David's New Book

Thursday, January 17, 2013

Antibiotics and Grants - 4

Continuing in our series on grant writing for the discovery, optimization and development of anti-bacterials (much of this holds for antivirals and antifungals as well), I would like to focus on the transition from screening to optimization especially thinking forward to preclinical and clinical development.

First – before you even start your program, as we have discussed earlier, develop a target product profile.  For the sake of today’s discussion, I am going to assume that you want to have both IV and oral bioavailability so that step down therapy is an option.

During your optimization program, starting from one or more lead chemical series, you are going to want to follow these general parameters.

Potency (MIC)
ClogP or logD
Protein binding  - this should probably initially be just an MIC shift in the presence of 50% pooled human serum. 

You want to prioritize compounds with high potency, reasonable solubility and a less than an 8 fold shift in MIC in the presence of serum.  But you also should be able to balance these characteristics such that you can give up a little on potency in order to optimize solubility and protein binding.  You want to have a solubility of around 10 milligrams (not micrograms) per ml of water or saline.  Obviously you can start with a lower solubility but this should be your goal. 

At some point, you need to think about early in vivo studies.  There are two schools of thought here.  Some prefer early PK to assure bioavailability while others just want to see early efficacy.  I am more in the latter category.  The first in vivo model I prefer is an ED50 in treatment (at least 1 hour post challenge) in a lethal sepsis (peritonitis) murine model using a single bacterial pathogen.  I prefer administering the drug subcutaneously.  The PK for this is similar to IV – but the Cmax is frequently a little lower and this avoids some of the tox you can see with bolus IV administration of large doses of a novel compound. If your drug is aimed primarily at Gram negatives – start with E. coli for your first model.  These days, you should allow for early sacrifice of moribund animals. I am biased to this model just because of my own experience and I understand how to translate from this to more precise efficacy models.  Some would recommend starting with a model where you can show decreased bacterial burden such as thigh or pneumonia models.  I shy away from those for a first try in vivo because  the numbers of animals tend to be higher and the workload is also higher.  But this is certainly a feasible approach and I have a number of collaborators and clients who go that way. 

You will need to look at oral efficacy or PK (see below) if you are planning for oral bioavailability. 

If you see no efficacy and need to understand why – you might want to check PK (which means you have to have an assay) to confirm that you are getting exposure to the compound.  You also might want to check protein binding with mouse serum since that is occasionally very different than what is seen with human serum.  If this is an issue – do equilibrium dialysis both with mouse and human sera.

If you see low plasma exposure, plasma instability and metabolism are common problems – and you need to look at this early on anyway.  Plasma stability is easy to assess and can sometimes be the first indication of chemical instability in an aqueous environment. Stability in gastric fluid can also be problematic. 

Ultimately you will need an estimate of oral bioavailability.  This is easily done with a simultaneous IV and oral PK study.  Most people would accept a minimum of 10-15% oral bioavailability in a rodent as a reasonable starting point. 

A chemist can frequently identify potential sites of metabolism and sometimes you can pick metabolites up early during your assay development anyway. Metabolism tends to be a greater problem for oral than parenteral administration. Early assays on CYP inhibition are also helpful.  If metabolism, plasma stability or CYP inhibition are problematic – bring these assays forward in your optimization program.

This then becomes the optimization program for your series.  Again, you can balance potency with favorable physicochemical properties, protein binding, plasma stability, metabolic stability, oral efficacy (bioavailability) etc.  There may be other specific assays related to your specific lead series or your target product profile that I have not mentioned. 

I know this is a short summary – but I’m not writing your grants for you.  

The next installment will deal with preclinical development leading up to first in man studies.