The problem is you (as a scientist) don't know that an experimental subject will have a better life, or a life at all, and you don't have enough data about the risks to fairly compensate a person for taking those risks on. You're just making the cost of collecting data much more expensive while shifting the risk (of suffering, disability, death) from animals bred for the purpose to poor people.
While the topic was about animal suffering, you are talking about suffering in human subjects. As far as I know(and that is not much), human studies are done after animal studies. This means the drug is most likely not toxic to animals and by extension humans.