Benchmarks are tricky because it all depends on workload and use case. If you are in a VR headset for example, it’s all about power envelope and GPU flops. If you are in a Desktop used for productivity it might be all about peak CPU performance.
When comparing fab processes, you wouldn't want performance of a whole processor but rather the voltage vs frequency curves for the different transistor libraries offered by each fab process.
Ooof. I really need to pick up another HD, these model sizes are killer.
Lacking a godly GPU, I will probably hold off for a quanitized version which has the potential to run okish on CPU or my modest GPU, but really appreciate the info.
If the LLM's are emulating you (and others) in every other sentence, perhaps you should be getting royalties or something.
That would be a bit silly in this particular case, but in general we ought to celebrate cases where somebody has authentically done something that millions of others find it useful to copy.
>the cancers were caused by the chemotherapy conditioning regimen for the treatment, not Lyfgenia itself.
I am certain some media group is going to conveniently leave this part out of the title of their article, and surely no one is gonna waste time reading the actual article and the rumors will take off.
The article says "patients must undergo a preparatory treatment with a chemotherapy drug to remove any native stem cells that might remain in their bone marrow." It doesn't make much difference to the patient if it's the Lyfgenia itself or the chemo drug, if the chemo drug is a requirement. Right?
There is active research on doing gene therapy without requiring chemotherapy (or radiation) first. It has been shown to work in mice, and they may eventually get it working in humans too. It would likely require a significant modification to these gene therapies though, since one approach is to alter the method of growing the stem cells to produce a significantly higher number, and then transplant that, with the hope that the transplanted edited cells outnumber and outnumber the original unedited ones. So very likely the FDA would treat that as a new therapy requiring a new approval process.
There is also ongoing research into immunotherapy for killing stem cells, as an alternative to the existing methods of chemotherapy and radiation. Potentially, immunotherapy could have significantly reduced secondary cancer risk.
I was thinking more that it could poison the reputation of gene therapy by causing folks to falsely associate cancer with gene therapy. And that false stigma could carry on even if one day chemo was no longer needed.
It's reasonable to expect small improvements in the risk profile, but I think blood cancer is going to be a side effect for any drug following this basic idea. You will always need some chemo to destroy the defective blood-making stem cells before replacing them with the genetically-modified blood-making stem cells, and any chemo that is strong enough to kill all of the blood-making stem cells in your body is necessarily going to have a risk of damaging healthy cells and turning them into pre-cancer cells. So the risk can be reduced but probably not eliminated.
> Does anyone know if changing the drug would require a new FDA approval for the entire regimen, or could the protocol be easily changed?
The FDA-approved prescribing information will recommend a particular chemotherapy regimen, but clinicians will be free to substitute alternatives if they believe those are clinically superior. They won't need permission from the FDA or the manufacturer to do that; clinicians deviate from the FDA-approved manufacturer recommendations all the time ("off-label prescribing").
If the manufacturer wants to update the official recommendations in the prescribing information, then they'll need FDA approval for that. But it is possible for clinicians to publish their own treatment guidelines (e.g. in medical journal articles), independent of the manufacturer, and the FDA has no control over those.
What a weird system: there's something better, but the manufacturer isn't allowed to tell you about it. What if they, like, slide the journal article across the desk, whilst holding their finger alongside their nose and winking?
> What a weird system: there's something better, but the manufacturer isn't allowed to tell you about it.
It is the way medicine works – not just in the US, in most countries worldwide. Not just about gene therapy, about all drugs and devices.
The FDA and its international equivalents (the EMA in the EU, the TGA in Australia, etc) regulate the manufacturers, not the clinicians. They control what the manufacturers sell and even what the manufacturers are allowed to say about their products (in product packaging, prescribing information, advertisements and marketing collateral). They don't control what the treating clinicians do with those products – to the extent that is regulated, it is the job of other regulatory agencies (e.g. professional licensing boards, civil courts hearing medical malpractice claims, etc)
> What if they, like, slide the journal article across the desk, whilst holding their finger alongside their nose and winking?
What they'll do instead: there will be a conference where (among other things) the journal article author will present their findings/recommendations, and the manufacturer will sponsor (and hence help pay for) the conference. They never actually said anything, they just made sure you were there to hear about it.
I'm not a doctor but my mother is. When I was a teenager, she'd be invited to these free dinners at fancy restaurants paid for by pharmaceutical companies, and a couple of times they allowed her to take me along (she was allowed to bring her spouse/partner to some of them, so she just asked "can I bring my teenage son instead"?). During the dinner, some academic would do a presentation on their research into how wonderful one of the company's drugs was, and also do some Q&A. So the manufacturer wasn't technically saying anything, everything was said by some academic (whose research they were funding). I didn't understand it all, but I found it rather interesting. Still didn't follow her footsteps into medicine though (although my younger brother has).
But, she tells me the regulators have cracked down on free perks from pharmaceutical companies, so they are forced to be a lot less generous nowadays than they were back in the 1990s. (This is not the US though, this is Australia.)
“If anyone had predicted three years ago when the ONISQ program began that Rydberg neutral atoms could function as logical qubits, no one would have believed it,” said Dr. Guido Zuccarello, a DARPA technical adviser who has supported the ONISQ program since its start in 2020. “It’s the DARPA way to bet on the potential of these less-studied qubits along with the more well-studied ions and superconducting circuits. As an exploratory program, ONISQ gave researchers the leeway to explore unique and new applications beyond just the optimization focus. As a result, the Harvard-led team was able to leverage much more of the potential of these Rydberg qubits and turn them into logical qubits, which is a very significant discovery.”
