Cancer Genomics Symposium

Posted 21 January 2014 by M. Wilson Sayres

I attended the Cancer Genomics Symposium, hosted at UC Berkeley ( http://qb3.berkeley.edu/ccb/cancer-genomics-symposium/). A lot of great research was discussed, but it really struck me that every talk highlighted the importance of understanding evolution if we are going to tackle cancer. My general take-home messages from the Symposium are: - Cancers evolve, so treatments need to keep up. - Next-generation sequencing is a sensitive diagnostic tool and allows for earlier detection. - Therapies can be fine-tuned as we understand patient-specific and tumor-specific signals. You can read the Storify here:

25 Comments

Robert Byers · 21 January 2014

Understanding biological evolution would be a dead end for cancer research.
If there is selection on this its just that. Its not crossing boundaries to create something new. Its just working within traits already within.
Creationism has no problem with that.
Its really the breakdown our bodies that allows cancer to take ahold. tHis is why it happen's in old age mostly.
There must be a way to stop cancer before old age. Thats when it should not be happening.
I think there must be some way to trigger the immune system to fight the breakdown of resistence to cancer in early ages.
lord help us.

Ian Derthal · 21 January 2014

Its really the breakdown our bodies that allows cancer to take ahold.

Presumably this happened after the fall ?????

tHis is why it happen’s in old age mostly

Nonsense

lord help us.

Ian Derthal · 21 January 2014

lord help us.

Why ? Real scientists doing real research ?????

phhht · 21 January 2014

Robert Byers said: Understanding biological evolution would be a dead end for cancer research.

You know what, Robert Byers, it just never occurred to me that you might know anything at all about cancer research. So tell us: are you an MD? No? A PhD? No? Do you even have a bachelor's in premed? No? How about a GED? No, Robert Byers, you're the guy who sweeps the corridors outside the labs. Right?

https://www.google.com/accounts/o8/id?id=AItOawkJ0_Ke3ZSt4RDHdFbEAoEX-DzWA0cEwuY · 22 January 2014

Can someone explain how do cancers evolve? Specifically, how does one cancer inherit information from another.

SWT · 22 January 2014

https://www.google.com/accounts/o8/id?id=AItOawkJ0_Ke3ZSt4RDHdFbEAoEX-DzWA0cEwuY said: Can someone explain how do cancers evolve? Specifically, how does one cancer inherit information from another.

A good starting point in understanding how evolutionary biology is relevant to understanding cancer and its treatment might be to take a look at a couple of the papers to which Dr. Wilson Sayres linked in the Storify -- for example, this one, helpfully entitled "Evolutionary Theory of Cancer."

Karen S. · 22 January 2014

The Howard Hughes Medical Institute did their latest Holiday Lecture Series for young people on Medicine in the Genomics Era and they discuss cancer. It's free, and you can stream it or get a free dvd. Folks confused by this post would do well do watch it.

DS · 22 January 2014

https://www.google.com/accounts/o8/id?id=AItOawkJ0_Ke3ZSt4RDHdFbEAoEX-DzWA0cEwuY said: Can someone explain how do cancers evolve? Specifically, how does one cancer inherit information from another.

Well in the multi-hit model, cancer develops from a succession of mutations. The first mutation, possibly inherited, predisposes the development of tumors which can then undergo mutation to become vascularized and increase in size. Then, further mutations can cause them to proliferate and spread to other tissues. So, in a sense, the tumors evolve by mutation and selection, just like bacteria or anything else. In order to detect and treat these tumors, it is important to know the order and timing of the mutations. The reconstruction of such events can be accomplished by means of phylogenetic trees, such as that presented in Figure 4B of the paper that SWT linked to. And of course this is just one way in which evolutionary theory can be helpful in the study and treatment of cancer. The article points out many other connections between the two fields. So modern medicine depends critically on a good understanding of evolutionary principles and techniques, which are increasingly being included in physician training programs. I wouldn't want to go to somebody like bobby byers in order to get treatment for cancer. WIllful ignorance doesn't seem to help much in such cases.

M. Wilson Sayres · 22 January 2014

https://www.google.com/accounts/o8/id?id=AItOawkJ0_Ke3ZSt4RDHdFbEAoEX-DzWA0cEwuY said: Can someone explain how do cancers evolve? Specifically, how does one cancer inherit information from another.

I'd echo the advice to read some of the papers. I also wanted to point out that cancer cells don't inherit information from cancer cells in other individuals (just like I didn't inherit information from you), but cancer cells can inherit mutations from the cells they derived from (like I inherited mutations that occurred in the germlines of my parents). A tumor is a population of cancer cells that each inherit a unique set of mutations depending on their progenitors.

Scott F · 22 January 2014

M. Wilson Sayres said:
https://www.google.com/accounts/o8/id?id=AItOawkJ0_Ke3ZSt4RDHdFbEAoEX-DzWA0cEwuY said: Can someone explain how do cancers evolve? Specifically, how does one cancer inherit information from another.
I'd echo the advice to read some of the papers. I also wanted to point out that cancer cells don't inherit information from cancer cells in other individuals (just like I didn't inherit information from you), but cancer cells can inherit mutations from the cells they derived from (like I inherited mutations that occurred in the germlines of my parents). A tumor is a population of cancer cells that each inherit a unique set of mutations depending on their progenitors.

What about the deadly cancer in the Tasmanian Devil population? My Wiki-level understanding is that that cancer is contagious, independent of the specific "host". Does that conflict with your statement? Or maybe I didn't understand the point you were trying to make? If nothing else, the "Devil" cancer seems to be an exception to the rule.

M. Wilson Sayres · 22 January 2014

The Devil cancer is fascinating, but is not an exception. In this example, some tumor cells from one individual are spread through to the next. I would think about it like a tumor metastasizing in the human body (instead of moving from one part of the body to colonizing a new location, in the Devils it is colonizing a different individual). Cancers can be studied this way, by taking a graft of a tumor from a human and placing it into an immune-compromised mouse (so it's immune system won't reject the tissue graft).

Most cancers do not spread between individuals like this. But also, humans don't regularly go around biting each other on the face. But, that brings up another interesting point. Something similar may be happening with HPV, which causes many forms of cervical cancer, and is now thought to cause a proportion of throat cancers (due to oral-genital contact between individuals).

harold · 22 January 2014

Robert Byers is on topic here and uses civil, non-threatening language to express his perspective. His perspective is completely wrong, however. With the caveat that there is little threat that Robert Byers could successfully mislead anyone into perceiving him as an expert on this topic, he deserves a reply.

Robert Byers said: Understanding biological evolution would be a dead end for cancer research. If there is selection on this its just that. Its not crossing boundaries to create something new. Its just working within traits already within. Creationism has no problem with that.

Actually, an evolution denier can be a technically competent provider of cancer diagnosis or therapy (Egnor). However, it takes severe compartmentalization. First of all, cancer is usually the result of mutation and natural selection, at the cellular level (I say usually because almost all cancer is associated with somatic mutations but some types of cancer cells may be transformed by epigenetic factors in the absence of somatic mutations). Second of all, cancer is a result of evolutionary history. Cancer occurs because multicellular organisms have evolved ways to regulate development and function, so that a (usually) single fertilized cell can give rise to massive numbers of phenotypically diverse (but genotypically approximately identical) cells, which work in concert as a single organism. This network of mutually dependent cells/organs/systems is surprisingly robust, but somatic mutations, or controversially, epigenetic effects, can disrupt it fatally.

harold · 22 January 2014

M. Wilson Sayres said: The Devil cancer is fascinating, but is not an exception. In this example, some tumor cells from one individual are spread through to the next. I would think about it like a tumor metastasizing in the human body (instead of moving from one part of the body to colonizing a new location, in the Devils it is colonizing a different individual). Cancers can be studied this way, by taking a graft of a tumor from a human and placing it into an immune-compromised mouse (so it's immune system won't reject the tissue graft). Most cancers do not spread between individuals like this. But also, humans don't regularly go around biting each other on the face. But, that brings up another interesting point. Something similar may be happening with HPV, which causes many forms of cervical cancer, and is now thought to cause a proportion of throat cancers (due to oral-genital contact between individuals).

I'm not sure whether I agree with this statement or not. Epidemiologically, the situation in Tasmanian Devils is very rare. It is occurring largely because of the lack of genetic diversity in the population. In a sense, it is actually similar to human metastasis. Human livers don't reject metastatic colon cancer cells, because the cells are from the same individual, recognized as self. Tasmanian Devils, in a sense, are all "self", due to the tragic loss of genetic diversity in this endangered species. Cancer cells themselves are directly spreading from individual to individual in an infectious way, but because the individuals receiving the cells aren't immunologically able to strongly recognize them as non-self. This is not analogous to grafting cancer into mouse models, because the mice are genetically engineered to lack key components of the immune system; that is the sole reason why they don't reject the grafted cancer cells. As you say. If the mice had functioning immune systems they would reject the grafted cells. Humans do the equivalent of biting each other in the face all the time. There are numerous blood-borne human diseases. Also, plenty of types of cancer have an indirect infectious epidemiology. Many cancers are associated with viruses (the molecular mechanisms are similar to those that underly non-virus-associated cancer). In most such cases, viral infection is common but the infection usually does not lead to cancer. For example Epstein-Barr virus is associated with quite a number of cancers, some of which occur only or more often in immunosuppressed individuals, but almost everyone is exposed to it and even immunosuppressed people usually don't get the cancers. This is sort of an example, although technically these are benign (non-invasive, non-metastatic) tumors http://en.wikipedia.org/wiki/Shope_papilloma_virus. However, in these cases, tumor cells could not be transmitted from individual to individual, except possible identical twins (and I have NEVER heard of even that happening). Rather, the virus can be transmitted (because viruses have mechanisms to partly evade the immune system). It can then lead to similar tumors in different individuals; however, each individual's tumor is based on their own unique genome.

Robert Byers · 22 January 2014

harold said: Robert Byers is on topic here and uses civil, non-threatening language to express his perspective. His perspective is completely wrong, however. With the caveat that there is little threat that Robert Byers could successfully mislead anyone into perceiving him as an expert on this topic, he deserves a reply.
Robert Byers said: Understanding biological evolution would be a dead end for cancer research. If there is selection on this its just that. Its not crossing boundaries to create something new. Its just working within traits already within. Creationism has no problem with that.
Actually, an evolution denier can be a technically competent provider of cancer diagnosis or therapy (Egnor). However, it takes severe compartmentalization. First of all, cancer is usually the result of mutation and natural selection, at the cellular level (I say usually because almost all cancer is associated with somatic mutations but some types of cancer cells may be transformed by epigenetic factors in the absence of somatic mutations). Second of all, cancer is a result of evolutionary history. Cancer occurs because multicellular organisms have evolved ways to regulate development and function, so that a (usually) single fertilized cell can give rise to massive numbers of phenotypically diverse (but genotypically approximately identical) cells, which work in concert as a single organism. This network of mutually dependent cells/organs/systems is surprisingly robust, but somatic mutations, or controversially, epigenetic effects, can disrupt it fatally.

I am indeed not an expert and completely ignorant of cancer research. my only observation is that its unrelated to Toe. mutations here are bad as they always are. Evolutionism teaches mutations can be positive and selection on them bring survival to a population. Cancer mutations are always bad. Anyways they don't cross or create important thresholds in new physical systems or parts. also that cancer hits the elderly does teach its because their immune system is not fighting off the problem. i don't think elderly uniquely get attacked by cancer mutated cells but rather they fail to fight them off. A decay is at fault. Even for young people its a issue of inheritance i understand. A special teat in the defence. Anyways I have nothing to add or help.

Keelyn · 23 January 2014

Robert Byers said: I am indeed not an expert and completely ignorant of cancer research.

You are indeed thoroughly ignorant about cancer research, Byers. What else is new?

A decay is at fault.

What the hell is that even supposed to mean?

Even for young people its a issue of inheritance i understand. A special teat in the defence.

At the least, that is partial wrong. My younger brother died at age 14 from acute myeloid leukemia (a rare disease that usually strikes much younger children – mostly it affects older adults) and although there is an inheritance risk, in this case it was not inherited; it was just shit bad luck. So, you “understand” wrong, or at least not fully correctly.

DS · 23 January 2014

oH brother, the old "there are no beneficial mutations" routine again.

let me axe you a question bobby boy, what if a mutation occurs in a mouse that causes the coat to become darker, is the mutation deleterious or beneficial?

what? you refuse to answer. ok let me tell you. In some environments the mutation would be deleterious and in some it would be beneficial. sO you see bobby boy, it is illogical and irrational to claim that there are no beneficial mutations, because it depends on the environment. aNd just so you know, there are indeed suppressor mutations that could be beneficial in the case of cancer. so next time you might want to think twice before spouting off about things you admit that you know nothing about.

harold · 23 January 2014

Robert Byers raises a few good points here. His views are incorrect but these are interesting issues.

Evolutionism teaches mutations can be positive and selection on them bring survival to a population. Cancer mutations are always bad.

Well, yes, cancer related mutations are always bad for the multicellular organism. Cancer related mutations cause cells to become deregulated. Cancer clones then proliferate in a parasitic way. They are selected for relative to normal cells. In a sense, every cancer clone is a population that ultimately goes extinct*, either by killing the host or, preferably, by provoking medical intervention that kills the cancer. (*The weird transient exception being that Tasmanian Devil cancer cell population, of course, which has a genetically-near-identical population to work its way through, at least for now. Also, technically, cancer clones preserved as cell cultures have not gone extinct. However, almost all human cancer clones that have ever existed have.) This supports the theory of evolution and argues against intuitive Lamarckism* and also raises problems for ID/creationism. (*As usual, no denigration of the critical achievements of Lamarck, an important and innovative early scientist, is intended by this usage.) Cell accumulate somatic mutations in a way that causes phenotypic variance, and some phenotypes are selected for. Evolution. Meanwhile, it is unclear why a mechanism sensing the "needs" of an organism, or a benevolent interfering designer, would deliberately generate somatic mutations leading to cancer. And in a sense, the cancer mutations are "good for" the cancer cells. Whether something is "good" or "bad" is a human value judgment.

Anyways they don’t cross or create important thresholds in new physical systems or parts.

This is sort of debatable, but let's say I agree with it. This would be an excellent argument against a theory that claims that all genetic mutations always immediately lead to such outcomes. However, the theory of evolution does not involve such a claim.

also that cancer hits the elderly does teach its because their immune system is not fighting off the problem. i don’t think elderly uniquely get attacked by cancer mutated cells but rather they fail to fight them off. A decay is at fault. Even for young people its a issue of inheritance i understand. A special teat in the defence. Anyways I have nothing to add or help.

None of this is terribly wrong, not even the use of the word "decay". It's actually not quite this simple. There are types cancer that predominantly hit children and young people. They are relatively rare. There are types of cancer associated with immunosuppression. Different types of immunosuppression may actually be associated with slightly different cancer risks. Virus driven cancers are especially associated with immunosuppression. There are many genetic syndromes, each one individually somewhat rare, which increase risk for various types of cancer. The cancers associated with these syndromes are diverse. In the parts of the world with good statistics, the most common type of cancer, by far, is a group known collectively as "carcinoma" - cancers in which the cells show development similar to that of some type of epithelial cells. In genetically and immunologically normal-seeming people, these tend to hit the elderly and the middle aged, with an almost linear age-risk relationship in many cases. Even this isn't terribly simple. Different developed nations have quite different rates of cancer types. In Japan, carcinoma in relatively older people is also the most common situation, but they have much lower rates of some carcinomas, relative to the Anglosphere and Europe, and much higher rates of some others (overall Japan has lower cancer rates but some types are more common in Japan). The immune system does change with age, but that is not only factor in the increasing incidence of cancer with age. There are also other factors which may help explain this.

eric · 23 January 2014

DS said: let me axe you a question bobby boy, what if a mutation occurs in a mouse that causes the coat to become darker, is the mutation deleterious or beneficial?

A more genetically specific example: Bobby, can nature produce the point mutation A->T or not? Just so I can't be accused of trying to trap you, I will tell you that according to science, (1) nature can do this, (2) when it happens on a specific spot on the HBB gene, it improves immunity to malaria, but (3) if you inherit two of these modified HBB genes from your parents, you get sickle cell anemia. So, its beneficial when you get this mutation once. Its not beneficial when you get it twice. I want to know how you think "nature" can only produce the nonbeneficial version since to do so it has to give the benefit first.

M. Wilson Sayres · 23 January 2014

harold said: Rather, the virus can be transmitted (because viruses have mechanisms to partly evade the immune system). It can then lead to similar tumors in different individuals; however, each individual's tumor is based on their own unique genome.

Thank you for correcting this. Yes, I was conflating cancer-causing viruses being transmitted between people with the actual cancer cells being transmitted between individuals (as happens in the Tasmanian Devils).

Ian Derthal · 23 January 2014

i don’t think elderly uniquely get attacked by cancer mutated cells but rather they fail to fight them off.

I always thought the elderly had a much better chance with cancer since the body's metabolism slows down with age, meaning the cancer progresses more slowly ????? I've no idea what Robert Byers is on about.

harold · 23 January 2014

Ian Derthal said:
i don’t think elderly uniquely get attacked by cancer mutated cells but rather they fail to fight them off.
I always thought the elderly had a much better chance with cancer since the body's metabolism slows down with age, meaning the cancer progresses more slowly ????? I've no idea what Robert Byers is on about.

Somehow you have been misinformed. Cancer overall rises with age. There is absolutely no tendency whatsoever for elderly people to do better, overall. There are many different types of cancer, of course, but overall, there is no such tendency. This part of Byers' comments was not particularly incorrect. Do you have a reference from where you got this information? I'm curious to take a look at it, if you do.

Robert Byers · 23 January 2014

harold said: Robert Byers raises a few good points here. His views are incorrect but these are interesting issues.
Evolutionism teaches mutations can be positive and selection on them bring survival to a population. Cancer mutations are always bad.
Well, yes, cancer related mutations are always bad for the multicellular organism. Cancer related mutations cause cells to become deregulated. Cancer clones then proliferate in a parasitic way. They are selected for relative to normal cells. In a sense, every cancer clone is a population that ultimately goes extinct*, either by killing the host or, preferably, by provoking medical intervention that kills the cancer. (*The weird transient exception being that Tasmanian Devil cancer cell population, of course, which has a genetically-near-identical population to work its way through, at least for now. Also, technically, cancer clones preserved as cell cultures have not gone extinct. However, almost all human cancer clones that have ever existed have.) This supports the theory of evolution and argues against intuitive Lamarckism* and also raises problems for ID/creationism. (*As usual, no denigration of the critical achievements of Lamarck, an important and innovative early scientist, is intended by this usage.) Cell accumulate somatic mutations in a way that causes phenotypic variance, and some phenotypes are selected for. Evolution. Meanwhile, it is unclear why a mechanism sensing the "needs" of an organism, or a benevolent interfering designer, would deliberately generate somatic mutations leading to cancer. And in a sense, the cancer mutations are "good for" the cancer cells. Whether something is "good" or "bad" is a human value judgment.
Anyways they don’t cross or create important thresholds in new physical systems or parts.
This is sort of debatable, but let's say I agree with it. This would be an excellent argument against a theory that claims that all genetic mutations always immediately lead to such outcomes. However, the theory of evolution does not involve such a claim.
also that cancer hits the elderly does teach its because their immune system is not fighting off the problem. i don’t think elderly uniquely get attacked by cancer mutated cells but rather they fail to fight them off. A decay is at fault. Even for young people its a issue of inheritance i understand. A special teat in the defence. Anyways I have nothing to add or help.
None of this is terribly wrong, not even the use of the word "decay". It's actually not quite this simple. There are types cancer that predominantly hit children and young people. They are relatively rare. There are types of cancer associated with immunosuppression. Different types of immunosuppression may actually be associated with slightly different cancer risks. Virus driven cancers are especially associated with immunosuppression. There are many genetic syndromes, each one individually somewhat rare, which increase risk for various types of cancer. The cancers associated with these syndromes are diverse. In the parts of the world with good statistics, the most common type of cancer, by far, is a group known collectively as "carcinoma" - cancers in which the cells show development similar to that of some type of epithelial cells. In genetically and immunologically normal-seeming people, these tend to hit the elderly and the middle aged, with an almost linear age-risk relationship in many cases. Even this isn't terribly simple. Different developed nations have quite different rates of cancer types. In Japan, carcinoma in relatively older people is also the most common situation, but they have much lower rates of some carcinomas, relative to the Anglosphere and Europe, and much higher rates of some others (overall Japan has lower cancer rates but some types are more common in Japan). The immune system does change with age, but that is not only factor in the increasing incidence of cancer with age. There are also other factors which may help explain this.

I've never read the word Anglosphere before. Its pretty good and I might steal it. Well you seem to know your stuff. My point was to question evolutions claim of being true because of minor selectionism going on here. I don't think you disagree its the elderly who get cancer and this because of decay in their bodies. They can't fight it off, just like other problems, and NOT that they pick it up more. Otherwise cancer seems to move through weakness in the fabric of the younger people. Anyways perhaps the aberrations can be hints to beating cancer especially before old age. My father died of cancer in his mere 70.s and so its a ugly thing to me.

Rolf · 24 January 2014

A propos cancer

harold · 24 January 2014

Rolf said: A propos cancer

Many thanks for that. Here is an excellent Wikipedia article. http://en.wikipedia.org/wiki/Canine_transmissible_venereal_tumor By a coincidence human tumors that resemble the normal cells known as "histiocytes" fall within my diagnostic subspecialty. A number of cell types have been called that name, but it applies properly only to the monocyte/macrophage family of cells. Whereas the Tasmanian Devil tumor is transmissible because of lack of host genetic diversity (by implication, you could do organ transplants between TD's without much need for immunosuppression, because they're all so closely related genetically), this appears to be a bizarre and rare example of tumor cells that have evolved some infectious-like mechanisms to prevent diverse canine hosts from recognizing them as "other". The value of studying these cells is potentially incredible. In addition to understanding strange cancer cells, their mechanism of evading the self-recognition mechanisms of the canine immune system (which are highly, highly similar to the mechanisms used by the human immune system) could provide insights for transplant medicine. Fascinatingly, histiocytes superficially resemble free-living amoebae at the phenotypic level. (However, they are not at all amoebae, human histiocytes have a human genome, dog histiocytes have a dog genome, etc, and there are numerous things they do that amoebae can't do, and vice versa.) They may be a very ancient type of differentiated cell. Cells resembling histiocytes are found in the most "primitive" unicellular organisms. A few years ago a group of physicists proposed that cancer might be a "reversion" to a unicellular condition. They used naively simplistic language and were appropriately ridiculed for some indulgence in Dunning-Krugerism. Nevertheless, their basic thought was interesting, albeit not original (as they mistakenly thought it was). Cancer cells do resemble infectious or parasitic cells in many ways; they do lose response to regulatory signals and proliferate and metastasize independently. The main difference is that cancer cells are nearly always confined to the single host and "go extinct" when the host dies. Whether these dog "cancer" cells are still "cancer" cells, or represent a species of infectious agent which evolved from a cancer clone is, fascinatingly, semantic. They could be seen as either one. (I leave it to the reader to decide whether this would be most compatible with the theory of evolution, or with sudden creation of all extant species.)

Rikki_Tikki_Taalik · 29 January 2014

Robert Byers said: I am indeed not an expert and completely ignorant of cancer research. my only observation is that its unrelated to Toe.

How is it that you can be "not an expert and completely ignorant of" a subject and then in the very next sentence assert that it is unrelated to evolution. This is completely contradictory Robert and some would call it arrogance of ignorance.

mutations here are bad as they always are. Evolutionism teaches mutations can be positive and selection on them bring survival to a population.

Not "bring survival to the population" but a mutation can potentially give an individual in the population a greater survival potential which in turn gives them an advantage in procreation. They will have more offspring. These mutations can be handed down to their offspring, and in turn their offspring, which is how they can become spread and fixed in a population. This is possible with beneficial, neutral, and "bad" mutations. I've seen this explained to you before and I do not believe that you are incapable of understanding this. At this point I think you are being deliberately obtuse Robert when you say there can be no "good" mutations.

Cancer mutations are always bad. Anyways they don't cross or create important thresholds in new physical systems or parts.

Nobody has claimed this Robert, this is your own misconception.

also that cancer hits the elderly does teach its because their immune system is not fighting off the problem. i don't think elderly uniquely get attacked by cancer mutated cells but rather they fail to fight them off. A decay is at fault. Even for young people its a issue of inheritance i understand.

It's not all inheritance through genetics, which apparently you can accept as long as the mutations are "bad," simple exposure to carcinogens can cause cancer.

Anyways I have nothing to add or help.

We'd rather you educated yourself and dumped the apologetic propaganda. Most of us think of you as a genuinely nice and good person Robert, though unfortunately misguided.