• wewbull@feddit.uk
    link
    fedilink
    English
    arrow-up
    12
    arrow-down
    1
    ·
    1 year ago

    Also sounds very hard to do a proper controlled trial on. Every treatment produces a different protein, so there’s no consistent factor to test except for the delivery mechanism.

    • Natanael@slrpnk.net
      link
      fedilink
      English
      arrow-up
      9
      ·
      1 year ago

      There’s still ways but not trivial. You have to do multifactor analysis, but it’s gonna have a ton of noise unless you have a large sample of different people with recurring “neoantigens”. It’s similar to how drug side effects are tracked for people who take multiple medicines, you compare against populations which share different combinations of the same factors.

      • wewbull@feddit.uk
        link
        fedilink
        English
        arrow-up
        3
        ·
        edit-2
        1 year ago

        Multifactor analysis still requires an underlying commonality. People taking multiple drugs are all still taking the drug being trialed. You’re removing the confounding factors. If every treatment is a unique cancer protein there is no common factor. The treatment is the confounding factor.

        To put it another way. A safety trial has to prove that any protein administered is safe.

        Edit: just realised you’re probably talking about efficacy trials, whereas I’m more concerned with safety.