> Li began collaborating with two professors at the University of San Francisco: Hani Goodarzi, also an incoming Arc Institute Core Investigator, and Laura Van’t Veer, a clinician who leads the I-SPY 2 Trial, a groundbreaking breast cancer trial.
These famous researchers are actually at UCSF, not University of San Francisco. UCSF is really a leader in breast cancer the world over, far more influential than Stanford's much smaller amount of faculty on the topic. So it's a bit weird to see such a basic mistake.
For those who are not familiar, the I-SPY 2 trial is quite remarkable. It is an adaptive trial design - in essence reinforcement learning with human experiments.[1]
For all the criticism FDA gets, they should also get credit for leaning into new approaches to testing drugs like adaptive trials. Part of why the US remains one of the most innovative countries.
It should be clarified that adaptive trial design was studied for several decades before being applied to games and called reinforcement learning by computer scientists. The multi-armed bandit papers all reference "Asymptotically efficient adaptive allocation rules" because they derived the optimal lower bound for regret [1].
I genuinely appreciate the link, but not so much the campaigning for the FDA or USA more broadly, which, my stance aside, feels like a trap leading to a fruitless discussion.
IIRC they recruited and actually signed a kid (Football maybe?) who later decommitted once he realized it wasn't in Florida. I'm not sure that's entirely accurate because I don't think you get to signing without an on-campus visit but maybe that isn't always true.
Maybe -- I don't watch that show, too close to real life.
I tried to find any reporting on this, and while apparently it's happened to a few students (they applied, got accepted, then discovered the school is in rural Ohio and not Florida) but not the specific case of it happening with an athlete (seems really unlikely).
My wife was telling me about a foreign student who got a full ride scholarship to Miami university and was so excited to move to Florida… only to realize it was in Ohio
As much as I dislike many aspects of living in most of both Ohios (NE (which, to me, should be Cleveland/Akron/Canton/Kent/Youngstown/PGH/Buffalo),vs the rest of the state), I am super impressed with the leniency (latitude?) tolerated in that place re: "academic freedoms," while not going [dipshit] (ok, I cannot describe or conceive of a more comprehensive term, my bad, my apologies).
Pennsylvania has both Indiana University of Pennsylvania (located in the town of Indiana, PA) and California University of Pennsylvania (in California, PA). Both were formally teachers colleges named after their respective towns.
>> Patients with high ENPP1 levels had low response to pembrolizumab and high chance of metastases. Those with low ENPP1 levels had a high response to pembrolizumab and no metastases. ENPP1 predicted both response to immunotherapy and likelihood of relapse.
I'm most interested in head and neck cancers, and pembrolizumab (Keytruda) is only effective in 20 - 30% of patients; see e.g. https://ascopubs.org/doi/full/10.1200/JCO.21.02198: "In the phase Ib KEYNOTE-012 study of pembrolizumab monotherapy in R/M HNSCC (N = 192), objective response rate (ORR) was higher in patients with PD-L1 CPS ≥ 1 than CPS < 1 (21% v 6%; one-sided P = .023)."
A lot of oncologists I've talked to also think PD-L1 expression is overrated as a marker of pembro effectiveness; obviously there's something going on, with higher PD-L1 expression being net better in terms of pembro utility overall, but the data are noisy and lots of exceptions abound. I'm also not sure how reliable PD-L1 expression tests are. My tumor was tested once by a well-known-in-the-field company called CARIS, and it showed a 20% Combined Positive Score (CPS). Another test done by the Mayo Clinic showed a 5% score. Who's right? Who knows. No one I've talked to. The attitude seems to be that no one is convinced the PD-L1 tests are highly accurate, and that once the CPS is above 1%, pembro's effectiveness is a crapshoot anyway. I got five doses of pembro and failed it, utterly.
From the article:
> This means that clinicians can use ENPP1 levels to better determine appropriate treatment for breast cancer patients. It also means that drugs that destroy the ENPP1 dam could make existing therapies more effective – and several ENPP1 inhibitors are already in clinical development.
They found a correlation between a protein (ENPP1) and breast cancer metastasis that they feel is a very strong signaler for pembrolizumab immunotherapy resistance. This would be helpful information to help get to the right dosage of medicine faster (or skip a class of drugs entirely), which might save weeks or months of taking a drug that effectively is neutralized by the resistance.
Additionally, hopefully new medicine will (after more research) go after this protein and see if attacking it allows the immune system to better resist cancer itself.
It's also worth noting that two of the researchers involved in this work own a company to try to utilize patents around this work.
>Facts can be seen as important without implying any specific interpretation of them.
This isn't even true. In the example you gave, you clearly remember the speed of light because you think it may be useful someday.
When someone believes a fact is important or "worth noting", they used some reasoning to arrive at that conclusion. Whether or not that reasoning is flimsy is another matter.
As an uninvolved bystander I can agree that the reason it is noteworthy can be separated from the fact of whether or not it is.
In this case I think there are a few reasons of note which were already alluded to, but I don't personally think it need call any questions about the validity of the experiments and the other reasons being around the potential future value of associated company, any opinions of which are for the individual and their financial advisors to consider.
I wasn't aware of the fact there was already a company created for it but it does say something about the confidence these researchers have placed in the results, even if that may not be noteworthy, per se, regarding the results.
Right. And additionally, it provides at least a trace of possibility that there's some sort of a vested interest in the outcome of the research, although it shouldn't necessarily be held against them that they believe in their work and want to see applications from it.
Of course it is. This is one of those 'only on HN' moments. You absolutely can be cognizant that a fact may give rise to any number of competing interpretations, and you can understand those interpretations to be significant if true, while not endorsing them.
>When someone believes a fact is important or "worth noting", they used some reasoning to arrive at that conclusion
Again, you can understand that a fact gives rise to certain interpretations without endorsing those interpretations, and this could nevertheless be enough of a reason to merit sharing the fact.
So I think there may be some equivocation of the term "interpretation" in this thread--whether intentional or not--but I understand the point being made here.
My point is that when one determines that a fact is notable, there is an underlying reason for making that determination, whether it involves several "competing interpretations" or not. It is perfectly reasonable to ask for that underlying reason so I can assess the validity of the statement. If someone refuses to give it, as we've seen here, then its reasonable to dismiss their assertion. I'm not responsible for justifying their own argument for them.
As for the 'only on HN' remark, I think that was unnecessary. At any rate, this will be my last comment on the subject.
I don’t think knowing c is going to particularly impact most people.
On one level it’s essentially random trivia you can easily lookup, but on another level it’s fundamental in a way that few things are. Individual mutations from random events wildly shaped human history making most facts people remember essentially happenstance. However, c is more closely tied to the underlying aspects of reality.
I let other people chime in, but to provide an answer myself:
I felt that it was worth noting because, depending on the viewpoint of the reader, that information could be seen as a positive or a negative, and could have an impact on their viewpoint.
1. The existence of the company suggests that there might be a meaningful or profitable future from what has already been learned.. commercialization of this information might be more readily possible.
2. Someone who's cynical might question the validity of the research and/or the motives of people involved.
I certainly have my own opinion, but I also have not deeply researched this subject, or the researchers involved. I was curious to get some high level information on this topic so I read the article. I felt it was hard to skim through so, after reading it, I wanted to provide the high level summary as I understood it for others here.
If you're thinking we could test pharmaceuticals in a simulation I think thats a long way off. The worm model is not sufficient to model chemical reactions/lifecycles.
'Inmiced' lol! Thanks for keeping the titles accurate, so important especially with cancer research results. Happy holidays and I really enjoyed reading the quote on your profile https://news.ycombinator.com/user?id=dang
idk why I see so many comments like this on the medical threads of HN lately. Laboratory mice are close enough to us to do testing on and have provided reliable results throughout their use. A basic Wikipedia search shows references for all of those claims and more: https://en.wikipedia.org/wiki/Laboratory_mouse
That's because mouse models suck for many diseases. Cancer very much included, for two reasons:
1. Mice _love_ to get cancer naturally. If you have 100 mice, it's quite likely that around 20 of them will die within one year of cancer. This makes it difficult to extract useful signals. BTW, that's why if you see a study that a "chemical X results in cancer in mice", you should take that with a grain of salt.
2. Mice are small, so tumors are necessarily small too, with several orders of magnitude fewer cells than typical human tumors. So many drugs can just cure mice of cancer entirely, by killing cancerous cells too quickly to allow them to evolve resistance.
> If you have 100 mice, it's quite likely that around 20 of them will die within one year of cancer.
If you have 100 people, about 40 of them will be diagnosed with cancer over their lifetime (39/100 females and 41/100 males). Note "diagnosed with" is very different from "die" and "over lifetime" is not the same as "within one year", but the probability of people getting cancer naturally is high as well.
Sure. But in general, cancer risk should scale with the number of cell divisions, which is fairly low for mice. And other small, short-lived mammals like weasels are not as susceptible to cancer. Naked mole rats (also rodents!) are downright cancer-proof.
This all means that mice don't have a good natural cancer-suppression machinery, and this in turn makes them somewhat awkward to use to discover new treatments.
I'm not sure it follows that high incidences of cancer means that it's unhelpful to use them for research to understand the mechanisms of cancer. And I'm not sure that either of your observations one or two are pertinent in this case as it applies to the on/off protein switch, as it is not about guaging the statistical frequency of cancers or about killing cancer cells.
It’s a common trope in the field that we’ve already cured cancer in mice. They are similar enough to be a useful model for study. However, they are still quite different to the point where you never quite know how treatments will react… especially in cancer.
The biggest issue is that the mice we use for research typically have no or a highly depleted immune system. One of the biggest breakthroughs in the cancer field (IMO) is the development of humanized mice. These are mice that have had their immune system genes replaced with the human versions of these genes.
This is incredibly important for work like this where you’re studying cancer cell-immune cell interactions.
> I like to get GPT-4 output for certain things like this to understand certain aspects of it (and not assuming everything it tells me is fact). I've been downvoted before for this but here goes...
You likely get downvoted because you post things verbatim that you acknowledge may not be true, without actually bothering to fact-check the output.
This is not an informed comment. New immunotherapies for cancers are real. It's just that they are very expensive. But it's not right to claim they are very far or imaginative.
While this particular treatment may go nowhere, cancer treatment has come a LONG way recently. Lots of theoretical “in mice” treatments I recall reading about 5-10 years ago are now actively being deployed clinically. By some measures we’ve had more than a 25% drop in age+health standardised cancer mortality over the last decade.
It’s easy to feel cynical when so many things seem like they get stuck for ages in trials and never make it to an actual treatment, but modern medicine is truly amazing. I have family who are sadly no longer with us, but who likely would have been cured completely if they had developed the same cancer today.
> our finding showed a significant increase in breast cancer mortality rate in the world during the past 25 years ... Statistical analysis showed a significant increase for breast cancer mortality rate in all super regions, except for High-income super region. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6745227/
and for women under 40:
> Female breast cancer mortality rates have stopped declining in women younger than 40 years, ending a trend that existed from 1987 to 2010. Conversely, mortality rates have continued to decline in women aged 40–79 years. https://pubs.rsna.org/doi/full/10.1148/radiol.2021203476
The conclusion I'd come to from those quotes is all these "amazing breakthroughs" might be in use sure, but not to the scale (and price) a layman would expect in order for it to be a properly viable solution for many sufferers.
My wife died from metastatic breast cancer, in 2011. Even then there were advancements. As much as I miss her and envy those who may get advances now, or a “cure”, I am excited to see progress. Maybe she’ll be one of the “last to pass” from this disease.
I agree that there's been some progress in therapies and robot-assisted surgery, but the reason for the improvement in mortality seems more to do with early detection. Cancer deaths overall are also improving due to fewer smokers, but that should be accounted for by "health standardized" measurements.
I think this paper explains it: https://pubmed.ncbi.nlm.nih.gov/27175568/ "the major factors that accounted for these favorable trends were progress in tobacco control and improvements in early detection and treatment"
I feel like the treatment progress for most metastatic solid cancers have not significantly improved. A significant improvement wouldn't be just improving overall survival by a year or so. What seems possible to achieve (someday soon perhaps) is making cancer either a manageable condition, like HIV now is, or curable.
ENPP1 is a known target that has been under active investigation for many years. This press release would like for you to think this Stanford lab discovered it, but it’s been known about for a while. We are just now starting to understand more about how it works in cancer (this work is a part of that). But treatments are already being worked on using this pathway.
These famous researchers are actually at UCSF, not University of San Francisco. UCSF is really a leader in breast cancer the world over, far more influential than Stanford's much smaller amount of faculty on the topic. So it's a bit weird to see such a basic mistake.