6 ms·

Looks like it's happening: "Modified blood stem cells reverse type 1 diabetes in mice" https://www.diabetes.co.uk/news/2017/nov/modified-blood-stem-cells-reve

Looks like it's happening: "Modified blood stem cells reverse type 1 diabetes in mice" https://www.diabetes.co.uk/news/2017/nov/modified-blood-stem-cells-reverse-type-1-diabetes-in-mice-96647921.html https://www.diabetes.co.uk/news/2017/nov/modified-blood-stem... And type 2 got cured by an extreme diet in 86% of patients: https://sciencealert.com/extreme-diet-reverse-type-2-diabetes-up-to-86-patients-remission-weight https://sciencealert.com/extreme-diet-reverse-type-2-diabete...

aaavl2821 9y ago
hate to be the bearer of bad news, but there are something like 400+ drugs that have prevented type 1 in mice and another hundred or so that have reversed it. none have worked in people the most common animal model for type 1, the NOD mouse model, is notoriously bad. studies have shown that loud music cures diabetes in these mice. mice raised in one lab get diabetes differently than mice raised in another. the immune systems of these animals are very fickle and almost any environmental perturberance can change an immune response
- test6554 9y ago
  Lab mice are truly living in a golden age of medicine!
  - sorokod 9y ago
    The humans in the 15th century lived in the golden age of the plague.
- ImSkeptical 9y ago
  It's some kind of philosophical ethics problem that we can't experiment on humans even though there would be a net benefit. I feel like there are morally unintuitive solutions that may help - perhaps criminals of certain kinds, who have damaged society, could repay their debt, in part, by volunteering for trials we can be sure they understand. We might also leverage The suicidal by offering, after therapy, a way for them to live in comfort while contributing to humanity as experimental subjects. I realize the mind recoils instinctively at these ideas, but it seems so frustrating and horrific that we have to delay life saving medical advanced due to insufficient experimental subjects.
  - aaavl2821 9y ago
    I understand where you are coming from logically, although I am of the belief that this is not ethically acceptable. Ethics aside, jumping straight to human testing without assessing safety wouldnt really save that much in terms of cost and time to develop a drug, compared to developing better in vitro and animal models of disease safety studies are not the largest driver of cost and duration of drug development. failing a later stage human study that assess effectiveness is a much bigger needle mover in terms of cost, as is the arduous and painstaking process of early drug discovery and development. human studies of effectiveness cost $15-500M+ and can take 3-7 years. putting a drug in humans that has a low probability of success is the most value destructive thing you can do in pharma. the other big cost driver is early research in drug discovery and development. understanding the biology and chemisty, developing assays, and screening compounds can cost $15-20M+ and take 5+ years. by comparison, animal safety studies cost maybe $5M and take a year or two For more context: There's an important distinction to be drawn between two main types of animal models: models assessing drug "safety", and those assessing "effectiveness". the models of effectiveness are very poor and of questionable value in some cases, although they are often the best we have. animal models of effectiveness are not literally required for FDA approval, although in reality they pretty much are animal models of safety, however, are required by FDA, and rightly so. they are of much greater value. generally you must do studies in a small species like a mouse as well as a non-human primate. these "toxicology" studies basically entail dosing animals with huge amounts of drug, way more than you'd dose in humans, and then seeing what doses are not toxic. this informs the first dose youd do in humans the initial human studies are not done to study effectiveness, but to study safety. based on the data from your tox studies in animals, as well as other studies, you gradually dose cohorts of patients with higher doses, watching carefully for safety signals.
    - tialaramex 9y ago
      IIRC There's already a special case where you can volunteer to try things if you're dying. If you have a rare cancer and doctors are like "Six months maybe less" and there's some crazy Hail Mary drug, which could work but isn't tested yet, you can volunteer to try it and see what happens. You can't pay (so it's for science, not a chance to get rich selling false hope), and you must be advised by doctors who have no financial interest, something like that. Judges were like, eh, you're dead anyway, what's the worst that could happen? This trial was phase I/IIa which I believe means it's combining safety in humans with initial efficacy data. The thinking is, the real treatment is a single dose. If we give participants less than a full size dose we don't learn much from that. If we give them a real dose, we might as well see if it works. So their endpoints were firstly how is the treatment tolerated, then secondly does it reduce the need to take clotting factors? Hence this good news story.

aaavl2821 9y ago

hate to be the bearer of bad news, but there are something like 400+ drugs that have prevented type 1 in mice and another hundred or so that have reversed it. none have worked in people the most common animal model for type 1, the NOD mouse model, is notoriously bad. studies have shown that loud music cures diabetes in these mice. mice raised in one lab get diabetes differently than mice raised in another. the immune systems of these animals are very fickle and almost any environmental perturberance can change an immune response

test6554 9y ago
Lab mice are truly living in a golden age of medicine!
- sorokod 9y ago
  The humans in the 15th century lived in the golden age of the plague.
ImSkeptical 9y ago
It's some kind of philosophical ethics problem that we can't experiment on humans even though there would be a net benefit. I feel like there are morally unintuitive solutions that may help - perhaps criminals of certain kinds, who have damaged society, could repay their debt, in part, by volunteering for trials we can be sure they understand. We might also leverage The suicidal by offering, after therapy, a way for them to live in comfort while contributing to humanity as experimental subjects. I realize the mind recoils instinctively at these ideas, but it seems so frustrating and horrific that we have to delay life saving medical advanced due to insufficient experimental subjects.
- aaavl2821 9y ago
  I understand where you are coming from logically, although I am of the belief that this is not ethically acceptable. Ethics aside, jumping straight to human testing without assessing safety wouldnt really save that much in terms of cost and time to develop a drug, compared to developing better in vitro and animal models of disease safety studies are not the largest driver of cost and duration of drug development. failing a later stage human study that assess effectiveness is a much bigger needle mover in terms of cost, as is the arduous and painstaking process of early drug discovery and development. human studies of effectiveness cost $15-500M+ and can take 3-7 years. putting a drug in humans that has a low probability of success is the most value destructive thing you can do in pharma. the other big cost driver is early research in drug discovery and development. understanding the biology and chemisty, developing assays, and screening compounds can cost $15-20M+ and take 5+ years. by comparison, animal safety studies cost maybe $5M and take a year or two For more context: There's an important distinction to be drawn between two main types of animal models: models assessing drug "safety", and those assessing "effectiveness". the models of effectiveness are very poor and of questionable value in some cases, although they are often the best we have. animal models of effectiveness are not literally required for FDA approval, although in reality they pretty much are animal models of safety, however, are required by FDA, and rightly so. they are of much greater value. generally you must do studies in a small species like a mouse as well as a non-human primate. these "toxicology" studies basically entail dosing animals with huge amounts of drug, way more than you'd dose in humans, and then seeing what doses are not toxic. this informs the first dose youd do in humans the initial human studies are not done to study effectiveness, but to study safety. based on the data from your tox studies in animals, as well as other studies, you gradually dose cohorts of patients with higher doses, watching carefully for safety signals.
  - tialaramex 9y ago
    IIRC There's already a special case where you can volunteer to try things if you're dying. If you have a rare cancer and doctors are like "Six months maybe less" and there's some crazy Hail Mary drug, which could work but isn't tested yet, you can volunteer to try it and see what happens. You can't pay (so it's for science, not a chance to get rich selling false hope), and you must be advised by doctors who have no financial interest, something like that. Judges were like, eh, you're dead anyway, what's the worst that could happen? This trial was phase I/IIa which I believe means it's combining safety in humans with initial efficacy data. The thinking is, the real treatment is a single dose. If we give participants less than a full size dose we don't learn much from that. If we give them a real dose, we might as well see if it works. So their endpoints were firstly how is the treatment tolerated, then secondly does it reduce the need to take clotting factors? Hence this good news story.