«

»

Jan
27

High Throughput Screen and screen some more

We live in what I can only call an ‘interesting scientific time period’ with seemingly endless resources.  I am fascinated by how we think we can still just screen our way to finding the definitive drug for a disease. I was reminded of this today when an email arrived from the TB Alliance announcing the 2013 annual report. I was drawn to the detail on their TB pipeline..in which they describe that they will screen the libraries of 3 Japanese pharmas. Lets just remind ourselves how many millions of compounds have been screened in say the last 5 years or so. First the SRI/NIAID screened over 300,000 compounds. The Broad Inst. routinely test over 300,000 compounds, GSK screened 2 Million, Novartis 2 million, AstraZeneca probably screened 100,000’s if not more. Then of course there are various academic groups that have screened libraries from the low 1000’s to >50,000. In all these cases interesting hits have been found and may lead to in vivo active compounds.

After all this screening for compounds against TB we have several thousands hits (over 1000 from SRI/NIAID alone, 177 from GSK, etc…). So I am a little surprised to see that the TB Alliance is funding yet more screening. I would be interested to ask what they expect to find. We are overwhelmed with the number of hits already in the public domain. Many for targets we have absolutely no idea about. Then of course I also see in the same breath they are progressing me-too nitroimidazoles, I wonder how possibly this can be useful. Why are they not attempting to find better leads/ drugs than we already have. A bedaquiline without the side effects for example. These compounds will not be found by more high throughput screening. Perhaps we need to take a page from the golden age of TB drug discovery in the 1950s when screening millions of compounds would have been a fantasy. Yet still during this period the first line drugs appeared and are still with us. Remember this was also a time before the Mtb genome was sequenced. For all our brute force, and immense knowledge of the bacteria, cannot replace finding in vivo active compounds in the  mouse that translate to humans? We have already 1000’s of compounds that are active in vitro. How many would be active in vivo and how many of these could progress to the clinic? Time for a HTS screening truce and push some compounds further along in the pipeline.

Daiichi Sankyo, Shionogi, and Takeda
Daiichi Sankyo, Shionogi, and Takeda
Daiichi Sankyo, Shionogi, and Takeda

2 comments

No ping yet

  1. cdsouthan says:

    We could cogitate on this jointly but I am converging to your thesis that deposition of yet more disparate and discordant sets of HTS output (for TB or other NTDs) into the public domain actually confounds data mining more that it enables “progress” (depending exactly on how this is defined). As you imply, there must be 100s to 1000s of leads “out there” in public data sets, papers, patents and lab books that have never been robustly, reproducably and comparatively cross-screened en mass. Ditto many never deconvoluted to their molecular mechanisms of action. The Malaria box was one approach to this problem http://cdsouthan.blogspot.se/2013/07/a-peek-into-malaria-box.html but even this does not solve the problem, partly because new leads are being published rapidly but, beyond a few internal standards, never cross-compared in detail.

  2. sean says:

    Many thanks Chris, I think as usual you hit the nail squarely on the head.

    On the flipside, I am aware of several groups looking at the initially reported SRI TB hits but not reporting their results on re-testing. I think because of variability in testing for activity some see hits and others do not. Big issue.

    Others have also followed up and done Lead optimization of some of the hits as far as I can tell – generally they do not really attribute to SRI set, so people do not see the connection or the value of what has been done.

    Then there is the data out of the big pharmas. A for effort, but not sure what community can do – other than raise head, say thank you and get back to what they were already doing. Compared to 1950s we are drowning in in vitro data but try finding the in vivo data….

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>