Cancer Vaccine Works 'Startlingly Well' in Mouse Model (medscape.com)

First sentence shows something is wrong:

>"It has recently become apparent that the immune system can cure cancer." https://doi.org/10.1126/scitranslmed.aan4488

Recently? This has been thought for quite awhile... https://www.ncbi.nlm.nih.gov/pubmed/?term=immune+surveillanc...

And really it is obvious that there are trade offs between immune surveillance of cancer and auto-immune disease. Next they will be surprised at the tradeoff between aging/organ failure and cancer.

I'm sure in an alternative universe there's a mouse complaining about how many vaccines seem to work so well on the humans they test but too few make it market.

Maybe we should skip the middleanimal? I'm sure some of those drugs that failed to work on surrogates are perfectly fine for humans.

As I understand it, many diseases that wild mice almost all carry have been eliminated in lab mice, using some fairly crazy measures. They also have very little genetic variation due to the way they have been bred.

I think there is a very interesting scifi novel in reading about how that stuff was achieved and what human society would be like after it was applied...

It is just easier to cherry pick results, p-hack, etc for mouse studies since there is less oversight. Once you get into even primate studies, where the data is expensive enough, the level of crap drops substantially (although is still very high in absolute terms, I mean like 99.99% to 99% junk).