Comment by epgui | Hacker News Reader

Comment by epgui | Hacker News Reader

Parent Full thread

Biochemist: agreed, "we now live long enough to suffer from it" works well to explain the rising prevalence of cancer, but not that of auto-immune diseases.

For auto-immune diseases, the consensus (if we can call it that) mostly revolves around the hygiene hypothesis. Refrigeration is one aspect of this, but other things that fall in this category are the use of soap, rising cleanliness standards, disinfectants, aseptic methods, positive pressure toilets and good sewage systems, vaccines, antibiotics, not breastfeeding as much (not directly a hygiene issue, but breastfeeding transfers passive immunity to a child with a developing immune system, which has an effect not unlike being exposed to a different spectrum of microbes), caesarian births, always eating very fresh meats and vegetables, etc.

There is a grain of truth to "what doesn't kill you makes you stronger" (but don't take that idea too far, and please do wash your hands and do get vaccinated, for god's sake), and the essence of the hygiene hypothesis, in comp sci / ML terms, is that it's a training set problem.

Edit: it's unclear to me why people are downvoting this. This is my area of expertise, so I'd be curious to know what people find objectionable in this comment.

Bringing this to the article, the authors are outside that consensus. They believe that Crohn's is an immune response directed toward a bacterium, rather than toward the self.

The hypothesis in the paper is that Crohn's isn't autoimmunity and that improving hygiene might prevent or treat it.

Edit: They do mention a training set factor, but at the level of population genetics instead of at the level of an individual developing immune system. Their evidence is that a history of the plague, rather than hygiene, promoted a genetic immune system weakness that made people more likely to survive the plague, but that Crohn's is a consequence of that today. A related example that comes to mind is the prevalence of sickle cell anemia in regions with malaria.

If it's a training set problem, wouldn't it be possible to give a "vaccine" consisting of a reasonable mix of antigens from time to time? Has this been tried?

It's not completely inconceivable that we will one day have therapeutic approaches in that line of reasoning, but it's also not quite that simple.

There is a critical period during infancy and early childhood when the immune system goes through a number of critical maturation processes. Afaik, once that's done there's no going back to a perfectly fresh slate.

After puberty, for example, your thymus (an important site for T cell maturation) begins shrinking and eventually all that remains of the organ is basically non-functional traces of scar tissue. That is not to say that T cell maturation doesn't occur afterwards, but it is a particularly striking example of the irreversible age-related change we can be talking about.

Another less-obvious example is that children born by cesarian have a different skin and gut flora through the rest of their lives than children who are born through the vaginal canal: the initial colonization of the skin with a different set of bacteria has some degree of permanence, though it is also somewhat plastic. Yet another example, children who are breastfed tend to have fewer food allergies or auto-immune diseases (atopy, generally) than children who are fed formula. Early conferral of passive immunity through maternal antibodies secreted in breast milk selects for different populations of gut microbes, and again there is some degree of permanence to this. In both cases, the differences are measurable throughout adulthood!

Everything epgui said above. Just to add another line of argument: thanks to sequencing, we also know there are a lot of germs that we simply can't culture. We used to suspect that, but now we know that. Most general hygiene methods are non-discriminitive, so as cleanliness goes up, opportune exposures to rare edge cases go down, and probably go down fast. We are becoming, in a sense, a species of oligoculture hosts.

There's people who engage in Helminth therapy (purposefully getting parasitic infections like a hookworm) for purposes of inflammatory disorders. I don't know much about it but enough people take it seriously

That's how many vaccines work... dead versions of pathogens.

Reply on news.ycombinator.com