3 ms·

That's a critically better health outcome though. It's like fixing a corruption bug when you detect it on one server, versus finding out there's corruption in e

• carlmr 5y ago
  Ransomware infecting one machine or all machines including backups would also be a good analogy.
• jacquesm 5y ago
  > That's a critically better health outcome though. Not necessarily. It depends on how old you are, what your genetic make-up is, whether the cancer is growing rapidly, whether it has easy access to other organs to spread to (or has already begun to do so) and so on. It's not a binary thing.
• Gatsky 5y ago
  I’m not saying it isn’t a great development. It certainly adds an option for early detection of some cancers where there previously were none. But it isn’t cancer prevention. It is a highly sophisticated technical undertaking which for educated and motivated patients in well resourced health systems will reduce cancer mortality and morbidity by some increment. Depending on many factors this could be a big or small increment. In the real world the efficacy will be less than what the trial shows. It is also likely that Grail (with north of $1 billion in VC funding) will charge as much as they can for this test. The size of the trials necessary to prove efficacy means that there will be highly constrained competition in this space. The distinction with prevention is important, because we ultimately must aim for preventing almost all cancers, and not be content with anything less than that. Disclaimer: I’m an Oncologist.