we have not found a physiological difference between the brains of addicts and non-addicts
people are more likely to get addicted to drugs when their lives are terrible; only focusing on biomedical angles on tackling drug addiction means that it's not considered "real" drug-addiction work to try to improve underlying social problems like poverty or injustice
in particular drug-war policies are often part of the problem, and biomedical addiction research can't critique laws
https://www.science.org/doi/abs/10.1126/science.abb5920 this one didn't make the cutoff for my success-story post (only 1/10 patients had a CR) but it's astonishing that it does anything at all; a fecal matter transplant resulted in a complete response (and two partial responses) upon reintroduction of PD1 immunotherapy, in metastatic melanoma patients who had failed it before.
i am so disillusioned with FMTs that i might still chalk this up to a fluke, but who knows
really high complete response rates in metastatic cancers almost only occur when you have a topical/intratumoral/etc treatment physically localized to the tumor, frequently using an innate-immune mechanism.
that's also the literal majority of all historical cases of spontaneous tumor regressions -- they tend to happen when there's an infection at the tumor site, causing a powerful (innate! fever, inflammation, sepsis!) immune reaction.
the innate immune system is potent, and it is nasty, which is why you want to confine it.
immune checkpoint inhibitors are real good for metastatic cancer:
https://www.nejm.org/doi/full/10.1056/NEJMoa2028485 here's an antibody-drug conjugate for metastatic breast cancer. not enough complete responses to make it into my post, but look at that sweet Kaplan-Meier curve.
https://link.springer.com/article/10.1245/s10434-018-07143-4 isolated limb perfusion for melanoma: get higher doses of chemo into the tumor than the patient could survive otherwise, by cutting off circulation to the limb. when this sort of thing is possible, it really, really works.
https://link.springer.com/article/10.1007/s10549-022-06678-1 I hate on growth factor-targeted therapies a lot, but there are exceptions. Herceptin is a real drug. Look at this. 69 HER2+ patients presenting with metastatic breast cancer and treated with trastuzumab as part of their initial treatment, 54% get a complete response. 41% survived 5+ years after diagnosis. This is really, really solid.
electrochemotherapy is injecting tumors with cytotoxic drugs and electroporating the tumor so the drugs get in better.
It's only possible when you can physically access the tumor, i.e. when it's on the skin or when you're operating anyway (but can't surgically remove the tumor, because if you could, you would just do that).
if you can prove your computer program does what it's supposed to -- for almost any reasonable interpretation of "what it's supposed to" -- you will, as a side effect, also prove it doesn't have common security flaws like buffer overflows.
people I looked up while reading Neal Stephenson's Baroque Cycle: