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

Rule of 5

Posted by kinasepro on August 24, 2007

>> Update: Here.
rule of 5 v 1.0

Well… I think its self explanatory, but thats probably because I’ve spent the last month compiling it. 😦 Basically, my list has 7 approved kinase inhibitors, 9 in phase 3, 31 in phase 2 and 33 in phase 1. There’s probably more, but this is what I have for current clinical trials. Structures for the approved and ph3 compounds are known, and 23/31 of the ph2 and 18/33 of those in ph1 are either known or I’ve made educated guesses for them.

From these structures I’ve calculated a handful of the properties people most commonly associate with oral activity, and tabulated them above. Molecular weight, LogD at ph 5.5, Molar Refractivity, # of freely rotatable bonds, # of hydrogen bond donors, and acceptors…

chemspider 4tw


13 Responses to “Rule of 5”

  1. pf01273 said

    Hi, I’m not sure why have you calculated log D at pH 5.5 and not 7.4?
    Any chance you could include clog P & PSA in there as well?



  2. kinasepro said

    cLogD: Intestinal pH is around 5 (5.4 is usually quoted, but chemspider calculates at 5.5? I’m not sure there’s a difference there.) I have clogp and psa – people actually use them?

    Here, have a spreadsheet.

  3. oleg said

    For BBB permeability ?

    logBB = –0.0148 PSA + 0.152 CLogP + 0.139

  4. tyrosine said

    Very interesting, thanks. I was wondering if you could also add standard deviation to the chart? It would be interesting to see which extremes are tolerated.

  5. pf01273 said

    Thanks for the spreadsheet; however I would say intestinal pH is around 7-8 in humans (slightly more acidic in other species such as rat). The pH of plasma is also around 7.4, hence why log D is usually calculated at this value.
    From a medicinal chemistry perspective I would say log P to be the property I consider most frequently.



  6. Kevin said

    You show the averages, but I’m wondering how many compounds at each stage of development violate the rules of 5. I know Tykerb has a MW >500. Do you think you could post that kind of breakdown?

  7. I have a general question; what percentage of all the drugs on the market currently violate Lipinsky (> or = 3 violations)?

  8. kinasepro said

    Thanks for your patience, and I think it’s safe to assume there will be several continuations of this post. I’m not terribly happy with it just now as some of the IV compounds seem to screw up the averages.

    For now have a primer on in silico Admet from some guy at organon.

  9. It’s worth pointing out that Rule of 5 makes no mention of ionization and uses logP rather than logD. There are a series of posts on Ro5 in The Crapshoot and interested folk might want to take a look there. It can be argued that logD is perhaps more appropriate than logP for rationalizing solubility and permeability but I’m not aware of anyone demonstrating this conclusively. It is more difficult to predict logD than logP because for acids and bases it becomes necessary to predict pKa. Personally I prefer to separate logP and pKa in QSAR approaches rather than bringing together two very different physical properties. Basically logD is what gets measured.

    Ionization is a complex factor in oral absorption. Ionization reduces the proportion of the dose that is neutral but typically increases solubility. As neutral form diffuses across the gut wall, it is rapidly replaced by conversion of ionized form (ionization equilibrium is typically fast).

    Finally a couple of comments on the BBB permeability equation above. First the equation suggests no dependence on ionization. Secondly both permeability and PSA have units and without these defined that equation is completely meaningless.

  10. […] Rule of 5 [image] […] […]

  11. kinasepro said

    I prefer clogd, but mostly just because it simplifies the picture, perhaps at the expense of some accuracy. Accurate or not, I don’t think that any single number is very meaningful – but all these numbers together may give an expanded view of a compound, lead, or class.

  12. Sanji Bhal said

    I’ve been trying to educate people about logD for years. As med chemists we’re so married to logP…but its only relevant if your compound is neutral in the pH range 2-9!! How many compounds have we made that fit that criteria? If we want to use these parameters to help guide discovery shouldn’t we be consider their real world behaviour? Chris Lipinski did a great thing trying to get us to think beyond efficacy but just because there weren’t many reliable predictors of logD when he did his study, or possibly because he was trying to simplify things…doesn’t mean we shouldn’t use the more appropraite term now!
    By the way…a personal gripe…clogP is a prediction algorithm belonging to a particular company, the general term for this property is logP…but what is clogD?? (and its linked to a paper I authored and I don’t remember using that term!)

  13. kinasepro said

    clogd and clogp have become the common man’s parlance for differentiating between a measured and calculated property… Thanks for pointing out that the term ‘clogp’ has a somewhat deeper meaning. I didn’t know that.

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: