When the treatment puts a large number of terminal patients into remission, there is no statistical reason for a placebo control. You know what percentage of su

When the treatment puts a large number of terminal patients into remission, there is no statistical reason for a placebo control. You know what percentage of such patients will die when offered a potentially revolutionary therapy that does nothing, because there are previous therapies that didn't work. If one third of your patients go into remission, one third show improvement and the treatment has no effect on one third, you know the treatment works better than a placebo. As for offering this to desperate, last chance sufferers, unfortunately the history of potentially revolutionary cancer treatments shows this is a bad idea. If the treatment kills 90% of people within hours or days of treatment (sadly this is a realistic outcome before dosing is worked out and in some cases, regardless) then it is morally unacceptable to roll it out to thousands of people. You have to do the trials first to establish whether it works, is safe, that it doesn't cause unnecessary suffering and hasten death and what the optimal dose and protocol is. You also have to figure out how to deal with patients who do have an adverse reaction to the treatment.