Patent Watch

March 25, 2013

Find better ways to make old and new epilepsy drugs. J. Surtees and co-inventors disclose alternative processes for making active pharmaceutical ingredients (APIs) that are used to treat epilepsy and seizures. One compound that can be prepared by their processes is the established drug levetiracetam (1, Figure 1), marketed under the trade name Keppra. Because 1 is now off-patent, there is obvious interest in new drugs.

The inventors also claim that seletracetam (2) and brivaracetam (3) (Figure 2) can be prepared by their processes. These drugs are apparently much more active than 1.

All of the drugs are used as single isomers, so a stereoselective synthesis is desirable. The inventors describe two routes for preparing the molecules; the first, shown in Figure 1, is the synthesis of 1 by the reaction between pyrrolidone (4) and chiral bromo amide 5 in the presence of a base. GC analysis showed that the conversion is 40.3% and that the product contains 51% of the (S)-enantiomer and 49% of the (R)-isomer. No details of their separation are given, although the use of chiral HPLC is discussed.

The same reaction is used to prepare derivative 6 of 1. Compound 7 is prepared from the corresponding hydroxy ester and then condensed with 4 to give 6. Chiral HPLC showed that the product is a mixture of 89.3% (S)-enantiomer 6 and 10.7% of its (R)-isomer.

The inventors do not describe the detailed preparation of 2, but they report that acid 8 is prepared in 41% yield from pyrrolidone 9 and acid 10 in the presence of NaH (Figure 2). Ammonolysis of 8 produces 2; no reaction details are provided.

In a reaction similar to the preparation of 8, acid 11 is prepared from 10 and pyrrolidone 12. The product is isolated in 77% yield and can be converted to 3 by ammonolysis. Again, no details are provided for this reaction.

The second route for preparing the substituted pyrrolidones does not start with simple pyrrolidones and is the subject of additional claims. The route involves a cyclization reaction, shown in Figure 3. The preparation of enantiomer 13 begins with the reaction of racemic salt 14 and optically pure bromo ester 15. This step produces intermediate 16, isolated as a yellow oil. The crude material is treated with 2-hydroxypyridine (2-HP) to cyclize it to 17. This ester is hydrolyzed to give acid 18. Conversion to 13 is carried out by adding ClCO2Et, followed by reaction with liquid NH3 in the presence of K2CO3. The overall yield of 13 is 32%.

This route is also used to prepare levetiracetam (1) by treating 5 with the HCl salt of amino ester 19 to give 20, recovered as its HCl salt in 49% yield. The salt is basified with Et3N and treated with 2-HP to cyclize it to 1, initially isolated as an oil. GC analysis showed 100% conversion, and chiral HPLC showed that the product contains 98.6% (S)-isomer and 1.4% (R)-isomer.

The inventors also prepared 1 and its (R)-enantiomer 21 by using a similar reaction scheme with alternative substrates to 5. Figure 4 outlines the route, which starts from protected hydroxy amide 22 and amino ester 23. When the reaction is carried out in the presence of Cs2CO3, the product is (R)-enantiomer 24, which is used without purification to prepare 21 by treating it with 2-HP. Chiral HPLC showed that the product is 94% (R) and 6% (S).

When the reaction between 22 and 23 is run with K2CO3, the product is (S)-enantiomer 25. This is used to prepare 1, but the product contains only 79% (S)-isomer.

The inventors do not comment on the apparent stereoselectivity of the carbonate salts in the reaction of 22 with 23. This is an intriguing finding and worthy of investigation. (UCB S.A. [Brussels]. US Patent 8,338,621, Dec. 25, 2012; Keith Turner)

What do you think of Patent Watch? Let us know.