Drug candidates stability testing

The model can be used to evaluate the metabolic stability from the 3D structure of drug candidates prior to experimental measurements. Thus, we have used a test set of 1346 compounds from Johnson Johnson [20]. 

Reasons for chopping clinical candidates at any stage of the drug approval process included (1) stability, formulation, or other pharmaceutical development issues, (2) renal toxicity or neurotoxicity, and (3) insufficient advantage over current chugs. Since 1997, the NCI has also operated a screen for compounds active against the cytotoxic effects of HIV in CEM cells. Of 80,000 compounds tested, 4050 (or about 5%) were active. Of the compounds tested, 2291 have included metals. Of those, 136 (about 6%) were active, and two became clinical candidates. Both were chopped due to toxicity problems. One clinical candidate was a polyoxometallate, and therefore about 80 other similar molecules were tested. These were found to be strongly active in vitro, but too toxic in animal models in vivo. If a way around the toxicity problem can be found, interest in these... 

---