Kinase Chemistry – Just a year and a half behind the times.

R-788 (Fostamatinib disodium)

Posted by kinasepro on November 9, 2007

R-788 is Rigel’s SYK / Flt3 inhibiting prodrug of R-406, and despite a positive sounding press release on their Ph2 ITP study, the market is reacting negatively so far by knocking 20% off of the stock price.

the name suggests disodium.


15 Responses to “R-788 (Fostamatinib disodium)”

  1. milkshake said

    looks very nonselective to me – just by judging from the structure

  2. xtallographer said

    Not to mention Cyp-chow

  3. Presumably this is intended for use an IV agent so GI effects are interesting. I seem to remember that IGFRK inhibition may have effects on blood pressure. If it is indeed hitting IGFRK there is a good chance that it’ll also hit IRK which has implications for control of glucose levels. Is it my imagination or will this compound really generate formaldehyde when the pro-drug breaks down?

  4. kinasepro said

    R-788 is oral, and yah the drawn compound will release formaldehyde. I think the phosphate thing falls off pdq. 100 mg bid = 10 mg / day H2CO

    Afaik, a selectivity panel has not yet been published for R-788, but recent applications have started calling it a Flt3-Syk dual inhibitor. If it is relatively non-selective the first place I would look at other kinases is Pdgf on the Flt-3 side, and Zap70 & Fak on the Syk side. From a discovery viewpoint the question on everyones mind is how selective is selective enough?

    As for the hypertension… It’s a fairly common side effect w/ TKIs and practically a biomarker for the Vegf inhibitors. The comment on neutrophil counts relates to chemo-agents often causing neutropenia

  5. If it’s oral, why the phosphate group? And when/where does that fall off?

  6. Formaldehyde although not ideal, in tiny amounts is tolerated and OK. I remember an antibiotic used as a prodrug that releases formaldehyde. I think it’s an ampicillin prodrug.

  7. kinasepro said

    I believe its less in the way of a classical prodrug, and more of a solubility aid. I’d expect the aminal to be hydrolyzed in the gut, but I’m not aware of any literature on this.

  8. pf01273 said

    This is an example of a ‘double prodrug’ with the phosphate group there to improve the aqueous solubility. Metabolic cleavage of the phosphate to the hemiaminal occurs which decomposes spontaneously to liberate the pydrido-oxazinone moeity.
    As Wavefunction mentioned in ampicillin similar strategies have been employed to mask the carboxylic acid group and improve oral absorption; in these cases ‘double esters’ were used eg Pivampicillin and Bacampicillin.

  9. Not sure how patients woull take to a drug that needs to generate something that is used in embalming, before it can work. I’m sure the marketing guys would figure out a way to put some positive spin on this.

    In response to pf01273, are we talking about metabolic cleavage in the gut?

  10. milkshake said

    You are too hung up on the formaldehyde thing – look at the MW of formaldehyde and MW of the whole molecule, then molest it no more. FYI formaldehyde-releasing pro-drugs are in the two most succesful Gillead phosphonate antivirals (originally discovered in Prague). The dosing of these two drugs is quite high and formaldehyde tox is apparently a non-issue. You are like those cranks that noticed a methyl ester in the aspartame molecule and worry about methanol poisoning from Diet Pepsi…

  11. pf01273 said

    GMC I don’t know exactly where hydrolysis of the phophate group would occur. My guess is that you would want it to be following absorption ie on first-pass metabolism in the liver rather than in the gut itself.
    Following on from milkshake’s previous post I think N-demethylation is another instance where formaldehyde can be produced – I’m sure no one is usually too worried about using that as a solubilizing group?

  12. OK, here’s a wild and woolly calculation that I did:

    The fatal dose of methanol is supposed to be about 100 ml (lower limit) which is 76 g (density 0.76 g/ml)

    A mole of CH3OH produces a mole of formaldehyde. So 76 g of methanol will produce 71g of formaldehyde (that’s a hell of a lot of formaldehyde) as a fatal dose. Now methanol does not kill only due to formaldehyde generation, it also kills by CNS depression. But still, it seems that the fatal dose of HCHO is on the order of few tens of grams by any standard.

    Now consider the drug. Let’s say its mol. wt. is 500. Clearly one mole of drug produces one mole of HCHO. So 500 g drug will produce 30 g HCHO. A usual dosing regimen might be a 100 mg tablet two times a day. So that’s 200 mg. Let’s take an upper dose of 500 mg just to push the limit.

    500 mg of drug corresponds to 30 mg of formaldehyde produced. The fatal dose cited above was 71 g.

    I am inclined to say that if this drug kills you, it probably won’t be because of formaldehyde generation.

  13. Formaldehyde is not an issue. As somebody once said, ‘the solution to pollution is dilution’. I did however think it rather droll to have a drug producing its own embalming fluid, even in tiny amounts. Apparently the Russian Pickling Team flew into Hanoi before Uncle Ho had passed away so as to be able to start their work promptly.

    The issue for me is the phosphate group because that’s not the sort of thing I’d start with to try improve absorption from the gut. It is the sort of thing I’d stick on an IV drug because solubility is quite critical for an IV agent.

  14. pf01273 said

    Phosphate ester prodrugs of Taxol have been synthesised in the past to try and improve it’s poor solubility. Apparently the phosphate groups are hydrolysed by alkaline phophatases in the brush border of the small intestine. Perhaps a similar mechanism occurs in this case.

  15. I do agree that it’s rather droll, and that’s a good word here!
    I carried out a simple calculation on the phosphate and the non-phosphorylated analog using the program QikProp developed by William Jorgensen’s group. It’s not highly accurate but what it does is it compares your structure with a database of drugs and then gives a list of metabolic properties (possible metabolism sites, metabolic products, GI tract absorption, ROF violations, polar SA, logP. HERG channel blockage, Caco cell permeability, serum protein binding etc.) and tells you how they compare with the drugs in its database.

    Result for phosphate:
    GI tract absorption 36% (<25% is poor)
    logP 3.5
    ROF violations 2
    log S (aq. sol.) -6.1

    Result for non-phosphate:
    GI tract absorption 90% (this just shoots up predictably)
    logP 3.7
    ROF violations 1
    logS (aq. sol.) -5.1 (range is from -6.5-0.5)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: