A recent study has examined whether senescence might allow cancer cells to survive oncogene withdrawal and become even more dangerous [1].
Some blocking techniques are temporary
Many cancers depend heavily on a particular oncogenic signal to maintain their growth. Drugs that block these drivers can, therefore, produce dramatic tumor shrinkage. A major clinical problem, however, is that a small population of cancer cells may survive treatment and eventually regenerate the tumor, which no longer responds to the same treatment and often grows in a more aggressive, invasive manner. The biological changes that allow these surviving cells to persist during prolonged suppression of the oncogene are not fully understood.
Therefore, these investigators used a genetically controllable system in which cancer cell growth is driven by the SV40 large T antigen (Tag). Expression of this oncogenic protein could be switched on or off by adding or removing doxycycline, respectively.
Doxycycline can be used either by directly adding it to a cell culture or by adding it to the drinking water given to animals who carry such modified cells. This provides a way to reproduce, experimentally, the situation in which a tumor suddenly loses the oncogenic signal on which it has become dependent. The researchers followed the cells after oncogene withdrawal both in culture and in tumors grown in mice.
They also examined whether similar phenomena occur in a human cancer model. For this purpose, they used A375 melanoma cells carrying the common BRAFV600E mutation and treated them with vemurafenib, a drug that inhibits mutant BRAF. This second model tested whether the observations from the engineered mouse system might also apply to a clinically relevant form of targeted therapy.
Oncogene loss drives cellular senescence
When the oncogenic driver was removed, tumor cells rapidly stopped dividing and developed morphological and molecular characteristics associated with senescence. They became larger and flatter and accumulated senescence-associated β-galactosidase, a well-known biomarker. At the same time, cell-cycle regulator expression changed in a way that signified durable proliferation arrest: the cells were no longer able to divide.
Interestingly, this process did not follow the classic pattern in which p16 is strongly induced. Instead, senescence induction relied largely on a p21-associated mechanism. The authors explained this unusual pattern in terms of the interaction between SV40 Tag and the tumor-suppressor proteins p53 and Rb. Once Tag was removed, the regulatory relationship between these proteins changed, producing a form of senescence that differs from the canonical pathway.
Senescent cells remain biologically active
Although the cells stopped proliferating, they did not become metabolically or functionally inert. They altered their gene-expression programs and began producing a range of secreted molecules associated with inflammatory signaling and tissue remodeling.
These cells also underwent substantial metabolic adaptation. Rather than simply reducing energy production, they increased activity in both glycolysis and mitochondrial respiration. This suggests that the surviving population enters an energetically demanding state in which multiple metabolic pathways are running simultaneously. Such flexibility may help cells remain viable while they are unable to divide.
This finding changes our understanding of residual senescent cancer cells. They may appear dormant because they are no longer proliferating, but they remain metabolically active and capable of substantially influencing their surroundings.
Senescence can be followed by tumor regrowth
The most important observation of the study was that tumor cells that experienced oncogene withdrawal were more capable of producing recurrent tumors than cells that had not undergone this state. Tumors initially regressed when the oncogenic driver was switched off, but a subset subsequently began growing again.
In some animals, tumors eventually emerged even though the original oncogene remained suppressed. Nine of twelve animals in one experimental group developed tumors after a long delay despite continued inhibition of Tag. This demonstrates that recurrence did not necessarily require restoration of the original oncogenic stimulus. Instead, some cells acquired new ways of sustaining proliferation.
Thus, this study suggests a two-stage process: oncogene removal initially forces cancer cells into a non-proliferative state, but the prolonged survival of these cells creates an opportunity for genetic and functional adaptations that can eventually restore tumor growth.
Genetic changes accompany this escape from senescence
The cells isolated from recurrent tumors were substantially different from the original tumor population. They displayed extensive chromosomal abnormalities and increases in chromosome number, indicating that genome instability had developed during or after the senescent period. Such abnormalities can generate genetic diversity, potentially allowing a subset of cells to find alternative routes around the growth restriction imposed by oncogene loss.
The recurrent cells also displayed altered metabolic programs. Pathways involved in nucleotide production, amino-acid metabolism, and folate metabolism became more active, consistent with the increased biosynthetic requirements of cells that had returned to proliferation.
One of the most notable molecular changes was the increased expression of Mdm2, a protein that suppresses p53 activity. This provided a plausible mechanism for overcoming the growth arrest that had followed loss of SV40 Tag. Importantly, cells from recurrent tumors were particularly sensitive to an Mdm2 inhibitor, whereas the original tumor cells were not. This suggests that Mdm2 became a new dependency during the transition from oncogene dependence to oncogene-independent growth.
Changes in the tumor microenvironment
This study also indicates that recurrence cannot be explained entirely by alterations within the cancer cells. The investigators compared the immune and stromal composition of tumors before oncogene withdrawal, during tumor regression, and after recurrence.
The composition of the tumor environment changed substantially during this process. Recurrent tumors contained more endothelial cells, consistent with renewed blood-vessel formation. They also had fewer conventional dendritic cells and more regulatory macrophages. Together, these changes suggest that the environment surrounding the recurrent tumor becomes less favorable for effective immune surveillance and more supportive of tumor growth.
The secretory activity of senescent cells may contribute to this remodeling. Molecules released by these cells can affect neighboring immune, stromal, and vascular cells. Consequently, a population that initially suppresses tumor expansion by ceasing to divide may simultaneously create conditions that make later tumor growth easier.
Relevance to BRAF-targeted therapy
The experiments with human A375 melanoma cells provide evidence that the phenomenon is not restricted to the engineered mouse model. Inhibition of BRAFV600E with vemurafenib generated cells with several features of senescence, including prolonged growth arrest and changes in cellular morphology and secretory activity. The authors therefore suggest that a similar response could occur when human tumors are treated with drugs that remove a major oncogenic growth signal.
However, the authors also emphasize that the molecular mechanism is likely to depend on the genetic background of the cancer. For example, Mdm2-based escape may be particularly relevant to tumors in which p53 remains functional. Tumors carrying TP53 mutations would be expected to use different mechanisms to bypass senescence, so the therapeutic implications cannot simply be generalized to all cancers.
These findings have potentially important implications for targeted therapy. A treatment that efficiently suppresses an oncogenic driver may nevertheless leave behind a population of viable cells with a capacity to adapt. Consequently, preventing relapse may require strategies that eliminate these surviving cells or block the mechanisms they use to escape growth arrest.
Literature
[1] Schmitt, P., Hönig, K., Norcia, M. T., Nogueira, M. F., Flore, V., Vesperinas, I. S., … & Blankenstein, T. (2026). Oncogene inactivation-induced senescence facilitates tumor relapse. Nature Communications, 17(1), 6244.
View the article at lifespan.io
