One Cancer Gene, Two Fates: Study Shows Senescent Cells Split Into Tumor Blockers and Tumor Seeds

A single cancer-causing gene does not push every cell down the same path. Researchers at New York University Abu Dhabi have found that when an oncogene is switched on in liver cells, the resulting precancerous cells split into distinct populations: some become permanently arrested in a tumor-suppressive state, while others retain the capacity to resume division and potentially seed malignancy. The discovery, published in EMBO Reports, adds a crucial layer of nuance to one of the most relied-upon concepts in cancer biology, cellular senescence, and suggests that therapies designed to clear senescent cells may be hitting only a fraction of the targets they need to reach.

Cellular senescence is often described as a biological emergency brake. When a cell sustains DNA damage, experiences intense stress, or detects aberrant oncogenic signaling, it can enter a state in which it remains metabolically alive but permanently loses the ability to divide. Because uncontrolled proliferation is the defining hallmark of cancer, this arrest is widely regarded as a powerful intrinsic defense mechanism, a tumor-suppressive function inherent to essentially all cells. For decades, researchers have observed senescent cells accumulating in premalignant lesions and interpreted their presence as a reassuring sign that the tissue has contained the threat.

The NYU Abu Dhabi team, working within the University’s Center for Genomics and Systems Biology, set out to test how secure that containment really is. Using a zebrafish model of liver cancer, they examined the earliest responses of cells to a cancer-causing gene, capturing the moment when a normal hepatocyte first confronts oncogenic signaling. Zebrafish are uniquely suited to this kind of question: their livers develop rapidly, their transparency and genetic tractability allow individual precancerous cells to be tracked over time, and the molecular circuitry governing proliferation arrest is broadly conserved with that of humans.

What the researchers found was not a single senescent state but several. Some of the precancerous cells entered what the team describes as terminal senescence, an arrest so deep that the cells can no longer generate tumors under any circumstances examined in the study. Others, however, entered a form of senescence that proved reversible. These cells could regain the ability to divide, re-entering the cell cycle and reacquiring the proliferative potential that, if left unchecked, could contribute to tumor development. In other words, the label senescent concealed two radically different prognoses for the same tissue.

Senior author Kirsten Sadler Edepli, Professor of Biology at NYU Abu Dhabi, framed the finding as a fundamental challenge to how senescence is assessed in both the laboratory and the clinic. Cancer, she explained, is not simply a matter of whether a cell stops dividing or continues to grow. Her team’s data show that there are different types of cellular senescence, and that the type acquired by a precancerous cell can make the difference between whether it ultimately becomes a cancer cell or not. She also emphasized that these different forms of senescent cells respond very differently to treatments, a point she described as critically important when developing and applying cancer therapies.

The therapeutic implications are immediate. Many commonly used chemotherapies do not kill cancer cells outright; instead, they damage them severely enough that the cells enter senescence. The tumor appears contained, but the senescent cells survive, lingering in the tissue. A growing body of oncology research argues that these therapy-induced senescent cells should be eliminated with a second class of drugs known as senolytics, agents designed to selectively kill cells that have entered senescence. The logic is straightforward: remove the arrested cells before they can recover, secrete inflammatory signals, or occasionally slip back into the cell cycle.

But the zebrafish experiments reveal a complication in that logic. When the researchers tested a common senolytic against the precancerous cells in their model, the drug proved effective against only some of the senescent cells and failed to eliminate others. The senolytic was not broadly useless; it simply discriminated between senescent states, clearing one population while sparing another. If the spared population corresponds to the reversible form of senescence identified in the study, then a treatment course that looks successful on paper, with senescence markers dropping and tumor burden appearing controlled, could still leave behind the very cells most capable of reigniting the disease.

This heterogeneity helps explain a persistent puzzle in cancer medicine: why tumors sometimes recur long after treatment has apparently eliminated all visible disease, and why senolytic strategies that succeed spectacularly in one experimental context underperform in another. If senescence is not a monolithic state but a spectrum of arrest programs with distinct molecular underpinnings, then senolytic drugs, which typically target survival pathways that senescent cells depend on, will only work against the subsets of senescent cells that share those dependencies. Accurate classification of senescent states may therefore become as important to treatment planning as the drugs themselves.

The study also carries a message about how precancerous lesions should be interpreted. A liver nodule densely populated with senescence markers has traditionally been read as a lesion that the body has successfully neutralized. The new findings suggest that within such a population, some cells may be irreversibly locked out of the cell cycle while others are merely paused, biding their time with the latent ability to divide again. Distinguishing between these populations at the molecular level could allow clinicians to identify which premalignant lesions warrant aggressive intervention and which can be safely monitored, sparing patients unnecessary treatment.

Beyond its clinical relevance, the work offers a striking example of research training in action. The project was led by Research Scientist Elena Magnani, who mentored several undergraduate students through their capstone research projects, and Sadler Edepli highlighted their contribution to the discovery, noting that students can play a valuable role in high-impact research when given the opportunity to engage directly with important scientific questions and dedicated mentors to guide their efforts. As the field moves toward senescence-subtype-aware therapies, the study’s central insight stands out: the same oncogene, acting in the same tissue, can produce cells that guard against cancer and cells that quietly prepare for it, and only therapies that respect that difference are likely to finish the job.

Subject of Research: Heterogeneous cellular senescence responses to oncogene activation in a zebrafish liver cancer model

Article Title: NYU Abu Dhabi study reveals how cells respond differently to a cancer-causing gene

Article References: NYU Abu Dhabi study reveals how cells respond differently to a cancer-causing gene. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: cellular senescence, oncogene, liver cancer, zebrafish model, senolytics, tumor suppression, precancerous cells, EMBO Reports, cancer therapy, NYU Abu Dhabi, cell cycle re-entry, chemotherapy-induced senescence