Drug-induced adaptation along a resistance continuum in cancer cells

Abstract

Advancements in rational drug design over the past decades have consistently produced new cancer therapies, but such treatments are inevitably countered through an adaptive process that fosters therapy resistance. Malignant cells achieve drug resistance through intrinsic and acquired mechanisms, rooted in genetic and non-genetic determinants. In particular, recent work has highlighted the role of intrinsic cellular heterogeneity in the emergence of transient drug-tolerant persister cells that survive drug treatment, as well as non-genetically driven cell plasticity toward stable resistance. However, these models do not account for the role of dose and treatment duration as extrinsic forces in eliciting cancer cell adaptation. Here, we show that these two components together drive the resistance of ovarian cancer cells to targeted therapy along a trajectory of cellular adaptation, that we denote the 'resistance continuum'. We report that gradual dose exposure and prolonged treatment promote a continuous increase in fitness, and show that this process is mediated by evolving transcriptional, epigenetic and genetic changes that promote multiple cell state transitions. The resistance continuum is underpinned by the assembly of gene expression programs and epigenetically reinforced stress response regulation. Using both in vivo and in vitro models, we found that this process involves widespread reprogramming of cell survival pathways, including interferon response, lineage reprogramming, metabolic rewiring and oxidative stress regulation. Together, the resistance continuum reveals the dynamic nature of cellular adaptation, and carries implications for cancer therapies, as initial exposure to lower doses primes cells over time for increased resistance to higher doses. Beyond cancer, such continuous adaptation exposes a basic aspect of cellular plasticity, which may also be deployed in other biological systems such as development, immune response and host-pathogen interactions.

Clonal differences underlie variable responses to sequential and prolonged treatment

“…In all three treatments, we found that prolonged treatment resulted in significant changes in pathway activity, most of which have been reported in the literature to be involved with treatment resistance or aggressive cancer phenotypes [34][35][36][37][38][39][40][41] . Thus, across all three different treatments, we found further evidence that the transcriptional states of the population change dynamically as a function of the duration of treatment 27 .…”

“…Our findings further support a growing body of evidence that resistance to treatment can not be explained by single mechanisms acting in isolation 10,49 . Rather, resistance reflects a complex and dynamical process in which diverse and continuously evolving cellular states are generated 9,27 . Our work contributes another layer of phenotypic complexity by demonstrating that the diverse clones generated under treatment have differential sensitivities to subsequent treatments.…”

“…Recent studies demonstrate that duration and dose of treatment can influence the transcriptional state of cancer cells 8,9,27 . Therefore, we asked whether the treatment history of multiple agents can be observed in the gene expression of surviving cells, and whether the first or second treatment has a greater effect on the final transcriptional state of a cell.…”

Peristromal niches protect lung cancers from targeted therapies through a combined effect of multiple molecular mediators

“…S6E). In summary, our results suggest that similar to the multifactorial nature of cell-intrinsic resistance mechanisms 14,15 , EMDR cannot be reduced to a single molecular pathway. Instead, it integrates the functional impact of multiple mechanisms enabling tumor cells to escape therapeutic elimination within peristromal niches and contributing to tumor growth relapse (Fig.…”

Toward a systems-level probing of tumor clonality

“…Cell-intrinsic non-genetic differences can either be pre-existing within a population or can be induced by the selective pressures (e.g., anti-cancer therapies). 6 , 7 , 8 , 25 , 26 , 27 , 28 These modes are not necessarily mutually exclusive and are often attributed to phenotypic plasticity, such as the well-established paradigm of epithelial-mesenchymal transition. 29 Collectively, genetic and non-genetic variabilities can impart fitness advantage to a cell, enhancing its survival, proliferation, and dominance over other cells in both pre-malignant lesions and cancerous tissues.…”