For GC antibiotics, the development path looks like a trip down the road less traveled, leading into some uncharted territory.
Let’s take Melinta’s statement [1] at face value according to which the single 900 mg dose of delafloxacin failed because of “insufficient efficacy”.
So, what then determines efficacy? Delafloxacin’s MIC90 was between 0.125-0.5 mg/L which should have made it an excellent drug. Besides organism MIC, one would think that PK / PD had something to do with organism eradication. As the primary efficacy endpoint in the recent FDA Guidance is a negative culture at Day 3-7 [2], issues related to relapse or reinfection won’t dilute the success rate.
Because GC treatment is one-time therapy, a drug with long half-life has a higher chance to deliver. As the gonococcus is an intracellular pathogen, tissue penetration may also be an important feature.
Besides delafloxacin, there are only 2 drugs at an advanced stage of clinical development: AZD-0914 and solithromycin/CEM-101. For both it will be important to get the PK/PD relationship right.
Some older literature have addressed the MIC / PK / efficacy relationship. Moran et al, based on an extensive data review, provides a simple formula:
What he terms ‘therapeutic time’, is in essence T> 4x MIC90 . His analysis concludes that for a drug to be effective against urogenital GC it has to provide exposure for >10 hours above 4x the MIC90 [3].
So, let’s put this formula to use with existing information for our candidate drugs:
It is clear that both ceftriaxone and delafloxacin have no problem achieving ‘therapeutic times’ much longer than 10 hours. So, the formula is not holding up here. Only when factoring in the unbound fraction, the T> 4x MIC90 drops below 10. But why would it affect delafloxacin and not ceftriaxone which – as we know – has retained full activity despite a very small unbound free active fraction?
What a pity: yet another formula which is doing well for retrospective analyses but not so much for forecasting. You could call it a validation failure. Or, for the more cynically inclined: Past performance is not a guarantee for future success.
WITH THIS MAJOR CAVEAT, let’s move on to solithromycin for which the formula paints a very dismal picture indeed: here the PK parameters predict only 4 hours of T> 4x MIC90. In addition, the drug has a very small unbound fraction. We already heard of tolerability concerns: The dose was recently reduced (from 1200mg to 1000mg) indicating that nausea/vomiting has been an issue, is an issue, and will continue to be an issue.
Our last attempt to apply Moran’s formula: AZD-0914. This new gyrase/topoisomerase inhibitor is now being developed by Entasis Therapeutics, an AstraZeneca’s spin-off . Insufficient PK information in the public domain makes it difficult to use Moran’s formula, so it is anyone’s guess how this drug compares to the other newcomers. But it looks better than solithromycin assuming the Cmax is > 4 mg/L (which would bring it over the 10 h threshold).
Which seems quite doable given the whopping dose of 3 g orally.
References:
[1] http://www.melinta.com/news.php?c=41
[2] FDA Guidance June 2014: Uncomplicated Gonorrhea
[3] J Moran CID 1995;20 (Suppl 1): S47
