No single drug or treatment reliably kills cancer stem cells on its own, which is precisely why these cells are considered one of the hardest problems in oncology. Cancer stem cells (CSCs) are a small subpopulation within a tumor that can renew themselves, generate the diverse cell types that make up the rest of the tumor, and survive the chemotherapy and radiation that destroy the bulk of cancer cells around them. Researchers have identified multiple strategies that can eliminate or neutralize CSCs in laboratory and early clinical settings, ranging from drugs that block the signaling pathways CSCs depend on, to engineered immune cells, to treatments that force CSCs to mature into ordinary cells that can no longer fuel tumor growth. The challenge is that CSCs have layered defenses, so the most promising approaches tend to attack them from several angles at once.
Why Conventional Treatment Leaves Cancer Stem Cells Behind
Chemotherapy and radiation work by targeting rapidly dividing cells. Cancer stem cells sidestep this by entering a dormant, slow-cycling state called quiescence. A quiescent cell is essentially hibernating: it is alive and functional, but it is not actively copying its DNA or splitting into daughter cells, so the drugs designed to catch cells mid-division pass right over it. This quiescent state has been shown to contribute to therapeutic resistance across multiple cancer types.1PubMed Central. Stem cell quiescence and its clinical relevance In glioblastoma, for example, a signaling molecule called BMP can push brain cancer stem cells into quiescence without stripping away their ability to self-renew or form new tumors. Those dormant cells then resist both radiation and the standard chemotherapy drug temozolomide, creating a hidden reservoir from which the tumor can regrow after treatment ends.2PubMed Central. BMP signaling mediates glioma stem cell quiescence and confers treatment resistance in glioblastoma
CSCs also protect themselves with molecular pumps embedded in their cell membranes. These pumps, known as ABC efflux transporters, actively eject chemotherapy drugs from the cell before the drugs can do their work. Researchers studying liver cancer stem cells found that blocking two of these pumps (called MDR1 and ABCG2) allowed doxorubicin, a common chemotherapy agent normally considered incapable of killing CSCs, to accumulate inside the cells and trigger their death. In that study, inhibiting the drug pumps essentially transformed an ineffective drug into what the authors called “a robust LCSC killer” in lab conditions.3Scientific Reports. The inhibition of ABCB1/MDR1 or ABCG2/BCRP enables doxorubicin to eliminate liver cancer stem cells This finding is still at the laboratory stage, but it illustrates how CSC defenses can sometimes be dismantled to let existing treatments reach their targets.
Blocking the Signaling Pathways CSCs Rely On
Cancer stem cells hijack a handful of developmental signaling pathways that are normally active during embryonic growth and wound healing. Three pathways come up repeatedly in CSC research: Wnt/β-catenin, Notch, and Hedgehog. When these pathways are abnormally switched on, they drive the self-renewal and survival properties that make CSCs so dangerous.4PubMed Central. Targeting Signaling Pathways in Cancer Stem Cells for Cancer Treatment The logic of pathway-targeted drugs is straightforward: shut down the signals CSCs need to maintain their stem-like identity, and the cells either die or lose the ability to regenerate the tumor.
The Wnt pathway has attracted the most drug candidates so far. Several compounds have been identified that target CSCs through this route. Niclosamide, an inexpensive antiparasitic drug already approved by the FDA for treating tapeworm infections, has shown the ability to suppress CSC populations in ovarian, breast, and colorectal cancers by reducing key Wnt pathway components. It reached a phase 2 trial in colorectal cancer, where its safety and effectiveness were evaluated. Another compound, WNT974, inhibits the proliferation of breast cancer stem cells, while ONC201 has suppressed CSC-related genes in prostate and glioblastoma tumors and is being tested in a phase I/II study for advanced cancers.5PubMed Central. Targeting the Wnt/β-catenin signaling pathway in cancer Small-molecule inhibitors of the Wnt pathway are now considered among the most promising therapeutic strategies for eliminating CSCs, in part because they can be designed to hit very specific molecular targets with fewer side effects than broad chemotherapy.6PubMed. Wnt/β-catenin Signaling Inhibitors
Combination approaches work on this front too. One experimental strategy combined ipafricept, a protein engineered to block Wnt signaling, with gemcitabine chemotherapy in a pancreatic cancer model. The combination reduced both overall tumor growth and the frequency of cancer stem cells more effectively than either agent alone.7Signal Transduction and Targeted Therapy. Wnt/β-catenin signaling in cancers and targeted therapies
The Notch pathway is another major target. Gamma-secretase inhibitors (GSIs) block the enzyme that activates Notch signaling. Growing evidence suggests that Notch activation drives the transformation of ordinary cancer cells into cancer stem cells, so shutting it down could both prevent new CSCs from forming and make existing ones more vulnerable to treatment.8PubMed. Inhibition of γ-secretase/Notch pathway as a potential therapy for reversing cancer drug resistance
Exploiting Metabolic Weaknesses
CSCs have a different metabolic profile from the bulk tumor cells around them. In acute myeloid leukemia (AML), the most primitive and treatment-resistant leukemia stem cells maintain unusually low levels of reactive oxygen species (ROS) and rely heavily on a specific energy-production process called oxidative phosphorylation in their mitochondria. This dependence turns out to be an Achilles’ heel. Researchers found that inhibiting BCL-2, a protein that supports mitochondrial function, selectively eradicated quiescent leukemia stem cells by crippling their energy generation and disrupting their ability to manage oxidative stress. The same BCL-2 inhibitors had far less effect on normal blood-forming stem cells, suggesting a therapeutic window.9Cancer Cell. BCL-2 Inhibition Targets Oxidative Phosphorylation and Selectively Eradicates Quiescent Human Leukemia Stem Cells This finding contributed to the development of venetoclax, a BCL-2 inhibitor now used clinically in AML treatment.
The low-ROS state that protects CSCs also points toward another strategy: deliberately raising oxidative stress inside the cells. Because CSCs maintain tight control over their ROS levels to avoid damage, flooding them with additional oxidative molecules could overwhelm their defenses. Research has proposed evaluating compounds that elevate ROS through external agents as a way to selectively eliminate CSCs.10PubMed Central. Reactive oxygen species in cancer stem cells
Triggering Ferroptosis
Ferroptosis is a form of cell death driven by iron-dependent damage to the fatty molecules in cell membranes. It is mechanistically distinct from the kind of cell death most chemotherapy drugs induce, which matters because CSCs have evolved strong resistance to conventional cell-death pathways. Researchers are now investigating whether ferroptosis can be weaponized against CSCs specifically. The key regulatory mechanisms involve iron metabolism, the buildup of damaged lipids in cell membranes, and the antioxidant defenses CSCs use to prevent that damage.11PubMed Central. Unlocking ferroptosis to overcome cancer stem cells-mediated treatment failure and immune evasion
The strategy centers on tipping the balance between CSCs’ antioxidant shields and the oxidative stress that triggers ferroptosis. Disrupting the system that supplies cystine and glutathione, the main antioxidant buffer that keeps lipid damage in check, can make CSCs acutely sensitive to ferroptosis. Similarly, inducing abnormal iron accumulation or accelerating lipid damage can selectively compromise CSC survival.12PubMed Central. Targeting ferroptosis in cancer stem cells: A novel strategy to improve cancer treatment This is still largely preclinical work, but it represents a fundamentally different angle of attack from traditional therapy.
Engineered Immune Cells
CAR-T therapy, in which a patient’s own immune cells are genetically modified to recognize and kill cancer cells, has shown the ability to target CSCs directly. Because CAR-T cells lock onto specific proteins on the cell surface rather than relying on the usual immune recognition machinery, they can be designed to seek out markers that CSCs display.13PubMed Central. CAR-T therapy: Prospects in targeting cancer stem cells
In glioblastoma, CAR-T cells engineered to recognize NKG2D ligands efficiently killed both bulk tumor cells and cancer stem cells in lab experiments, releasing high levels of the molecules immune cells use to destroy their targets.14PubMed Central. T cells expressing NKG2D chimeric antigen receptors efficiently eliminate glioblastoma and cancer stem cells A separate approach targeted a protein called GRP78 on the surface of glioblastoma stem cells. CAR-T cells designed against GRP78 specifically killed tumor cells and glioblastoma stem cells in co-culture experiments and, in animal models, reduced the number of stem cells while suppressing overall tumor growth.15PubMed Central. Chimeric antigen receptor T cells targeting cell surface GRP78 efficiently kill glioblastoma and cancer stem cells
The appeal of immunotherapy-based approaches is that they could work against quiescent CSCs, since immune cells do not care whether their target is dividing. The difficulty, however, is that CSCs are not passive targets.
How Cancer Stem Cells Hide from the Immune System
CSCs actively suppress the molecular “danger signals” that would normally alert immune cells to their presence. In a study of 175 patients with acute myeloid leukemia, researchers found that leukemia stem cells suppress NKG2D ligand proteins on their surface. These are the proteins that natural killer (NK) cells use to recognize and destroy abnormal cells. By keeping those signals turned off, the leukemia stem cells become effectively invisible to NK cells. The study found that leukemia stem cells achieve this by producing high levels of PARP1, an enzyme that blocks the production of NKG2D ligands.16Nature. Making cancer stem cells visible to the immune system This discovery also suggested a countermeasure: PARP inhibitors, drugs already used in some cancers for other reasons, could potentially restore those danger signals and make CSCs visible to NK cells again.
A similar evasion strategy has been observed in liver cancer. Liver cancer stem cells with high levels of the surface marker EpCAM upregulate a protein called CEACAM1, which directly inhibits NK cell killing. When researchers silenced CEACAM1 on the cancer stem cells or neutralized it on the NK cells, the immune cells regained their ability to destroy the CSCs.17PubMed Central. EpCAM-high liver cancer stem cells resist natural killer cell–mediated cytotoxicity by upregulating CEACAM1 These findings suggest that effective immunotherapy against CSCs may require not just engineering better immune cells but also stripping away the cloaking mechanisms CSCs use to hide.
Differentiation Therapy
Instead of trying to kill cancer stem cells outright, differentiation therapy forces them to mature into ordinary cancer cells that have lost the ability to self-renew. The concept borrows from a principle in normal biology: stem cells naturally give rise to specialized cells that can no longer act as stem cells. If CSCs can be pushed down that same path, they lose the properties that make them dangerous.
All-trans retinoic acid (ATRA), a derivative of vitamin A, is the best-known differentiation agent. It has been used clinically for decades in acute promyelocytic leukemia, where it transforms leukemia cells into nearly normal white blood cells. Researchers are now investigating whether the same approach can work in solid tumors. In glioblastoma, five days of ATRA exposure reduced the expression of key stemness markers and also lowered levels of MGMT, an enzyme linked to temozolomide resistance, in stem-enriched glioblastoma cultures.18PubMed Central. All-Trans Retinoic Acid Induces Differentiation and Downregulates Stemness Markers and MGMT Expression in Glioblastoma Stem Cells That dual effect is significant: it suggests ATRA could both strip CSCs of their stem identity and simultaneously make them more sensitive to chemotherapy.
Colorectal cancer research has explored the same idea, with ATRA showing the ability to push colorectal CSCs toward terminal differentiation, effectively converting them from tumor-initiating cells into cells that can no longer seed new growth.19PubMed. Pleiotropic effects of all-trans retinoic acid in attenuating the hallmarks of colorectal cancer- challenges and scope of differentiation therapy The challenge is that solid tumors are more complex environments than leukemia, and ATRA does not always penetrate tumor tissue effectively on its own.
Oncolytic Viruses
Oncolytic viruses are viruses engineered or selected to infect and destroy cancer cells while sparing normal tissue. They kill cells through mechanisms fundamentally different from those of chemotherapy or radiation, which means the resistance pathways that protect CSCs from conventional treatment do not necessarily protect them from viral infection.20Molecular Therapy. Oncolytic Virotherapy as a Novel Approach to Target Cancer Stem Cells Several virus families are being investigated, including herpes viruses, adenoviruses, vaccinia, and reoviruses. Beyond directly killing infected cells, these viruses trigger immunogenic cell death, meaning they cause the dying cancer cell to release signals that attract immune cells and promote a broader anti-tumor immune response.21PubMed Central. The role of oncolytic virotherapy and viral oncogenes in the cancer stem cells: a review of virus in cancer stem cells
A particularly clever aspect of oncolytic virus design is that the viruses can be genetically engineered to exploit the same pathway abnormalities that CSCs rely on. Because CSCs have altered their signaling to support self-renewal and immune evasion, those alterations can inadvertently make them hospitable hosts for a virus designed to replicate in cells with exactly those features.22STEM CELLS. Concise Review: Targeting Cancer Stem Cells and Their Supporting Niche Using Oncolytic Viruses
Nanoparticle Delivery and Targeted Surface Markers
Getting drugs to CSCs in sufficient concentrations is a problem in itself. CSCs often reside in poorly accessible parts of the tumor, sheltered by a surrounding microenvironment. Nanoparticle-based delivery systems are being developed to carry anti-cancer agents directly to CSCs by coating the nanoparticles with antibodies or molecules that bind to surface markers unique to or enriched on CSCs.23PubMed Central. Nanoparticles for Targeted Drug Delivery to Cancer Stem Cells: A Review of Recent Advances
In gastric cancer, researchers constructed nanoparticles conjugated with antibodies against both CD44 and CD133, two common CSC surface markers. These dual-targeted nanoparticles loaded with ATRA were delivered more efficiently to gastric cancer stem cells than nanoparticles targeting only one marker or no marker at all, and they produced a stronger growth-inhibitory effect.24PubMed. The promotion of nanoparticle delivery to two populations of gastric cancer stem cells by CD133 and CD44 antibodies Newer nanoparticle designs allow co-loading of multiple agents, such as pairing a drug that induces cell death with a differentiation-inducing agent, so that CSCs are hit simultaneously from two directions.25PubMed Central. Targeting Cancer Stemness Using Nanotechnology in a Holistic Approach: A Narrative Review
Disrupting the Protective Microenvironment
CSCs do not survive alone. They depend on a local microenvironment, sometimes called a niche, that provides signals promoting their survival and stemness. One of the most important features of this niche is low oxygen (hypoxia). In gliomas and neuroblastomas, CSCs express high levels of a protein called HIF2α, which helps cells thrive in oxygen-poor conditions. When researchers knocked out HIF2α, cancer stem cell proliferation and self-renewal dropped significantly.26PubMed. Hypoxia and hypoxia inducible factors in cancer stem cell maintenance Targeting the hypoxic niche, whether by disrupting the blood vessel abnormalities that create oxygen-poor zones or by directly inhibiting the molecular responses to low oxygen, represents yet another route to undermining CSC survival.
Why Combination Strategies Are Considered Essential
Given how many defense mechanisms CSCs employ, the field has largely moved toward the view that combination therapy will be necessary. The idea is to pair a conventional treatment that kills the majority of tumor cells with a CSC-specific agent that eliminates the resistant minority. Preclinical studies combining chemotherapeutic drugs with CSC-targeting agents have consistently shown benefits over either approach alone.27PubMed. Combination of chemotherapy and cancer stem cell targeting agents: Preclinical and clinical studies
In triple-negative breast cancer, a notoriously aggressive subtype, researchers developed a hybrid nanoparticle loaded with both paclitaxel (a conventional chemotherapy drug) and verteporfin (which targets pathways important to CSC survival). In patient-derived tumor models, this combination simultaneously repressed multiple signaling pathways involved in cancer growth and CSC maintenance, and outperformed either drug alone.28PubMed. Co-targeting Bulk Tumor and CSCs in Clinically Translatable TNBC Patient-Derived Xenografts via Combination Nanotherapy Similar logic has been tested in breast cancer more broadly, where pairing the CSC-selective compound salinomycin with standard chemotherapy agents showed improved killing, though the response varied across different breast cancer subtypes.29PubMed Central. Combination therapy targeting both cancer stem-like cells and bulk tumor cells for improved efficacy of breast cancer treatment
The Plasticity Problem
Perhaps the most unsettling discovery in CSC biology is that the cancer stem cell state is not always permanent. Under certain conditions, ordinary cancer cells that are not stem cells can revert back into a stem-like state, a process called de-differentiation. In prostate cancer cell lines, researchers found that growing cells in androgen-deprived or drug-supplemented conditions increased the population of cells bearing stem cell surface markers, suggesting that the very treatment designed to fight the cancer could push non-stem cells toward a stem-like, more aggressive identity.30Cancer Research. Efficient conditional reprogramming of differentiated prostate cancer cells back to a stem cell-like state with increased aggressive properties
Stress conditions within the tumor can drive this transition even without treatment. In lung cancer, when ordinary tumor cells experienced nutrient and oxygen deprivation, nearby non-malignant stromal cells became activated and released signaling molecules, including IL-6 and G-CSF, that drove the tumor cells to de-differentiate into CSCs capable of initiating new tumors.31PubMed Central. Stroma-derived IL-6, G-CSF and Activin-A mediated dedifferentiation of lung carcinoma cells into cancer stem cells This plasticity means that even a treatment that successfully eliminates every CSC in a tumor at one moment could be undermined if the remaining cancer cells regenerate new CSCs under stress. It also helps explain why relapse can occur months or years after apparently successful treatment, and it underscores why strategies that target only existing CSCs, without addressing the microenvironment signals and stress conditions that create new ones, may not be enough on their own.
Identifying Cancer Stem Cells in the First Place
Before you can kill CSCs, you have to find them, and that remains a genuine bottleneck. CSCs are identified primarily by surface markers like CD44, CD133, and EpCAM, but no single marker reliably identifies CSCs across all cancer types, and marker expression can shift over time. The most common isolation methods use flow cytometry and fluorescence-activated cell sorting (FACS), where fluorescently labeled antibodies bind to cell surface proteins and cells are physically sorted based on which markers they carry.32PubMed Central. Strategies for isolating and enriching cancer stem cells: well begun is half done Magnetic bead-based sorting is another widely used method.33Cancer Pathogenesis and Therapy. Biomarkers, isolation methods, and therapeutic implications of breast cancer stem cells
The practical limitation is that these techniques work well in research labs but are difficult to apply at scale in clinical settings. A tumor biopsy yields a mixed population of cells, and CSCs typically represent a small fraction. Improving the speed, accuracy, and clinical accessibility of CSC detection is considered one of the key steps toward making CSC-targeted therapies practical. Without reliable identification, clinicians cannot easily monitor whether a treatment is actually depleting the CSC population or whether residual CSCs are poised to drive a relapse.