Not to denigrate the scientific effort and achievement of the people creating this, but if even a fraction of what our medicine can do in mice were at all workably, usefully applicable to humans, we'd be living dream lives of medical therapy by now. Too bad so much that works in said mice would instead simply kill us stone dead, or just do nothing.
I don't get why there's a comment like this in every research related thread on HN.
Yes, they don't translate in the sense that "works in mice != works in humans". But it's a pretty iterative and gradual process by which we make those discoveries, and the preclinical work is an important first step.
A lot of applied medicine, especially recently around tumor immunology, immune checkpoint inhibitors, and such started with preclinical investigations in mice and then worked towards humans. The models certainly have limitations but they are known by the researchers and either factored into the interpretation of the studies or improved upon to create better models.
(I did liver cancer research in the past, both in clincial trials and pre-clincial work where we tried therapies in mice / developed new mouse models that better replicated the tumor microenvironment for therapies).
A non-zero fraction of what our medicine can do in mice, in fact, works in humans. That is why testing of mice hasn't been abandoned, unlike, say, Galen's theory of bodily humors. Yes, that fraction is lower than you or I would like it to be.
Mice are short-lived and cancer-prone, no wonder that the discrepancy between them and humans is big. They even do not get some of the diseases that we do (e.g. the murine models of ALS are pretty useless). But they are cheap, too, which counts.
It is possible that, in the future, a lot of the testing will be done with organs on chips, much closer to actual human tissue. But we don't really have anything more efficient right now.
Delivery systems are just as important as part of the cure, or the actual substance killing the tumor cells. If we had better targeted delivery for platinum-based chemo, we could try a bit more things for different patients (increased dosage, low-dose infusions for long-term plans).
5 comments
[ 3.0 ms ] story [ 22.3 ms ] threadYes, they don't translate in the sense that "works in mice != works in humans". But it's a pretty iterative and gradual process by which we make those discoveries, and the preclinical work is an important first step.
A lot of applied medicine, especially recently around tumor immunology, immune checkpoint inhibitors, and such started with preclinical investigations in mice and then worked towards humans. The models certainly have limitations but they are known by the researchers and either factored into the interpretation of the studies or improved upon to create better models.
(I did liver cancer research in the past, both in clincial trials and pre-clincial work where we tried therapies in mice / developed new mouse models that better replicated the tumor microenvironment for therapies).
Mice are short-lived and cancer-prone, no wonder that the discrepancy between them and humans is big. They even do not get some of the diseases that we do (e.g. the murine models of ALS are pretty useless). But they are cheap, too, which counts.
It is possible that, in the future, a lot of the testing will be done with organs on chips, much closer to actual human tissue. But we don't really have anything more efficient right now.
More on lipid nano-particle delivery: https://www.nature.com/articles/s41578-021-00358-0
Looks like the authors started a company: https://www.hopewell-tx.com/
Delivery systems are just as important as part of the cure, or the actual substance killing the tumor cells. If we had better targeted delivery for platinum-based chemo, we could try a bit more things for different patients (increased dosage, low-dose infusions for long-term plans).
Albumin-bound paclitaxol is an example of this.