I believe that most of the time patients are enrolled in these trials, they've already exhausted all other options. It isn't quite "life-or-death" but rather "death-or-death-or-maybe-another-chance-at-life"
It doesn't excuse the doctors for not being direct about the choices at play: you are trying to get into an experiment where there's a 50% chance you'll get the same treatment you'd get anyway (one hopes -- it's what is supposed to happen, but some trials have been less-than-ethical), or a 50% chance you'll get a drug that might well be worse than the control.
Particularly for phase 1 trials, that last part needs to be emphasized. You're not doing it to survive. You're doing it because it's the ultimate altruism -- using your own life to find an answer that might help the next patient.
Patients absolutely enroll in ph1 trials because they want to live, and doctors enroll them because they want them to live too.
A lot of time and money goes into picking drugs that companies and doctors think will perform better than the stand of care. That's the whole reason the trial exists. It exists because there is a plausible argument it will be better than the alternative.
If the currently-available treatments mean I'm going to die, "try something that might kill me or cure me" is absolutely a thing I would consider "to save my life".
Unfortunately, by the time most people reach the "no hope" stage, the disease has processed to the point where no intervention would help.
Conversely most interventions work best at the "early detection" phase. Which (ironically) is when you're best off with proven stuff, not new experimental stuff.
This catch-22 is precisely why its so hard to get improvements in this field.