I hate to be the bearer of bad news but people must know that even medicines that have been extremely well tested and obtain a sales licence still occassionally turn out to have a dangerous side-effects.
Of the RCs, I reckon I've looked at the GC-MS / NMR instrumentation of well over 2000 samples and the class where I noted the most impurities were the synthetic cannabinoids. I kept seeing compounds that had a MW almost exactly x2 or x3 that of the expected compound. These turned out to be dimers and trimers. The problem is that with such a high MW, they won't vapourize ahead of a flame-front i.e. in a joint. Instead they will be pyrolized and the pyrolysis products are extremely carcinogenic. I fully expect to witness a peak in cancer cases among people who used that class of RC.
But we are also seeing cathinones that are growing ever further from that which is known. It's not so much a matter of if, but simply a when one of those turns out to have an unwanted side-effect.
Even if it's only a patent, if a drug has been tested in animal models, at least we have a little information although to be clear, animal models can fail in quite spectacular ways. One is that rodents appear to be much more sensitive to opioid analgesics that humans. So when a novel opioid it claimed to have some ridiculous potency, don't forget that the original animal (rodent) studies of BDPC stated that it was >10000x more potent than morphine. In man it turns out to be x504. My best guess on that one is that rodent pain models are much more subject to NOP (norciceptin receptor) ligands than man. Certainly I know someone who made the 1,3,8-triazospiro class opioids Janssen had listed as being x450 morphine in animal models. Let's just say the truth was that in man, it wasn't inactive, but it wasn't good enough to be worth the time and effort.
To be fair, I did find and send them the later patents that demonstrated that they were in fact NOP ligands but they wanted to be certain. But as I've said, you can expect to make a dozen novel opioids before you find one that is actually 'good'. Becuase nobody really knows why one opioid is euphoric while another is not. As a rule-of-thumb, the more potent the opioid, the less euphoric it is.
Hence H, dextromoramide, dipipanone, levorphanol, oxymorphone and so forth still being sought by users. Yes, U-47700 was a lucky find, but it wasn't the first or even the tenth compound tried.