Hacker News new | past | comments | ask | show | jobs | submit login
Crispr Gene-Editing May Offer Path to Cure for HIV, First Published Report Shows (npr.org)
84 points by pseudolus 4 days ago | hide | past | web | favorite | 48 comments
 help




Many times when I see this kind of articles about cure for HIV and cancer I do some research, get to the page of the project that work on the cure and understand that we are still really far from getting to a point wherer there is a cure and it's just a marketing tricks they do to raise funds for their research

I support this kind of innovation and understand that the marketing buzz is necessary for the projects to raise funds

just keep in mind that those titles sometimes are pretty far from reality


Unlike cancer we do know the mutation which protects against HIV and about 2-4% of the population has it so it’s somewhat much simpler if you can actually use it.

That particular receptor which the mutation affects is also what some HIV drugs especially the post exposure ones target.

Edit: looks like 2-4% of Europeans have the mutation world wide population is closer to 1%

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


Exactly. A Chinese doctor was ostracized for effectively programming a zygote to be immune to HIV. Theoretically this should work for human adults. And not everyone need it too once a community achieve herd immunity.

Yeah. It would be nice if journalism degrees in general would offer classes to help students understand how to give the public a standard and expected way to understand where research really is, because at current they only seem to be able to describe that research on something is happening, and even then not always very well.

Or maybe a greater degree of ethics? But classes like this might not have any affect unless there are teeth in laws that prosecute knowingly publishing falsehoods.


"May Offer Path to cure HIV"

The title seems to be pretty ok in my opinion


Yes, almost everything may offer a path to cure HIV.

That's how research works

> "It's not a home run at this point, but getting to first base is really critical for this technology," says Carl June , a professor in immunotherapy at the University of Pennsylvania.

Three challenges with CRISPR:

- find right change to make

- make that change in target cell without any off-target changes to other genes.

- efficiency - being able to change enough of the cells to have the desired effect.

The last one is probably the hardest if the cell type isn't something you can easily take out of the body, change, grow up and put back.


In this case it would be "remove, treat, and re-implant all of the bone marrow" I think. Better than trying to CRISPR your brain or heart I'd imagine, but much harder than, say, skin.

I read somewhere efficiency is 10%-20%. That may be fine to fix a condition where you need to add a new protein, e.g. muscular dystrophy. However, maybe not so well where you need to subtract. If you can pose an HIV therapy as the former- say make 10% of the T-cell stem cells infection-proof, then that could work. On the other hand, hoping the remove all the retro-inserted HIV DNA probably wont work.

What are the "high score" for efficiency? Did researchers tried and were successful at modifying a whole organ, a whole body?

The human body contains 35-40 trillion cells.

By far the best way to make a whole body change would be to change one cell ( the embyro/egg/sperm ) and let nature take it's course - however that has the 'playing god with human evolution' concern.

Delivery mechanisms are key - you have to get your therapeutic inside each cell - this is hard and even harder to do in a cell type specific way.

It's much easier to do ex-vivo than in-vivo: ( see the diagram on this page ) http://www.crisprtx.com/gene-editing/therapeutic-approach

How many cells you need all depends on the biology of what you are trying to do - there is no universal answer.

So you might need a small number cells if you are changing the messages the changed cells send to other cells through control networks.

In other cases, where the change only changes that cell - you might need to change a large number to see an effect.


> 'playing god with human evolution' concern

Don't mean to derail this entire thread, but even as someone who believes in God, I find it increasingly harder to understand what objectively is wrong about these purported ethical dilemmas (e.g. cloning)


I was using that as a short hand. The real ethical concern is about potential misuse of germline or embryo mods.

ie how do you choose between the ethical ones and the unethical ones?

So if I found a mutation I could make to make people of a compliant disposition I could make an army of slaves.

Note this is someone changing people before they are born ( this is to some extent the playing god bit - you are making a choice for somebody else - not yet born ).

Most people who consider that to be not acceptable.

Or how about a mutation that made people better soliders - one of the problems with warfare is the soliders can be mentally damaged by the things they have done or seen - surely it's better to make a solider that is unaffected....

ie the opportunities for good and evil are enormous - but until you have a mechanism to decide between those, it's easier to blanket ban.

Even something that, on the face of it is benign - like a mutation that reduces the risk of heart attack - what happens if you later find out that it increases the risk of something else? You can't undo it later ( only for subsequent children by the same procedure ) - for the people already impacted.

Again, the cautionary principle applies.


> I find it increasingly harder to understand what objectively is wrong about these purported ethical dilemmas (e.g. cloning)

Believer in God here, and, like you, I find these "playing God" concerns to be largely unfounded, and often based in fear-of-the-unknown more than objectivity.

In the 18th and 19th centuries, some religious groups opposed surgery because it was playing god; "disfiguring the divine image." But today, we religious people almost invariably welcome such medical advances: the good they do far outweighs any theoretical fear-based concerns. I think the same will happen with modern medical advances.

And, theologically in the Jewish and Christian Bible, God has given humans dominion over nature. I see medical advances -- including surgery, blood transfusions, vaccines, and yes, gene editing -- as exercising that dominion to help heal people and reduce suffering.


Gene editing on an embryo involves making a choice for an unborn person without possibility of consent - and their children, and their children etc.

That assumption of knowing better than the person affected is 'playing god' in a way that doing a surgical procedure on a consenting person isn't.

Obviously that stuff is already done today - abortion or medical intervention at an early stage where the parent acts on behalf of the person.

One of the things that's different is the scope and power of these techniques.

Look at the diversity of domestic dogs, from great danes to toy dog breeds - that all came through selective breeding from the same wolf ancestor in a few thousand years.

Also remember that humans are 99% the same as chimps at the DNA level and about 85% with mice.

Direct, untrammeled genetic modification of humans could unleash huge changes in a very short time.

Until we are clear about how we want to manage that, it's easier to have a moratorium.


Since HIV is a virus, does that mean the same tools could be used for other viruses? Is there any hope of wiping out influenza or the common cold?

HIV is specifically a retro-virus, a class of virus that's very hard to get out of a person's system. This specific cure, though, is making it so that the key that HIV uses to get into a certain door[1] to your cell won't work. Since most viruses get into cells in different ways this won't have any impact on influenza.

There's been some work[2] on broad spectrum anti-virals but I don't know where that is right now.

[1]https://en.wikipedia.org/wiki/CCR5

[2]https://en.wikipedia.org/wiki/DRACO


CRISPR is used against a wide range of viruses all the time -- by bacteria.

There isn't a reason why it couldn't be used against other viruses, although it may be more economical to sip chicken soup for a cold and keep your flu shots up to date.


I remember reading something one time where people were discussing if the cold could be eradicated by having everybody on the planet stay home for 2 weeks. I don't remember what the answer was though...

It looks like there are some old Reddit and Quora threads where people discuss this. Some answers there seem to be

* no because of non-human animal hosts for some viruses

* no because everyone would die if everyone actually stayed home for 2 weeks

* no because some people might be long-term carriers of some viral strains (maybe because of compromised immune systems?)

* no because people's homes aren't properly isolated from one another (e.g. within an apartment building)

* no because a virus could slowly spread between unlucky members of a single household and thereby persist within the household for many weeks at a time


This particular study is about inserting a gene variant that confers immunity to HIV into bone marrow. It's a pretty drastic step that was only happening because the patient had leukemia and was going to get a bone marrow transplant anyway.

To do it for other viruses, you'd have to know a specific genetic variant that confers immunity.


I know this question might not apply to the article, but in general, how does Crispr work on adults? Does it change the DNA of all the cells in the body?

Scientists were not sure if it's safe to use Crispr on people, the chinese tried it and now we know that someone has tried it. - that's about all of the info in the article

> the chinese tried it

"The chinese", really? That was one single rogue Chinese scientist.


> The first trial of CRISPR in humans took place in China in 2016, when Lu You, a physician at Sichuan University, put gene-edited cells into a lung cancer patient. Since then, other Chinese trials have moved forward, though not much is known about them.

> Last year Chinese researcher He Jiankui caused a global outcry ...

Ethics aside, it’s definitely more than one Chinese scientist working in this area.

https://www.scientificamerican.com/article/despite-controver...


"Chinese scientists have published the first report in a scientific journal of an attempt to use CRISPR-edited cells in a patient--a 27-year-old man who is HIV-positive." is the third sentence of the article.

Perhaps the title should have been "Two rogue scientists are checking out CRISPR"


Oh wow this reminds me so much about that article I read some time ago about a chinese scientist who had edited genes of babies (he actually altered a gene in IVF embryos): https://www.sciencemag.org/news/2019/08/untold-story-circle-...

So editing genes of adults is ok but babies is not? This article sounds so positive and supportive of the act ... I'm a bit confused.


>So editing genes of adults is ok but babies is not?

Adults can consent.


true, that makes sense ... I'm not a gene or biology specialist, but if you change the gene of a human and he has kids, the kids will inherit the modification, right? so my problem with all these gene manipulations is that the more we do and the more it will impact future generations and who knows what the impact on those will be? So for me having the adult consent is not enough as the impact is not just on him but potentially much greater.

Gene manipulation of an adult likely would target a subset of overall tissue. Although, ethics of modifying reproductive tissue could be an issue.

The difference is what happens if the edit in an adult is unsuccessful. I don’t think people are objecting to edits that are net benefits as much as preforming highly risky procedures.

Though like most ethical issues people don’t have universal objections.


If you modify bone marrow specifically, it won't be passed on.

When do babies consent to being born?

they dont!

it's not about consent. Or, even better, consent is a very weak argument. If you use the consent argument, then you need to apply it for other areas, like being against vaccines for example. But you probably don't use it there because you know that babies can't consent anything (or maybe you do, if you are an antivax).

The very brief discussion about it in genome.gov even mentioned that.


There is also a big difference between editing the germline and somatic genome - the former can be passed on to children and descendents - clearly a much bigger change than one person.

For a genetic change to be passed on it has to appear in the male or female gametes ( sperm/egg ). So if I change your skins cells that change isn't in your sperm/egg so can't be passed on.

When you edit an embyro at an early stage you are effectively changing all cells - including gametes - so the change has impacts beyond the immediate individual.


This was designed to cure a disease he already had. The babies were being modified to confer resistance to a disease they didn't have yet.

It's one thing to fix a disease in utero, or avoid the disease in the first place (as in mitochondrial transfers for genetic mitochondrial diseases). It's another to try and blithely "improve" a baby.


1) the father and mother had sida so the child would have had too. 2) it does not matter, improving a baby maximize baby happiness and does not affect negatively anybody else, such trivial utilitarism shouldn't be dismissable without solid arguments. That scientist is now in prison for making us closer to go beyond our pathetic human condition (e.g ageing)

You can reduce the risk to the baby quite considerably using existing therapies.

https://www.who.int/hiv/topics/mtct/en/


Interesting but reduce is less interesting than eliminating it.

Sure, but CRISPRing a fertilized egg is not exactly eliminating risk, is it? There's a whole set of unknown risks you're introducing.

It's one thing to enable a viable birth for a child that would otherwise be 100% guaranteed to have a life-limiting genetic disease, but modifying a gene- a gene we do not know that much about- on the chance of reducing a 5% risk of contracting a virus that can be effectively treated- is pretty well beyond what most people would consider ethical.

To put it another way: there would be a 95% chance any CRISPR modifications would have not prevented the baby contracting HIV because 95 out of 100 such babies wouldn't have contracted HIV anyway. And those 95 babies (and at least some of their children) would live with that genetic variant forever.

If the baby was going to be nonviable and you had the chance to eliminate the disease by knocking out a faulty gene or replacing it with the version of the gene everyone else has, that's one thing. But this is a rare gene variant. There may be a reason why only a small proportion of people have it! It could well be that the side-effects of lifetime ART are actually better than the unknown side-effects of this gene.


From what I remember of the discussion at the time, CRISPR wasn’t good enough for human gene editing to be ethical. It might still not be, but science does have a way of progressing so I can easily believe CRISPR is one step on a path that eventually gets us to where it isn’t a cancer risk.

Yes I assume it is like you just said ... but what scares me, are those consequences the editing might have in a far future, yes we cured cancer by editing a gene, but now we created another problem XY and we are fucked because we can not edit it back, something like that ... you know like oh we have a problem that plants are infested by lice, so lets import frogs from another country, oh but now we have an overpopulation of frogs and they have created a new problem

We’ll always be able to edit it back, the thing to be worried about is someone using the same tech to engineer a new plague — there are 27 post-WW2 genocides listed on Wikipedia, and no particular reason to think that rate will abate before the technology becomes cheap enough and widely accessible for it to be weaponised.

So the headline talks about HIV but the article states it cured cancer and didn't do anything to HIV.

Crispr didn't cure the cancer, the bone marrow transplant did; bone marrow transplants (without Crispr) are the standard of care for treating this cancer. The fact that the bone marrow transplant was necessary for the cancer gave the researchers an opportunity to also try out this HIV treatment, which requires a bone marrow transplant that would ordinarily be too high-risk to be justifiable on its own. This was a very early-stage trial, probably not expected to be curative; they were just seeing if the method in general was safe, and did anything. It apparently was safe, and successfully conferred the desired mutation to at least some of the patient's immune cells (so it did something), but not enough of them to be curative. It sounds like future trials will focus on improving efficacy.



Applications are open for YC Winter 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: