Chemotherapy kills cancer cells, but it also leaves behind something troubling: damaged cells that stop dividing yet refuse to die. These senescent cells, sometimes called “zombie” cells, linger in tissues for months or years, pumping out inflammatory signals that can erode healthy tissue and, according to a growing body of preclinical research, may help tumors come back. Now, as of spring 2026, researchers are closing in on ways to hunt those cells down, repurposing existing cancer drugs and even engineering immune cells to finish what chemotherapy started.
The problem chemotherapy leaves behind
When cells sustain severe DNA damage from chemotherapy or radiation, many die outright. Others enter a state called senescence: they halt their own division, which prevents them from becoming cancerous, but they remain metabolically active. In that suspended state, they release a cocktail of inflammatory molecules known as the senescence-associated secretory phenotype, or SASP. A 2017 study published in Cancer Discovery by Demaria and colleagues established that chemotherapy-induced senescent cells contribute to systemic inflammation and adverse outcomes in mouse models, with direct implications for understanding why some patients relapse or suffer lasting side effects after treatment.
That finding reframed a basic question in oncology. Doctors have long focused on killing tumor cells during treatment and then monitoring for their return. But if chemotherapy itself seeds the body with inflammatory zombie cells that reshape the tissue environment, then “finishing treatment” may not mean what clinicians once assumed.
Drugs that kill zombie cells already exist
The first breakthrough in targeting senescent cells came in 2015, when Zhu and colleagues published a study in Aging Cell mapping the survival pathways that keep senescent cells alive, called senescent cell anti-apoptotic pathways, or SCAPs. That work identified dasatinib combined with quercetin as a drug pair that could selectively push senescent cells into programmed death. Dasatinib was already an approved leukemia drug; quercetin is a plant-derived flavonoid. Together, they became the first recognized “senolytic” combination, a term coined to describe drugs that selectively kill senescent cells.
A parallel line of research zeroed in on the BCL-2 family of proteins, which act as survival switches in many cell types. A 2016 study in Nature Communications showed that senescent cells could be preferentially killed by inhibiting BCL-W and BCL-XL, two anti-apoptotic proteins, using a tool compound called ABT-737. BCL-2 family inhibitors were already a major drug class in oncology. Venetoclax, a selective BCL-2 inhibitor, is approved for chronic lymphocytic leukemia and acute myeloid leukemia. Its cousin navitoclax (ABT-263), which inhibits both BCL-2 and BCL-XL, never gained approval as a standalone cancer drug partly because it causes thrombocytopenia, a drop in platelet counts, since platelets depend on BCL-XL to survive. But that broader activity against BCL-XL turned out to be exactly what made navitoclax effective against senescent cells.
In 2016, Chang and colleagues reported in Nature Medicine that navitoclax acts as a senolytic in living animals, clearing senescent cells and rejuvenating aged hematopoietic stem cells in mice. The drug has since been tested in a scenario directly relevant to cancer treatment: a preclinical study indexed on PubMed demonstrated that cisplatin, a widely used chemotherapy agent, induces senescence in head and neck tumor models, and that navitoclax can eliminate those senescent tumor cells after treatment. This two-step sequence, chemotherapy to shrink the tumor and induce senescence, followed by a senolytic to clear the zombie remnants, is the direct cancer pathway that has generated the most excitement in the field.
Engineered immune cells join the hunt
Small-molecule drugs are not the only option. In 2020, Amor and colleagues published a study in Nature showing that CAR T cells, the engineered immune cells already used against blood cancers, could be redirected to selectively eliminate senescent cells by targeting a surface protein called uPAR. In mouse models of liver fibrosis and lung adenocarcinoma, where senescence had been induced by MEK and CDK4/6 inhibitors, the uPAR-targeted CAR T cells cleared senescent populations and improved tissue outcomes.
An expert commentary in Signal Transduction and Targeted Therapy placed the Amor et al. findings in context, noting the progression from chemical senolytics like dasatinib-quercetin and navitoclax to cell-based senolytic therapies that exploit immune recognition. The advantage of a CAR T approach is precision: rather than broadly inhibiting survival proteins across many cell types, engineered T cells can be designed to recognize molecular flags that distinguish harmful senescent cells from normal tissue. The disadvantage is complexity and cost, challenges that have dogged CAR T therapy in oncology more broadly.
The first human evidence
Only a handful of senolytic trials have reported results in people, and none so far have focused on cancer. The most cited is a 2019 pilot study published in EBioMedicine, in which Hickson and colleagues gave intermittent doses of dasatinib plus quercetin to patients with diabetic kidney disease. The treatment reduced measurable markers of senescent cell burden in adipose tissue and skin, providing the first proof that senolytics can hit their target in living patients, not just in mice. The intermittent dosing schedule, three days of treatment followed by extended breaks, was designed to limit toxicity while still catching senescent cells during their vulnerable window.
Other early trials have clustered around aging-related conditions rather than malignancies. A Phase 2 trial of UBX0101, a senolytic drug candidate developed by Unity Biotechnology, was registered on ClinicalTrials.gov for osteoarthritis of the knee, but the trial failed to meet its primary efficacy endpoint, a setback that underscored how difficult it is to translate senolytic biology into clinical benefit. No publicly available registry entry confirms an active, large-scale oncology-specific senolytic trial as of the latest available data.
Why this is harder than it sounds
The central tension in senolytic therapy is biological. Senescence evolved as a tumor suppressor: when a cell detects dangerous DNA damage, it enters senescence precisely to stop dividing and avoid becoming cancerous. Clearing senescent cells indiscriminately could, in theory, remove a brake on early-stage tumors or premalignant clones lurking elsewhere in the body. The field has not yet resolved how to target only the harmful, therapy-induced senescent cells while leaving protective senescence intact.
Strategies under discussion include timing senolytics in narrow windows after chemotherapy, when the burden of therapy-induced senescent cells is highest and the risk of disrupting protective senescence is lowest. Surface markers like uPAR could help distinguish pathological senescence from its beneficial counterpart. And combining senolytics with molecular monitoring or surveillance imaging could provide early warning if clearing senescent cells inadvertently promotes new tumor growth.
Safety in the post-chemotherapy setting is another open question. Navitoclax’s tendency to lower platelet counts is manageable in otherwise healthy mice but could be dangerous in cancer patients whose bone marrow is already stressed from treatment. Whether intermittent dosing can thread that needle remains untested in published human data. No primary biomarker study yet links senolytic clearance to reduced inflammation specifically in chemotherapy patients, so assumptions about improved recovery or treatment tolerance are still speculative.
Durability is uncertain, too. In mouse models, a single course of senolytics can produce lasting improvements in tissue function and reduced fibrosis. But human tissues may behave differently over years. If senescent cells re-accumulate quickly after treatment, patients might need repeated senolytic cycles, compounding safety concerns. If a brief cleanup after chemotherapy is sufficient to reset the inflammatory environment, the risk-benefit calculation looks far more favorable.
Then there is the question of how to measure success. Relapse-free survival would be the most clinically meaningful endpoint for an oncology senolytic trial, but it requires large patient cohorts and years of follow-up. Shorter-term readouts, such as changes in inflammatory cytokines, SASP markers, or quality-of-life scores, could provide faster answers but may not correlate tightly with whether a patient’s cancer actually stays away.
What comes next for senolytic oncology
The hypothesis is now well-defined: chemotherapy and targeted agents induce a wave of senescence in both tumor and normal tissues; those senescent cells secrete factors that promote inflammation, tissue damage, and potentially tumor regrowth; and a second wave of senolytic drugs or engineered immune cells clears this residual population, improving recovery and lowering relapse risk. The animal data around navitoclax and uPAR CAR T cells show that each step in this chain is biologically plausible. What is missing is rigorous human evidence that the chain holds together in real-world oncology practice.
If that evidence arrives, senolytics could become a standard addition to cytotoxic and targeted therapies, changing how remission is consolidated and maintained. If safety or efficacy concerns prove too steep, the field may pivot toward more selective approaches, sparing the tumor-suppressive role of senescence while blunting its contribution to chronic damage. Either outcome would reshape how oncologists think about what “finishing treatment” actually means, and what still lingers in a patient’s tissues long after the last infusion drip runs dry.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
