My favorite uncle died from pancreatic cancer when I was a teenager. I remember at the time I quickly looked up the survivability rates of pancreatic cancer and was immediately sad to see it was nearly zero [1].
For the next few months, there was a roller coaster of hearing about his new treatments, sometimes seeming effective where we would convince ourselves that he's going to get better, only for him to take a turn for the worse. This kept happening until, of course, he died. I was sad he was dead but the combination of the cancer and the chemo seemed really miserable so I was at least happy he wasn't in pain anymore.
This happened around the time Patrick Swayze died from pancreatic cancer, and a bit before Steve Jobs, and it was at least a little reassuring that this disease didn't spare even obscenely wealthy people. I hope we figure out better treatments for it, and this seems promising.
[1] At least when caught late, which it often is and was for him.
Very early results, but it's exciting stuff. I am less confident in this specific trial working out than I am in some therapies in this field making significant progress in the next few years. (As usual, I am reminded how terribly slow research is for people that need treatments now.)
splittydev
This article very much reads like AI, but it's great that we're making progress towards improving survival rates for pancreatic cancer. Hopefully there's gonna be a lot more of that stuff coming.
I was trying to figure out what the actual intervention was, and what clinical trial this is from based on the extremely limited information, and have not been able to yet.
In the mean time, here's 25 clinical trials for oncolytic viral therapies for pancreatic cancer:
Adenoviruses are very easy to make with custom contents, and are used for many many purposes. However they rarely make it all the way to become a therapy in practice.
It's always fascinating what hits the front page of HN when it comes to bio!
> The latest report adds follow-up, not a final verdict
Is there a "label" for this type of AI slop writing? I see it A LOT in most of my sessions with the format:
[Some positive assertion of something that is], not [something that is not/doesn't hold].
That's a sure sign of something being AI written, and I battle to make AIs avoid it because it doesn't add anything of value.
cosmotic
Is it normal to not report how many people received the treatment and just report the number of successes? Is this better than placebo or doing nothing?
show comments
phendrenad2
I remember hearing about this way back in 2014. Science moves slowly.
My favorite uncle died from pancreatic cancer when I was a teenager. I remember at the time I quickly looked up the survivability rates of pancreatic cancer and was immediately sad to see it was nearly zero [1].
For the next few months, there was a roller coaster of hearing about his new treatments, sometimes seeming effective where we would convince ourselves that he's going to get better, only for him to take a turn for the worse. This kept happening until, of course, he died. I was sad he was dead but the combination of the cancer and the chemo seemed really miserable so I was at least happy he wasn't in pain anymore.
This happened around the time Patrick Swayze died from pancreatic cancer, and a bit before Steve Jobs, and it was at least a little reassuring that this disease didn't spare even obscenely wealthy people. I hope we figure out better treatments for it, and this seems promising.
[1] At least when caught late, which it often is and was for him.
Relevant recent paper from the lead researcher: https://doi.org/10.1158/1535-7163.MCT-25-1231
Very early results, but it's exciting stuff. I am less confident in this specific trial working out than I am in some therapies in this field making significant progress in the next few years. (As usual, I am reminded how terribly slow research is for people that need treatments now.)
This article very much reads like AI, but it's great that we're making progress towards improving survival rates for pancreatic cancer. Hopefully there's gonna be a lot more of that stuff coming.
An excellent source, Kurzgesagt covered the use of virology to cure cancer recently: https://youtu.be/0NY2gAftzJE?si=yeYfT4yc8_ffGmDD
I was trying to figure out what the actual intervention was, and what clinical trial this is from based on the extremely limited information, and have not been able to yet.
In the mean time, here's 25 clinical trials for oncolytic viral therapies for pancreatic cancer:
https://clinicaltrials.gov/search?cond=Pancreatic%20Cancer&i...
And here's a review article on the general concept from the lead physician-researcher in this article:
https://pmc.ncbi.nlm.nih.gov/articles/PMC6186423/
Adenoviruses are very easy to make with custom contents, and are used for many many purposes. However they rarely make it all the way to become a therapy in practice.
It's always fascinating what hits the front page of HN when it comes to bio!
https://www.youtube.com/watch?v=B3xY6Ffy_wE
A site that actually functions:
https://www.newscientist.com/article/2528235-pancreatic-canc...
> The latest report adds follow-up, not a final verdict
Is there a "label" for this type of AI slop writing? I see it A LOT in most of my sessions with the format:
[Some positive assertion of something that is], not [something that is not/doesn't hold].
That's a sure sign of something being AI written, and I battle to make AIs avoid it because it doesn't add anything of value.
Is it normal to not report how many people received the treatment and just report the number of successes? Is this better than placebo or doing nothing?
I remember hearing about this way back in 2014. Science moves slowly.
Still no cure for... wait, what?