WATSON
Original assignee: Savant Neglected Diseases, Llc
Priority date: 2016-05-25
Filing date: 2017-05-25
Publication date: 2017-11-30

Abstract

A method of making benznidazole by reacting 2-nitroimidazole (, Formula 1) with a haloacetate ester (Formula 4) and a base in an R1-OH solvent in a stirred reaction mass, wherein X is chosen from Br, I, and Cl, and wherein R1 is chosen from methyl, ethyl, isopropyl, neobutyl, isobutyl, and tertbutyl, cooling the reaction mass, adding benzylamine to the reaction mass, stirring the reaction mass, adding water to the reaction mass and stirring at low temperature for 2 hours, filtering to obtain a cake of benznidazole, dissolving the benznidazole cake in a mixed solvent solution, and crystallizing the benznidazole. A method of making benznidazole by producing benznidazole with a single stage reaction with low-toxicity alcohol solvents (R1-OH). A method of crystallizing high purity benznidazole by dissolving crude benznidazole in a solvent combination of acetone, methanol and water, and crystallizing benznidazole.

Description

METHOD OF MAKING BENZNIDAZOLE

BACKGROUND OF THE INVENTION

1 . TECHNICAL FIELD

[0001] The present invention relates to methods of making benznidazole. The present inventi relates to methods of maximizing yield and purity of benznidazole.

2. BACKGROUND ART

[0002] As shown in FIGURE 1 (marked Prior Art), benznidazole (Formula 3)

Formula 3

antiparasitic drug derived from 2-nitroimidazole (Formula 1),

Formula 1

used in the treatment of Chagas disease and was formerly marketed by Hoffman-La Roche. The original synthetic route employed by Roche is outlined in FIGURE 1. Stage 1 of the synthetic route requires the use of dimethylformamide, high temperatures (110-153 degrees C), and two instances of distillation to dryness, which are undesirable in larger scale manufacture. An alternative synthetic route to the Stage 1 imidazole methyl acetate (Formula 2) in which Rl = methyl employs more favorable conditions,

Formula 2 although a distillation to dryness is still required (Dyes Pigm., 2011, 89, 915, FIGURE 2, marked Prior Art). In both procedures, the isolation of the Stage 1 acetate (Formula 2, Rl = methyl) incorporates filtration of the inorganic salt byproducts and crystallization of the filtrate.

[0003] Stage 2 of the Roche procedure involves reaction of the imidazole methyl acetate (Formula 2, Rl

= methyl) with benzylamine in methanol at room temperature for 16 hours with no agitation. Similarly, the procedure in the Dyes and Pigments paper involves reaction of Formula 2 with an analogous amine (ethanolamine, FIGURE 2) in methanol, at room temperature overnight with agitation. Both procedures involve subsequent recrystallization from ethyl acetate.

[0004] There remains a need for a method of synthesizing benznidazole with a scalable procedure that maximizes the yield and purity of benznidazole.

SUMMARY OF THE INVENTION

[0005] The present invention provides for a method of making benznidazole, (Formula 3), by reacting

2-nitroimidazole ( , Formula 1) with a haloacetate ester (Formula 4) (Formula 4) and a base in an Rl-OH solvent in a stirred reaction mass, wherein X is chosen from Br, I, and CI, and wherein Rl is chosen from methyl, ethyl, isopropyl, neobutyl, isobutyl, and tertbutyl, cooling the reaction mass, adding benzylamine to the reaction mass, stirring the reaction mass, adding water to the reaction mass and stirring at low temperature for 2 hours, filtering to obtain a cake of benznidazole, dissolving the benznidazole cake in a mixed solvent solution, and crystallizing the benznidazole.

[0006] The present invention provides for a method of making benznidazole by producing benznidazole with a single stage reaction with low-toxicity alcohol solvents (Rl-OH).

[0007] The present invention provides for a method of crystallizing high purity benznidazole by dissolving crude benznidazole in a solvent combination of acetone, methanol and water, and crystallizing benznidazole.

DESCRIPTION OF THE DRAWINGS

[0008] Other advantages of the present invention are readily appreciated as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings wherein:

[0009] FIGURE 1 is a schematic of Stage 1 and Stage 2 reactions to produce benznidazole by a prior art process; [00010] FIGURE 2 is a schematic of Stage 1 and Stage 2 reactions to produce an analogue of benznidazole by a prior art process;

[00011] FIGURE 3 is a schematic of a proposed base catalyzed ester hydrolysis affording carboxylic acid

(Formula 2, Rl = H);

[00012] FIGURE 4 is a schematic of a base catalyzed trans-esterification of Stage 1 methyl acetate

(Formula 2) in ethanol;

[00013] FIGURE 5 is a graph of Stage 1 reaction progress by HPLC Area%;

[00014] FIGURE 6 is a table (TABLE 27) of a summary of Stage 1 critical parameter investigations;

[00015] FIGURE 7 is a table (TABLE 28) of a summary of Stage 2 critical parameter investigations;

[00016] FIGURE 8 is a table (TABLE 29) of a summary of Stage 3 critical parameter investigations;

[00017] FIGURE 9 is a schematic of the synthetic method of making benznidazole of the present invention; and

[00018] FIGURE 10 is a table (TABLE 30) of a comparison of the method of the present invention with prior art methods.

DETAILED DESCRIPTION OF THE INVENTION

[00019] The present invention generally provides for improved methods of synthesizing benznidazole, as shown in FIGURE 9, which provide increased purity and increased yield.

[00020] Most generally, the Stage 1 reaction of 2-nitroimidazole (Formula 1) and a haloacetate ester

Formula 4 (preferably ethyl bromoacetate, X = Br, Rl = ethyl) with an Rl-OH solvent (preferably ethanol, Rl = ethyl) is performed with a reduced reaction temperature and an increased reaction time to provide a higher yield of the Stage 1 product ethyl acetate (Formula 2, Rl = ethyl).

[00021] Carboxylic acid formation (Formula 2, Rl = H) is minimized during this stage by using ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) and ethanol (Rl-OH, Rl = ethyl) with increased equivalents, and by preferably operating under anhydrous conditions. The Stage 1 product is not isolated and a telescoped reaction is performed into Stage 2 by adding benzylamine directly into the Stage 1 product. An increased yield and purity of the Stage 2 product benznidazole (Formula 3) can be obtained by reacting the Stage 1 product (Formula 2) at 50 degrees C with 3.0 equivalents of benzylamine. Stage 2 also uses a 2 hour low temperature stir-out post isolation, and a cold water wash of the isolated cake. In the Stage 3 crystallization, benznidazole (Formula 3) is dissolved in a mixed solvent system instead of a single solvent and then crystallized.

[00022] Most preferably, the method of the present invention is performed as follows (with example amounts given). Stages 1 and 2 are performed. 2-nitroimidazole (Formula 1, 5.00 g), potassium carbonate (6.72 g) and ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) (8.12 g, 5.39 mL) are heated to 70 degrees C in ethanol (40 mL) for 110 minutes. This reaction mass is then cooled to 50 degrees C and benzylamine (14.21 g, 14.49 mL) added, and the suspension stirred for 16 hours. Water (50 mL) is then added, and the reaction mass is cooled to 0 to 5 degrees C, stirred for two hours, filtered, and washed with cold water (25 mL). A solid cake is obtained of benznidazole (Formula 3). The solid cake is dried under vacuum at 50 degrees C for 16 hours and benznidazole (Formula 3) is obtained as a pale yellow solid, 10.01 g (87% over two stages).

[00023] For Stage 3, benznidazole (Formula 3, 10.01 g) is heated in acetone:methanol:water (49.9 mL :

49.9 mL : 5.3 mL) until dissolution occurs. Two volumes of acetone/methanol are then removed by distillation at atmospheric pressure (20 mL), and water (20 mL) is added dropwise to the flask over 5 minutes. The resulting slurry is then cooled to 0 to 5 degrees C and stirred for 2 hours. The solid is then filtered and washed with ice cold methanol (50 mL), dried under vacuum at 50 degrees C for 16 hours and benznidazole (Formula 3) is obtained as an off-white solid, 9.11 g (91%). The Examples below detail the development of this method.

[00024] The method is significantly improved over the prior art in using alcoholic solvents of low-toxicity and environmental impact that enable integration of two separate reactions into a single integrated process. The selected alcoholic solvent avoids the need for phase transfer catalyst, strong bases, and isolation of intermediates resulting in higher yields and fewer impurities. Those impurities that are formed can be separated with a novel crystallisation process to provide consistently high purity product in high yield.

[00025] Therefore, the present invention provides generally for a method of making benznidazole by producing benznidazole with a single stage reaction with low-toxicity alcohol solvents ( l-OH), and crystallizing high purity benznidazole. The present invention also provides for benznidazole recovered by the methods described herein.

[00026] The prior art for making benznidazole (Formula 3 (FIGURE 1)) describes the use of known toxic solvents and harsh conditions that result in undesirable impurities and low overall yields. The synthesis of imidazole methyl acetate (Formula 2 Rl = methyl), is described in GB01138529 (Roche process). The reaction employs DM F solvent. The use of DM F solvent is undesirable as it has known carcinogenic properties and is on the "undesirable" list of solvents. The use of DM F at the high temperatures described is particularly problematic as it is known to decompose to give toxic by-products. The reaction also requires a moderately strong base, sodium methoxide, used as a methanol solution that dilutes the process and makes it less productive. The yield of this reaction is only 7.3%. Another prior art reference for making (Formula 2, Rl = methyl), [Dyes and Pigments 2011, 89, 9-15], describes the use of boiling acetonitrile solvent, TBAI phase transfer catalyst and inorganic base, and reports a higher yield of 81%. Using the conditions described, two repeats of this reaction achieved product yields of only 41.5%, 56.7%. Furthermore in order to carry out the second stage reaction, as described in GB01138529 an exchange of acetonitrile to methanol solvent is required, which is problematic on- scale, due to the inability to strip solvents below the agitator, along with their low boiling points. A further disadvantage of the method described [Dyes and Pigments 2011, 89, 9-15] is the use of the tetrabutylammonium iodide phase transfer catalyst, which increases costs, creates new impurities, and with similar organic solvent solubility to the product, makes it difficult to separate.

[00027] The synthesis of benznidazole (Formula 3) described in GB01138529 employs conditions that are unfavorable for scale-up. There is no agitation during the addition of benzylamine that is required to ensure homogeneity and control exotherms. The crystallization of benznidazole (Formula 3) is long and uncontrolled without seeding or agitation and requires multiple solvents, with crystallization from methanol, recrystallization from ethyl acetate, and the final recrystallization from ethanol. The process is dilute and the yields are low.

[00028] In the present invention it has been found that replacement of DM F by environmentally acceptable alcoholic solvents, ( l-OH), is beneficial in improving the yield. Rl is selected from methyl, ethyl, isopropyl, neobutyl, isobutyl, tertbutyl, and preferably methyl or ethyl, especially ethyl.

[00029] The methyl and ethyl esters of imidazole acetate (Formula 2, Rl = methyl or ethyl) have been found to be particularly sensitive to transesterification. Therefore, Rl-OH is used preferably in combination with the corresponding haloacetate ester (Formula 4, Rl = methyl, ethyl; X= Br, CI). It has also been found that if water (Rl = H) is present in the reaction, hydrolysis occurs and A/-2-nitroimidazoleacetic acid (Formula 2, Rl = H) or after neutralisation with the sodium carbonate present forms sodium A/-2-nitroimidazoleacetate (Formula 2, Rl = Na) is formed, which results in lower yields and makes purification of the final drug product more complex. It is therefore desirable to minimise hydrolysis by ensuring a moisture-free environment with dry solvents. If the level of water is controlled at low levels this improves the overall yield, and keeps hydrolysis products to minimum acceptable levels that can be removed during the crystallisation process.

[00030] In the present invention, it has been determined that the phase transfer catalyst is not required as the yields are identical to those where it is part of the reaction mass. The benefit of removing the need for TBAI is to reduce the cost, avoid additional impurities, and make the product easier to separate.

[00031] In the present invention, a process that employs a solvent Rl-OH enables that both reactions to be integrated into a single stage. This improves the overall yield, reduces the impurities and simplifies operation of the process. The single-stage process involves contacting 2-nitroimidazole (Formula 1), optionally added as a solid or diluted in Rl-OH solvent, where Rl is hereinbefore defined, with a compound of Formula 4, optionally neat or diluted in Rl-OH solvent, in either a batch or continuous flow reactor. In the compound of Formula 4, X is chloro-, bromo-, iodo-, preferably chloro- or bromo-, and especially bromo-, and Rl is as hereinbefore defined. Reactant (Formula 4) is used preferably in a molar ratio of 0.9-1.5 equivalents, preferably 1-1.2 equivalents and especially 1.1 equivalents. The reaction is preferably carried out in the presence of a base, such as an organic or inorganic base, preferably sodium carbonate or potassium carbonate. The temperature of the reaction is preferably at or below the boiling point of the solvent l-OH. The reaction time is 0.0167 to 4 hours, particularly 0.5-2.5 hours, and especially 2 hours. The reaction is mixed in batch or continuous flow with an impellor with sufficient power to maintain good mixing of the slurry, or if in continuous flow with a static mixer. When the alkylation is determined to be complete between 1 and 4 molar equivalents of benzylamine are added to the stirred reaction mass at a temperature lower than the boiling point of Rl-OH, preferably 50 degrees C, and stirred for between 0.5 and 55 hours, preferably 16 hours. A further benefit in this integrated process is the avoidance of the screening and washing of inorganic salts that contribute to loss of yield and generation of impurities. The overall yield of crude product (Formula 3) using this integrated process is 87%.

[00032] Also, in the present invention the conditions are identified for increasing the purity of the benznidazole (Formula 3). The crude product is dissolved in a solvent combination of acetone, methanol and water, preferably in a volumetric ratio of 1:1:0.1. Preferably, two volumes of 7:1 acetone/methanol are stripped by vacuum distillation at ambient temperature, and replaced with a similar volume of water. The reaction mass is cooled to preferably 0-5 degrees C, optionally seeded with pure (Formula 3) and stirred a further 1-10 hours, preferably 2 hours to crystallise the pure product. The solid is filtered, washed with 1-10 volumes, preferably 5 volumes of methanol at -10-5 degrees C, preferably 0 degrees C, dried using warm nitrogen in an enclosed filter dryer, or in an oven at 20-70 degrees C, preferably 50 degrees C to obtain the product in >91% yield with purity measured by HPLC against authentic standards at >99.5%.

[00033] The invention is further described in detail by reference to the following experimental examples.

These examples are provided for the purpose of illustration only, and are not intended to be limiting unless otherwise specified. Thus, the invention should in no way be construed as being limited to the following examples, but rather, should be construed to encompass any and all variations which become evident as a result of the teaching provided herein.

[00034] EXAMPLE 1

[00035] Stage 1 Development

[00036] A. Process familiarization and identification of key issues

[00037] The initial familiarization experiments of Stage 1 followed the prior art Dyes and Pigments paper as a route to the imidazole methyl acetate (Formula 2, Rl = methyl). The key issues identified from these early trials were the low yield obtained from the reaction, the presence of fine inorganic salt byproducts following filtration, and the low recovery from recrystallization with ethyl acetate (see TABLE 1). TABLE 1 - Results from initial investigations of Stage 1 reaction

[00038] B. Removal of inorganic salts

[00039] During the Stage 1 reaction, potassium bromide is formed as a by-product, which is insoluble in organic solvents. In addition, excess potassium carbonate, TBAI, and the unreacted salt form of 2-nitroimidazole are also present at the end of the Stage 1 reaction and are insoluble in acetonitrile, leading to a fine suspension. The Dyes and Pigments paper indicates that the solid material should be filtered and the remaining filtrate concentrated and recrystallized. However, following filtration through sintered glass (pore size 40-90 μιη), a significant amount of fine particulate material still remained in the solution, and filtration through fine pore filter paper (11 μιη) was also unable to remove all the fine particulates. In addition, qualitative analysis of the resulting filter cake by thin layer chromatography revealed the presence of the Stage 1 acetate, which may account for some loss of yield.

[00040] i. Reaction component solubility

[00041] It was proposed that the filtration step be substituted for an aqueous wash to aid the removal of the inorganic salts and unreacted starting materials. A qualitative study was performed to assess the solubility of each reaction component in water and acetonitrile, to determine those which would be expected to be removed by an aqueous wash (see TABLE 2).

TABLE 2 - Assessment of the solubility of reaction components in water and acetonitrile

[00042] The solubility study shows that although the majority of the reaction byproducts would be removed via an aqueous wash, the slight solubility of the Stage 1 acetate in water would lead to a loss of yield. It was determined therefore that the isolation of the inorganic salts and their subsequent removal by aqueous wash should be postponed until the end of the Stage 2 reaction. [00043] C. Reaction solvent screen

[00044] It was believed that the poor solubility of 2-nitroimidazole and the moderate solubility of the

Stage 1 product (Formula 2) in acetonitrile was a major contributing factor to the observed low reaction yield. Therefore, a study was carried out to examine the conversion of 2 -nitro imidazole to Stage 1 when heating at reflux in a range of solvents for a fixed period of time (see TABLE 3).

TABLE 3 - Conversion of 2-nitroimidazole to products in various solvents

*Unknown peak, later identified as the corresponding ethyl acetate (Formula 2, Rl = ethyl).

[00045] Acetonitrile, ethyl acetate, isopropyl alcohol, and acetone displayed poor consumption of 2- nitroimidazole after a reaction time of 1 hour under the respective reflux conditions. Methanol showed much greater consumption despite the lower reaction temperature employed, whilst reaction in ethanol led to rapid consumption of 2-nitroimidazole, although less than 2% of the observed products were the desired Stage 1 methyl acetate (Formula 2, Rl = methyl). Reactions with methanol and ethanol also led to the formation of an additional compound, which was later identified as the corresponding carboxylic acid (Formula 2, Rl = H) via analysis by mass spectrometry (m/z 172).

[00046] As the polar protic solvents methanol and ethanol demonstrated the highest consumption of 2- nitroimidazole, these solvents were chosen to investigate as reaction solvents in future experiments. In addition, as both of the standard Stage 2 reactions outlined by the Roche process and the Dyes and Pigments process employed methanol as the reaction solvent, the potential for developing a telescoped Stage 1/2 procedure without intermediate isolation was investigated.

[00047] D. Carboxylic acid formation in protic solvents

[00048] The extent of the formation of the corresponding carboxylic acid (Formula 2, Rl = H) in Stage 1 reactions in methanol was monitored over time (see TABLE 4). The carboxylic acid was observed to form at a comparable rate to the Stage 1 acetate (Formula 2, Rl = ethyl) over the first hour, and then no significant conversion of 2nitroimidazole to products was observed thereafter. The Stage 1 acetate (Formula 2, Rl = ethyl) was also observed to decrease in intensity over time whilst the carboxylic acid (Formula 2, l = H) was observed to increase. Following an overnight stir at room temperature, the Stage 1 acetate (Formula 2, Rl = ethyl) had completely converted to carboxylic acid.

TABLE 4 - Conversion of 2-nitroimidazole to products over time

[00049] This data shows that the Stage 1 acetate (Formula 2, Rl = ethyl) converts to the carboxylic acid

(Formula 2, Rl = H) on longer reaction times, likely via base catalyzed ester hydrolysis (see FIGURE 3). In order to minimize the loss to carboxylic acid in future reactions, the presence of water should be avoided as far as possible by using oven-dried glassware and nitrogen blanketing.

[00050] E. Methyl bromoacetate (Formula 4, X = Br, Rl = methyl) visibility by HPLC

[00051] It was noted during in-process sampling and analysis of the Stage 1 reaction by HPLC, that the presence of methyl bromoacetate (Formula 4, X = Br, Rl = methyl) could not be detected. It was initially suspected that during those reactions conducted at 65-80 degrees C, the low boiling point of methyl bromoacetate (Formula 4, X = Br, Rl = methyl) (51-52 degrees C) led to evaporation of this reagent when the vessel was opened for sampling. It was proposed that this loss to evaporation may be responsible for the poor conversion of 2-nitroimidazole to product. However, when the Stage 1 reaction in methanol was repeated below the boiling point of methyl bromoacetate (Formula 4, X = Br, Rl = methyl) (50 degrees C), the conversion of 2-nitroimidazole and the formation of products occurred at a similar rate to those reactions at 65 degrees C, and methyl bromoacetate (Formula 4, X = Br, Rl = methyl) was not observed by HPLC in either case (TABLE 5).

TABLE 5 - Comparison of reaction rates of the Stage 1 reaction and the detection of methyl bromoacetate

(Formula 4, X = Br, Rl = methyl) at 50 and 65 degrees C

[00052] Therefore, it was concluded that loss of methyl bromoacetate (Formula 4, X = Br, l = methyl) by evaporation was not likely to be the major contributing factor to the poor rate of conversion in the Stage 1 reaction in methanol, and was attributed to the inherent lower reactivity in this solvent.

[00053] F. Identification of imidazole ethyl acetate (Formula 2, Rl = ethyl)

[00054] Stage 1 reactions performed in ethanol led to the observation of more than one product by

HPLC analysis (see TABLE 6). These compounds were analyzed by mass spectrometry and the observed parent peaks corresponded to the carboxylic acid (Formula 2, Rl = H; m/z 172), Stage 1 methyl acetate (Formula 2, Rl = methyl; m/z 186), and a Stage 1 ethyl acetate (Formula 2, Rl = ethyl; m/z 200). In order to further confirm the identity of the major product, ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) was investigated as the Stage 1 acetate source. A single Stage 1 product was then observed by HPLC analysis, which was consistent with the imidazole ethyl acetate (Formula 2, Rl = ethyl) observed previously.

TABLE 6 - Stage 1 reaction results in ethanol with methyl- and ethyl-bromoacetate sources

*RRT of both compounds 0.87

[00055] It was proposed that the Stage 1 methyl acetate (Formula 2, Rl = methyl) obtained from reactions with methyl bromoacetate (Formula 4, X = Br, Rl = methyl) undergoes trans-esterification with ethanol in situ catalyzed by excess base, to give the Stage 1 ethyl acetate (Formula 2, Rl = ethyl).

[00056] Although reaction of methyl acetate (Formula 2, Rl = methyl), carboxylic acid (Formula 2, Rl =

H), and ethyl acetate (Formula 2, Rl = ethyl) with benzylamine in Stage 2 would be expected to afford the desired API, the occurrence of this in situ trans-esterification would potentially lead to a mixture of both acetates as impurities. Therefore, it was concluded that future reactions should be carried out with either methyl bromoacetate (Formula 4, X = Br, Rl = methyl) in methanol or ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) in ethanol to prevent this mixture.

[00057] Comparison of the Stage 1 reaction yields under both conditions indicated that a higher conversion of starting material to product is achieved with ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl), and lower levels of carboxylic acid are also observed (TABLE 7). In addition, there is a reduced concern relating to the loss of ethyl bromoacetate (Formula 4, X = Br, l = ethyl) at higher temperatures due to evaporation, as the boiling point (159 degrees C) is twice that of the maximum reaction temperature.

TABLE 7 - Comparison of Stage 1 reaction yields under methyl- and ethyl- conditions

[00058] It was therefore concluded that in order to minimize the number of potential by-products, and to achieve lower levels of carboxylic acid and higher levels of Stage 1, future Stage 1 reactions would be performed with ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) in ethanol.

[00059] G. Ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) equivalents

[00060] The number of equivalents of bromoacetate required in the Stage 1 reaction was investigated.

The Stage 1 reaction was performed with 1.0, 1.1, and 1.2 equivalents of ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) (see TABLE 8), and the conversion of 2-nitroimidazole (Formula 1) to products was observed. The results obtained indicate that an excess of ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) would be required for full consumption of 2-nitroimidazole (<1% remaining). In addition, 1.2 equivalents of ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) were shown to lead to an unreacted excess in the reaction mixture. This excess is unfavorable as it may result in further reaction and the generation of unwanted byproducts.

TABLE 8 - Comparison of Stage 1 reaction with variation of ethyl bromoacetate (Formula 4, X

equivalents

[00061] In light of this, 1.1 equivalents of ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) will be employed in future Stage 1 reactions to maximize reaction completion whilst minimizing unreacted starting materials. [00062] H. Requirement of TBAI

[00063] Tetrabutylammonium iodide (TBAI) is utilized in the Stage 1 reaction as a phase transfer catalyst.

The quaternary ammonium salt coordinates to the deprotonated 2-nitroimidazole starting material and facilitates its dissolution into the organic solvent, where reaction with ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) occurs. Phase transfer agents of this type are particularly useful when employing non-polar, aprotic solvents.

[00064] The proposed Stage 1 synthesis is to be carried out in ethanol, which is a polar, protic solvent. It is feasible therefore, that reaction of 2-nitroimidazole and ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) in ethanol would proceed without the addition of TBAI. In addition, the proposed benznidazole isolation method involves removal of unreacted Stage 1 components, byproducts, and impurities via an aqueous wash. The low solubility of TBAI in water (see TABLE 2) can result in poor removal by an aqueous wash and so lead to contamination of the isolated benznidazole. Therefore, it would be beneficial if TBAI could be removed from the Stage 1 process.

[00065] Reactions were performed to examine the effect of removal of TBAI on the observed yield of

Stage 1 product (see TABLE 9). The results indicated that the Stage 1 reaction in ethanol would progress to near complete consumption of 2-nitroimidazole (<1%) without the need for a phase transfer catalyst.

TABLE 9 - Comparison of Stage 1 reaction with and without TBAI

[00066] It was therefore concluded that Stage 1 reactions in ethanol do not require the use of phase- transfer agents, and so TBAI would be excluded from future reactions.

[00067] I. Reduction of reaction solvent volume

[00068] Following the observation of low recovery from the Stage 2 reaction, it was proposed that a more concentrated Stage 1 reaction be employed. The standard Stage 1 reaction incorporates 10 volumes of ethanol, and so a series of reactions were undertaken to examine more concentrated reactions and assess the effect on reaction progress (see TABLE 10). TABLE 10 - Comparison of reaction rate when varying solvent concentration

[00069] The results obtained indicated that a comparable reaction rate could be obtained for the Stage 1 reaction with a reduced number of solvent volumes, with full consumption of 2-nitroimidazole (<1% remaining) after 1 hour reaction time. However, lower solvent volumes can present an issue with regards to stirring capability in the Stage 2 reaction, and so a reduction to 8 volumes would be preferable.

[00070] J. Reaction time and temperature

[00071] Previous experiments have indicated that the Stage 1 reaction proceeds very rapidly at 78 degrees C in ethanol, giving complete consumption of 2-nitroimidazole in 1 hour, although a significant percentage of carboxylic is generated. It was proposed that a more selective reaction may be possible if the reaction was performed at lower temperatures over longer reaction times.

[00072] A series of experiments at reduced temperature (70 degrees C) were devised to determine the reaction end-point and the relative concentrations of each component were monitored via in-process sampling (see TABLE 11 and FIGURE 5).

TABLE 11 - Stage 1 reaction progress monitored over time at 70 degrees C

[00073] The Stage 1 reaction was observed to require 100 minutes at 70 degrees C to achieve adequate completion (2-nitroimidazole <1%), whilst the relative carboxylic acid concentration remained low throughout the course of the reaction (max. 1.3%). It appears from these results that a lower temperature is beneficial in terms of reducing the carboxylic acid concentration, although longer reaction times are required to reach completion.

[00074] Stage 1 reactions will therefore be carried out at 70 degrees C for a minimum of 100 minutes to ensure completion and to minimize the formation of carboxylic acid. Following the minimum reaction time, a completion sample will be analyzed by HPLC to ensure the correct conversion has been achieved.

[00075] K. Summary of Stage 1 Development

[00076] In summary, familiarization of the Stage 1 reaction was performed following the experimental method from Dyes and Pigments (2011, 89, 9-15) and the reaction was found to be practically difficult due to the presence of fine particulate by-products, and afforded low yields (42-57%).

[00077] A synthetic route to the Stage 1 acetate was developed in the present invention in which isolation of the Stage 1 product was removed, the reaction temperature reduced to 70 degrees C, and the reaction time increased to 110 minutes, which led to a more selective reaction in higher yield. The reaction solvent was concentrated to 8 volumes and changed from acetonitrile to ethanol, as use of a polar, protic solvent led to improved yields and allowed the removal of TBAI from the process.

[00078] The bromoacetate source was altered to use ethyl bromoacetate (Formula 4, X = Br, l = ethyl), and the number of equivalents was increased to 1.1, which led to higher Stage 1 yields and minimized carboxylic acid (Formula 2, R = H) formation. In addition, it was found that carboxylic acid formation could be further minimized by operating the reaction under anhydrous conditions in oven-dried glassware.

[00079] EXAMPLE 2

[00080] Stage 2 Development

[00081] A. Process familiarization and identification of key issues

[00082] The initial familiarization experiments of Stage 2 followed the Roche prior art process as a route to benznidazole (Formula 3) from imidazole methyl acetate (Formula 2, Rl = methyl). The key issue identified from the initial trial was the low recovery following isolation (see TABLE 12).

TABLE 12 - Results from initial investigations of Stage 2

[00083] B. Suitability of Stage 1 development in Stage 2 reaction [00084] The present invention's development of Stage 1 resulted in a number of changes to the Stage 2 procedure outlined in the Roche prior art process. Therefore, the principle Stage 2 investigations examined whether these changes had a detrimental effect on the yield and purity of the Stage 2 reaction.

[00085] C. Substitution of methyl bromoacetate (Formula 4, X = Br, Rl = methyl) with ethyl

bromoacetate (Formula 4, X = Br, Rl = ethyl)

[00086] One of the major changes in the developed Stage 1 process was the substitution of methyl bromoacetate (Formula 4, X = Br, Rl = methyl) for ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl). It was necessary, therefore, to examine the reactivity of the resulting Stage 1 ethyl acetate (Formula 2, Rl = ethyl) towards benzylamine. Both acetates (Formula 2, Rl = methyl and Formula 2, Rl = ethyl) were examined with 3.0 equivalents of benzylamine and the resulting Stage 2 yield compared (see TABLE 13).

TABLE 13 - Comparison of methyl and ethyl acetate intermediates in Stage 2 reaction

*Stage 2 reaction performed with pre-optimized conditions; stir reagents for 16 hours at room temperature and filter.

[00087] It is evident from these experiments that imidazole acetate (Formula 2, Rl = ethyl) has greater reactivity towards benzylamine than the methyl counterpart (Formula 2, Rl = methyl). These results are encouraging as acetate (Formula 2) was also found to be the more favorable Stage 1 intermediate due to the higher reaction selectivity observed. Ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) was therefore considered a suitable substitute in the Stage 1 reaction.

[00088] D. Telescoping of Stages 1 and 2

[00089] During Stage 1 development, it was proposed that omission of the isolation of the Stage 1 acetate would be beneficial to avoid a significant loss of yield on filtration. The proposed developed reaction method would involve a telescoped reaction procedure in which benzylamine is added directly to the reaction mixture, following completion of the Stage 1 reaction. The Stage 2 reaction would then be stirred for 16 hours at room temperature, filtered, and washed with water.

[00090] This developed reaction procedure was examined, and the results confirmed a higher yield of benznidazole could be obtained by avoiding isolation of the Stage 1 acetate (see TABLE 14). TABLE 14 - Comparison of the Stage 2 yields arising from an isolated and telescoped Stage 2 reaction

[00091] E. Isolation of Stage 2

[00092] Preliminary investigations of the Stage 2 reaction revealed the isolation method detailed in both the Roche prior art process and the Dyes and Pigments prior art process gave poor recovery of benznidazole (see TABLE 12). During the development of the Stage 1 reaction, it was noted that a yield loss was observed during the filtration to remove inorganic salt byproducts. It was proposed that isolation of Stage 1 be removed from the process, and the inorganic salts be removed via an aqueous treatment in Stage 2. The starting materials and by-products of both the Stage 1 and 2 reactions are soluble in water, whilst benznidazole is virtually insoluble (see TABLE 2). Therefore, an isolation method was developed which included the addition of 15 volumes of water following the Stage 2 reaction, and a wash with 5 volumes of water following filtration.

[00093] The efficacy of the isolation method was examined following a telescoped Stage 1/Stage 2 reaction. The isolated solid was examined for HPLC purity and assay and shown to be highly pure, which a benznidazole (Formula 3) assay greater than that of the purchased reference standard (see TABLE 15). However, HPLC assay of the mother liquors also revealed a 6.7% loss of benznidazole (Formula 3).

TABLE 15 - Analysis of isolated solid and mother liquors following proposed Stage 2 isolation

[00094] Examination of the mother liquors by HPLC also revealed the successful removal of the key impurities, namely the Stage 1 imidazole acetate (Formula 2, Rl = ethyl), carboxylic acid (Formula 2, Rl = H), benzylamine, and ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) (see TABLE 16 for details on process impurities). TABLE 16 - Mother liquor composition (HPLC area%)

[00095] In order to maximize the recovered yield of benznidazole, the Stage 2 isolation method was further developed to introduce a 2 hour low temperature stir-out with a 5 volume wash with ice-cold water on filtration. Due to the insolubility of benznidazole in water, a reduction of the water volumes from 15 to 10 was also introduced although this was not expected to have a significant impact on yield but would improve the process volumes and allow an increase in batch size during manufacture.

[00096] The improved isolation method was shown to increase the observed yield to 84.7% over two stages, and consistently give material of excellent purity in high yield with byproducts, impurities, and unreacted starting materials being successfully removed (see TABLE 17).

TABLE 17 - Comparison of HPLC purity of Stage 2 reaction before and after isolation

[00097] F. Stage 2 reaction temperature and benzylamine equivalents

[00098] The Roche process employs 3.4 equivalents of benzylamine in the Stage 2 reaction, whilst the

Dyes and Pigments process utilizes 4.0 equivalents of the corresponding amine. Both procedures incorporate a 16 hour reaction at room temperature. It was proposed in the present invention that the number of equivalents of benzylamine could be reduced, in conjunction with an increase in temperature, without a detrimental effect on yield or purity.

[00099] I. Benzylamine equivalents in the Stage 2 reaction [000100] The Stage 2 reaction was examined with various equivalents of benzylamine during a 16 hour reaction at room temperature. In each case, the extent of reaction progression was monitored by HPLC (see TABLE 18).

TABLE 18 - Reaction progression with varying benzylamine equivalents

[000101] This data revealed the Stage 2 reaction did not reach completion with 3.5 equivalents at room temperature, having undergone 89% conversion of Stage 1. A further study was undertaken to examine the Stage 2 reaction with 2.0 equivalents of benzylamine at a range of temperatures (see TABLE 19).

TABLE 19 - Reaction progression with varying reaction temperature

[000102] These experiments also revealed that the Stage 2 reaction was not complete after 16 hours at 78 degrees C, although 94% conversion of Stage 1 material was observed. Based on these results, investigation of two alternative Stage 2 procedures were proposed. The first method involved carrying out a Stage 2 reaction at the lower temperature of 50 degrees C and employing 3.0 equivalents of benzylamine, whilst the second method involved employing 2.0 equivalents of benzylamine at the higher reaction temperature of 78 degrees C.

[000103] A Stage 2 reaction at lower temperatures would be preferable in terms of selectivity and the avoidance of degradation, although higher equivalents of benzylamine are required for better conversion which would have implications on vessel fill volumes and cost. Experiments were undertaken to compare the two proposed reaction conditions and examine the effect on yield and purity of the Stage 2 material

(see TABLE 20). TABLE 20 - Examination of proposed Stage 2 reaction conditions

Experiment Temp (C) Time (h) Benzylamine equivalents Isolated Yield (%) Stage 2 Purity (HPLC Area%)

671-57 50 16 3.0 87.9 99.85

671-60 50 16 3.0 82.0 99.68

671-58 78 16 2.0 55.1 95.41

671-59 78 16 2.0 50.9 98.25

[000104] These results show high purity material is obtained in good yield from reaction at 50 degrees C with 3.0 equivalents of benzylamine, whereas yield and purity are affected when carrying out reactions at 78 degrees C. Future reactions would therefore be conducted with the lower reaction temperature of 50 degrees C and 3.0 equivalents of benzylamine, and an in-process completion sample would be taken to ensure adequate conversion prior to isolation.

[000105] G. Benzylamine impurity and source selection

[000106] During development work employing a batch of benzylamine from Alfa Aesar, an unknown impurity was noted in some batches of Stage 2 at T 1.09 with a peak area of 0.08%-0.10% by HPLC. Analysis of the batch of benzylamine by HPLC indicated an impurity with a peak area of 3.36% at a retention time of 13.706 minutes.

[000107] Investigative work was unable to identify the source of the impurity in benzylamine and a corresponding peak was not found by mass spectrometry, which suggested it did not ionize. Potential impurities in the benzylamine manufacture process including benzonitrile, and potential byproducts arising from reaction of ethyl bromoacetate (Formula 4, X = Br, Rl = ethyl) and benzylamine were examined and all failed to match the retention time of the unknown impurity.

[000108] Alternative sources of benzylamine were investigated and it was found that samples from Camida (West Yorkshire, UK) and BorsodChem (Marianske Hory, Czech Republic) did not contain the unknown benzylamine impurity and produced Stage 2 material that was free of the unknown impurity at RRT 1.09.

[000109] H. Summary of Stage 2 Development

[000110] In summary, familiarization of the Stage 2 reaction was performed following the experimental method from the Roche process as a route to benznidazole (Formula 3) from imidazole methyl acetate (Formula

2, Rl = methyl), and the reaction was found to give low recovery following isolation (27-32%).

[000111] The impact of the Stage 1 reaction modifications on the Stage 2 reaction, namely the use of ethanol as the reaction solvent, the removal of the Stage 1 isolation and the substitution to ethyl bromoacetate

(Formula 4, X = Br, Rl = ethyl), were investigated and found to have a positive impact on the reaction purity and yield. [000112] The reaction temperature and number of equivalents of benzylamine employed were also investigated and higher reaction yields and purities were found to arise from reactions at 50 degrees C with 3.0 equivalents of benzylamine. In addition, it was found that introducing a 2 hour low temperature stir out post isolation and introducing a cold water wash to the isolated cake could improve the yield further.

[000113] EXAM PLE 3

[000114] A. Process familiarization and identification of key issues

[000115] The development of an effective, scalable crystallization procedure for benznidazole was required, and initial familiarization experiments closely followed the Roche process. Benznidazole was dissolved in the minimum volume of refluxing ethanol, then cooled and filtered to collect crystals of purified benznidazole. This method required 14 volumes of ethanol due to the low solubility of benznidazole in ethanol, and the mass recovery was poor (see TABLE 21).

TABLE 21 - Results from initial investigations of Stage 3 crystallization

[000116] B. Development of Stage 3 crystallization solvent

[000117] i. Analysis of benznidazole solubility in ethanol

[000118] It was proposed that the poor recovery of benznidazole from crystallization in ethanol can be improved by the addition of an anti-solvent in which benznidazole has low or no solubility. The use of water as an anti-solvent was investigated, and the solubility of benznidazole in mixtures of water and ethanol was examined (see TABLE 22).

TABLE 22 - Solubility study of benznidazole in ethanol/water mixtures

[000119] These results indicate that benznidazole has very low solubility in ethanol and water, and that ethanol does not appear to be a viable solvent for crystallization.

[000120] C. Solubility solvent screen of benznidazole

[000121] In order to identify a suitable solvent for crystallization of benznidazole, its solubility in a range of solvents was examined. In each case, 0.3 g of benznidazole Stage 2 was fully dissolved in the minimum number of volumes of each solvent, and those solutions which indicated good solubility were allowed to cool to room temperature and crystallize. The precipitate was allowed to settle, and the liquors sampled and analyzed for benznidazole content by HPLC assay. The samples were then further cooled to 0 degrees C (with agitation) and sampled again. The potential yield was calculated by subtraction of the benznidazole content in the liquors from the input quantity (see TABLE 23).

TABLE 23 - Solubility and potential recovery of benznidazole in a range of solvents

[000122] i. Mixed solvent crystallization trials

[000123] Small scale trials indicated that benznidazole has reasonable solubility in acetone, acetonitrile, and methanol, with good recovery on cooling. These trials were then repeated on a 10.0 g scale, and the purity and yield of the recovered benznidazole was examined (see TABLE 24). The Stage 3 HPLC purity was also compared to the Stage 2 HPLC purity to determine the effectiveness of the crystallization. On examination of the clarity of each of the reaction mixtures, it was determined that each contained a persistent haziness which was not affected by further solvent addition. TABLE 24 - Purity and recovery of benznidazole in single solvent systems

[000124] These results indicated that those solvents which gave good solubility of benznidazole led to poor recovery following crystallization. It was proposed therefore, that a mixed solvent system with the use of water as an anti-solvent can provide more favorable yields of recovery.

[000125] A further series of experiments were then conducted to examine the solubility and recovery of benznidazole in a mixed solvent system. It was thought that the addition of water can be beneficial, as although it is an anti-solvent for benznidazole, it can help to dissolve the residual inorganic salts, starting materials, and byproducts which can be responsible for the persistent haziness (see TABLE 25).

TABLE 25 - Purity and recovery of benznidazole in mixed solvent systems

[000126] It was evident from these experiments that the crystallization trials gave Stage 3 material of excellent purity, although it would be preferable if a higher yield could be achieved. The number of solvent volumes required to achieve dissolution varies between each mixture, with the lowest number of volumes required for solvent mixtures containing acetone.

[000127] The addition of further anti-solvent at this stage can be beneficial for increasing the recovered yield of benznidazole, although it would be disadvantageous to greatly increase the vessel fill volume. In addition, it is likely that cooling of the crystallization mixture to 0-5 degrees C can also aid yield recovery.

[000128] Therefore, the proposed Stage 3 crystallization method involves dissolution of benznidazole Stage 2 in 10.5 volumes of mixed solvent (Acetone/Methanol/Water, 9.5 : 9.5 : 1) at reflux, followed by the distillation of two volumes of acetone, and the addition of two further volumes of water to act as an antisolvent. The reaction mixture would then be cooled to 0-5 degrees C and stirred for 2 hours to allow full crystallization, filtered, and washed with 5 volumes of cold methanol (see TABLE 26).

TABLE 26 - Examination of proposed Stage 3 crystallization conditions

[000129] These results indicate that the proposed crystallization method gives high purity benznidazole in excellent yield.

[000130] EXAMPLE 4

[000131] Critical Components Analysis

[000132] TABLE 27 in FIGURE 6 shows a summary of Stage 1 critical parameter investigations. TABLE 28 in

FIGURE 7 shows a summary of Stage 2 critical parameter investigations. TABLE 29 in FIGURE 8 shows a summary of Stage 3 critical parameter investigations.

[000133] A route was successfully developed for the synthesis of benznidazole which produces material of very high purity in high yield (the scheme shown in FIGURE 9). During the course of the development, the synthetic route was significantly altered from those methods described in the prior art processes of Roche and Dyes and Pigments, which were used as a starting point. A comparison of the method of the present invention alongside the Roche and Dyes and Pigment methods is shown in TABLE 30 in FIGURE 10. The manufacturing yields were 87% for Stage 2 and 91% for Stage 3. The processing parameters for the current invention are listed below in two different alternatives in TABLE 31 and TABLE 32.

TABLE 31 - Stage 1 and 2 parameters

Water/eq vol 10.00 10.00

Total ethanol/eq vol 9.41 8.00

Total water/eq vol 10.00 10.00

Cool back rate/deg Chr^"1 12 NS

Isolation Temperature 0-5 0-5

Stir out time/hrs 2 2

Water wash/eq vol 3.82 5.00

Drying Temperature/deg C 50 50

TABLE 32 - Stage 3 (Crystallisation)

[000134] Throughout this application, various publications, including United States patents, are referenced by author and year and patents by number. Full citations for the publications are listed below. The disclosures of these publications and patents in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.

[000135] The invention has been described in an illustrative manner, and it is to be understood that the terminology, which has been used is intended to be in the nature of words of description rather than of limitation.

[000136] Obviously, many modifications and variations of the present invention are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims, the invention can be practiced.

Claims

CLAIMS What is claimed is:

1. A method of making benznidazole, including the steps of:

reacting 2-nitroimidazole ( Formula 1) with a haloacetate ester (Formula 4)

and a base in an Rl-OH solvent in a stirred reaction mass, wherein X is chosen from the group consisting of Br, I, and CI, and wherein Rl is chosen from the group consisting of methyl, ethyl, isopropyl, neobutyl, isobutyl, and tertbutyl;

cooling the reaction mass;

adding benzylamine to the reaction mass;

stirring the reaction mass;

adding water to the reaction mass and stirring at low temperature for 2 hours;

filtering to obtain a cake of benznidazole;

dissolving the benznidazole cake in a mixed solvent solution; and

crystallizing the benznidazole.

2. The method of claim 1, wherein X is Br and Rl is ethyl.

3. The method of claim 1, wherein the haloacetate ester is present in a molar ratio of 0.9-1.5 equivalents.

4. The method of claim 1, wherein said adding step is performed at a temperature of at or below a boiling point of the Rl-OH solvent with 1 to 4 molar equivalents of benzylamine.

5. The method of claim 4, wherein said adding step is performed at a temperature of 50 degrees C.

6. The method of claim 1, wherein said reacting step is performed at a temperature of at or below a boiling point of the Rl-OH solvent for 0.0167 hours to 4 hours.

7. The method of claim 1, wherein said reacting step is performed at a temperature of 70 degrees C for 110 minutes.

8. The method of claim 1, wherein said cooling step is further defined as cooling the reaction mass to 50 degrees C.

9. The method of claim 1, wherein said stirring the reaction mass step is performed for between 0.5 and 55 hours.

10. The method of claim 9, wherein said stirring the reaction mass step is performed for 16 hours.

11. The method of claim 1, wherein the low temperature in said adding step is further defined as 0 to 5 degrees C.

12. The method of claim 1, wherein said stirring at low temperature step is performed for 2 hours.

13. The method of claim 1, wherein said filtering step further includes the step of drying the cake under vacuum at 50 degrees C for 16 hours.

14. The method of claim 1, wherein the mixed solvent solution is further defined as acetone:methanol:water.

15. The method of claim 14, wherein a volumetric ratio of the mixed solvent solution is 1:1:0.1 acetone:methanol:water.

16. The method of claim 1, wherein said crystallizing step further includes the step of removing two volumes of 7:1 acetone/methanol by distillation, adding water to the solution, cooling a resulting slurry to 0 to 5 degrees C, stirring for 1-10 hours, filtering and washing a solid with 1-10 volumes of methanol at -10 to 5 degrees C, and drying the solid under vacuum at 20-70 degrees C for 16 hours.

17. The method of claim 1, wherein said reacting step is performed by a process chosen from the group consisting of a batch reactor and a continuous flow reactor.

18. The method of claim 1, wherein the base is chosen from the group consisting of sodium carbonate and potassium carbonate.

19. The method of claim 1, wherein a yield of benznidazole in the cake is 87%.

20. The method of claim 1, wherein a yield of benznidazole crystallized is over 91%.

21. Benznidazole recovered from the method of claim 1.

22. A method of making benznidazole, including the step of:

producing benznidazole by performing a single stage reaction with low-toxicity alcohol solvents of the formula l-OH.

23. The method of claim 22, wherein the benznidazole produced has reduced impurities.

24. A method of crystallizing high purity benznidazole, including the steps of:

dissolving crude benznidazole in a solvent combination of acetone, methanol and water; and crystallizing benznidazole.

25. The method of claim 24, wherein said crystallizing step further includes the step of removing two volumes of 7:1 acetone/methanol by distillation, adding water to the solution, cooling a resulting slurry to 0 to 5 degrees C, stirring for 1-10 hours, filtering and washing a solid with 1-10 volumes of methanol at -10 to 5 degrees C, and drying the solid under vacuum at 20-70 degrees C for 16 hours.