Hacker News Reader

Comment by dekhn | Hacker News Reader

Your own immune system already kills cancer, too.

http://en.wikipedia.org/wiki/Cancer_immunology

Your body is constantly surveilling for cancer cells and eliminating them; in doing so, it selects for cells that can evade the immune system. By the time you have large tumors, the cells are already quite good and evading the immune system, at which point your adaptive immunity system can't help you. These scientists are trying to do first-order engineering on the immune cells to make them more effective.

We've seen this before in science, with antibiotics. A few major, early wins followed by the long silence of failure, due to second-order adaptation by cells.

View on HN

"We've seen this before in science, with antibiotics. A few major, early wins followed by the long silence of failure, due to second-order adaptation by cells."

But the key difference is that cancer does not pass from person to person, unlike bacteria. All cancer cells die before or with the patient, they don't stick around to "remember" how to defeat the hack.

If we do get some major, early wins, they are likely to be very reproducible, indefinitely.

Tumor cells form populations; within the person, there are many generations of expansion and adaptation. Within that time, they definitely "remember" how to defeat the hack.

Yes, but other people are unaffected by that one person. That doesn't mean 100 percent chance of cure, but the chance remains the same for every person.

you're just pointing out a minor difference in my analogy. in cancer, the "population" is the tumor cells in a single human's body, while with bacteria, the "population" includes invading cells that infect multiple people and carry resistance between them. In both cases, the dividing cells propagate their inherited traits that allow them to evade the immune system or treatment.

When you treat a person with a cancer drug, the cancer tumor population evolves (selection and natural variation), when you treat a population with an antibiotic, the bacterial population evolves. That was the point of my analogy.

The analogy is valid, but the difference is not minor.

With infectious disease, if there are ten possible mutations to make the pathogen resistant to treatment, each with a 1% chance of happening within the course of the disease in one patient, then pretty soon all mutations would be present in most of the whole pathogen population on Earth, because of natural selection.

With non-infectious diseases such as cancer, in the same scenario, there would be about a 10% chance of the cancer cells developing one or more of the mutations in each patient, period. A tragedy for those 10% of patients, but a cure for the other 90%.

you're missing the point. cancer is internally infectious. Primary tumors don't even really kill people that much- patients almost always die from metastases. These metastases typically are more resistant.

Please read the entire book "Molecular Biology of Cancer". I'm just explaining small parts of it.

I guess the question then becomes: can the cancer cell population in a single patient evolve fast enough to evade the newly improved immune system, before it gets killed off completely?

It seems unlikely. The genetic diversity was already created. Even when people report "we cannot detect any cancer cells in this person's body, so they're cured", they aren't saying there are no repositories in the body that harbor some resistant cells, whcih will then go walk up the exponential growth curves every aggressive tumor always does.

I do not intend this as a contradiction but merely a tangential tidbit: Devil facial tumor disease [0] is an example of an actual, transmissible cancer in mammals, showing that it is at least possible.

[0] http://en.wikipedia.org/wiki/Devil_facial_tumour_disease

Or oncogene transfection, which likely occurs in nature, not just the lab.

We also know that cancer can be spread / caused by virus, from one person to another (eg HPV, and there are probably others waiting to be discovered).

And we know that you can intentionally force the formation of tumors / cancer through introduction. We do this in mice as a regular course of study now.

I suspect that over the coming decades, we're going to find cancer is a far more communicable disease than previously thought.

> spread / caused by

That seems like a giant conflation to me.

HPV is also an oncovirus, responsible for cervical cancer and very likely many oral, throat, and esophageal cancers. And there's a vaccine for it.

[Edit: should be noted however that the vaccine is for the virus, not from the cancer that can arise from its infection.]

Almost every thing we do is passed on to and alters the next generation we give birth to, even if only in a very subtle way.

My question would be, why wouldn't cancer continue to evolve with our immune systems? We're not abolishing the process that leads to cancer, or making cancer impossible. This therapy kills it after it comes into existence. What would prevent a modified immune system from being passed on, and what would prevent an evolution on the kind of cancer that said immune system is susceptible to? That is, why wouldn't the next generation of cancer become that much more challenging to fight, requiring constantly evolving the therapy?

Not asking to be pro or con, but genuinely curious about this.

Perhaps because most people reproduce before their body evolves a superior cancer?

> What would prevent a modified immune system from being passed on

The fact that modifying your immune system doesn't automatically modify the part of your DNA that encodes how to create an immune system. Children don't inherit your experiences, they just inherit your DNA. (For the same reason, cutting your leg off doesn't cause your offspring to have a missing leg.)

This is incorrect in several ways.

1) Acquired antibodies are passed from mother to child.

2) Epigenetics

This is incorrect in several ways.

1) Ignores obvious intention of the author

2) "several"

How exactly is cancer adaptive anyway? Think about how natural selection works: genes that are better at surviving and propagating themselves survive and propagate themselves. Bacteria evolve because they're an independent genome that infests other organisms but cancer is just a failure mode of biology.

Cancer is adaptative in the same way. Tumors are populations of cells with variation (forget the part where your teachers said tumors are composed of clones, that's garbage). When you apply a selection, a small number of the cancer cells survive. Usually there are several different subpopulations. Then they grow, you end up with tumors that are mixed populations of different subtypes. Then the cells start growing again.

Saying cancer is just a failure mode of biology isn't given cancer enough credit. It's more like cancer is a side effect of complex organisms, which is nearly impossible to eliminate, without also eliminating a number of highly advantageous features of cells.

But they're not transmitted from host to host so the only way it would be passed from patient to patient is if a patient himself had genes for resistant cancer and that somehow made him more likely to spread his genome.

>We've seen this before in science, with antibiotics. A few major, early wins followed by the long silence of failure, due to second-order adaptation by cells.

I would hardly call antibiotics a "long silence of failure" - they are still pretty goddamn effective, even if they aren't quite as effective as they once were.

There are also very few new ones, and nearly all the ones we have remaining are becoming less effective.

I'm not denying we got some easy wins with penicillin back in the day, or that people can't be treated with antibiotics.

But I've been in the medical industry long enough to maintain a very high level of skepticism when it comes to cancer treatment.

Are you saying that antibiotics were a failure?

While anti-retroviral therapy for HIV/AID does not cure the disease, it has transformed a diagnosis that would otherwise have been a death sentence into a manageable illness.

If a "first order" treatment for cancer only extends lives by a few years or even decades, how is that not progress?

First order treatments typically get you about 6 months, not years or decades.

Reply on news.ycombinator.com