Malignant entities such as pathogens and cancer cells must improvise, adapt, and overcome efforts that thwart their growth and proliferation. One of these defenses is developing drug resistance. Inspired by findings that bacteria became increasingly resistant to higher concentrations of antibiotics, systems biologist Itai Yanai at New York University sought to explore whether cancer cells showed a similar adaptive process.1
“If we’re fighting cancer with drugs, it’s crucial to know your enemy. It’s not just transitioning to one resistant state, but it could keep transitioning,” said Yanai. “It’s like a moving target.”
In a new study, Yanai and his team assessed progressive changes in cancer cells when dosed with increasing drug concentrations over a year. They observed that the cells developed resistance in a stepwise manner, with each level marked by distinct characteristics—they described this pattern as a resistance continuum. Their findings, published in Nature, suggest that the emergence of distinct adaptive states could be leveraged to develop more targeted drug combinations tailored to specific cellular states.2
First, Gustavo França, a postdoctoral researcher in Yanai’s team and coauthor of the study, treated ovarian cancer cell lines with a ramp of the poly(adenosine diphosphate-ribose) polymerase (PARP) inhibitor, olaparib. Drug-naïve cells initially received a relatively low one micromolar dose. After about one month, when the cells adapted and reached confluency, França seeded them onto a new plate with a higher dose, eventually reaching 320 micromolar of the drug. Over 12 months, França generated 10 different adapted cell populations.
Using single-cell RNA sequencing (scRNA-seq), the team compared the 10 cell populations and defined five major transcriptional states, S1 to S5. Along the resistance continuum, the S1 state included less drug-resistant cells while the S5 state indicated highly drug-resistant cells.
“The notion is that it’s like a skipping stone. When you skip a stone on the water, every skip is in a different state. By the end, the cancer [cell] is very resistant,” said Yanai. Each state’s underlying gene expression was distinct. For instance, the expression of lineage-defining transcription factors and epithelial markers decreased along the continuum, which indicated dedifferentiation. The researchers also observed morphologic changes as cells underwent an epithelial to mesenchymal transition (EMT).
“We see EMT early on, but it’s only one step along the way. It’s not the be-all and end-all. It seems that putting the cell into a more plastic state can resist the drug, but to achieve higher orders of resistance, there are other changes built on top of that,” explained Yanai. The resistant states showed increased fitness to the drug, and in the most fit states, the researchers observed metabolic rewiring.
The team wondered whether these adapted cell states were genetic. Using scRNA-seq and whole-exome sequencing, they observed copy number alterations associated with the states. The researchers found that during drug adaptation, chromatin reprogramming led to increased accessibility for stress response transcription factors and decreased accessibility for lineage markers, indicating dedifferentiation.
“This work provides important mechanistic insights, including the role of the EMT, into the evolution of therapeutic resistance in cancer cells,” said Beverly Emerson, a molecular biologist at Oregon Health and Science University, who was not involved in the study.
To study this process in vivo, the team generated two mouse models: one received an acute short-term treatment, while the other received prolonged long-term treatment. The acute regimen elicited a partial reprogramming while the long-term model revealed reprogramming that overlapped with the in vitro adapted resistant states. “[These findings] contribute much needed detail to our knowledge of evolutionary principles of survival at micro to macro scales—from cells to organisms to species,” remarked Emerson.
By mapping this dynamic resistance continuum, Yanai posits that researchers can exploit these defining features to develop drugs specific to particular resistance states. “The big, new frontier of how they adapt is not mutations, but rather that [cancer cells] somehow have this intrinsic ability to reprogram themselves,” said Yanai.
- Baym M, et al. Spatiotemporal microbial evolution on antibiotic landscapes. Science. 2016;353(6304):1147-1151.
- França GS, et al. Cellular adaptation to cancer therapy along a resistance continuum. Nature. 2024;631(8022):876-883.
