Oral cancer drug kills 100% of tumors in vitro and in animal models | Hacker News Reader

Oral cancer drug kills 100% of tumors in vitro and in animal models

twitter.com

234 pointsflinner102 comments

Related recent discussions:

A New Mode of Cancer Treatment - https://news.ycombinator.com/item?id=36982813 - Aug 2023 (205 comments)

New targeted chemotherapy able to kill all solid tumors in preclinical research - https://news.ycombinator.com/item?id=36978199 - Aug 2023 (32 comments)

'Cancer-killing pill' is now being tested on humans - https://news.ycombinator.com/item?id=36969500 - Aug 2023 (33 comments)

Cancer pill AOH1996 shows promise in annihilating all solid tumours - https://news.ycombinator.com/item?id=36960895 - Aug 2023 (19 comments)

Cancer pill AOH1996 shows promise in annihilating all solid tumours - https://news.ycombinator.com/item?id=36960292 - Aug 2023 (16 comments)

Edit: @dang: Thanks, incorporated your addition.

https://en.m.wikipedia.org/wiki/AOH1996

Look at that molecular structure. Incredible.

Two amide bonds, a completely unremarkable aryl ether and an unsubstituted naphthalene ring system. No fussy weird metals or bonds that need pampering, no chiral carbon atom in sight.

You could probably pay someone a very modest amount of money to end up with kilograms of this stuff, and it would probably not degrade at all under ambient conditions.

I wouldn't be surprised if we soon see desperate, sick, rich people try this out on themselves, skipping FDA approval.

Beautiful, uplifting news. Let's hope the clinical trials proceed without issue.

> Look at that molecular structure. Incredible. Two amide bonds, a completely unremarkable aryl ether and an unsubstituted naphthalene ring system.

Something about this delivery reminds of the business-card scene from Psycho [0].

> Look at that subtle off-white coloring, the tasteful thickness of it--Oh my God, it even has a watermark..."

[0] https://www.youtube.com/watch?v=cISYzA36-ZY

Now let's see Paul Allen's molecule.

*American Psycho. Dude in Psycho didn’t use a business card.

Only it's the inverse: no subtlety, no surprises, all plain.

I'm not sure I'd describe that scene as actually containing amazing surprises of printing, typography, or subjective stylishness.

I see it as mockery of some maladjusted status-obsessed men as they fawn and suffer over trivial differences in pretty-much-identical plain white business cards. Every card says the same thing except for the name, and many of the things being described aren't even true. (For example, not embossed, no watermark, etc.)

> Two amide bonds, a completely unremarkable aryl ether and an unsubstituted naphthalene ring system.

I much prefer your description of the molecule than the one found on its Wikipedia page.

Thanks! Even though I've been in software instead of chemistry for the last 8-or-so years, a structural diagram of a molecule still sings to me of bonds, torsion and symmetry :)

Can you explain to a layman why works to kill tumors and not healthy cells?

The idea is that cancer cells have this mutant DNA clamp[0], which rallies DNA repair mechanisms in cells. The molecule targets the mutant clamp, disabling repair mechanisms for tumor cells but not healthy cells.

We all should hope that it has high specificity for the mutant PCNA, and doesn't affect healthy cells... clinical trials will investigate this.

[0] https://en.m.wikipedia.org/wiki/Proliferating_cell_nuclear_a...

I don’t think this is targeting a mutated version of PCNA… that would assume that all cancers had the same mutation.

Instead, this would target both healthy and cancerous cells. However, the idea is that the cancer cells are more susceptible to the drug. Healthy cells will be affected, but cancer cells will be affected more, and are more unstable, so losing PCNA will have more of an effect on the cancer cells.

The main idea is to do as much damage to the cancer cells while doing as little to normal cells as possible.

Edit: after reading the paper, the drug is targeting a cancer specific isoform of PCNA. This isn’t a mutation, but a slightly different form of the existing protein. Cancers tend to rely on this specific version, while healthy cells can use a different isoform (or both?). A selective inhibitor for this isoform would be cancer cell specific. There would still be some effect on healthy cells, but not as much as with a traditional cytotoxic chemotherapy.

It is written "no discernible side effects"... So we will see...

I read it selectively initiates apoptosis based on that protein and only affects cancer cells but that might have been relayed incorrectly

Edit: exactly -> "It was created to target a specific variant of PCNA found only in cancer cells. This protein is crucial in the body for DNA repair, but targeting it was difficult because of its role in healthy cells. By going after only the variant, it may be possible to selectively target only cancer cells"

From Wikipedia

Hold up, we've identified a mechanism that all cancer cells posses and healthy cells don't? That's almost unbelievable.

I wonder how something like that comes about? Is there something special about that isoform that is necessary for tumor growth or what?

>"Moreover, the identification of a distinct isoform of PCNA associated with cancer cells has potentially opened a novel avenue for the development of new chemotherapeutics"

It is targeting a mutated version.

https://www.cell.com/cell-chemical-biology/fulltext/S2451-94...

A distinct isoform does not necessarily mean a mutation.

The coding sequence for an isoform is part of the normal gene DNA sequence, but the cells can “choose” which version to make. In this case the caPCNA isoform is the form predominantly expressed in cancer cells.

(There is a mechanism where novel splicing isoforms can be produced by a mutation, but this doesn’t seem to be one of those.)

The linked thread that follows explains it in layman's terms.

The drug developers noticed that (a class of) cancer cells involve a shape change in the molecule that clamps on to DNA strands for the purposes of replication and repair. The molecule targets that shape change and thus prevents (a class of) cancer cells from being able to maintain and replicate their DNA.

Cancel cells die and can't replicate = no cancer.

> Look at that molecular structure. Incredible.

I first thought it was a joke because of the two "HN" bonds hah, pardon me.

Derek Lowe had a similar take:

"AOH1996 is a very unremarkable-looking molecule - to be honest, it looks like the sort of stuff that you used to see in old combinatorial chemistry libraries in the late 90s and early 2000s, a couple of aryl-rich groups strung together with amide bonds."

https://www.science.org/content/blog-post/new-mode-cancer-tr...

A key sentence:

“I hope that human cancers will prove vulnerable to this new mode of attack in the clinic, and that they are not able to mutate around it with new forms of caPCNA too quickly, either.”

Even if this does prove effective in humans, eventually cancer cell defenses will evolve to get around it.

Goal is to die for some other reason before cancer evades this

>Even if this does prove effective in humans, eventually cancer cell defenses will evolve to get around it.

Cancer becomes tolerant to drugs separately for each patient. I.e. initially it isn't resistant, but the drug used on a patient puts evolutionary pressure on the cancer to work around the drug. What this means is that people have more months or years to live, which is great. (I am not a doctor).

I don't think that evolve is the right word here. It is not that cancers generally will develop to the point where this is no longer useful in humans. In a single human it is possible that some collection of cancer cells mutates in a way that makes them resistant or immune from this effect (e.g.: by not having the mutated PCNA), and so that strain will become the dominant cancer in that one person. It is also possible that this pathway is pretty common, and this drug does not help out much because of that. But because cancer is not spread (outside of some rare cases), this is not really going to "evolve" like viruses do.

Good clarification, thank you.

It's possible that this drug doesn't wipe out all cancer cells if effective, thus leading to the possibility of the cells eventually mutating. But it's not like cancer is some bacteria that can become universally resistant.

The ABVD regime starts with the same baseline on all hodgkin's lymphoma patients, and 'cures' cancer permanently in quite a large number of them.

I used to bullseye DNA clamps with AOH1996 back home.

> I wouldn't be surprised if we soon see desperate, sick, rich people try this out on themselves, skipping FDA approval

I know nothing about the FDA but for things like terminal cancer treatment, shouldn't there be a blanket exemption for "if the patient so wish, they can take whatever drugs they want even if unapproved"? When facing certain death, what's the worst that can happen? You die a bit sooner?

The illegal part wouldn't be putting the chems into your body, but selling or giving such chems to someone when you know they're planning to ingest them

It seems like a great way to allow a system that it's known to exploit desperate people to do so selling all kinds of miraculous treatments that don't have to even have a relic of science on them.

Having cancer already sucks enough, having businesses and whatnot vulturing over your condition to sell you shit that doesn't work wouldn't really make it better?

There is compassionate use: https://www.cancer.org/cancer/managing-cancer/making-treatme....

The worst that can happen is a industry springing up around giving terminally ill scared desperate people suger pills at a 10000% markup

There sort of is, in a couple of ways. First, off-label use. This is very common in medicine, and means drug X (or machine X) approved for Y purpose can be used for Z purpose if a physician judges it a good idea. So if a heart drug is showing promise in treating sleep apnea in early papers, your doctor could read the papers and decide it was worth a go. Or if you asked them to; trick there is they are professionally liable for it being a supportable decision. There is a form of this for compassionate use which means most of the downsides are ignored in favor of quality of life (e.g. I'm not too worried about getting you addicted to opiods if you're only going to live a few weeks).

The second way is that drugs/devices are considered against their intended use. So the argument for using more dangerous things is much easier if you are treating brain tumors than common colds. There even are fast-track (e.g. mRNA vaccine for covid) but won't be pulled out for niche uses.

What this doesn't cover is things produced in uncontrolled ways, or "i read this on the internet and want a prescription". I can see how this is frustrating for people who are willing to try anything, but I can also see the liability side.

> "When facing certain death, what's the worst that can happen? You die a bit sooner?"

Part of the problem is you certainly can make things worse, and you can make them systemically more expensive. It's one thing to say "I'm dying anyway and I want to try this" but quite another to add "and if it goes badly I expect you to try and deal with it". No easy answers, really.

The concern isn't about killing people already dying of cancer, but rather allowing literal snake oil salesmen to target people dying of cancer and ready to extract all possible wealth from dying people.

This sounds very exciting, and I liked reading your intuitive, almost tangible appreciation of the particular molecular structure. You speak of the components like they are old familiar friends! As someone who doesn't know much at all about chemistry or biology, I had what I thought was a very fascinating discussion about your comment with GPT4 which you might find cool:

https://chat.openai.com/share/28d9b203-1cb5-405b-a109-0f5042...

Basically I wanted to see if your comment, along with the detailed description of the molecular structure, could lead the AI to guess how the compound could be useful. I then proceeded to drop more and more hints and asked it to use those hints along with the structure to guess at a potential mechanism of action. In any case, it certainly helped explain to me what you were talking about!

This was my reaction, as well. So visually and structurally unremarkable, it's actually elegant in how 'un-exotic' it is. Love to see it!

Okay, forgive me for this stupid question but my biology level isn’t that great.

But, if we know which cells to target and what their structure is, can’t we just extract some healthy cells, reproduce them, kill all cells which would dock to that pill, insert reproduced healthy cells?

So with that we don’t have to care if it attacks healthy cells, or is that reproduction idea not possible?

Two speculative substances, this and LK99, are both simple and cheap the synthesize. If either pan out it’s a miracle for man kind. But, it leaves me to wonder, what shenanigans will corporations go through to make it inaccessible and expensive?

Absolute clickbait fraudulent title.

From the article in Cell:

"Given orally, AOH1996 suppresses tumor growth but causes no discernable side effect" is a VERY different result.

Clickbait? Maybe. But fraudulent? Hyperbole much?

Hyperbole expands a truth; that title states a falsehood (see the paper) as true.

IN ANIMALS (mice and dogs)

And?

Well, it doesn't mean it will 1:1 map to humans. We have observed promising treatments for Alzheimer's-like conditions in mice multiple times, but alas, we have very little for humans still.

Suppressing tumor growth is how you kill cancer, is it not? Cancer cells themselves, like the original cells they mutate from, don't live forever. Depending on which type of cell we're talking about, they can die within hours (e.g. skin cells)/days (e.g. blood cells)/months (e.g. liver cells). The issue with cancer cells is they are reproducing like crazy way faster than they die, which is why the tumors grow bigger and bigger. Chemo and surgery attempt to kill or remove the tumors immediately, yes, but my understanding is that's because we have no way to stop the growth and the only hope is to eradicate them entirely before they grow too big to treat, which is why the cancer could come back if any parts are missed.

You stop the cancer cell reproduction and overall tumor growth, you also kill the cancer.

suppressing growth does not equate to 100% regression.

the growth rate could be suppressed to match the attrition rate and turn into a steady state, rather than eradication of the mutant cell lines (and their progeny).

If it turns into a controlled steady state, does it still fit the medical definition of cancer?

good question, but cancer is generally replicating aneuploidy, so it's curious for sure

> The affected cells show cell-cycle arrest, replication stress, apoptosis [!], and so on. And application of AOH1996 along with other known chemotherapy agents made the cells much more sensitive to those, presumably because they couldn’t deal with those on top of the problems that AOH1996 was already causing.

https://www.science.org/content/blog-post/new-mode-cancer-tr...

The story is beautiful, too: https://nypost.com/2023/08/02/new-cancer-hope-as-pill-kills-...

Named after a girl that died. Resulted in fundraising for Neuroblastoma, and this medicine.

That's really touching and admirable that they chose to name it after her.

I guess there's nothing stopping people from just shamelessly making up data on Twitter even when they're linking the actual article. This is very cool and very promising research but "almost completely inhibits the growth of xenograft tumors without causing toxicity to experimental animals" is not the same thing as "kills 100% of tumors." I do understand that there's an incentive to lie like this on the internet - it's not just magical internet points when people get paid real money as creators for getting clicks, but come on, really? Is it worth poisoning the discussion around some actually incredibly cool and promising research while driving false hope (which can actually hurt people!) Just for a few extra pennies from twitter's creator fund?

I am a total layperson in medicine but I try to challenge myself to read Derek Lowes blog posts and comments here from people like isoprophlex who do understand it. I really appreciate the simplicity of this solution - it doesn't need to be customized for each person, it doesn't need years of work on targeting, it's stable, potentially with few side affects of users. I know even this is a long shot but if it works, the power to change society! I imagine that it would even encourage cancer screening earlier for people to make for more effective treatments.

A working treatment for many cancers will bump many more preventable forms of deaths up the charts - obesity, cars, weapons. I hope that people will be inspired to take more significant actions to reduce those when we've tackled these really really hard ones.

And due to the potential low-cost, easy administration, and storage requirements, I am reminded a bit of how antibiotics ended up working in developing countries. Doctors prescribe them for every pain, making them ineffective at best and spawning antibiotic-resistant bugs at worst. I hope the researchers are able to consider this angle (overprescription) before these medicines become a widely distributed thing in other parts of the world.

It's an exciting decade for medicine. I hope that if we don't see the efficacy we want with this one, that more "simple" formulations will come around that do work.

Infectious diseases are different because they jump from host/reservoir to host/reservoir and can take their "experience" with them. Cancers spawn off from normal cells that mutate randomly. So the general scheme is that cancers (and each person's cancer's drug-resistance) develops from scratch in isolation within each person. Basically there's no real mechanism for one person's cancer to translate/share its experience against drugs with another person's cancer. Some cancers are related to viruses by increasing mutation rates which in turn increases the chance of a cancer developing. So basically I wouldn't expect overprescription to produce drug-resistance the way it can for anti-virals and anti-bacterials.

Please don't rely on a x-twitter headline for a de-facto statement on this. Please see other threads posted here on HN.

Or read Derek Lowe's response on AOH1996:

https://www.science.org/content/blog-post/new-mode-cancer-tr...

Or read about the drug that came before it, but didn't metabolize well:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6279569/

Or read about other new, interesting anti-cancer drugs:

https://med.stanford.edu/cancer/about/news/rewiring-cancer-c...

https://www.nature.com/articles/s41586-023-06348-2

Derek Lowe's take is, as usual, reasonable and factual. And even he is really optimistic and happy about this result!

We need some good news these days.

It seems to be getting more popular here to link to a twitter blurb, which links to the actual article. In this case, the blurb adds zero to the discussion, if anything adding blatant misinformation.

This is the actual article:

https://www.cell.com/cell-chemical-biology/pdfExtended/S2451...

I recommend replacing the link and the title with what appears on the Cell website.

Question from a layman:

>100% of solid tumors

Solid tumors == all cancer, or would something like a lymph node cancer not be included?

I'm a layman too but I think this distinguishes them from blood cancer, which does not form solid clumps of cancerous mass...

I’m pretty sure you are correct. At least from the way I’ve seen it used (I took cancer biology lessons but it was more on a cellular level)

leukemia

liquid cancers = blood and bone marrow cancers. Lymphomas are classified as solid.

If it passes all the tests, how long will it take to get it to the market/hopsitals? I know there are multiple phases but if it works, is there fast track or something?

I'm not versed in reading biomedical studies, but I'm curious about this. A successful in vitro test would not necessarily mean that the medicine could overcome other biological challenges such as the blood-brain barrier for delivering medication to brain tumors. Is my assumption correct?

> in vitro test

Means nothing for human treatment. It means something to the people working on the drug.

https://xkcd.com/1217/

https://www.cell.com/cell-chemical-biology/pdfExtended/S2451...

Also discussed at https://news.ycombinator.com/item?id=36982813

The patent appears to belong to City of Hope[0], a non-profit cancer hospital.

Conspiracy idea: such a simple drug is not what Big Pharma wants, thus was never investigated.

Hopefully, if effective, it's affordable for people worldwide.

[0]https://patents.google.com/patent/WO2017049206A1/en

That kind of conspiracy has never held water. Pharma makes bank off of cheap and simple drugs.

Temper expectations, in vitro tests are very different than in vivo tests.

Arsenic also kills 100% of tumors in vitro.

In vitro doesn't answers the question of "can this drug be taken safely" nor "can this drug be delivered to the cancer cells"

Obligatory xkcd https://xkcd.com/1217/

I admit I’ve never been convinced that a handgun would be especially effective. I would certainly not expect a bullet to make a dish full of airborne pathogenic bacteria or viruses safe.

Well the comic is about cancer cells....

I agree that we should still temper expectations (always, forever), but you seem to have missed the other half of the headline: it also seems to work in animal models with no discernible side-effects. Can't say that for arsenic.

Sure, and most miracle drugs don't pan out. Yet there is some reason to at least hope this is a little better than arsenic - it didn't kill the mice nor dogs, and there are various plausible sounding claims about toxicity made (to a non-expert, I just googled the various methods and showed the paper to cancer-researching friend that didn't immediately cry foul).

I mean, the track record of such drugs still suggests it'll likely not live up to the hype of the title here, but that's the point of further research, which does appear to be happening. It's quite likely there's some unforeseen limitation, but hey, we can always hope.

What’s the process for this to be tested and moved onwards toward approved use?

Phase 1: basic safety, some amount of dose-finding. Small number of patients.

Phase 2: does it appear to work as intended, tested in a larger number of people

Phase 3: is it better than alternatives.

Then approval, followed by monitoring of the drug and side effects.

There's also an additional phase of getting insurance to pay for the new drug, which can be the most difficult.

https://www.cancer.org/cancer/managing-cancer/making-treatme...

Nice summary! From my limited knowledge in case someone’s more interested. Don’t know if it’s different in oncology.

Phase 1 is pretty big on safety: sentinel dosing, then gradually increasing doses until the established max is reached or until serious side effects show up. Generally also includes PK/PD aspects: what is the half-life, how long it’s detectable in the body, does this match the modeling? How is the drug eliminated? All this is used to refine the dosing (if need be) but also provides a lot of the information needed for the safety documentation.

Phase 1 trial scheduled to end in March 2024

https://classic.clinicaltrials.gov/ct2/show/NCT05227326

Tbh Since it's a small molecule you can just hire a lab to make a custom synth , I've seen it done a few times,

So…what kind of labs can you hire to make this and how much would it cost?

It depends on the difficulty of making it and the amount required.

https://pastebin.com/0pyuDBX1

Formulation is pretty important too. Do they give any detail on that?

Given the importance of this medication and it's promising initial data I have no idea why it's not given "warp speed" clearance. We cleared mRNA technology almost overnight in response to the pandemic. I feel like society as a whole has collective amnesia on how this happened. A relatively novel strategy to rapidly produce a vaccine in response to a pandemic was trialed essentially on a global scale. The phase 1/2/3 trials may as well have not existed.

It seems strange to me that we can't give such an important novel drug the same treatment. The population size is smaller than global and the effect would be huge. I would seriously doubt anyone stricken with cancer would hesitate to agree to the same "experimental treatment" paperwork we all got when we were vaccinated.

Agreed; it should at least match the energy devoted to superconductivity.

Well if that's what you're looking for, Martin Shkrelli is spitefully publishing a bunch of related molecules to deny prior art claims. https://twitter.com/wagieeacc/status/1687137621699436552

The reason is that the PR department of most bio labs say they have cured cancer every other week.

We have MOUNTAINS of things that stop cancer in mice and sometimes other animals, that don't or won't pan out.

The superconducting results are much closer to if this paper was a small scale human study. Then you would likely see the same kind of "open all the floodgates" ramp up of effort.

If you are unfamiliar with biotech or FDA, Phase 1 announcement is like OpenAI announcing their mission in 2016. Ambitious, but very low probability of success and will take 5+ years, if not 10.

[dupe]

You really think this twitter rando is the first source on this topic?

Bunch more discussion over here: https://news.ycombinator.com/item?id=36982813

One of a number of posts over last few days

This comment is a dupe of https://news.ycombinator.com/item?id=37004911

Amazing if true, and can't be released soon enough. In the US this will likely be a million dollars or something insane, because its the US.

Medical tourism

Or, people will set up home labs. Seems like it's not a complex molecule (but I have no real background).

The molecular compound is about as tricky to synthesize as viagra (which is to say, very simple as far as drug syntheses go), maybe even easier.

All it might take is some instagram cancer influencers, a lab in india or china and some darkweb transactions to DIY your cancer away... if this is so incredibly effective and safe in humans, too.

"All it might take is some instagram cancer influencers, a lab in india or china and some darkweb transactions to DIY your cancer away... "

Or DIY yourself away accidentally. But I guess that would make it a tie, since the cancer dies as well...

Worth it if it works equally as well in humans. We pay way more for cancer drugs that do way less.

If it costs pennies a pill and you are trying to sell it for a million dollars, that is a surefire way to get murdered by a mob. I would just sell the patent to the government for a hefty profit and be done with it.

$100/vial insulin and $700 epi-pens would like a word.

Thalidomide used to cost $6 per capsule in the '70s when it was given for morning sickness (yes, I know, birth defects) but as soon as they found out it cured cancer the price went up to $18,000 and that's on GoodRx lol. [1] Harvoni for Hep C is $94,500 for a 12-week course.

Drugs aren't priced based on the cost to manufacture but based on the projected savings as compared to existing treatments for the medical system - or based on the value the provide to customers. Curing cancer provides value -> price moons.

Unfortunately healthcare is not and cannot by definition be a free market as a free market requires the voluntary exchange of money for goods and services. "Pay me or die of cancer" is not a voluntary exchange but a coerced one - hence the failure of pricing and availability of necessary care you see in the US.

[1] https://www.goodrx.com/classes/thalidomides

I get your point, but these medicines are all bought by insurance companies for some negotiated price they don't disclose. According to GoodRX a Medicare patient will pay between $42-$198 for thalidomide.

Which is still absurd for something that recouped all the R&D decades ago and costs a penny to manufacture.

Reply on news.ycombinator.com