Catalyst Trying to Improve Drugs by Lowering Doses

Edwin Madison, PhD, chief scientific officer, Catalyst Biosciences, says the company is developing protease therapeutic agents that work via a catalytic rather than a stoichiometric mechanism of action. This enables Catalyst to develop therapeutics that could maintain efficacy while using much lower doses of drugs.

For a given drug effect, the duration of action of a catalytic drug will be substantially longer than that of a stoichiometric drug and that in turn should lead to the ability to dose much less frequently, which might be particularly advantageous in situations such as, for example, a drug that has to be delivered by intravitreal injection,” Dr. Madison says. “We also think there may be an advantage to targeting C3 versus other proteins in the complement cascade. And that is because C3 is the one point in the cascade where one can block or inhibit the production of all anaphylotoxins as well as the membrane attack complex whether they are generated by any of the three arms of the cascade.” Catalyst is currently pursuing anti-C3 leads based on two proteases, u-PA and MTSP-1.

Participant:

Transcript:

Edwin Madison, PhD: Catalyst BioSciences has focused on the development of protease therapeutic agents that work via a catalytic mechanism of action. And the most obvious advantage to a catalytic mechanism of action versus a stoichiometric one is that the catalytic drug will be able to maintain efficacy at orders of magnitude lower concentrations. Therefore, for a given drug effect, the duration of action of a catalytic drug will be substantially longer than that of a stoichiometric drug. And that in turn should lead to the ability to dose much less frequently, which might be particularly advantageous in situations such as, for example, a drug that has to be delivered by intravitreal injection. Now we also think there may be an advantage to targeting C3 versus other proteins in the complement cascade, and that is because C3 is the one point in the cascade where one can block or inhibit the production of all anaphylatoxins as well as the membrane attack complex, whether they’re generated by any of the three arms of the cascade. And so consequently, we used our proprietary selection/counter-selection protease technologies to create orthogonal anti-C3 leads based on two distinct human proteases. Now the target product profile that we set for the program is listed on the slide. The efficacy and toxicity profiles are the second two points. In addition, we required that the ultimate lead contain 6 or fewer mutations, and that may have considered a little bit conservative for a drug that’s administered into an immune privileged site. But we’d rather be safe than sorry. So we then took two variants that we selected for each of the scaffolds, and we put them into a fairly comprehensive single dose escalation toxicity study in a non-human primate model, specifically the cynomolgus monkey model. And the good news from the study was that for one of two molecules for each of the scaffolds, we saw no observations at any dose. So we chose those two molecules to go forward into a study of the pharmacokinetics and pharmacodynamics of these two scaffolds or the variants of these two scaffolds. Now, what I show you here on this slide is some pharmacokinetic data for the small form, the short form of these two proteases. So these are isolated protease domains. Their mass is about 25 KD. So they’re about half the size of Fab so as expected, they have relatively short half-lives in the monkey vitreous, with a half-life of 1.7 to 1.8 days. Based on data published by the Genentech Roche group, we would expect that to correspond to a half-life slightly over 5 days in a human system. Now because of the relevance for the data that I’ll show you on the next slide, I want to point out that the two experiments with the UPA lead were highly consistent both with respect to the half-life measured for the drug as well is the in vivo recovery. And that’s in contrast to the MTSP lead, where again the measured half-life was very consistent. In fact, we got the same number, 1.7 days, in both experiments. However, the in vivo recovery in the two experiments varied dramatically. So in the first experiment, we observed a 100% in vivo recovery. In the second experiment we recovered only 18% of the dosed drug. So we suspect that that’s to do with a problem with the dosing solution in the second experiment. But since the experiment was done off site, we’ll obviously have to repeat it to be sure. But the pharmacodynamics data is shown on the next slide from these experiments, and you can see for both of the leads, either the UPA or the MTSP, there was a total inhibition of C3 activity out to beyond one week following the single injection of the drug. We clearly had lost all inhibition at the 28-day point, which would correspond to about 84 days in the human system with the UPA. However, the MTSP, for the reasons I mentioned before, was more difficult to interpret. There appeared to be a modest, albeit not statistically significant inhibition of C3 levels, even at this 28-day time point. However, again, for the reasons I mentioned, we’ll need to repeat this experiment. We also want to include a 20-day time point, which would correspond to every other month dosing in man. And this is because you would expect significantly greater efficacy because of the short half-life and the monoexponential decay. There would be about 25-fold more protease at the 20-day time point after injection than at the 28-day time point. So in conclusion, the anti-C3 protease leads that we have created are potent, stable and well tolerated in the non-human primate model. We believe that this anti-C3 protease approach can become significantly differentiated from antibody or small molecule approaches. And that’s based on coupling the potent catalytic turnover of C3 with a strategy to significantly improve the half-life. So for example, as illustrated by the figure at the bottom of the slide, we could put the protease domain back into the much larger full length construct. That would be one strategy that would not increase the antigenicity risk. Another strategy that would not increase the antigenicity risk would be to simply PEGylate the protease domain, and both those strategies have obviously been used successfully with either nucleic acid aptamers or antibody fragments by other investigators. So in conclusion, we hope that those strategies will allow us to create a drug that will demonstrate significantly greater inhibition of the progression of geographic atrophy, and yet require dosing only once every two to four months. Thank you.