ASU PSOC Workshop, Thursday February 9th – Friday February 10th 2012

Evolution, Development and Cancer: Connecting the Dots

Cancer is widespread among eukaryotes, and can be successfully tackled only by understanding its place in the story of life itself – especially the evolution of multi-cellularity. There is general agreement that the roots of cancer date back hundreds of millions of years. The ancient genes responsible for cancer are retained because they play a key role in embryo development. Normally these genes are subsequently silenced, but if they become re-awakened in the adult form, cancer is the result. The workshop is an ambitious attempt to connect the dots of evolutionary biology, developmental biology and cancer biology, bringing together three distinct communities in the search for hidden links that may form the basis of a radically new theory of cancer.
Listen to Audio Interviews and Read Transcripts

Workshop Photos

Workshop attendees


Paul sets the scene for the meeting


Alvin Kho presenting


Jean Paul Thiery and Stuart Newman


Vanessa Baack and Alvin Kho


Luis Cisneros and Amanda Wilber


Carl Pilcher and Sarah Walker


Break time interaction


Christine Chaffer and Aurora Nedelcu


Athena Aktipis, Carlo Maley and Tomislav Domazet-Loso


Paul and Pauline Davies


Audience engrossed in a session


Athena Aktipis summing up the workshop


Christine Chaffer and baby girl Alice

Audio Interviews and Transcripts from the Workshop

  • Paul Davies gives scene setting opening to the workshop Principal Investigator of the Center for the Convergence of Physical Science and Cancer Biology at ASU[audio:|titles=Paul Davies – Introductory Talk]


  • Interview with John Condeelis a professor in the Department of Anatomy & Structural Biology, Co-Chair of the Department of Anatomy & Structural Biology and Scientific Director of the Analytical Imaging Facility for Albert Einstein College of Medicine.
    Interview Transcript
  • Interview with Tomislav Domazet-Loso an evolutionary geneticist the Ruđer Bošković Institute in Croatia who is interested in micro-evolutionary patterns and dynamics at different levels of biological organization told a fascinating story.
    [audio:|titles=Tomislav Domazet-Loso]
    Interview Transcript
  • Interview with Jean Paul Thiery of the Cancer Stem Cells Programme, Cancer Science Institute of Singapore. Jean Paul is French and has lived in Singapore since 2006.
    [audio:|titles=John Paul Thierry]
    Interview Transcript
  • Interview with Stuart Newman, Professor of Cell Biology and Anatomy, New York Medical College who now realizes that his own work on embryo development can be applied to cancer research.
    [audio:|titles=Stuart Newman – for Web]
    Interview Transcript
  • Interview with John Pepper, a cancer biologist from the National Cancer Institute.
    [audio:|titles=John Pepper for web]
    Interview Transcript
  • Interview with Christine Chaffer, a Post Doc Researcher for Robert Weinberg, at the Whitehead Institute, Boston, Massachusetts
    [audio:|titles=Christine Chaffer interview for web]
    Interview Transcript
  • Interview with Aurora Nedelcu, Professor of Biology at University of New Brunswick, Canada
    [audio:|titles=Aurora Nedelcu.edited]
    Interview Transcript
  • Interview with Athena Aktipis, at the Center for Evolution and Cancer at UCSF and in the Psychology Department at ASU, kindly agreed to give a summing up talk, reminding us of the main points of the workshop.
    [audio:|titles=Athena moderating summation/overview of the workshop]
    Interview Transcript

Interview Transcripts

Interview with John Condeelis

(Back to Audio)

John Condeelis: John Condeelis, Albert Einstein College of Medicine, and I’m involved in directing the Gruss-Lipper Biophotonic Center which is a biophontonics biophysical approach to the study of living matter by interacting matter with photons.
Pauline Davies: Well at the workshop at ASU, you were talking about metastasis and the metastatic process and you had some very original thoughts on that. Can you tell me, in a very brief summary, what did you say?
John Condeelis: Yeah. Briefly I think the insight that comes from the work that we’ve been doing looking at tumor behavior at very high resolution has lead us to the conclusion that the ability of tumor cells to actually break free of and then migrate within and then leave the primary tumor is just a re-expression of their embryonic morphongenic patterns. So in this sense tumor metastasis is really embryonic morphogenesis replayed.
Pauline Davies: Now how do you make that connection?
John Condeelis: By looking at the behavior of tumor cells at single cell resolution in live animals, not dead biology where you’re actually looking at tissue sections of you’re looking at the pieces of tissue that have been removed from the mouse, rather you’re looking at the behavior of the tumor cells while they’re inside the living animal inside the living tumor, neither of which that have been dissected and you’re doing it in real time and it’s single cell resolution, those four components are very rare capabilities and that’s where these very exotic microscope technologies become really important.
Pauline Davies: Right, well there’s a number of issues there that I’d like to speak to you about. Let’s start off with the very clever technology that you’re using. So what are you doing exactly and how are you actually studying this process in live animals?
John Condeelis: Well I think the biological problem, that is the study of the single cell behavior in real time in a live animal in single cell resolution, requires special kinds of imaging microscopes. You can’t use ultra-sound, MR, PET, CT, any of the standard imaging technologies that are used with solid tissue because the resolution is not sufficient, but if you can figure out a way to get photons into solid tissue and then form images with them you can generate the kinds of resolution that are required to see single cells and to see sub-cellular structures within those single cells and so our whole attention then became shifted to accomplish those goals toward using optical microscopy and obviously you can’t use visible light because solid tissue is opaque to visible light so one is forced to go out into the far red and if you get out far enough into the far red you can no longer excite fluorescent molecules because you’re so far out. So how far out is that, a thousand nanometers or 1500 nanometers compared to visible light which is around 450 to 550 nanometers. So you’re at least two to almost three times longer wavelengths; so how do you excite these floras? Well what you do is you collide two photons simultaneously and take advantage of a non-linear quantum mechanical phenomenon called multi-photonic citation. So if we could bring two photons to interact simultaneously with chromo-four that usually excites at 500 nanometers we can put in two 1000 nanometer photons and get the same event. Now the tissue is transparent to a 1000 nanometer photon, so now we can excite the 500 nanometer florescent protein with a photon that is completely passed by the tissue, and this allows us now to do the imaging. So this is why multi-photon microscopes have really become essential tools in doing these kinds of studies in vivo.
Pauline Davies: And I think you are actually doing this in mice, is that right?
John Condeelis: We’re doing this in mice but it’s also possible to do it in human tissue that’s been removed from patients in the operating room. So we can take that tissue and look at it immediately or better yet we can implant it into the mammary gland, if it’s breast tissue, we can implant it into the mammary gland of a immunosuppressed mouse, and that mouse will then host that tissue and that human tissue can be imaged for some weeks.
Pauline Davies: And so you’d be watching what was happening to individual cells in the tumors.
John Condeelis: Yeah we’re watching what’s happening to individual cells and the next surprise was well we thought we would just be looking at tumor cells because after all they’re the cells that are generating the tumor mass and then spreading the disease. It turned out that the real action is the interaction between tumor cells and non-tumor cells, and if you don’t have that interaction the tumor cells are not actually very dangerous at all.
Pauline Davies: Because I was just reading from your presentation macrophages are involved.
John Condeelis: Yeah so it turns out that the macrophages, which are supposed to be suppressing tumor growth and tumor cell migration and all of these things because they are immune cells that attack non-self, so if the tumor cells mutates to the point where it’s recognized as non-self, macrophages will simply eat it. So for many years people thought that these large macrophage infiltrations that were seen in solid tumors in people were the body’s response to these tumors and that the immune system was trying to kill the tumor. It turns out that that’s not true, what’s going on is that the macrophages are invading the tumor because they’re being asked to come there by the tumor cells. And what the tumor cells are then doing is they’re forming pairs with these macrophages, they’re pairing up, it’s a buddy system and these pairs of cells then go on to stimulate each other to form morphogenetic fields, and these morphogenetic fields recapitulate what one sees in a normal developing breast. That was a big surprise. So the stromal cells are actually participating with the tumor cells to reconstruct a morphogenetic pattern.
Pauline Davies: And what is making these macrophages, the protective cells, go rouge and actually collude with the cancer cells?
John Condeelis: Well they’re not going rogue, it turns out that what they are doing is they’re precisely repeating a normal pattern of gene expression that they’re hardwired to express when they’re triggered just in the right way. How do we know that, okay so you can take a step back and say ‘how can you say any of this with certainty.’ Well we can say this with some certainty because we’ve watched morphorgenisis in normal breast tissue in developing mice and there’s also a lot of histology of normal human breast as it undergoes its normal developmental cycles. So if you put all this information together plus the live imaging during mouse development in the mammary gland you begin to realize that what’s going on is the normal epithelial cells, one of their great tasks in morphogenesis is to invade the mammary fat pad, so these epithelial cells are inherently invasive, and they have to be able to invade and migrate long distances. The macrophages that they recruit actually help them do this and this is what we see being repeated in the breast tumor. It’s not a rogue issue at all it’s actually a reawakening of a perfectly normal issue that these cells have to face.
Pauline Davies: Okay so the macrophages were absolutely essential in helping form the breast in the first place and then they go on and fulfill other roles, do they?
John Condeelis: They go on to fulfill other roles in the adult, but what happens is when a breast tumor or actually potentially any solid epithelial-derived tumor begins to grow in normal tissue that tumor generates microenvironments or patterns of cytokine signaling that are not normally seen in normal tissue and in some cases what appears to be the case is that these signaling patterns that get setup were actually last seen in the embryo and at some point you achieve just the right amount of that signaling and now these macrophages which are normally present in other cell types, other stromal cells that are normally present in the tissue, go from being stromal cells that support the normal, adult gland to switching their gene expression pattern to becoming embryonic versions of themselves and then they support the morphogenesis of the epithelial cells which, now in this case, are tumor cells instead of normal epithelial cells. So another way to say that, a shorthand way to say that is the microenvironment of the tumor basically generates signals that reawaken these patterns of morphogenesis.
Pauline Davies: Is there any way that that could be used in some form of treatment, do you think, some way of changing the macrophages?
John Condeelis: Yeah so that’s the inspiration that one can act on having made this realization and connection you can go ‘well you know ordinarily if you are trying to kill or attack in some way or suppress the behavior, let’s say the spreading of an epithelial cell which is now a tumor cell, you’re trying to suppress that tumor cell from spreading, what could you possibly use to do that that would not also effect other cells types in the body?’ Because after all these are epithelial cells, they were normal cells just a few generations ago and it’s really a formidable problem. How do you do that? And in fact nobody can do that, there’s no drug company that’s made a drug that’s specific for spreading tumor cells. In fact most of the drugs that are currently available have very low specificity in that regard, I mean you poison the tissue yes, you poison the tumor cells but you also poison the normal cells and the surroundings and other organ systems as well. So that’s the great problem with cytotoxic drugs. So how can you possibly use this insight we just talked about to address a novel therapeutic approach? Well the inspiration is that ‘hey the cells that are actually spreading think they’re embryonic.’ So is there something about embryonic cells that distinguishes them from their adult counterparts and if those characteristics are absolutely unique to the embryonic state, those become targets that are highly selective for the tumor cells that think they’re embryonic, because if you are an adult, you have no embryonic tissues in your body other than that tumor, unless you’re pregnant. So now you have an opportunity to go after these cells with great selectivity because now you’re just targeting the embryonic expression pattern. So the trick now is to be able to grab the cells that are undergoing the morphogenetic movement and at that moment determine what their gene expression pattern is to see if it’s really distinctive and if it distinguishes them from tumor cells that are not undergoing this morphogenetic movement pattern and distinguishes them from normal cells that are also in the same tissue and that has been done. So machines have been built to collect these tumor cells as they migrate into these streams and just grab them as a log population and measure the phenotypes in vitro and then do gene expression pattern analysis on them and what pops out are a series of four pathways that basically are responsive in their activation to EGF, which is one of the cytokines that drives this process of morphogenesis. EGF stands for epidermal growth factor, so it’s an epithelial growth factor that determines the activation and growth of normal glands, so it’s no surprise it would be involved in tumor growth and spreading. But these pathways work off of the EGF signaling pathway, they work directly off of the receptor and it turns out that one of the proteins that binds directly to the receptor that’s uniquely up regulated in these migratory tumor cells is the protein Mena, standing for mammalian ENA, and so we looked at this protein in detail because it’s regulating the receptor signaling directly and found that the form of Mena that was being expressed in the migratory tumor cells was different from the form of Mena that was being expressed in the non-migratory tumor cells and was again different from the normal cells in the tissue. And so it looked like it was only being turned on in the tumor cells that were migrating and we were really struck by this and when we went back and looked at this in some detail with our collaborator Frank Gurtler who discovered this family of proteins and found that the form of Mena that was being expressed uniquely in the migratory tumor cells was the embryonic form which distinguished these cells now from all the other cell types that are in that same tissue at that moment which express the adult form of Mena. And the embryonic form of Mena has an extra axon in it which can be identified by PCR or with antibodies and could be a drug target. So now you have your magic bullet target if you can make a drug that can basically stop the version of the protein that has that embryonic Mena axon in it, you could stop what’s special about those cells in terms of their signaling from the EGF essentially. So that’s how you get to a magic bullet.
Pauline Davies: So you’re not targeting all tumor cells, you’re targeting the tumor cells which would go on to spread around the body.
John Condeelis: So you’re not killing tumor cells by stopping them from growing, what you’re doing is you’re stopping the tumor cells from spreading.
Pauline Davies: Right because of course it’s the spreading tumor cells that kill people.
John Condeelis: That’s right.
Pauline Davies: Now this seems like a very ancient sort of pathway, the interaction between the macrophages and the tumor cells, you say it’s involved in the formation of the normal human breast and is it also involved in embryogenesis?
John Condeelis: Yes this particular isoform of Mena, embryonic Mena, is seen in embryonic development of the brain, in neurons, and in glandular tissue, colon, breast and so on. So the five solid tumors that plague humans at this time in history are all adenocarcinomas, which means epithelial-derived tumors, and all the adenocarcinomas that are undergoing metastasis, we believe, have to express this form of Mena to do it. It’s going to be a common characteristic of all these adenocarcinomas.
Pauline Davies: And this form of Mena is it turned on by the interaction with the macrophages?
John Condeelis: No it’s not, that’s interesting. It’s turned on by the tumor microenvironment. The tumor microenvironment generates signals that cause certain embryonic, the factors that turn on genes, the so called transcription factors like zeb, twist, snail, these are well known, they were given those names because they were discovered in the organisms. When you play with these transcription factors they generate particular shapes in the mutant flies or the mutant worms and so they’re given those names like twist, right, the embryo is all twisted up. So these are the cytokine-dependent transcription factors that regulate embryonic gene expression. And these transcription factors get turned on by microenvironment signals in the embryo. Well it turns out that the tumor is generating these kinds of cytokine signals that turn on these embryonic transcription factors, those in turn turn on the expression of the embryonic Mena.
Pauline Davies: Okay and how is it that certain tumor cells get turned on and others don’t, differences in their particular microenvironments?
John Condeelis: Yeah it probably depends on where you’re standing. If you happen to be in a region where the tumor is growing rapidly and it’s not generating any embryonic-like cytokine signals then you’ll just sit there and you’ll be a rapidly growing epithelial cell, tumor cell, but if you happen to be standing in a region where the cytokine signals are being turned on that mimic this embryonic stimulatory package, because it’s not just one factor it’s probably a whole mixture of cytokines that turn on this particular pattern of growth factors, then you get the full shot of information and when these transcription factors come on you actually begin to turn on these embryonic genes that then make these embryonic proteins and that’s probably what’s going on. Now you’re going to ask me a question how anticipated. The embryonic Mena is not a different protein from Mena, so it’s not just a matter of you’re expressing protein A versus protein B, no Mena is a protein and you can make it as an adult form or an embryonic form and the way you do that is by doing differential gene splicing. So what’s really going on here is you’re differentially splicing the gene from an adult form to an embryonic form and that’s the tricky part because that means there is no way to attack this phenomenon. This protein is not unusual it’s just a slight rearrangement of the gene product that’s causing this phenomenon. So this is a recent invention after the lowest eukaryotes to be able to splice mRNAs to generate multiple versions of a protein called isoform, and these multiple isoform patterns, they’re never in isolation. If you have one gene undergoing a splicing of its mRNA it’s usually linked to twenty or thirty other genes so these whole patterns they’re cassettes come on together. So the activation of one twist transcription factor generates a whole cassette of splicing patterns and these genes now they’re spliced in the embryonic form work together for a period of time, they generate a shape and then they go away.
Pauline Davies: You’ve just shown some amazing movies where you have a tumor cell sitting around doing not very much and then you have a macrophage on its own sitting around not doing very much but put them together and they wiz together and then they line up.
John Condeelis: That’s right. So the ability of the tumor cell and macrophage to interact and sustain that interaction to lead to a productive, amplified migration pattern is autonomous to these cells. You can actually put these cells in isolation together and they will interact and they will generate these rather very high speed streams or trains of cells moving and this is unexpected because the expectation we had actually imaging the behavior of these cells in tissue was that the microenvironment would have to be present to sustain the cell interaction throughout its functional lifetime. And the reality is it looks like all you have to do is get the two cell types together and have the tumor cell expressing this embryonic pattern and after that the cells are completely autonomous. So they can leave the microenvironment that generated that pattern of interaction and pattern of gene expression and be independent of it now and that’s fascinating, that also indicates the danger of the disease that you can generate a change in these cells and they’ll maintain it even after they leave their local environment.
Pauline Davies: As you say it’s fascinating. And here you’ve got this most amazing slide; this is a bud in the normal breast tissue development. What’s going on?
John Condeelis: The terminal end bud is invading the fat, which is otherwise acellular, I mean it’s got fat cells in it but it doesn’t have any of the parenchyma of the gland in it and what that terminal end bud is doing is it’s crawling, being pulled, all these things are going on simultaneously to pull that long string of epithelial cells out a great distance into the actual fat pad to form a glandular tree.
Pauline Davies: So this would be the normal milk gland.
John Condeelis: This is the normal milk gland. So here for example, just to orient you, the nipple is way down here and these strands of epithelial cells have moved out this far into the mammary fat pad which is otherwise empty and as you can see these distances are huge, this is millimeters, these glands have pulled out and they’re like a tree occupying the fat and each one of these fingers or epithelial strands has a tip on it called the terminal end bud and that’s where the action is, where all this force for migration is being generated by these macrophages.
Pauline Davies: So you can just see very vividly here just how important they are.
John Condeelis: It’s remarkable. What this illustrates is how important it is not to look at that biology. If you actually go and watch the embryo form I think you’ll see all the various forms of cancer recapitulated.
Pauline Davies: So do you think that there should be much more interaction between embryologists and cancer biologists and developmental biologists?
John Condeelis: I think it’s key to have the embryologists, the mammalian embryologists, and the cancer biologists working in parallel. It’s more important than any other possible multi-disciplinary interaction at this point of time. If you’re interested in how cancer stem cells generate tumors, the rules by which tumors grow and the rules by which tumors disseminate and metastasize it’s absolutely crucial to have embryologists and cancer biologists working together, and they have to mammalian embryologists, I’m not talking about somebody studying sea urchin, because the rules that are being followed are species-specific and obviously, developmentally they’re in a very narrow window so you don’t want to look at organisms that are great model organisms for development but aren’t mammals, you’re going to get the wrong answer or even dramatically diverse mammal species, you really want to get something close to human. So a mouse is good, a mouse works. I think that’s an urgent need, I think it’s a greatly overlooked need right now.
Pauline Davies: Well thank you very much for talking to me.
John Condeelis: It’s a pleasure. Thank you.

Interview with Tomislav Domazet-Loso

(Back to Audio)

Pauline Davies: And in fact you were talking about cancer in hydra, is that correct?
Tomislav Domazet-Loso: Yeah that’s right.
Pauline Davies: Well first of all describe hydra for me, I seem to remember it from my early biology days as a school girl.
Tomislav Domazet-Loso: Hydra is a cnidarian very similar to corals by phylogeny. It’s a species which lives in ponds in fresh water. It’s a tiny animal, a couple of millimeters, and it’s quite easy to maintain in the laboratory. It was very popular in regeneration research. Nowadays a lot of sophisticated genetic techniques are available for researching in hydra.
Pauline Davies: I seem to remember it photosynthesizes, is that correct?
Tomislav Domazet-Loso: Yes that’s just for green hydra, but there are a lot of species of hydra; there is hydra that contains symbiotic bacteria, but there are many species actually without, and the hydra I was using was hydra oligactis, more or less a cosmopolitan species, and that one doesn’t have a green color or symbiotic bacteria inside.
Pauline Davies: So how does it get its energy?
Tomislav Domazet-Loso: It has tentacles and the tentacles have specialized cells called the nematocytes and these are highly derived cells which can kill some small baby shrimps or things like that, and then the hydra swallows them and survives.
Pauline Davies: Right then, so it can be an aggressive predator in ponds.
Tomislav Domazet-Loso: It is, it can even catch a small fish, baby fish; some bigger species of hydra, yeah. They are very, let’s say so, aggressive.
Pauline Davies: So it’s a multi-cellular organism, but quite small.
Tomislav Domazet-Loso: Yes it is a multicellular organism, but by organization it resembles early stages of multicellular organization in animals. There are other groups like sponge, which are even more, let’s conditionally say, primitive. But hydra is also very basal in the tree of life of animals. And in that respect is very interesting for evolutionary studies.
Pauline Davies: So what’s the story with it and cancer?
Tomislav Domazet-Loso: Most of the cancer biology textbooks do not answer the question: “Which animals can and which animals cannot get cancer?” So that’s a basic question but there is no answer to that, and so I was wondering if there is some report in literature about the cancer in basal metazoan but it was very hard to find anything. Last year I was lucky to meet Thomas Bosch, he’s a professor in Kiel, Germany, he is one of the very prominent hydra developmental biologists, and he told me that he keeps in his lab a very strange strain which seems to have something what we could call a tumor. So in cooperation with him and his crew we were able to indeed show that his hydra strain has real bonafide tumors. And to my knowledge, this is the first report of tumors in basal metazoan, so we are about to publish the paper soon.
Pauline Davies: Well that’s very intriguing, and how did these tumors manifest? Are they just a lump on the side of the hydra?
Tomislav Domazet-Loso: The hydra is a very simple organism, it looks like a tube, but these cancers are really making distortions of the morphology; they look completely abnormal, bigger than the normal hydra. The other thing that is obvious is some malfunctioning because some of the cells are entering the body cavities and making a mess there, so the animal is very sick and it hardly reproduces so the hydra has two modes of reproduction – one is sexual reproduction, another one is budding. It has problems with budding and sexual reproduction. We keep the culture by cutting these hydras into halves and the hydra has a great rejuvenation potential so these two halves recover quickly and then we have two hydras with tumors and then we have to repeat this every couple of days to actually maintain the culture. That’s one of the biggest problems with this research for hydra that have tumors.
Pauline Davies: Well it could be not just a problem; it could be a positive asset. You can do all sorts of studies, can’t you, if you’ve got a hydra that you cut in half and you can decide whether or not to include the tumor, or see if a tumor develops newly in the half that doesn’t have one.
Tomislav Domazet-Loso: The hydra is great because you can do many classical experiments and developmental biology. For example you can cut a piece of the diseased hydra or hydra with a tumor and transplant this to the normal hydra individual and we did these experiments and often transplantation to the normal hydra [led to that animal] also receiving these cells and [they then] got tumors. So you can basically transplant the tumors, so this is one also very interesting feature.
Pauline Davies: Did you try to analyze the tumors and find out if they’re the same sort of things that are going wrong in the hydra tumor as there are in higher organism genus.
Tomislav Domazet-Loso: Yeah that’s an excellent question, this is exactly what we are trying to do but it’s not so easy. At the cellular level these tumors resemble ovarian tumors mammalian species. The reason is that hydras also have oogenesis and the production of eggs, so it seems this is a problem with the stem cells which are determined to make eggs. So they can’t stop dividing for some reason and that might be the major issue. We also made a series of experiments with the gene expression. We used micro-rays to do that, to find out which genes are over expressed or under expressed and it seems to be a really great mess, around four thousand genes have significant differences in expression compared to the normal animals. And now we are struggling to understand which of these differences or information is actually important for tumor genesis, but it will take some time.
Pauline Davies: And are any of the differences the sort of differences we see in cancer cells in humans?
Tomislav Domazet-Loso: We haven’t compared these directly, we are in the early phase of the comparison, but at the moment only on this cellular level this somehow resembles the situation in the human body. But we hope actually to find some resemblance in some because this might be a good model for studying tumors, because it’s very cheap to keep hydra cultures this costs almost nothing, you know, you can do it on a hobby level, compared to mice which are extremely expensive for the research.
Pauline Davies: And quick as well to get the division.
Tomislav Domazet-Loso: That’s true.
Pauline Davies: Now before you did your work, what was the most primitive creature that people had found cancer in?
Tomislav Domazet-Loso: I don’t know, this is hard to say. There are tumors in the drosophila, so fruit flies, I think also in nematodes but the information, it’s very hard to get that information as to who can get a tumor, who cannot. This is to my knowledge but I know also that some species that have great regeneration potential like amphibians or even hydra, the tumors seems to be very rare in these species for some reason and this is unclear why.
Pauline Davies: Well I guess if they had tumors every time that they regenerated, well my goodness the species would die out.
Tomislav Domazet-Loso: Yeah probably.
Pauline Davies: Well thank you very much
Tomislav Domazet-Loso: Thank you

Interview with Jean Paul Thiery

(Back to Audio)

Jean Paul Thiery: So my talk is based on thirty-five years of studies in embryonic development, trying to understand morphogenic processes and particularly cell migrations and trying to figure out whether the pathways that are being utilized by those normal cells migrating to the embryo are going to be co-opted by the cancer cells when they want to metastasize.
Pauline Davies: As you say you spent thirty-five years studying the subject, so did you learn anything new in this meeting?
Jean Paul Thiery: So I learned that obviously I am happy to see that a lot of people now are accepting this idea that in fact cancer cells are co-opting a number of mechanisms that embryonic cells have developed over a long period of evolution, starting apparently with very old species, like diploblasts, sponges, like jellyfishes that a very simple organisms. Still these organisms utilize already molecules and targets that potentially we would like to utilize in cancer treatment, so it is very intriguing to see that those mechanisms have been selected extremely early in evolution and this is where this meeting was particularly interesting, to have a melting pot of people more directed at research for evolutionary aspects. I’m not an evolutionary person so I think I learned quite a bit, even from the astrobiologists, so this is very rewarding to come here for this particular kind of assembly of people that you normally wouldn’t see speaking together in the same meetings. We don’t meet astrophysicists or astrobiologists.
Pauline Davies: Do you think it will change either the way you do research or think about the problem of cancer in the future?
Jean Paul Thiery: Yes because there were certain notions I did not quite accept or consider in my previous research so now I will be much more carefully looking into whether certain types of mechanisms that I was not considering potentially interesting are now to be put into play so I think this meeting is, as usual interdisciplinary meetings are very useful, and I like these ones much better than the specialized ones. In other words, if you are with your brother and sister going to the cancer meeting you always talk about cancer, cancer, cancer, caner and so it’s not putting you in a challenging situation of considering that in fact you could find the truth elsewhere. And I was very happy to see that, at least some people propose somewhat iconoclastic mechanisms but interesting ones to consider in the near future.
Pauline Davies: Can you give me any example that sticks in your mind? I know that you said you learned a lot, but anything in particular?
Jean Paul Thiery: Well, for example these modules that are not completely understood but developmental biologist and evolutionary people would consider them as important modules, but that have never been targeted before. For example the hydra, I was shocked my hydra. Hydra are making tumors. So hydra are very simple cnidarians and even though it is probably a very rare event, the very fact that it was shown I think is amendable now to I think experimentations and I think putting that into the context of targeted therapeutic screening method. I wish we could, I think that Paul Davies would like to do a radiation of hydra now and see whether or not they can really augment the probability to develop cancers, but this is very intriguing because these are stem cell-like. So this is something that we never had an opportunity to study with mice or obviously with human patients.
Pauline Davies: That certainly is exciting, isn’t it? So maybe in the long run some new ways of approaching, problems maybe new treatment.
Jean Paul Thiery: Yes I think, usually treatments often come from people that are quite outside of the field and have not a very biased view of what you should target, probably naive in some ways, but probably less contaminated by dogma and so being less contaminated by dogma have a much better chance to find something that other people have been working with for twenty, thirty years, are not able to anymore really consider. I think the vision should be broad and I think we should take any opportunity that is offered by developmental biology, by evolution to target something that we would never think of to do.
Pauline Davies: So it could be that the roots of cancer go back a long, long way
Jean Paul Thiery: Oh yes. In the old days when I went to Egypt and went to the pyramids and went to the Temples I was searching the hieroglyphs for the oldest descriptions of among other things bladder cancer. Bladder cancer was very frequent in Egypt three thousand years ago because the Nile was contaminated with schistosomiasis. So at this time the people treated patients with mercury, which is a very toxic agent obviously, but they were already trying to do some chemo in some way, so they understood the disease in some way. So I thought that cancer old and that Egyptians even though they were living in quite different ways, had already quite a lot of cancers. But in fact if you look outside the human species, if you look in the vertebrate, if you look at mammals and the vertebrates you start to see that fishes have cancers, frogs have cancers and now you think that invertebrates would never have cancers, but in fact drosophila, the fruit fly has many cancers. And then obviously I ignored that hydra could have cancers. That’s quite astonishing to me and I like this idea because it puts immediately the mechanism in a completely different perspective. And then archaebacteria, even the unicellular prokaryotes are having already the genes for cancer in some ways. So that’s quite amazing, how was that made in such a way that, obviously they are not going to develop cancer, but they have genes that can potentially do something else, but then they are unfortunately re-shuffled and re-adapted in some ways to perhaps protect the embryos but at the same time to hurt the adult and so it’s very intriguing to understand the theory about that. Unfortunately I am not Albert Einstein so I certainly will not be able to decode all of that; I call that the da Vinci code of cancer. But it would be very intriguing to put together these evolutionary people and cancer biologists together, but it will take time, you would probably need several sessions to match your good ideas, and then at the end what I would like to see is that people would write some kind of theoretical articles together.
Pauline Davies: Yeah, I think that would be the ultimate dream of all of us here at the cancer workshop. I was quite surprised that drosophila could get cancer because they must be going undergoing a relatively few cell divisions to get cancer.
Jean Paul Thiery: No they get different types of cancers. They get the so-called, you know the pupal stage they get imaginal cancer, they form gigantic tumors out of very small imaginating disc. The wing imaginal disc is the primordial of the wing that can suddenly transform with a gene that is also found in humans, by the way, it’s called the FAT gene. So it makes a cancer which is a about that guy from Vietnam who was recently operated on with a 92 kilogram tumor, it’s like a haemotome and so it’s a very intriguing situation that you find that already in drosophila. Now drosophila have much more malignant tumors than that one, which is some kind of central nervous tumors that are glioblastoma like. They are extremely aggressive and they finally kill the drosophila. And they can be transplanted tumors. So clearly we have something that is not terribly different from what we see in humans in terms of glioblastomas and I’m sure there are plenty of other insects, eukaryotes, but I think we should search for more primitive animals like the case of hydra, but I am sure they are probably the starfish or very low metazoans that probably would offer us ways perhaps to study that. You know the cell cycle was discovered in a sea animal?
Pauline Davies: No I didn’t know that.
Jean Paul Thiery: yes so the cell cycle was discovered there and in fact in all these marine molluscs are sources of anti-proliferatives in fact also. So young people still think that we may get one day natural products, or derivatives of natural products would be much more efficient than the current chemo therapy. People are still hoping that we could take advantage of the natural defense of these animals against excessive proliferations to perhaps find better drugs then what we have now for resting cell cycle.

Interview with Stuart Newman

(Back to Audio)

Stuart Newman: Well I’ve learned even though I never worked on cancer, that in some ways I’ve always been working on cancer because there is a real resemblance between the cancer phenotype and the embryonic phenotype and I learned it and I hope my own talk contributed to this discourse.
Pauline Davies: So what did you talk about?
Stuart Newman: I talked about the genes and physical processes that were prevalent in the earliest stages of multicellular evolution and how these cooperated together to give rise to complex forms and then over time these complex forms turned into organisms that had programs of development, but the programs of development came about slowly but the forms were there very early. So when cancer forms I believe it’s a bit of a reversion to this sort of primitive proto-embryonic state where the cells can do things like invade other tissues and move and create abhorrent forms.
Pauline Davies: Has this conference influenced what you’re going to be doing in the future?
Stuart Newman: Well I think I am going to read the cancer literature much more closely and look for parallels between my own work and work on cancer. It seems to me that the things we’ve been working on could helpful in promoting new ways to look at cancer therapy.

Interview with John Pepper

(Back to Audio)

Pauline Davies: John, what did you get from this meeting?
John Pepper: I got a lot of interesting new ideas and fascinating conversations, lots of useful tidbits of information too, but the interaction was the best part.
Pauline Davies: And when you said you got lots of new ideas, what sort of ideas?
John Pepper: I guess reoccurring themes. There was a lot of discussion of sort of what metaphors are used to envision cancer as an atavistic reversion to a single cell stage form of life or as a developmental atavism to cells having characteristics of fetal cells.
Pauline Davies: Whose talk stood out in your mind most of all?
John Pepper: I was very of intrigued by some of the data that John Condeelis shared. He had fascinating imagery of single cells behaving in vitro inside a mouse with cancer and interactions between the cells that are usually only in the realm of speculation, but there they were in living color moving around the screen in front of us doing intricate dances between different cells types. Great fun to watch! I was also intrigued to hear that wholesale vaccines developed from fetal stem cells have been very effective in preventing cancer in mouse models. I think that’s a fascinating early result.
Pauline Davies: Thank you very much John.
John Pepper: Thank you.

Interview with Christine Chaffer

(Back to Audio)

Christine Chaffer: My name is Christine Chaffer. I am a cancer biologist currently completing my post-doctoral work in the lab of Bob Weinberg in Boston at the Whitehead Institute.
Pauline Davies: So you just gave a fascinating presentation about how some types of cells can be identified in the lab and go on to be identified as stem cells, and then some other cells do something very peculiar, they don’t seem to be stem-like but then they become stem-like and go on to cause trouble. Is that a very quick and accurate summary?
Christine Chaffer: Yeah, that’s correct. So we’ve identified that there’s variation amongst cells that we think of as being differentiated or as being in a fixed, solid stage, and what we found is that there are some populations of cells that can re-activate programs and can transitions to a less differentiated state and the consequence of that is that they acquire properties that we do associate with normal stem cells and in the context of cancer we can identify that these cells transition towards a state that allows them to become much more aggressive and metastatic.
Pauline Davies: And all your work, it came from a discovery didn’t it?
Christine Chaffer: It did. It came from a kind of serendipitous observation in the lab and that was just that in culturing these, in working with these human mammary epithelial cells I noticed that at some point that there were cells that were actually floating within the culture media and normally we consider these cells to be dead and certainly my boss was very curious as to why I would even look at these theoretically dead cells, but what I found was that if I collected these cells and replayed them, they were actually alive and I was able to identity that the cells that were floating and that I could re-culture under normal conditions had very different biological properties.
Pauline Davies: So what did you find when you were able to re-culture these cells.
Christine Chaffer: What I found was that there was a sub-population of cells that we would consider to be quite a differentiated cell-type that had the ability to spontaneously switch or transition to a stem-like state. So this was happening without any known activation or expression of transcription factors.
Pauline Davies: So you didn’t add anything to them?
Christine Chaffer: Correct, yes we didn’t add anything to these cells, they able to do this spontaneously. This is something that the bulk population of cells didn’t do. So what happened was that we noticed that even within these floating cells that I’d collected, that there was still two populations of cells; one population of cells that had all the characteristics of being a stem-like cell, and a second population of cells that through our initial tests looked to be very similar to the differentiated, bulk population of cells, however when we looked at this more differentiated population and within the floating cells we actually found they had quite surprising characteristics and that they were able to spontaneously switch to the stem-like state.
Pauline Davies: And how do they compare to the cells that had been stem-like from the beginning?
Christine Chaffer: So we put these de novo derived stem cells, so stem-like cells that were being derived from this differentiated population, we put these through all of the normal stem cells assays that we do in the lab and for all intents and purposes they were the same as the pre-existing stem-like cells within the culture.
Pauline Davies: And then what happened?
Christine Chaffer: So what we are identifying is that there are many different cell-types that make up the mammary gland, which is the system that we’re working in, and that even though these cells are all classified as differentiated cells, they are varying in degrees of differentiation, and that some of these cells that are differentiated are epigenetically different from their partners, their neighboring cells, and that those cells, even under normal conditions can spontaneously transition into a stem-like state, whereas other cells cannot. So in the context of cancer, what this means is that if the cells that we know can spontaneously transition to a stem-like state receive an oncogenic insult, those cells are likely to form a more aggressive cancer because the addition of that oncogene accelerates the rate in which they are able to transition to a stem-like cell. And in cancer that means that the outcome is highly aggressive cancer cells. In contrast, if cells that are differentiated but are not poised to respond to micro-environmental stimuli and to transition to the stem-like state, if those cells receiving an oncogenic insult their outcome is that they still form a cancer but that the cancer is not as aggressive because it can’t spontaneously transition to the aggressive cancer cell state. So depending on the type of cell that is activated or receives an oncogenic insult within the mammary gland, you can generate a variety of different cancer cell types. And perhaps this is one way in which the heterogeneity that we see within breast cancer is generated, and that is to say that the cell that originally receives the oncogenic insult is a determinant of how aggressive your cancer could be. So individuals might have a different type of breast cancer depending on the type of cell that their cancer originated in.
Pauline Davies: So really that’s bad news, isn’t it, for people who are studying cancer because we don’t anymore have to just think about one horrible population of stem cells that go on to cause trouble, but there are a whole lot of naughty cells that we didn’t suspect could be trouble-makers.
Christine Chaffer: Yeah this is true and the broader implications I think of this work is that we do have to think about the heterogeneity that exists within tumors and perhaps broaden our minds to look at the idea that some of the cells that are less aggressive are not locked in that state and that they can in response to contextual signals from the micro-environment offered by the tumor cells, that they can then switch to this aggressive state. So we do start to think that is a very important interplay between the different types of cells within the tumor, and this gives us another set of pathways that we need to start thinking about in what governs these cells ability to transition from this less aggressive to more aggressive cancer state and the upside of that is that we might have some other potential therapeutic targets. For example, if this turns out to be a concept that happens broadly in the cancer setting, where the less aggressive cancer cells can generate cancer cells, this gives us a different way in which to treat cancer in that we can look for ways to block or to lock cells in this less aggressive state without focusing on trying to kill sub-populations of cells. So it might give us another avenue to explore because, you know, current therapeutic approaches looking at just chemo-prevention and radiation seem to be quite unsuccessful in stopping cancer reoccurrence.
Pauline Davies: And Christine you are here with your baby now, only two weeks old?
Christine Chaffer: Yes I just had my second daughter, she is two weeks old and made the trip here to sunny Arizona and we’re enjoying it.
Pauline Davies: Well congratulations.
Christine Chaffer: Thank you.

Interview with Aurora Nedulcu

(Back to Audio)

Aurora Nedelcu: My name is Aurora Nedelcu; I’m a professor at the University of New Brunswick in Canada.
Pauline Davies: And you study unicellular organisms, don’t you?
Aurora Nedelcu: Actually I study both unicellular organisms and simple multicellular organisms specifically to try to understand the transition from the unicellular lifestyle to the multicellular lifestyle, what’s cell differentiation.
Pauline Davies: So tell me what you were mentioning in your talk today about how the little organisms that you study, what were they called?
Aurora Nedelcu: So the group of organisms I study they are called Volvocine algae and they have basically unicellular species called Chlamydomonas and multicellular organisms with germ and somatic cell-type called Volvox.
Pauline Davies: Yes and you were mentioning that these organisms actually can produce germ cells and so reproduce only if their somatic cells, the ordinary cells, actually grew in size.
Aurora Nedelcu: So they have two cell types. The somatic cells are supposed to stay small and have their flagella work and basically move to help the whole colony swim while the reproductive cells are supposed to just grow bigger and bigger and then divide and make babies of the same structure as the mother.
Pauline Davies: So tell me a little bit more in detail, how does that system work? How does the somatic cells, the ones that have the flagella and swim around, turn into germ cells and reproduce? What’s happening?
Aurora Nedelcu: So normally during development the germ cells divide many times to the point that a population of cells become very small and the small cells become somatic cells and the large cells stay reproductive. However, they are mutants in the way a gene is affected such that the somatic cells, although they start small and they should stay somatic, like skin cells in our body, later on they gain the ability to reproduce and by that I mean they start growing, loose the flagella and start making babies on their own.
Pauline Davies: What connection can this have to cancer?
Aurora Nedelcu: So in a sense the somatic cells regain their immortality, meaning they are not acting just for the interest of the colony but rather they gain the ability to be immortal and regain their own individual level fitness, they start proliferating and they produce a mass of cells and by doing that, they negatively affect the fitness of the colony; in that respect they are quite analogous to cancer cells, which pretty much do the same thing, they proliferate uncontrollably and in the end they affect the fitness of the organism.
Pauline Davies: Because of course cancer cells almost always start in somatic cells, don’t they?
Aurora Nedelcu: Yes, so in that respect there is definite analogy. These mutants I’m discussing, the mutation is expressed in the somatic cells, so it is a somatic cell that starts proliferating the same way a cancer cell does.
Pauline Davies: And why do you think it’s worth studying volvox, your model, rather than just studying cancer cells to understand cancer cell’s biology?
Aurora Nedelcu: One possibility is that this is a much simpler system. It’s simpler because there is only one type of somatic cell; it’s also simpler because the genetics underlying the cell differentiation is simpler. So it might be easier to get to the basic principals in a system in which fewer genes are involved then try to unravel all the interactions between genes that are involved in the differentiation of the somatic cell in complex organisms like ourselves. So sometimes it’s easier to find the roots of a process in a system that is less complex, so there is less complexity to unravel.
Pauline Davies: And in Volvox is there any way of turning back the clock and making the germ cells behave themselves and revert to somatic cells if you want them to?
Aurora Nedelcu: This is what I am working on. We’re very hopeful that we could find and we already have some indication that we can affect the intracellular status of these mutant cells in a way that would prevent them from continuing to proliferate. So yes.
Pauline Davies: So in what sort of ways are you doing that?
Aurora Nedelcu: So we first have to find ways to induce the gene in a condition that is outside the developmental context. So we were able to do that and now we are trying to find environmental conditions that can simulate that signal.So we hope that by finding the right change in environment, we could turn back the gene to being a cooperative gene.
Pauline Davies: Well it would be great if we could do that in cancer as well.
Aurora Nedelcu: In principle that is the idea behind these experiments, to find the way to manipulate cancer cells by changing the environment, not necessarily by killing them or by affecting their genes, but just changing the expression of the genes in different conditions.
Pauline Davies: And even in humans and other mammals you were mentioning that cells that move around and have cillia and flagella cannot reproduce and that is something similar to what you’ve been finding in your unicellular organisms.
Aurora Nedelcu: Yeah, so there is a constraint that applies to both my group of algae, but also the ancestors of metazoan, so animals, and it’s also present in our body. There’s a constraint that has to do with a particular structure in the cell that cannot be used simultaneously for two functions, so that structure is called microtubule organizing center and ir organizes the microtubules that are responsible for the flagular activity. But the same structure is also involved in organizing microtubules in the mitotic spindle when the cells divides. So this brings back a very general principal in biology, the idea of trade-offs, whether it’s a functional trade-off or a life history trade-off. In many cases there’s a constraint that the use of one structure cannot be used for two functions at the same time, and the solution is to basically use the structure for one function in one cell-type and for the other function in the other cell-type. So this is supposed to be at the basis of the vision between somatic cells and reproductive cells because a cell cannot both reproduce and be motile. So the solution is that you now have cells that can only reproduce and cells that can only be motile.
Pauline Davies: So the general public might be thinking, well okay you’ve got sperm cells but of course the tail falls off doesn’t it, and it’s not involved itself in reproducing?
Aurora Nedelcu: The sperm cells never divide. To make more sperm cells you divide some cells that do not have tails and then they grow their tail,
Pauline Davies: And what allows the microtubules to decide which way they want to go?
Aurora Nedelcu: The position of the microtubule organizing center. So the position of the microtubule organizing center in the cell is going to determine whether the microtubules are gonna be used for growing the flagella or for division.
Pauline Davies: And is that itself determined very early on when things are newly created or does it change throughout life?
Aurora Nedelcu: So normally a cell ‘decides,’ under quotation marks, to divide when conditions are right, when the cell reaches a size large enough to divide, then some signals tell the cell, ‘okay now we’re ready to divide.’ When the cells gets those signals, then the microtubule organizing centers leave the flagella and they position themselves in the cell in two opposing poles and they start organizing microtubules into the mitotic spindle . So there is signal coming from the cells saying, ‘okay now we are ready to divide, leave the flagella, which is now good to have it working while we are looking for resources, so that was the survival phase, the growth phase, now we’re ready to divide, come and take care of the mitotic spindles.’
Pauline Davies: Okay so they always start off as moving cells and then they turn into reproducing cells, but never the other way around.
Aurora Nedelcu: Yes, so the unicellular life cycle starts with a growth phase which involves the flagella being active and moving the unicellular organism searching for light if it’s a photosynthetic organisms or for resources in general, than growth means increasing in size which triggers cell replication. So for cell replication to proceed and the chromosomes to be equally distributed between the two daughter cells then you need the microtubule organizing center to be involved informing the mitotic spindle.
Pauline Davies: And what other traits to cancer cells have in common with unicellular organisms?
Aurora Nedelcu: One of the odd things that cancer cells do, they escape cell death. So normally somatic cells should trigger a program of cell-suicide when they are damaged or when mutations affect them in a way that would be detrimental to the multicellular organism. Unicellular organisms also have this cell-suicidal system in place which should be turned on when the population fitness of the population is increased. However, there are mutants in unicellular population that escape this programmed cell-death fate., so they continue to stay alive and divide when they should in fact induce cell-suicidal. So there is another analogy between cancer cells and mutants in unicellular populations. Death evasion, so they just escape or they do not trigger cell-suicide when the should.
Pauline Davies: How long have you been interested in cancer?
Aurora Nedelcu: Not that long, probably just for the last couple of years. I do have an interest in multicelluarity and the evolution of multicellularity that goes back to maybe ten years ago but cancer is something that I just recently became interested in.
Pauline Davies: So you see the link between cancer and the origins of multicellularity, do you?
Aurora Nedelcu: Absolutely, although some of the traits or genes I can see them traced back to unicellular lineages.
Pauline Davies: Well thank you very much.

Transcript of Athena Aktipis’s summary and resulting dialogue

(Back to Audio)

Athena Aktipis kindly agreed to give a summing up talk, reminding us of the main points of the workshop. Below is a transcript of Athena’s talk and also the discussion session that followed.


Athena Aktipis: I am both at the Center for Evolution and Cancer at UCSF and here in the Psychology Department, and I’m going to try to sort of summarize what I think are some of the key big questions that we are dealing with, hopefully catalyze a little bit of discussion, and then I’ve left some space in the talk for us to actually talk about some of these questions as we go along.
These are the main ones – the first two are really are what I see as the main big issues that we’re dealing with. One is, “Is cancer a reversion to a unicellular phenotype?” Is that an approach that we agree is a good one that will yield important insights that we want to try to promote as an approach? And also the question of whether cancer is sort of a reversion to an embryonic phenotype in some ways. And then I am going to ask the sort of broader question of how do the metaphors that we use for cancer help or hurt our approach in terms of the research and treatment, and that’s something that I want you guys to think about as we’re kind of going along and once we get to the end hopefully we can have some discussion about that, because I think to the extent that we’re taking these approaches there are certain elements of them which are the reality and then there are certain elements that might be more of a sort of metaphor, or that could be useful and could be harmful .
Athena’s own interests
But before I do that I’m just going to tell you a little bit about what I’ve done and what my background is. I sort of have three main research areas. I started out really as a theoretical evolutionary biologist and evolutionary psychologist interested in how cooperation is maintained and how cooperation evolves in the first place. My main method for addressing those questions has been agent-based modeling. I also do some experimental economics and survey research and things like that. But more recently, I’ve been working a lot in cancer and evolutionary medicine, and trying to understand how we can use insights from social evolution and more generally social principles to inform cancer biology, and perhaps medicine more generally. Finally I have more recent work now looking at how the principles of competition and coordination in humans have been shaped by our ancestral environment, but I’m not going to talk about that. These are some recent and in prep papers that are in the realm of cancer research, so you’ll see this first one is the one that John mentioned about dispersal evolution, then we have some work on looking at the cancer literature and the extent to which evolution is in it. Hopefully I’ll have the chance to give you a little bit of that toward the end of the talk. More generally interest in how our way of thinking about cancer, or way about thinking of medicine in general, may bias us toward not taking a more integrative approach that takes into account evolutionary processes. Then this paper down here about cancer and stem cells – I’ll briefly talk about it. Then I also have some work on actually looking at what’s going on in the human time scale in terms of how social environments and reproductive decision making influence cancer incidence, for the case of breast cancer, and that’s some work in prep.

Is cancer an atavism, a reversion to unicellular phenotype?

Alright, so our first open question, and here I am not going to make any attempt to summarize everything that we’ve said, and I apologize in advance for leaving anything out that you think is important, but just chime right in. I just sort of want to give a sense of the big theme, so in terms of whether cancer is a reversion to unicellular phenotype, I think there were a couple things that were discussed that were relevant and interesting here. One is the question of atavism, so are we kind of uncovering old genes from our very ancient ancestors, and there is just the sort of more practical question of what’s the phenotypic similarity between cancer cells and unicellular organisms, especially early unicellular organisms. Then I think there is a related question about multi-cellular architecture, so how has multi-cellular architecture been important in cancer suppression.
Some of the themes that come with this atavism idea and this reversion to the unicellular phenotype are that what happens when cells start evolving in the body is that their individual level fitness becomes the relevant issue rather than the organismal fitness in terms of the kinds of outcomes that occur, and in doing this, these cells start opening up this old tool box that is suited better for unicellular survival and reproduction then for the survival and reproduction of the entire organism. That of course ties in with the phenotypic similarity question, so some things that look similar between unicellular organisms and cancer cells are higher rates of proliferation than you would expect. This focus on the individual survival rather than like in multi-cellular organisms where we have apoptosis as a very natural part of how the system works. We see altered metabolism, higher metabolism is cancer cells; there’s also I think some interesting questions we haven’t worked through entirely about shift to the Warburg effect and what that might mean, and how that relates to more ancient form of life but clearly I think there’s some interesting questions around this issue of metabolism.

Why do cancer cells move out of home?

Then of course there’s the cell motility question which is close to my own heart, it’s one of the topics I work on, and you know, if you are part of a multi-cellular body you had better do your job and stay where you’re supposed to be so that the multi-cellular body can function well. But this kind of reversion to more unicellular phenotype, perhaps one of the key issues there is that cells that otherwise would not be motile become motile. And a number of the talks over the last two days touched on that issue with the epithelial mesenchymal transition, and just sort of the diversity of the ways cells might move, including both individual level migration and collective migration of various sorts.
Another theme of course is the issue of plasticity. I think Aurora’s work touches on this as well, that for unicellular organisms, they will go through a period of being motile and then shift to reproduction, but they’re not necessarily doing them at the same time, but when you start building a multi-cellular body you sort of specialize and have cells do one or the other. So this sort of de-differentiation and regaining of this ability to change phenotype for a given cell also seems to be a theme that we’re thinking about more and thinking about the connections across the talks that address that.

Multicellular architecture – do cancer cells create it?

In John’s talk he discussed multicellular architecture quite a bit and this was a theme that came up a little bit more too, how do multicellular organisms inhibit cancer?
Then there’s sort of another question I wanted to put out there, based on some on some of my work with my colleagues, which is whether you actually potentially get emergence of something like multi-cellular architecture during cancer progression? So is it not just that cancer cells look like unicells but perhaps that cancer cells begin to evolve phenotype that involves multiple cells and involves perhaps dividing labor or some sort of separation of the cells that will be the diving cells and the cells that will be the supporting cells. I think that’s an open question and an interesting one, so me and Kathleen and Carlo and Jerry and Aurora, who’s here, have a model about stem cells and this notion of multi-cellularity and it sort of came from the bigger question of how can you have non-proliferative cells maintain a portion of the population just sort of on basic evolutionary grounds or you could rephrase it as “Why do non-stem cells exist if it’s the stem cells that are the proliferating cells?” So one of the possibilities for the existence of non-stem cells is that the stem cell progeny are actually enhancing the fitness of the stem cell from which they came and perhaps creating kind of a proliferative unit or sort of a proto-multi-cellular entity. We tested this possibility in a model and it’s viable but whether or not it actually happens is an open question but I think it’s one that we should consider and think about. Should we talk about cancer as a reversion to unicellularity and if we do, do we also want to talk about there being a transition to mulitcellularity in cancer.
Carlo Maley: When you say proliferative, I think you mean indefinitely proliferative or self-renewing. Not that some stem cells are not proliferative, they just have a finite number of divisions they can go through
Athena Aktipis: Yeah exactly.
Alvin Kho: I don’t know the definition of stem cells that they can proliferate continuously, because I do know that people are aware of dormant stem cells, what proliferates actually is the daughter that does all the work and it’s the daughter that goes crazy
Athena Aktipis: Yeah so I think there’s a lot of different way people talk about what constitutes as a stem cell and I’m happy to talk about it more later but I think I’d rather focus on the bigger picture
Carlo Maley: I think there is confusion in the terminology. I think what you’re talking about is the trans-amplifying cells which proliferate like crazy, and what Athena talks about is self renewal that those trans-amplifying cells are not self-renewing so they have a short – they are a dead end in evolution, they have short life span.
Athena Aktipis: So one other point with regard to is cancer a reversion to unicellularity is, you should all know that second bi-annual Evolution and Cancer Conference at UCSF is going to be on a closely related theme which is, ‘What kinds of insights can we derive about cancer from looking at unicellular organisms?’ So if this is a topic that you’re interested in and you want to explore further and be exposed to more or give a talk on, put it on your schedule and get on our mailing list so we can let you know once we have the final dates and get you signed up.
Fertility, cancer and the embryo
Alright next question. Is cancer a reversion to an embryonic phenotype? That’s sort of this large, general question about the relevance of development. I am going to mention just a couple of things that are relevant here. I think the phenotypic similarity is a large part of why we are talking about this question because it looks like there is a similar phenotype. As we also saw it looks like there are major similarities in terms of gene expression so I think it’s clear that there is an emerging picture that the genes that are involved in development are important. And then I also want to talk about fertility, cancer susceptibility trade offs just a little bit, to put a couple of pieces of information out there that I think relate to these kinds of issues about how cancer and development, especially early development, might be related. Some of the similarities that people have talked about, and again sorry for leaving anything out that’s critical, just speak up, the fast proliferation of cells early in development seems to be one of the major similarities, and then of course the invasion of tissues. One thing that blastocytes are really good at is invading into the placenta and creating a supportive environment that will let them grow. To the extent that a maternal physiology is susceptible to an embryo, are there also aspects of maternal physiology that might make her more susceptible to cancer? I think that’s an open question. Other things that seem similar are these issues of immune invasion, and then of course also the plasticity question. So early on in development you have much more de-differentiated cells that can have a diversity of fates.
Go back to this question of whether there might be any trade off here between fertility and cancer. I just sort of want to get everybody doing a little thought experiment about. If it is the case that there is variation among females in their likelihood of having a blastocyst implant, which it seems like there is because some women end up in fertility clinics and some women don’t, then there’s probably some physiological basis for that. That is likely to be based in their genetics. And those differences among women might affect the susceptibility of their tissues, perhaps especially their reproductive tissues, to other kinds of growths, to other kinds of entities that might be trying to extract resources or develop a little niche inside those tissues. So there’s some evidence that something like this might be going on in susceptibility of getting breast cancer.
It’s been found in BRCA one and two mutations are associated with higher fertility and also lower survival, presumably from the breast cancer. This first graph here where we have a really low P-value, this is women born before 1930 and basically outside of a culture of birth control in any way, and then women who were born in between 1930 and 1975 show a slightly smaller effect but there’s still this clear advantage in women who were born before 1930 in that if they have the BRCA mutation they have on average about two more kids. That’s a huge effect.
It also looks like P-53 may be important. So we all know that P53 is really, really important in maintaining genomic stability and suppressing tumors. It’s sort of the conical tumor suppressor gene, but P53 also plays a role in blastocyst implantation and reproduction more generally through LIF, and it’s been found that this codon P72 is actually associated with poor implantation and pregnancy rates for women who are less than 35 years old. So that’s an interesting finding as well.
Metaphors for thinking about cancer – street gang, alien invader, enemy
Alright, so on to our final question, which I kind of want a little audience participation on. How do the metaphors that we may be using help or hurt our approach? And I think embedded in this question is, to what extent are these metaphors and to what extent are they description of what the system is. I think it’s important to recognize that to a certain extent we may be using as a metaphor in other ways we may be using it as an actual description of the underlying structure and the underlying processes.
So there are three aspects of this metaphor or three kinds of metaphors that I want to talk about. One is this notion of cancer cell as unicells. Secondly the notion of cancer cells as embryonic cells. Then third, I want us to talk a little about the metaphor for the war on cancer and what role that plays in our approach to research and treatment.
So at this point I think I want to take about five minutes, for this and for the next slide, for a little bit of discussion about what the strengths and weaknesses are for thinking about cancer cells as unicells. And if there’s anybody who is willing to take some notes it might be interesting to have that as a kind of product that we can share. So let’s starts with the strengths. What are some of the strengths of thinking about thinking of cancer cells as unicellular entities?
Alvin Kho: I was going to ask you what is a unicell?
Paul Davies: Well I can say the sort of strengths that people mention is that they stop listening to the signals of their tissue micro-environment and they’ve got a sort of go it alone agenda. But whilst I am speaking, the weakness that we see is lots of evidence of cooperative behavior in the neoplasms, and these needn’t necessarily be the same time; it could be that they sort of start out as selfish cells but as there is a community of them they become more like a street gang with a sort of loyalty among thieves or something like that. So these are all metaphors, so we go on to these – is this useful or not. So I never felt that going right back to the unicell is the right way to go. This Metazoa 1.0 is an attempt to capture them in their phase before the full organ tissue type differentiation where there is a sort of rudimentary amount or organization but it is sort of loose, and it’s easy to believe that that was just a very short transitory phase. But I think we’ve seen that it might be five or six hundred million years of life on Earth when what we might now call Metazoa 1.0 reigned, ruled supreme. So this is a non-trivial phase in evolution so, that for me is what I am identifying cancer with. But then we’ve heard that maybe this stuff goes all the way back to before LUCA.
Athena Aktipis: Someone over here did you have something to say John? You had your hand up.
John Pepper: Yeah, well forgive me for repeating this from my talk, but to me we could make a very long list of ways in which cancer cells are like unicells, ways in which cancer cells are different from unicells, but to me the critical similarity is the level of selection that is shaping their phenotype and their genotype. Unicells are subject to selection among cells primarily; in things like biofilms there may be also higher levels of selections, but primarily, when we say unicells we’re implying that the selection among individual cells is the primary evolutionary force shaping their traits, and that’s an important way in which they’re similar. In cancer as far as we know, the important evolutionary force is selection among individual cells.
Luis Cisneros: The way to specify the specific stage of cancer were talking about, like if we’re talking about cells in the primary tumor or different cells or that sort of thing.
Athena Aktipis: Yeah, well I think that’s an issue to raise, at what stage Is this metaphor or approach most useful, and perhaps it’s not at all stages.
Paul Davies:It may be metastatic because a primary tumor is like an ecosystem and the metastatic cells have escaped and it looks like they’re going their way
Jean Paul Thiery: So if a cancer cell will be like a unicell, a protist, they will never survive. Cancer cells are … So I’s like in Tokyo when you gout outside with neighbors they don’t know very well so they don’t talk to the very often but the neighbors talk to the cancer cells a lot. So I bet if you would destroy the stroma of cancer cells, then never a cancer cell would progress. Never. Even the most traumatic cancers have the most strona. Pancreatic cancer kills patients in 6 months – It is 97{236bd5e292587b885399ce1fe93b84c86ca4f34851d3c4bf06f3f0da35a3ccbb} stroma, 3{236bd5e292587b885399ce1fe93b84c86ca4f34851d3c4bf06f3f0da35a3ccbb} cancer cells.
Unhappy single cancer cells vs life for a lone bacterium
Athena Aktipis: Carlo, did you have something to say?
Carlo Maley: I was actually going to ask Jim – so Jim, I wanted your perspective on bacteria, the question here is, we know from working in cancer cell culture. We know from work in cancer that if you put a single cancer cell in a tube it doesn’t grow very well at all, it needs neighbors and stroma often. And I was wondering, in bacteria world, if you put a single bacterium on its own, is it similarly sick or is it just fine?
Jim Shapiro:well the ones we use in the laboratory have been cultured and selected, but there are many ones where the ability to grow depends on adequate population.
Carlo Maley: And is there an analogy dependent on the stroma to cells of a different phenotype?
Jim Shapiro:Well you have co-metabolism systems where two organisms are complimentary, reactions that are necessary to work in concert to get it to proliferate. And so you actually end up growing a mixed population and sometimes it’s actually how they order the pairs themselves.
Carlo Maley: Yeah so I still think your given the diversity of multcellularity interactions that you see in the unicellular world and these kinds of things that there may be important, fruitful analogies here, metaphors here, things to learn from these other systems.
Paul Davies:Well I think biofilms and the relationship to the extra-cellular matrix is one that came up with an earlier workshop, we didn’t do anything with it, but I think it’s a good clue.
Athena Aktipis: Yes so this actually kind of creates a connection between what you guys were just saying and what John was saying. John was saying, “Well, the key issue is the level of selection and that selection is going on at the level of the individual.’ But these cases where survival is dependent on neighbors, that actually means that there are other levels of selection that are relevant even for unicellular organisms”. So it may not be just that it’s selection going on at the individual level but that it’s primarily at the individual level or at least initially so, and we may get some fruitful insights from thinking about how higher levels of selection function on entities that are unicellular from these other systems.
Carlo Maley: Part of the value of any metaphor is what good does it predict, what can you get out of it. I would then wonder are there ways that we’ve learned to disrupt these communities and bacteria and other systems that we could import into cancer and ask, well first I’d have to ask how much cooperation is going on in cancer, but if we disrupt those interactions do we get a therapeutic benefit?

Confusing terminology

Jim Shapiro:Can I ask you something provocative? I think the two terms de-differentiate and unicellular are absolutely pernicious, especially in cancer
Paul Davies: Have you got better terms?
Jim Shapiro:Well I think there is no such thing at any level of development, at any stage in human reproduction of an un-differentiated cell. They are all differentiated and specialized in ways that allow them to do the jobs that they do. And this idea of the de-differentiated cell I think is a misleading term.
Paul Davies:Are you happy with totipotent or pluripotent?
Jim Shapiro:No, those are specialized to be totipotent or pluripotent; it’s a special stage of differentiation. And cells are capable from changing from one to the other and at some stage these are terminal
Paul Davies:but there is an arrow of time isn’t there because you differentiate by switching things off
Jim Shapiro:Let me finish the provocative statement and the argument because I think it gives you this idea that there’s, again it relates to what I said earlier about, that there’s something common among cancer rather than there’s something special about each one. And the same thing with the unicellular organisms, the idea is that there is some way where we can treat each cell as an independent entity. And all I know is that, looking at e.coli for many years, as soon as you have the first cell division you have cells that are interacting with each other in quite specific ways. I think it leads us down a wrong path because, the reason I say it’s pernicious is because it’s the reaction, it’s the ability of the cancer cells to interact with the complex architecture of the body and create a tumor that will be not just in the cell or able to metastasize. Those are the things that are critical and important, and I think we lose sights of those when we use the metaphor of an image with a dedifferentiated or unicellular or some sort of primitive state of the cell. That’s the provocative point.
Athena Aktipis: Thank you. Unfortunately I think we should move on. I have a bunch of slides after this too and that was probably more like ten minutes. So I now want to invite some discussion about the metaphor of cancer cells as embryonic cells. So what are the strengths of that metaphor and what are some of the weaknesses of that metaphor.

Can cancer cells be thought of as embryonic cells?

Stuart Newman: Well I think it relates to sort of Jim’s points that cancer cells really have not been evolutionarily adapted to perform particular functions so but they do represent cells that are differentiated in that they do particular things. So what do they do, they do things somewhat similar to embryonic cells. They back up a little bit. They don’t back up necessarily to an analogical state of development but they divested themselves of some differentiated functions, and they maintain some of the toolkit that allows cells to participate in embryonic development. I think that cancer cells as embryonic cells is kind of quasi-embryonic cells is relevant. They don’t represent a particular stage of embryogenesis, but they can do many of the things that embryonic cells can do.
Paul Davies:But that’s not what we just heard from Alvin – or John either.
Stuart Newman: They may acquire antigens and protein and gene expression characteristic of early stages, I’m acknowledging that. But a breast cancer tumor cell doesn’t represent a stage of breast cancer development, it maybe has affinities to some earlier stages but it’s not that particular growing stage. Like, if there are a thousand things going on in development, and then later only five hundred of those things are prevailing, and there are five hundred more, you back up and find two fifty of what the previous stage was.
Carlo Maley: That was the nature of my question to Alvin. Do these tumors land right on the trajectory of Bell or do they land nearby or somewhat further away?
Alvin Kho: No, this is just an association, you are correct, in the large it looks like, that’s all I’m saying, I’m not saying it is, but it looks like there is a correlation.
Jim Shapiro:But Jon earlier was saying that it is.
Paul Davies: Which Jon?
Jim Shapiro: Jon Widom. About the breast cancer, the films he was showing of the migrations, and what you have there in fact is development of breast tissue in inappropriate places in ways that are kind of a perversion of a normal developmental system, and that why he said that cancer is equal to embryology or embryogenesis.
Stuart Newman: But he just showed that the breast cancer cell itself just reacquired the embryonic MENA spliced form.
Jim Shapiro:No he talked about the interactions with the immune system cells.
Jean Paul Thiery: The macrophages.
Jim Shapiro:Not macrophages.
Paul Davies:Yes they were.
Jim Shapiro: Macrophages – that this was very similar to what you saw in the embryonic development, and much farther than just the similarity of an RNA profile, and these were transient stages which then led to formation of breast-like tissue in metastatic locations, and that’s the thing that kills you, it’s not the proliferation of the primary tumor.
Stuart Newman: So it is the breast-like tissue that forms it?
Carlo Maley: So I think one of the strengths here is that it makes a prediction, that if you vaccinated with embryonic stem cells you might actually prevent cancer and we got some data that suggests that’s a good way to go down, or at least were getting some positive results on that, so I mean I think in terms of theories being valuable for making predictions and being useful in the clinic that’s a strong argument for this metaphor, for this analogy.
Jean-Paul Thiery: I have spent forty years studying development, I can tell you that Melanoma are certainly quite trying to repeat what melanoblasts were doing in early stage development, so probably re-acquire most of the goodies and other modules that you really need to execute all these metastatic processes because melanoblasts are extremely metastatic. If we injected a melanoma blast to a chicken embryo it would populate the entire thing. It is as effective as a melanoma cell. Now if you take breast cancer or all other kinds of cancers, they only reacquire some of the modules of early embryonic stage, not necessarily the breast cells, they might require something else, the breast cells, but they need at some stage to acquire unicellular-like features that will migrate, but most breast cells are not single cells even though he showed a few cells at the terminal but most develop as epithelial and branch, so it’s not as unicellular. So if you look at a colon, when the colon develops, there is no unicellular system there, it’s a tube, an epithelial that differentiates, there is nothing under that; however, when colon cancer forms, you’ve seen that you have cycles of appearance of single cells and so on and so on, so this something which is being borrowed to other systems. It’s not that tissue of origin will initially act in its embryonic history, all the phenotypes and all the modules in action may take that somewhere else.

‘War on Cancer’ – Is it a helpful metaphor?

Athena Aktipis: Okay, I’m going to go over our last metaphor and then we have some slides after it. Just a couple minutes of discussion here. The metaphor of the war on cancer is everywhere and even those of us who might not necessarily agree with it can find ourselves using words like “fighting cancer,” “the battle against cancer,” here you see language like “doctors are the combatants in the war on cancer,” it’s everywhere. So one of the questions I am interested in for my own work is how this metaphor can be useful, but also how it might be harmful in understanding cancer from the research prospective and then in treating cancer. So maybe just take about two or three minutes here for some views and then I’ll present a couple more slides that I feel are related.
John Pepper: I think it’s empirically well established that killing cancer cells is consistently the way to get very transient benefits and a very poor way to reduce mortality. If war on cancer means our job is to kill cancer cells, then we really need to drop that idea. If we take it very literally, it’s like the war on terror, the more we go out and kill individuals the more terrorist there will be and the harder it will be to get them under control. The more we go out and kill cancer cells, the harder it will be to get the cancer under control; it’s not a good approach.
Stuart Newman: Well there is a research paradigm associated with Mina Bissell. It basically looks at how you contain tumor cells by introduction of extracellular matrix and other cell types to kind of domesticate them and make them revert to normal morphology. This is very different from the war metaphor.
Jim Shapiro:I just want to make another provocative remark, which is that I think the word ‘cancer’ itself is kind of counterproductive because it gives us the idea that we are talking about a unitary thing.
Participant: So we should call it influenza?
Jim Shapiro:No I think if we called it oncogenesis or transformation we would not have the same idea as when we call it cancer.
Athena Aktipis: In fact some of the slides later talk about this notion of essentialism that we think about. Or one of the ways that people often think about cancer is as a unitary entity and a very uniform entity and I think that could be a problem.
Jim Shapiro:Right, that’s precisely the problem that I think the name contributes.
Alvin Kho: I have a question. I don’t remember what John Condeelis said, but from his talk I came away thinking that cancer is almost a default state, my whole body is settling into this state, so it’s a war against myself, is that true?
Paul Davies: I think it is a default state, people have this idea that they got cancer and they have been invaded by some alien entity and they got to get rid of it, it’s got to go, and I think we all agree that it’s really just a seamless transition from a new born baby with perfectly healthy operating cells to an old person with all forms of defective cells, and there isn’t a line being drawn that says you got cancer now and you didn’t have it then. If you let people think about it that way. War is good for getting money – that’s the only thing!
Athena Aktipis: Yeah, let me go through the rest of the slides because some of the points are beginning to be brought up here. If we look at the history of this metaphor, when Nixon signed the national cancer act, he didn’t actually use the language of this metaphor, but it was picked up by the press and by pretty much the whole machine of cancer research because it works, it works for getting money, it works for getting attention, but there’s some issues with this and one them is this bias toward highly aggressive approaches, but another issue may be that when we think about the out-group, and this is a finding from social psychology, what we think about is a big group of individuals who are really not that individual; they’re all the same, the all behave the same, they’re like each other, they don’t possess unique characteristics. This is called the out-group homogeneity error. So it may also be that by thinking of cancer as the enemy, we’re also thinking about cancer as more homogenous than it actually is.
Jean Paul Thiery: These guys can say nasty things, in about 5 minutes the secret police or FBI is going to take me out of here and send me back to France or Singapore, and I am going to tell you it’s very nasty. When the second world war was starting, well the second world war was okay because you came in and saved us, but then the Vietnam war, don’t talk about Iraq, don’t talk about Afghanistan, okay. All failures, absolute failures, wars are failures, so I think cancer, the ‘War against Cancer’ is by definition going to be a failure.

(Participants laugh and clap)

Jim Shapiro:You left out drugs, and terror, and poverty as well.
Paul Davies: Is there something even more provocative which is that all these wars made the mistake that you can spend your way to a solution just by throwing massive resources.
Jean Paul Thiery: That’s right.
Paul Davies: Whereas what we really need is some clear thinking.
Jean Paul Thiery: Absolutely.
Athena Aktipis: Also act intimidating, that supposed to happen. So I think that one of things that John eluded to was this notion that if we’re taking this very aggressive approach, we may be neglecting other ways that we could be approaching cancer, that may turn cancer into a chronic disease, that may ultimately extend life rather than killing cancer. And really what’s more important, killing cancer or having the patient live? I think we would agree that having the patient live is more important than killing the cancer.
Jean Paul Thiery: This is exactly what happened in the clinic, chronic mylo leukemia – whereas before you would die in a year or two, going through a blast crisis. With Gleevac you lost 10 or 15 years in relatively good shape, and this is the way the hope is to have a chronic disease, like rheumatism, diabetes and anything else.

Horror of the invader – risk appreciation distortion

Athena Aktipis: So we have a little bit of data on cancer metaphors that you guys may be interested in, as a part of some surveys we’ve been doing with undergraduates and some medical students. So one of the things that we found was interesting was that thinking about cancer as part of yourself was negatively correlated with thinking about cancer as an invading barbarian. So this notion that cancer is from your own cells and all of that does put an extent to this metaphor of “Is cancer an out-group, is it something that is invading from the outside?” And then also interestingly, endorsement of evolutionary theory was positively correlated with thinking about cancer as a criminal element in society but not as an invading barbarian. And this notion of sort of criminal gangs, which Paul also eluded to, may be a slightly more useful metaphor than the war metaphor because it suggest that it’s maybe your own kids that are in the gang, it’s part of society but it’s kind of dysfunctional. They may be doing some harm, they may be exploiting some people, but ultimately we don’t want to go out and shoot them all, we want to try to rehabilitate them or get them to behave better, or at least have them keep each other under control, at the very least.
Paul Davies:Keep them quiet.
Alvin Kho: Isn’t this what Stuart just, it’s what I call differentiation therapy – snake oil – everyone here knows what differentiation therapy is, right? You throw in some growth hormones or whatever to these tumors and then suddenly they become gentle and differentiate.
Paul Davies: Is this snake oil?
Carlo Maley: It’s useful in one form of leukemia, it seems to react really well to the differentiation factor, but it’s been hard to do in general.
Athena Aktipis: So just a couple more points that I think have started to be brought up here about other way that our thinking about cancer might be important for how we are approaching it as researchers, and also what kind of treatment options we consider. I think essentialism is a very clear and important way that our cognitive biases may be influencing how we are approaching cancer and by “we” I don’t necessarily mean the people in this room. So I think the essentialist view of cancer is really “The Blob,” you remember that movie from the 80s right? It’s this entity that is just dangerous and is out to get you, and from the essentialist view, if you reduce the size of the blob you’re getting rid of the problem because it’s a unitary entity and it’s static, other than potentially reducing its size. But that neglects the heterogeneous dynamic and adaptive nature of cancer. And as we all know, if you have a heterogeneous entity or a set of entities and you apply a selective pressure, like for example chemotherapy, you get the evolution of resistance and I think the essentialist view can really block reasoning about the evolution of resistance. There’s also this issue of risk eversion which is a strange one in this context because it may be that risk eversion actually leads to more aggressive treatment because people want to get rid of all of the risk, they want the risk totally gone, which means curing cancer, getting rid of it, rather than living with having cells in your body that may be dangerous, may be harmful, that may harm you in the future. It’s almost as if they would rather die from dangerous treatment than live with the risk that cells in their body may at some point kill them. So I think that this is a tricky question but it’s an important one for us to consider, especially on the clinical side.

Patient’s perspective:

Pauline Davies: You know when you’re talking about names you should call these diseases, I actually think that you should be asking cancer patients. You can call them something else, but you know you’ve got to respect those people who are suffering because they know what’s coming. Going down the chemotherapy route knowing that it’s going to be good for some time, but you know you’ve got to ask patients because there’s no point in changing the names of diseases when they know what’s coming.
Athena Aktipis: We totally agree and in fact part of this project is actually talking to breast cancer advocates and working with them as we develop these surveys and figuring out which alternative metaphors to include when we are trying to understand how the metaphor might be affecting people’s reasoning. I think you’re exactly right that we don’t just want to be focusing on the researchers or the clinicians; we also want to understand how patients are seeing cancer.
Pauline Davies: In a sense I was thinking about who are you talking to try to change this definition, are you trying to change it with researchers, because that might change the way they think if they use different language, I completely agree. The doctors who interact with the patients have got to use language that the patients find meaningful.
Athena Aktipis: I think those are complicated questions but really important ones to ask.
Luis Cisneros: It is a matter of communication; the media – it’s a social problem.
Pauline Davies: Well you know when they have cancer there’s so many resources, there’s so many people talking to them they soon get the hang of what’s going on.
Paul Davies: But there is the whole problem about the drug companies that we’ve heard so much about that it seems in their interest to say, “We’ve got an empire and we can come in and annihilate your cancer” and they might buy you a few more months, and most people will think well okay.

Psychology of treatment

Athena Aktipis: So I think just two or three more slides here. So the disgust response, I think this is really important on the patient side and also on the clinician side potentially as well. So we know that humans have this disgust elimination response evolved to minimize our exposure to pathogens. But this response may be activated by thinking about a tumor as a foreign entity. It’s not me, it’s this other, it’s inside me; I have to get this out. It’s a risk by virtue of it not being me. As well as any other harm it might do to our bodies and to the extent that this sort of instinctive response may be leading to a desire to excise the tumor, we should know and understand the role of the psychological response in the desire of patients to choose various treatment decisions and we should I think be very careful about how cancer is being framed especially in situations where you. There may sort of a spectrum of how serious cancer is. There’s this issue with breast cancer now that ductal carcinoma in situ is starting to be thought about as perhaps not being breast cancer and that maybe we don’t have to go after it aggressively, maybe surgery isn’t necessary and I think to the extent that people see cancer of more of this spectrum as you were saying from having no somatic mutations to having a lot, that may be helpful in reducing this desire to just get rid of the cancer, understanding that it’s not so clear cut that this has now become something that is not you, but that there is instead more of a spectrum there. I sort of want to come back again to this issue of essentialism because I think it’s relevant to both the evolutionary prospective and developmental prospective of cancer. It’s relevant to the two other analogies that we’ve been talking about; cancer as a unicell and cancer as basically an embryo or as an embryonic cell. For both of those metaphors you have the dynamic nature, you have the plasticity, the mutability and the heterogeneity among cells. So I think that the bias of thinking of essentialism may interfere with both of those metaphors that we are seeing as useful for certain aspects of approaching cancer. And so we are interested in the psychological side of this and we have some ongoing survey research to get at these issues and I am happy to talk to you more about that, we’re in the process of developing another grant right now to fund this work both with undergraduates, with medical students, and eventually we’ll be looking at cancer researchers and cancer patients in more depth as well as clinicians. So right now we’ve looked at undergraduates and medical students and have been working with the breast cancer advocates.


So just to sort of recapitulate what we’ve been discussing the last forty-five minutes, we have these questions about whether cancer is sort of a reversion to a unicellular phenotype and what value we can get from thinking about cancer as a unicell. We have the question of whether cancer is a reversion to a sort of embryonic phenotype and what value we can get from that as a metaphor and then we sort of generally talked about how metaphors might be hurtful or helpful and some other ideas that are sort of related to the metaphors like essentialism, potentially like the disgust response, and then also this broader question of the metaphor “the war on cancer” and how that might limit us in some ways to exploring what might be viable alternatives for treatment. So thank you, and that’s it.

Workshop Agenda

Participant List

Leave a Reply