You don't - you compare it to "standard of care" (antidepressants, CBT, etc.). If we can compare an inferior treatment to placebo, and we can compare a breakth

You don't - you compare it to "standard of care" (antidepressants, CBT, etc.). If we can compare an inferior treatment to placebo, and we can compare a breakthrough treatment to this established standard, we can evaluate the new treatment.