Hacker News Reader

Comment by alextheparrot | Hacker News Reader

I'm not certain where the premise that our medicine must "Do no harm" comes from. Chemotherapy, for example, is known for being extremely toxic and harmful, yet the justification is that it often yields better outcomes than no treatment. Why can we not use the same justification for editing DNA? Honestly some therapies cause mutations by design, as some cancers are less capable of DNA repair.

I do see one argument, hinging on the difference between an embryo and an adult aged many years. If your argument is that better techniques may exist at the embryonic stage, then I submit to the medical community to ascertain the appropriate course. However, I don't think offsite mutations pose a significant enough risk that one can say that, on aggregate, gene editing won't "do some good".

View on HN

Maybe because chemo currently gives you 10% chance, while gene editing at the whole embryo level has 100% chance of resulting in mosaicism.

This more recent work by Mitalipov's group dealt with single cell embryo and still ended up with mosaicism, which I was a bit surprised at first. But when you think about it, this single-cell embryo's genome is actively being replicated, so not 100% of the alleles will be edited.

A way around it, would be to reactivate Cas9 after every cell division event, or maybe applying Cas9 gene-drive to embryo development. Also, I have yet to see a CRISPR paper with ensemble sequencing showing no off-targets. Even all the enhanced specificity Cas9 variants had a small percentage of off-targets.

And is mosaicism inherently bad or harmful? Some people have more than one genome and don't even know about it...

Many of the worst disorders are lack of function mutations where mosaicism would indeed fix it.

I would be more worried about unwanted edits than missing ones.

Mosaicism is not inherently bad. It is definitely case specific. The current technology just doesn't allow much control over how much mosaicism you will get, on top of the off-targets, which is your point basically.

The same mosaicism phenomenon which cause tabby cat's coat (especially callico) can be seen in human and would cause that individual to sweat by plaques, instead of the entire body. So you could imagine a scenario where someone would like to select a specific eye, or hair, or skin pigmentation, and end up with inequal pigmentations or a patchwork. That actually sounds cool...

In the case of an autoimmune disorder, like type-I diabetes or a blood disorders like beta-thalassemia, you would absolutely need to hit the proper germline which is responsible for creating your diseased cells. So that's a hit or miss, there is no inbetween. If you do have a hit, then while this germline is dividing and differentiating, it would need to reach a threshold at which the disease becomes asymptomatic. I don't what that number is for any of two diseases I just mentioned, but I would think it is higher than 50% of your cells that need to be corrected, so you would need to successfully hit more than one allele in your succesfully hit germline population that will become blood making cells.

It's a lot of moving parts, which needs to align correctly. That being said, I am optimistic that we will eventually get there. I have some doubt about the scientific pertinence of that "breaking news", i.e. I just don't believe they have accomplished anything that wasn't obvious, but being an optimist, I am sure that not all the ink will be a waste.

Nice question.

> I'm not certain where the premise that our medicine must "Do no harm" comes from.

It comes from the hippocratic oath. "I will apply dietetic measures for the benefit of the sick according to my ability and judgment; I will keep them from harm and injustice."

Where it comes from:

https://en.wikipedia.org/wiki/Hippocratic_Oath

But in that sense it's not literal at all. Again, chemotherapy.

And surgery, and injections, and stitching wounds, etc.

Obviously it's net harm, not no harm at all.

It's not simply 'net' harm, as we would regard surgeons as breaking this principle, if they were to murder a homeless man and distribute his organs to 5 otherwise terminal patients.

Medical ethics is its own field precisely because overeager and overconfident solutions are fraught with problems.

If the principle is understood to be applicable with regard to each individual, that would seem to accord with opposition to organ theft.

net harm per person then :D

But what about if somebody donates a kidney...? Hm

Sure, but read the link.

> It is often said that the phrase "First do no harm" (Latin: Primum non nocere) is a part of the Hippocratic oath. The phrase as such does not appear in the oath...

It’s a myth that doctors even take this oath. There might be some modern version taken, but it is a symbolic pledge, not a legal requirement. Laws and committees cover issues of ethics these days.

>It’s a myth that doctors even take this oath.

Not really:

"In a 2000 survey of US medical schools, all of the then extant medical schools administered some type of profession oath. Among schools of modern medicine, sixty-two of 122 used the Hippocratic Oath, or a modified version of it."

We do not take the original Hippocratic Oath.

Like I said in my previous post, some schools use a modern rendition. The modern versions don't forbid surgery, the use of poison, abortive measures, or compensation for medical training--necessary features of modern healthcare.

I challenge you to find me one med school that uses the original oath.

Medical ethics these days follows a principle-based system. The most common framework employs four basic moral principles: respect for autonomy, beneficence, non-maleficence, and justice. The current system of ethics is far more robust than any of the pledges taken. The pledges are more about personal significance and have no legal or procedural weight.

"Do no harm" speaks to intent, as in harm can't be the intent. Making great discoveries doesn't justify doing harm. Mengele (the Nazi) comes to mind.

Chemotherapy is more of an exception than the norm when it comes to what treatments are acceptable. Most other drugs have to be proven safe to use even before showing tangible benefits. And even in oncology there are Phase I studies to ensure that your treatment is not Too toxic either. So yeah, Do Not Harm is very much a thing.

> Most other drugs have to be proven safe to use even before showing tangible benefits.

What? I encourage you to read about prescription drug side effects. If a drug were safe, it would be over the counter rather than prescription. Prescription drugs are exactly the kind of harm/benefit bundle GP is describing.

As someone with chronic migraines, the question is very much about benefits outweighing costs. Also largely ineffective.

the benefit should always outweigh the risks, that goes without saying.

It's not exactly an exception; it's just on one end of the spectrum. The more serious the illness, the more serious the treatment side effects we tend to tolerate.

My apologies, my previous post was not clear on what I was really questioning. I am attempting to hold up chemotherapy as an example of pragmatism which exposes how the premise of "Do not harm" is not a helpful rule. I believe we may be conflating "proven safe" with "does absolutely does no harm", which are much different standards.

Assessing treatments on thereputic merit - does a treatment create significantly more benefit than it does harm (As chemotherapy is a standard bearing example), is, based on the evidence I've observed, the preferable approach.

Reply on news.ycombinator.com