Cancer, the malignant evil that corrodes fatally, is supposed to start in one cell. In appearance and behavior, this cell and its daughters are so different from their “normal” predecessors and counterparts that they appear to represent a new species. In this essay, I suggest that the transformation of a non-malignant cell into a frankly malignant state accompanied by all the biologic changes that define cancer as a disease (expansion, angiogenesis, metastasis) may follow the rules of evolutionary biology during speciation. In the strictest sense, speciation refers to reproductive isolation, which is obviously not the case here; subsequently I will use “clones” of cells in lieu of species. How this clone develops a growth advantage over its surrounding neighbors and at the same time, manages to suppress the growth of its normal counterparts, is a subject which is not well understood. The conventional approach of most scientists to such a problem is that of reductionism where an attempt is made to break the cell down into its individual components, and concentrate on identifying abnormalities that could explain the malignant characteristics. Reductionists would view the initiation and subsequent expansion of a cancer cell into an overwhelming clone as being driven by events related predominantly to the cell itself; for example the dysregulation of genes by mutations or deletions. Although, the reductionist method constitutes the backbone of solid science, transformation of a normal cell into a frankly malignant one is a gradual process involving multiple steps, making it difficult to apply the reductionist approach to the problem. These steps are not confined to the cell alone, but also involve a dynamic microenvironment which affects, and is in turn, affected by the expanding population of the abnormal cells. Thus the cell and its microenvironment, or the seed and the soil, constitute a complex system, and pluralists would argue that complex systems cannot be reduced to simple properties of their individual components. Or, to paraphrase Einstein, one can reduce the problem to its simplest possible solution, but no simpler.
Thousands of putative cancer cells are produced in the body each day, but die without further expansion because they are not well equipped to survive in an environment optimized for the support of normal cells. An ongoing interaction between a potential cancer cell and its micro-environment is therefore a necessary requirement for their co-evolution towards a malignant disease state. In other words, even as thousands of cancer cells are produced in the body on an annual basis, the clinical disease with all its malignant manifestations does not appear unless the cancer cell has had a chance to “evolve”. In fact, the situation has many parallels with the ongoing lively debate between the two groups of evolutionary biologists regarding speciation. The orthodox Neo-Darwinians (Maynard Smith, Richard Dawkins, Daniel Dennett) are reductionists who believe that natural selection is the sole engine driving evolution. The proponents of the punctuated equilibrium hypothesis (Niles Eldridge, (the late) Stephen J. Gould and Richard Lewontin) see evolution as being more complex so that natural selection may be the primary but not the exclusive source of modification. They are the pluralists. Application of the broad principles of evolutionary biology to carcinogenesis may define the sequence of events involved in the development of a malignancy, thereby locating therapeutic targets where intervention is likely to lead to an arrest, if not a reversal, of the process.
Let us take the example of the human bone marrow which is an exceedingly dynamic compartment with billions of cells of many different varieties being produced, as well as being programmed to die on an hourly basis. Deviancy is not well tolerated in this high throughput factory. Darwinian tenet would hold that natural selection acts to maintain stasis in a population by jettisoning the anomalous. Survival of a potential cancer cell is clearly incongruous in this background, since it should have been weeded out long before its daughters were able to overwhelm the marrow, but not if the initiation of cancer is a serendipitous phenomenon. Within every population, there are cells with minor variations; some cells are more “fit” to survive than others. Cellular proliferation in the bone marrow, occurs in “niches” where the balance between the negative and positive growth signals is tilted towards the latter. Imagine that a population of cells happened to become isolated in a microenvironmental niche that provided less than ideal support (for example, a slightly hypoxic environment) for the growth of normal cells. Some of the trapped cells may have been better able to survive in this abnormal environment as compared to normal cells that would have died perhaps because they were smaller in size, or they divided faster, or could withstand hypoxia better or lacked a surface protein necessary for recognition by a death effector for elimination. In short, cancer cells may be able to survive and outnumber normal cells in certain “abnormal” microenvironments precisely because of their inability to compete with normal cells in the “normal” microenvironment. The abnormality is best manifested as a growth advantage. If a cancer cell enjoys even a slight growth advantage, it will outnumber its normal counterparts within a few generations, something that can happen in a matter of weeks or days as far as the human body is concerned.
I would like to posit that at least in some instances, the initiation of cancer involves isolation and entrapment of variant cells in a microenvironmental milieu that is not conducive to the proliferation of normal cells. Any variation that enhances the likelihood for survival and reproduction will then be passed from one generation to the next simply as a result of natural selection. Accumulation of even subtle genetic changes over many generations could eventually have a dramatic effect.
An example is that of fatty foods causing gastrointestinal cancer. In rather simplistic terms, there is a burst of secretion of bile acids in the gut following the ingestion of a fatty meal. These bile acids perform their metabolic function efficiently, but as a side effect, also induce programmed cell death in the surrounding mucosal cells. With frequent fatty meals and repetition of this cycle, the stressed cells facing the bile acid assaults fight back by developing survival strategies in this noxious environment. Eventually, one cell will either be selected for survival because of its “differential fitness” or because it has silenced the genes that mediate programmed cell death. An epigenetic mechanisms that cancer cells have been widely shown to employ for silencing genes for death and differentiation is that of hyper-methylation. Simply by adding methyl groups to the cytosines (CpG islands) in the promoter sites of critical genes, the cell can block transcription of that gene. This cell develops the ability to thrive in a microenvironment which is killing its normal counterparts. A survival phenotype is a cancer phenotype.
Chance factors could operate to facilitate the survival of a variant clone of cells, slightly different than the normal cells, but it is still natural selection that does the rest of the work. The role of natural selection is to improve the “fit” between an organism and its environment. Expansion of the clone of cells must be accompanied by co-evolution of the seed (cells) and the soil (microenvironment). Take the following example. Cancer cells may proliferate continuously either because the soil is providing these “growth factors”, or the cell is constitutively turned “on” because of a genetic mutation. The cancer cell must not only divide and expand its own population continuously, it must also shut off the proliferation of normal cells. One way this is accomplished may be by developing the ability to proliferate in response to signals that are inhibitory to the normal cell as illustrated in the following example.
Cells communicate and transmit signals through proteins called cytokines. Tumor necrosis factor or TNF is a cytokine that induces normal cells to undergo programmed cell death. Some leukemia cells on the other hand are stimulated to proliferate by TNF. Let us go back to our statement that within every population, there are cells with minor variations; some cells are more “fit” to survive than others. Now imagine what happens when there are a number of stem cells with varying “fitness” trapped in a microenvironmental niche which had a higher than normal level of TNF. The “normal” cells will be inhibited from proliferating while the slightly “abnormal” one will begin to proliferate. With time, the more TNF is produced, the better the abnormal cell fits the environment and expands its population at the expense of its normal counterpart. In fact, the abnormal cell itself may start producing TNF to enhance its own growth while at the same time suppressing that of the normal cells.
The microenvironment of cancer cells in the body not only consists of stromal cells capable of producing cytokines such as TNFa, but in addition harbors components of the immune system as well as newly formed blood vessels which directly affect the growth and perpetuation of the abnormal clone of cells. An important implication of these biologic insights is that the “cause” of cancer as a disease entity is at least in part related to the changed microenvironment and not something restricted to the intrinsic properties of the cancer cell. Consequently, strategies directed at eliminating the malignant cell alone, no matter how efficient, will only solve part of the problem at best, and be successful temporarily. Even if 99% of the abnormal cancer cells are destroyed but the microenvironment is left intact with all its abnormal features, then normal cells would not be able to survive for long in that setting, resulting in the redistribution of the growth advantage back to a “more fit” or abnormal cell causing relapse. This scenario is unfortunately all too familiar in the treatment of most cancers. Chemotherapy can produce striking complete remissions, but the cancers relapse eventually, and the second time around, they are more resistant to therapy as the cells causing a relapse have followed the Darwinian selection process of having survived in the presence of the noxious drug in the first place. In order to obtain complete and durable responses, both the seed and the soil would need to be targeted.
Developing models like this is not just of theoretical interest, but there are immediate and practical applications of these to the human condition. The conclusion is that it should not be a case of “either/or” in terms of the reductionist versus pluralistic view of cancer, but a combination of the two views as far as planning effective treatment is concerned. In order to kill the seed or the cancer cell, a reductionist approach must be used to identify the key steps involved in the perpetuation of the clone. Targeted therapies should be developed to interfere with the specific intracellular steps, for example an abnormal protein being produced by a mutated gene. In addition, with the pluralistic view of cancer in mind, the extracellular components should be targeted simultaneously, for example blood vessels or cytokines such as TNF. The future success of cancer treatment will depend on how rapidly and how effectively we learn to combine therapies which simultaneously attack several targets in the cell as well as the microenvironment. Studying cancer cells in isolation without their natural in vivo microenvironment, or through artificial mouse models will only yield limited information.
In summary then, cancer initiation could be the result of the serendipitous presence of an abnormal cell in an abnormal microenvironmental niche. Natural selection then works to improve the fitness between the seed and the soil, making both increasingly abnormal. The rate at which this occurs depends at least in part on the body’s ability to mobilize the immune system to mount a counterattack, and that of the cells to expand their clone, for example through the formation of new blood vessels. Thus, the time from initiation to actual disease manifestation could vary considerably depending on the forces driving the fitness landscape. The famous quip by a Neo-Darwinist (who believe that evolution is a gradual process) criticizing the punctuated equilibrium theory that he “did not believe in evolution by jerks” was answered by the Gould group (who suggest that periods of stasis are punctuated by sudden proliferation of species) with the retort that they “did not believe in evolution by creeps”. The evolution of cancer is probably best described by both jerks and creeps.