Tumor Heterogeneity Is Fundamental to the Tumor Ecosystem

September 15, 2014

It may be necessary to broaden our concept of the malignant process beyond that of a disease to be attacked-to one that reflects a deeper understanding of the fundamentals of living systems.

In this issue of ONCOLOGY, Drs. Allison and Sledge describe the ways in which interpatient and intratumor heterogeneity present obstacles to the selection of effective personalized cancer therapies based on the characteristics of a cancer.[1] In their view, heterogeneity is seen as a problem to be solved-perhaps by better delineation of the genetic underpinnings of neoplastic development or by more intense sampling of an individual’s cancer in time (over the disease course) or space (from different metastatic loci) with ever more precise technologies. They present a useful review of the recent literature on tumor heterogeneity and describe the potential consequences of therapy-induced heterogeneity.

With regard to intratumor heterogeneity, that is, the heterogeneity among cancer cells within a tumor, Allison and Sledge present the treatment issues through the lens of Darwinian evolution, seeing disease progression as a natural selection of the most “fit” subclones, with systemic therapy sometimes able to alter the tumor composition by providing a mutagenic stimulus that promotes selection of resistant/mutant clones and thus changing the mix of tumor phenotypes over time.

Stepping back from the perspective of the clinical consequences, we would like to explore the phenomenon of tumor heterogeneity from a more fundamental point of view, namely, by exploring the question of how heterogeneity reflects the biologic principles that drive all living entities, whether at the cellular, organismic, or population level. Is heterogeneity an unfortunate consequence of a pathologic process and our efforts to control it, or is it rather a basic feature, essential to life itself?

Much effort has gone into, and continues to be poured into, attempts to reduce cancer development and growth to specific mechanisms-growth factors, receptors, genetic/genomic alterations-and the ways in which these mechanisms are integrated into networks of cause and effect that result in unwanted disease. The hope is to discover the means by which tumor progression can be thwarted-to find the wrench that can be thrown into the works to bring the process to a halt. The idea is to somehow reduce the heterogeneity so as to produce a more uniform target that will be amenable to treatment. But, as is too often demonstrated by the re-emergence of treatment-resistant disease, cancer is an “ecosystem” that has built-in redundancies and safeguards and, importantly, inherent variability at the most fundamental levels; these features protect it from assaults by specific treatments. Ultimately, these features result in the phenotypic heterogeneity that confers treatment resistance, a type of “immortality,” at least until the host unit can no longer function.

Seen this way, cancer is a metaphor for organismal life, for Darwinian evolution in real time. But what if cancer is not a metaphor, but rather an exemplar, of life itself? We suggest that without heterogeneity-that is, variability-there would be no life as we know it. Without the capacity for change, we would never have gotten beyond the primordial ooze. We believe that cancer is a microcosm of fundamental biologic principles played out over a time course of years rather than millennia. Without a whole new approach to treatment, we will not escape this biologic reality.

Allison and Sledge have described tumor heterogeneity in terms of Darwinian evolution, and indeed this is a common way of viewing malignant progression. In our own work, we have focused on another aspect of cancer growth, namely that of the “societal” relationships among cancer subpopulations, which enable these to reciprocally influence each other’s behavior in areas such as growth rate, metastasis, immune sensitivity, and chemoresponse.[2] The mechanisms by which cancers do these things are numerous and idiosyncratic to the interacting subpopulations. The “bottom line,” however, is that tumor behavior is not solely determined by its most aggressive parts but by a composite of the diversity within it. A recent article by Crespi et al[3] considers tumor heterogeneity in the light of “Hamiltonian medicine,” drawing on the concept of “inclusive fitness” to describe how subclone heterogeneity may mediate cancer growth and treatment response. This concept focuses not just on the variability among cancer cells but also on their similarity. The authors describe the dynamic equilibrium among cancer cells in terms of “cheaters” who draw on the capabilities of other cells to advance their own growth (for example, a clone that relies on the angiogenic capacity of other cells) and “helpers” who by their growth abilities (for example, the ability to produce a particular growth factor) enable the proliferation of otherwise deficient relatives. In this view, the balance between cooperation and competition among the subclones, rather than the individual potential of a particular subclone, is what drives the ultimate outcome. Thus, both the similarities and the differences among the participants play major roles in the control of the whole. Crespi et al point out that this way of thinking has heretofore been the domain of behavioral and evolutionary scientists; they suggest that concepts gleaned from research in these disciplines might prove fruitful to the development of new directions for cancer therapeutics.

So, where could this way of thinking lead cancer therapeutics? Would it be possible to use the societal nature of cancer cells as a way to control malignancy, if not to cure it? Would it be better to preserve heterogeneity than to seek ways to knock out particular components based upon some identifiable vulnerability? We suggest that it may be necessary to broaden our concept of the malignant process beyond that of a disease to be attacked-to one that reflects a deeper understanding of the fundamentals of living systems.

Financial Disclosure:The authors have no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.


1. Allison KH, Sledge GW. Heterogeneity and cancer. Oncology (Williston Park). 2014;28:772-8.

2. Heppner GH.Tumor heterogeneity. Cancer Res. 1984;44:2259-65.

3. Crespi B, Foster K, Ubeda F. First principles of Hamiltonian medicine. Philos Trans R Soc Lond B Biol Sci. 2014;369:366-80.