Biodefense - Science topic

@ Huyen  You need to at least do your research and correctly spell the method. It's Bredig's method dating from 1898

See slides 34 onwards in : February 18th, 2016 Metal colloids - their preparation, application and characterization

And also:

September 2nd, 2015 Silver colloids and invisible ink

Thank you very much Leonidas...

It's not going to be easy.

I was involved in an EU-wide project (FASTVAC) over the last few years looking at exactly this question: how do we produce vaccines faster in response to new biological threats?

There are two aspects - first you need to actually identify the pathogen, and generate a response that is protective. In the case of a new pathogen, that might not actually be very obvious - look at how long it took to work out what a protective vaccine against HIV might look like. Anyone who thinks that all you need to do is identify a few immunodominant antigens and inject them is pretty much doomed to failure: vaccine development history is littered with immunogenic antigens that offer no protection at all, when used as vaccines and a few which actually make the disease worse.

That leads to the second aspect - regulation, release and safety. Vaccines are different from almost all pharmaceuticals in that they are potentially given to millions or even hundreds of millions of perfectly healthy people. As a result, you cannot ignore rare side effects: even a side effect in the "very rare" category could lead to thousands of cases if a vaccine is broadly used. That's not going to be acceptable even in an epidemic situation.

So any new vaccine is going to have to go through clinical trials, both for efficacy and for safety. Typically this can take 10 years or more from first application, and clearly that's no good for emerging threats. It can be done faster though: we routinely make new influenza vaccines every year, and researchers were able o produce a protype SARS vaccine ready for trials in about 6 months. In both cases, though that is possible because we already know a lot about the pathogen in question (or related pathogens) and immunity to these types of virus, and because there was already a process in place for making that type of vaccine.

So while I have great respect for Annie and the Epivax team, I don't buy the idea that we can (yet) create vaccines on demand. So far, computational identification has not proven to be a success in terms of rapid vaccine design, simply because the vast majority of targets identified are not protective. We've been playing with epitope identification for vaccines since the 1980's (it's what I did my PhD on, in fact!) and so far, we don't have a single success to point to. It does speed the process of identifying potential targets, but those targets then have to be tested in vivo in the old-fashioned way.

The Bexsero vaccine is a good example of this. Computational analysis identified 570 potential antigens which should be immunogenic - these were cloned and tested. 28 of them showed some degree of protective ability. That's a 5% hit rate, which is actually pretty good. Of those 28, only 7 showed actual vaccine potential - for a final hit rate of 1% and that's fairly typical, based on what's been published and my own experience. But to work that out, and then demonstrate safety and efficacy took 12 years. Identifying targets was by far the fastest and easiest part of that process. Computational analysis can speed that first step, but it doesn't remove the need for the subsequent steps, which is actually what takes the most time.

So if we genuinely want to speed up vaccine development, we need:

1) better understanding of biological processes, so we can predict what's protective, not just what's seen by the immune system. This is going to be a major challenge for computational biology in the years ahead

2) better processes for approval (and preferrably pre-approval) of vaccine production, so that we can avoid having to plod step-by-step through the process in an emergency.