The invention relates in particular to an industrial process for the manufacture of 17- Oxoestra-1,3, 5 (10)-trien-3-yl sulfamate (II), an intermediate for the synthesis of the pharmaceutical active 17ß-Hydroxyestra-1, 3,5 (10)-trien-3-yl sulfamate (I11).

Polycyclic compounds bearing a sulfamate function, in particular steroid sulfamates have been reported to be sulfatase inhibitors (W093/05064).

Estra-1,3, 5 (10)-triene 3-yl sulfamate such as 17 (3-Hydroxyestra-1, 3,5 (10)-trien-3-yl sulfamate (W096/05216, W096/05217) have been described to exhibit an increased systemic and reduced hepatic estrogenicity at oral application (J. Steroid Biochem. Mol.

Bio. 1995; 55, 395-403).

Sulfamate are obtained by the reaction of alcohols or phenols with sulfamoyl chloride.

The known processes all apply excessive amounts of the sulfamoylating agent (sulfamoyl chloride). When dichloromethane or acetonitrile is used as solvent, additional base is required (Steroids 1999,64, 460-471 ; J. Prakt. Chem 1999,341, 574-583; J.

Med. Chem. 1999,42, 2280-2286). Nevertheless, even under these conditions the conversion rates and thus the yields are low. Due to the high reactivity the sulfamoylating reagent tends to react with solvents. In the case of the most commonly

used solvent N, N-dimethyl formamide (DMF), the formation of an undesired DMF- adduct was observed (Steroids 1996,61, 710-717).

Sulfamoylation is the key step in the synthesis of 17ß-Hydroxyestra-1, 3,5 (10)-trien-3-yl sulfamate, a potent estrogen after oral administration. 17-Oxo-1, 3,5 (10)-triene-3-ol (estrone) is reacted with sulfamoyl-chloride to-yield--17-Oxoestra-1 3, 5 (10)-trien-3-yl sulfamate, which is in turn converted to 17ß-Hydroxyestra-1, 3,5 (10)-trien-3-yl sulfamate by means of a complex hydride reagent. According to this invention, the preferred complex hydride reagent is sodium borohydride.

Due to the problems associated, the sulfamoylation processes of the state-of-the-art did not prove suitable for a large scale industrial process for the manufacture of sulfamates, in particular for that of 17p-Hydroxyestra-1, 3,5 (10)-trien-3-yl sulfamate. Sulfamoylation on a small scale is described in Tetrahedron Lett. 2000,41, 7047-7051.

It was necessary to find an improved process for the manufacture of sulfamate from alcohols and phenols, for example from estrone, in which higher conversion rates can be reached with only a slight excess of sulfamoylating reagent and side-product formation can be reduced or avoided.

It has now been found that these problems are avoided if N, N-dimethyl acetamide (DMA) or 1-methyl-2-pyrrolidone (NMP) are used as a solvent for the sulfamoylation reaction. Moreover, by use of the solvents DMA or NMP, the amount of sulfamoyl chloride can be dramatically reduced from 5-6 equiv. in the known procedures to as low as 1,0-2 equiv. with no base present. The preferred range of the amount of sulfamoyl chloride is 1,0-1, 5 equiv. Under these conditions a complete conversion to the sulfamate without formation of detectable by-products is achieved.

This procedure according to the invention has turned out to be applicable to other phenols and alcohols as well. This invention thus provides an efficient and economic method to cleanly convert a hydroxyl group to a sulfamoyloxy group using a minimum amount of the reagent in DMA or NMP.

This process allows the manufacture of sulfamate on industrial scale. Table 1 shows that the laboratory process (0,185 mol) according to this invention could successfully be scaled up to 25,5 kg (94 mol) of estrone in the pilot plant.

However, the pre-requisite for the industrial scale sulfamoylation of estrone was a safe industrial scale method for the production of sulfamoyl chloride, which is the key reagent to convert alcohols to sulfamates. This invention relates therefore to a safe production process for sulfamoyl chloride as well.

Several processes for the manufacture of sulfamoyl chloride are known. According to known processes, chlorosulfonyl isocyanate dissolved in an apriotic solvent such as an aliphatic, aromatic or chlorinated hydrocarbon, diethyl ether or acetonitrile reacts either with water (EP 0403185; Aldrichimica Acta, 10, 2,1977, 23-28) or with formic acid (EP 0403185) to form sulfamoyl chloride. This reaction is highly exothermic and goes along with boisterous gas evolution. Attempts to scale up these laboratory methods for pilot plant production failed. In some cases, experiments on a scale as small as of 100 g showed an uncontrollable run-away behaviour with explosive gas evolution. Extensive thermal safety measurements confirmed a tendency of heat accumulation rendering the reaction of chlorosulfonyl isocyanate with formic acid on a larger scale hazardous. An uncontrolled onset of gas evolution and thereby an abrupt pressure jump would have a devastating effect on the industrial scale apparatus (scheme 1).

The risk of heat accumulation marked by the non-dosage-controlled gas evolution, and thereby the risk of a run-away reaction is in the given example particularly high because the reactive anhydride intermediate is formed faster (rate constant kl) than it could in the course of a consecutive reaction cascade with the loss of carbon monoxide (rate constant k2) and carbon dioxide (rate-constant k3)-. collapse-into-the reaction product sulfamoyl chloride (scheme 2). SOaCI OZCI 1 I I HN k 2 HN k 3 N t1 0 CI-S-1H2 /CO rn HCOH chCoros2 chlorosulfonyl isocyanate anhydride intermediate carbamic acid intermediate sulfamoyl chlorid The enormous hazard potential of this process and at same time the strict safety measures made the present invention of a safe industrial method absolutely necessary.

For developing an industrial scale process for the manufacture of 17 (3-Hydroxyestra- 1,3, 5 (10)-trien-3-yl sulfamate, it was therefore crucial to invent a process by which the sulfamoyl chloride reagent could be provided on a large scale safely.

According to the present invention, a safe process, characterized by the dosage- controlled gas evolution is achieved by adding formic acid to a solution of chlorosulfonyl isocyanate in dichloromethane in the presence of catalytic amounts of carboxamides (typically 0,1-20 mol% relative to chlorosulfonyl isocyanate). According to the present invention, preferred catalysts include N, N-dialkyl carboxamides, more preferably N, N- dimethyl formamide or N, N-dimethyl acetamide. The preferred range of the amount of catalyst is 0,5-2 mol%. During the addition of formic acid to chlorosulfonyl isocyanate, the reaction temperature is kept in the range of 35-45°. The catalyst needn't be present in the solution of chlorosulfonyl isocyanate before formic acid is added. More preferably, the catalyst is mixed to the formic acid, which is added continuously to the reaction mixture.

The amide catalyst is thought to act by accelerating the collapse of the reactive anhydride-type intermediate into carbon monoxide and the corresponding carbamic

acid, and subsequently the fragmentation thereof, resulting in carbon dioxide and sulfamoyl chloride.

Thermal safety measurements confirmed a smooth, dosage-controlled gas-flow with no indication of thermal accumulation in the presence of 0,5-2 mol%] N, N-dimethyl acetamide (scheme 3).

By means of the present invention, sulfamoyl chloride could safely be produced in pilot- plant batches, starting from 20 kg of chlorosulfonyl isocyanate (tables 1 and 2).

Procedure I (Synthesis of 17-Oxoestra-1, 3,5 (10)-triene-3-yl sulfamate) A mixture of formic acid (284,92 mmol) and N, N-dimethyl acetamide (3,03 mmol) was added to a stirred solution of chlorosulfonyl isocyanate (277,47 mmol) in dichloro- methane-(87, 5 ml) at 42°C within a period of 3,5 hours. The mixture was heated to reflux for 15 minutes and cooled to ambient temperature. The resulting mixture was added to a stirred solution of estrone (184,93 mmol) in N, N-dimethyl acetamide (625 ml) at ambient temperature within a period of 20 minutes. The mixture was stirred at ambient temperature for 17 hours, and then poured onto water (1875 mi). After a while a white precipitate was formed. The suspension was stirred at ambient temperature for another 2 hours, and filtered. The crystals were washed twice with water (250 ml), and used without further drying and purification in the next step (Procedure II). Yield : 87,8 g (wet material) white crystals, purity (HPLC): 99,12 % (table 1).

Procedure II (Synthesis of 17ß-Hydroxyestra-1, 3,5 (10)-trien-3-yl sulfamate) To a cold (2-5°C) suspension of moist 17-Oxoestra-1, 3,5 (10)-trien-3-yl sulfamate (171,7 mmol, from procedure 1) in ethanol (450 ml), a solution of sodium borohydride (156,23 mmol) in water (94 ml) was added at 2-5 °C within a period of 20 minutes. The resulting mixture was stirred at 2-5°C for another 20 hours. A solution of citric acid (495,86 mmol) in water (668 ml) was then added at this temperature within a period of 15 minutes to adjust the pH to 3. Water (668 ml) was added. The resulting crystalline solid was filtered and washed three times with water (500 ml). The wet crystals were suspended in acetone (2010 ml) and the resulting suspension was heated under reflux until complete dissolution. Removal of acetone (1700 ml) by destillation resulted in a suspension, which was cooled to ambient temperature and filtered. The crystalline solid was washed with cold acetone (50 ml) and dried in vacuum. Yield: 45,06 g white crystals, purity (HPLC): 99, 61% (table 2).