Principles and mechanisms of non-genetic resistance in cancer

Abstract

As well as undergoing genetic evolution, cancer cells can alter their epigenetic state to adapt and resist treatment. This non-genetic evolution is emerging as a major component of cancer resistance. Only now are we beginning to acquire the necessary data and tools to establish some of the underlying principles and mechanisms that define when, why and how non-genetic resistance occurs. Preliminary studies suggest that it can exist in a number of forms, including drug persistence, unstable non-genetic resistance and, most intriguingly, stable non-genetic resistance. Exactly how they each arise remains unclear; however, epigenetic heterogeneity and plasticity appear to be important variables. In this review, we provide an overview of these different forms of non-genetic resistance, before exploring how epigenetic heterogeneity and plasticity influence their emergence. We highlight the distinction between non-genetic Darwinian selection and Lamarckian induction and discuss how each is capable of generating resistance. Finally, we discuss the potential interaction between genetic and non-genetic adaptation and propose the idea of 'the path of most resistance', which outlines the variables that dictate whether cancers adapt through genetic and/or epigenetic means. Through these discussions, we hope to provide a conceptual framework that focuses future studies, whose insights might help prevent or overcome non-genetic resistance.

Fig. 1

The potential adaptive modes for non-genetic resistance. Resistant states can be either pre-existing or acquired. If there is a pre-existing resistant state, then resistance can emerge through simple Darwinian selection and outgrowth. This mode of adaptation is completely dependent on epigenetic heterogeneity and depends on the pre-existing state being relatively stable. Acquired non-genetic resistance can theoretically arise through either gradual Darwinian selection or Lamarckian induction. Gradual Darwinian selection could occur by selecting for gradually increasingly resistant cells or cells that have an increasingly stable resistance programme. For this mode to apply, the initial cells selected cannot revert spontaneously back to the initial state. In addition, the next generation of cells has to stochastically acquire a more resistant or more stable state. Put differently, the normal distribution of resistance must gradually shift towards resistance with each generation stochastically. Therefore, it is currently unclear how this mode of resistance would be possible. Acquired resistance can also arise through Lamarckian induction. An initial subpopulation in the appropriate epigenetic state is capable of initiating epigenetic changes in response to the drug, which results in the cells moving to a new cell state. We propose that, in general, these drug-induced changes are due to compensation. In some instances, these epigenetic changes could result in the cell transitioning into a new stable cell state, therefore resulting in stable non-genetic resistance.

Fig. 2

The path of most resistance. Drug resistance can arise through purely non-genetic changes, purely genetic changes, initial non-genetic changes then genetic changes, or non-genetic changes within a particular genotype. Cancer cells will use the adaptive pathway that results in the most rapid or highest degree of drug resistance. The factors that are likely to contribute to which pathway is followed are epigenetic plasticity, epigenetic heterogeneity and genetic stability. Higher plasticity, epigenetic heterogeneity and genetic stability will favour non-genetic adaptation, whereas lower plasticity, epigenetic heterogeneity and genetic stability will favour genetic adaptation. Genotoxic therapies might also promote genetic resistance.