Preparation of Low Cost (R)-(-)-Praziquantel

29 Oct
Published by CreativeChemistry

Preparation of Chiral Praziquantel

Creative Chemistry/Development
Chemicals Ltd

 
Following on from a project for the preparation of low cost
chiral mefloquine, we were asked by WHO to have a brief look at praziquantel in
order to see if we had any ideas for a low cost synthesis of the chiral
material.
 
 
Current Industrial
 Process
The first objective was to determine the process which is
currently in use for the preparation of low cost racemic PZQ. The key
intermediate that is used is benzoyl-PZQ (structure 1) which is sold by Chinese
producers for $50-55/kg. We had initially thought that the transformation to
PZQ involved a selective hydrogenation, but it transpires that the process is a
simple hydrolysis followed by re-acylation. The sales price of formulated PZQ
is $80/kg, which indicates a cost of less than $30/kg for these later processes
(this will be reasonably fixed for either chiral or racemic material).

 
 
 
Examination of the product lists of the Chinese suppliers
indicates that the Bz-PZQ is prepared from the Reissert compound followed by
high pressure hydrogenation then cyclization (the old Merck process).
 
 
Initial Proposals
We considered several processes for the chiral synthesis,
but the lowest cost route still seemed to be one which utilised a resolution
procedure involving one of the current industrial intermediates. The one which
we considered gave the best potential was the benzoyl intermediate (Structure
2).
 
 
 The reasons for this choice were 

  • Used in the current process and potential low
    cost
  • Chiral centre adjacent to the amine
  • Benzoyl confers higher crystallinity than
    cyclohexyl
  • Potential for recycle using the acidic benzylic
    centre

 
A review of the literature indicated that this resolution
had already been patented, US 3,993,760
. Also, the cyclohexyl derivative and
the diamine resolution were reported in US 4,362,875. The detail in these
patents is quite scant, though the indication is that the benzoyl resolution is
superior in that multiple crystallisations are not required and that it
utilises the cheaper tartaric acid (the cyclohexyl requires the more expensive,
though this is still only $8/kg in bulk).
 
 
Resolution Procedure
Despite multiple attempts to purchase the required starting
materials (directly, Chinese agents, PZQ-producers) we failed. Consequently,
due to the lack of high pressure hydrogenation facilities, we were forced to
prepare the benzoyl stating material through a circuitous route.
 
 
Reaction of the crude benzoyl material with L-tartaric acid
in methanol gave a 37% yield of the salt which contained 3.2% of the undesired
isomer (15mmole scale; chiral analysis below-racemic, resolved, mother liquors).
This material was carried though to chiral PZQ by cyclization, hydrolysis and
re-acylation (quant, 85%, 80%- second HPLC trace).
 
 

 
 
 

 
Optical rotation confirmed that this is the (-)-PZQ.
Additional proof that this was the desired enantiomer was
conducted through the “taste test”!
 
 
Conclusions
Given a probable production cost of less than $40/kg for the
benzoyl intermediate, even without recycle, this would fulfill the requirement
for the production of chiral PZQ at below $200/kg.
It has also been shown that, if required, the mother liquors
can be racemized simply by heating in toluene with palladium on carbon as
catalyst.
 
It should be noted that "Kilomentor" suggested this approach a few weeks ago while we were waiting to get permission to post this work. 
 
Please excuse formatting errors etc- tricky enough getting the pictures attached!
 
 
 

Comments

MatTodd's picture

This is awesome Bill, thanks for posting this summary. I'm hoping we can unpack the detail of what you've done and post everything. You did the taste test!? That's commitment.

I forgot to mention that the (-)-PZQ obtained was 98% ee, so there has been a slight enhancement in purity during the subsequent steps, which is what we had anticipated.