Iron Metabolism as a Therapeutic Vulnerability in Stem Cell-Like Castration-Resistant Prostate Cancer | bioRxiv

Iron Metabolism as a Therapeutic Vulnerability in Stem Cell-Like Castration-Resistant Prostate Cancer | bioRxiv

Iron: The Hidden Fuel Driving a Stubborn Subtype of Castration-Resistant Prostate Cancer — And a New Way to Turn It Against the Tumor

Researchers discover that "stem cell-like" CRPC cells depend on iron to survive — and that disrupting iron balance triggers a lethal form of cell death in those cells while leaving others largely untouched.

BLUF — Bottom Line Up Front

A new laboratory study published in February 2026 from the University of Bern (Switzerland) and McGill University (Canada) has identified a potentially exploitable weakness in one of the most treatment-resistant subtypes of castration-resistant prostate cancer (CRPC). The "stem cell-like" subtype (CRPC-SCL), which accounts for roughly 25% of CRPC cases and resists standard hormone therapy, is uniquely dependent on iron. Cells in this subtype accumulate excess iron to maintain their aggressive, stem-like behavior. When a natural compound called Brusatol is used to block a key iron-regulating protein (NRF2), the resulting iron overload becomes lethal to those cells — a process called ferroptosis. In animal models, Brusatol shrank tumors selectively, while cells without the stem-like traits were comparatively spared. The research is pre-clinical (conducted in laboratory cell cultures and mice) and is not yet ready for human trials, but it illuminates a concrete new target for a subtype of CRPC that currently has few good options.

Why This Research Matters to Patients

If you have been told that your prostate cancer has progressed to castration-resistant disease — meaning it no longer responds reliably to hormone therapy — you are likely familiar with the frustration of finding that the disease seems to always find a way forward. One major reason is that not all CRPC cells are alike. In recent years, scientists have learned that CRPC comes in at least four biologically distinct subtypes, each with its own driver mechanisms and, importantly, its own vulnerabilities.

One of those subtypes — the stem cell-like subtype (CRPC-SCL) — has been especially difficult to treat. These tumors are defined by cells that behave more like primitive stem cells than like adult prostate tissue, and they show poor responses to the hormonal drugs that work better for other CRPC subtypes. A 2022 landmark study by Tang and colleagues in Science first defined these four CRPC subtypes through chromatin profiling, and it found that the SCL subtype has a particularly aggressive clinical course. Until now, no one had a clear plan for how to specifically attack it.

The new study from Bern and McGill may be a meaningful step toward changing that.

What the New Research Found: A Three-Act Story

Act 1 — CD44 Marks the Most Dangerous Cells

The research team, led by Marianna Kruithof-de Julio and colleagues, focused on a protein called CD44 — a molecule sitting on the outer surface of cells that is known to be elevated in cancer stem cells. CD44 has long been recognized as a marker of prostate cancer stem cells, cells that resist castration and can re-seed tumors after treatment. The scientists confirmed that CD44 is especially prominent in the CRPC-SCL subtype.

Within CRPC-SCL tumors, not all cells express the same amount of CD44. The team sorted cells from a patient-derived tumor model (LAPC9) into CD44-high ("CD44hi") and CD44-low ("CD44low") groups. What they found was striking. CD44hi cells formed larger and more numerous cancer colonies in a dish. When injected into mice, they grew into significantly larger tumors and grew faster. Just as important, even when CD44-low cells were injected alone, they eventually converted themselves into the CD44-high state during tumor growth — suggesting the tumor actively maintains this dangerous subpopulation as a kind of self-defense strategy. Clinical data from real patients supported the relevance of this finding: in prostate cancer tissue samples, patients whose tumors had higher CD44 expression had worse long-term outcomes and faster progression of disease.

What This Means for Patients CD44hi cells appear to be the "bad actors" within CRPC-SCL tumors — the most aggressive cells, the ones that survive treatment and regrow the tumor. Finding a way to specifically target them, while leaving other cells unharmed, would be a meaningful advance.

Act 2 — Iron Is the Secret Ingredient

The second major discovery is that CD44hi cells are iron addicts. The researchers found that CD44hi cells maintain significantly higher levels of iron inside the cell compared to CD44-low cells. This isn't accidental. CD44 itself acts as a shuttle for iron into the cell — it binds to hyaluronic acid (a common molecule in the tissue environment) which carries iron, and escorts it inside.

Why does this matter? Because iron, at the right level, does something very useful for these cancer cells. It activates a family of enzymes — called histone demethylases, specifically one called KDM3A — that strip away chemical "silencing tags" from genes, including the gene for CD44 itself. In plain terms: the iron helps the cancer cells keep themselves in a stem-like state. When the researchers used a drug called deferoxamine (DFO, an iron-chelating agent that removes iron from the body) to deplete iron in the cells, CD44 expression fell, the cells became less aggressive, and — critically — CD44-low cells could no longer convert themselves into the more dangerous CD44-high state.

This creates what scientists call a positive feedback loop: more CD44 means more iron uptake, which activates the enzymes that maintain CD44 expression, which pulls in more iron. The tumor essentially locks itself into a stem-like, drug-resistant state using iron as the key.

Act 3 — Disrupting the Iron Balance Kills the Cells

Having identified this iron dependency, the researchers then asked the obvious next question: can we use iron against these cells?

This is where ferroptosis enters the picture. Ferroptosis (from the Latin "ferrum" for iron) is a form of programmed cell death driven by iron-catalyzed damage to cell membranes — specifically, the peroxidation of lipids (fatty molecules) in the cell's outer layers. Think of it as the cell membrane rusting from the inside. Normal cells have a robust defense against this, anchored by a protein called NRF2 (Nuclear factor erythroid 2-related factor 2), which acts as a master switch for the cell's antioxidant defenses and helps manage iron storage in ferritin (a kind of cellular iron-safe).

The research team found that CD44hi cells — precisely because they carry so much iron — have very high NRF2 activity. NRF2 is essentially working overtime to protect the cells from the toxic consequences of all that iron. This is their Achilles heel.

When the scientists treated cells with Brusatol, a natural compound derived from the plant Brucea sumatrana that inhibits NRF2, the results were selective and striking. Inhibiting NRF2 reduced the cells' ability to safely store iron in ferritin, causing the free iron pool inside the cell to surge. This iron overload drove massive lipid peroxidation, and the CD44hi cells died via ferroptosis. CD44-low cells, which carry less iron and have lower NRF2 activity to begin with, were comparatively unaffected.

In head-to-head competition assays — where CD44hi and CD44-low cells were mixed together and then treated with Brusatol — the CD44hi cells were preferentially eliminated. In living mouse models bearing human CRPC-SCL tumors, Brusatol treatment significantly reduced tumor size, lowered tumor weight, and markedly decreased the proportion of CD44hi cells within the tumor. Tissue analysis confirmed the tumors had undergone ferroptosis. When the researchers tested Brusatol across four different CRPC organoid subtypes (SCL, neuroendocrine, Wnt-driven, and androgen receptor-driven), the SCL subtype was by far the most sensitive.

The Core Mechanism in Plain English CD44hi stem-like cells hoard iron to stay aggressive. They need NRF2 to keep all that iron from killing them. Block NRF2 with Brusatol, and their iron stores become a lethal weapon against themselves — the cells essentially rust to death.

Clinical Relevance: What Patient Data Show

The researchers didn't stop with laboratory models. They validated their findings using two large real-world patient datasets: The Cancer Genome Atlas (TCGA), which includes RNA profiling of 418 prostate cancer patients, and the DKFZ prostate cancer cohort, which includes 242 patients.

Across both datasets, patients with tumors that scored high on what the team called a "FerroScore" — a combined measure of CD44, iron storage proteins (ferritin light and heavy chains), iron transporter (SLC11A2), the demethylase KDM3A, and NRF2 (NFE2L2) — had significantly shorter progression-free survival and higher rates of biochemical recurrence (PSA rising after treatment). The statistical separation was highly significant (p < 0.0001 in both cohorts), suggesting this iron-stem cell axis is not just a laboratory curiosity but a clinically meaningful driver of aggressive disease.

Placing This Research in Context: The Growing Field of Ferroptosis in Prostate Cancer

The Bern/McGill study does not stand alone. It joins a rapidly growing body of research exploring ferroptosis as a therapeutic vulnerability in prostate cancer, particularly in drug-resistant forms.

A January 2026 review in Prostate Cancer and Prostatic Diseases by Chen, Lyu, and Gao catalogued multiple ferroptosis-inducing agents — including erastin, RSL3, and FIN56 — and noted that regulators of the PI3K/AKT/mTOR pathway and the enzyme DECR1 play significant roles in determining whether CRPC cells are susceptible to ferroptosis-based killing. That review concluded that ferroptosis represents a legitimate new therapeutic direction for CRPC, and that ferroptosis-related gene signatures may serve as prognostic biomarkers.

Separate research published in January 2026 in Advanced Science identified a completely different mechanism by which CRPC cells protect themselves from ferroptosis: through a protein called ZDHHC2, which destabilizes a lipid peroxidation enzyme (ACSL4) and thereby suppresses ferroptotic death. Researchers developed a small-molecule ZDHHC2 inhibitor (TTZ1) that restored ACSL4 function and reactivated ferroptosis, reversing enzalutamide resistance in cell lines and patient-derived tumor models. This convergent finding — a second entirely separate group independently arriving at ferroptosis reactivation as a way to overcome drug resistance in CRPC — adds credibility to the overall concept.

A January 2026 study in Cellular & Molecular Biology Letters found that the compound ilicicolin A (ili-A) sensitizes enzalutamide-resistant CRPC cells to ferroptosis by blocking the RORC/GPX4 signaling axis. GPX4 is one of the cell's primary defenders against lipid peroxidation; RORC is a transcription factor that activates it. This represents yet another ferroptosis-related entry point for overcoming resistance.

In 2025, a study in the Journal of Experimental & Clinical Cancer Research developed a PSMA-targeted nanoparticle system (GUL@LsiYY1@MZ) that used the prostate-specific surface antigen (the same target used by Lutetium-177 PSMA therapy) to deliver iron-containing nanoparticles and gene-silencing agents directly to prostate cancer cells, triggering ferroptosis with high spatial precision. This convergence of precision targeting (PSMA) with ferroptosis induction represents a potentially powerful future combination strategy.

A comprehensive review published in Nature Reviews Drug Discovery in 2025 by Zhang and colleagues reviewed thirty years of NRF2 research, concluding that NRF2 represents a major therapeutic opportunity across many cancer types, but that achieving clinical translation requires better tools for patient selection, more selective inhibitors, and improved delivery systems.

Good News Context: Multiple independent research groups, using entirely different approaches and different cell types, are converging on the same conclusion: ferroptosis is a real vulnerability in drug-resistant prostate cancer, and tools to exploit it are advancing. This is a hallmark of a maturing scientific concept, not a single laboratory finding.

What About Brusatol Itself? Limitations and Cautions

Brusatol is not currently approved for any clinical use in cancer. While it has shown potent anti-tumor activity across a range of laboratory cancer models — including leukemia, lung cancer, pancreatic cancer, colorectal cancer, and now CRPC — it faces several challenges before it could be used in human patients.

First, Brusatol is not perfectly selective for NRF2. A 2019 review in Cell & Bioscience noted that while NRF2 inhibition is the primary mechanism, Brusatol may also inhibit protein synthesis more broadly and suppress the oncogene c-Myc, and a bioinformatics analysis predicted that up to 464 proteins could theoretically be targeted by the compound. This "off-target" activity could cause side effects in normal tissues. Early experimental observations have noted potential toxicities including hypotension, nausea, and vomiting.

Second, a 2025 study in Cancers (MDPI) found, to researchers' surprise, that Brusatol actually promoted cell growth in one specific bone cancer (osteosarcoma) cell line, underscoring that NRF2 has both protective and growth-promoting roles in some tissues. This context-dependent effect is a caution that cancer-type specificity matters enormously — what works in CRPC-SCL cells may not be universally applicable.

Third, the current study was conducted entirely in cell lines and mouse models. While the findings are compelling and mechanistically rigorous, human biology is significantly more complex. Drug delivery, toxicity at therapeutic doses, and immune system interactions all require extensive clinical validation.

Important Caution for Patients: Brusatol is a research compound, not an approved drug. It is not available for patient use and should not be self-administered in any form. Patients interested in novel therapeutic approaches should ask their oncologist about clinical trials targeting ferroptosis pathways or NRF2 in CRPC.

The Broader Picture: Understanding Your CRPC Subtype

One of the most important takeaways from this body of research is that CRPC is not a single disease. It includes at least four molecularly distinct subtypes — androgen receptor (AR)-driven, Wnt-driven, neuroendocrine (NE), and stem cell-like (SCL) — and these subtypes likely require different treatment strategies. This matters for you as a patient because not all treatments will work equally well across all subtypes.

Currently, molecular subtyping of CRPC is primarily a research tool, but as the evidence base builds, it may eventually be used clinically to guide treatment selection. If a test can determine that a patient's tumor has the SCL subtype — characterized by high CD44 and elevated iron metabolism signatures — that patient might one day be triaged toward ferroptosis-targeting therapies rather than (or in addition to) the conventional second-line hormonal agents that may work less well for that subtype.

This is the direction toward which precision oncology in prostate cancer is pointing, and the new research is a meaningful contribution to that trajectory.

Questions to Ask Your Oncologist

If you are dealing with castration-resistant prostate cancer, here are some questions this research might prompt you to explore with your care team:

1. Has my tumor been molecularly subtyped? Most community oncology practices do not currently perform CRPC subtyping, but some academic medical centers and research programs do. Comprehensive genomic profiling (e.g., through Foundation Medicine, Tempus, or Caris Life Sciences) may offer some insights into whether your tumor has stem-like or neuroendocrine features.

2. Are there clinical trials targeting ferroptosis or cancer stem cell markers in CRPC that I might be eligible for? The ClinicalTrials.gov database is the best resource for finding open trials. Search terms such as "ferroptosis prostate cancer," "NRF2 inhibitor prostate cancer," or "CD44 prostate cancer" may yield relevant results.

3. Is there any role for assessing my CD44 expression or iron metabolism in the context of my current treatment? These are research questions at this point, but some academic centers conducting prostate cancer research trials may have protocols that include such profiling.

Summary

A February 2026 pre-clinical study from the University of Bern and McGill University has mapped a new therapeutic vulnerability in the stem cell-like subtype of castration-resistant prostate cancer (CRPC-SCL). The key discovery is that the most aggressive cells in this subtype — those expressing high levels of CD44 — depend on elevated intracellular iron to maintain their stem-like identity and tumor-initiating capacity. Iron does this by activating an enzyme (KDM3A) that chemically removes gene-silencing marks from the CD44 gene, creating a self-reinforcing loop that locks cells in a dangerous state. These same iron-loaded cells upregulate a protective protein (NRF2) to avoid being killed by their own iron excess. When NRF2 is inhibited by the experimental compound Brusatol, the cell's iron stores become toxic, triggering ferroptosis — an iron-driven form of cell death — selectively in the most dangerous cells. In mouse models, this approach shrank CRPC-SCL tumors significantly and reduced the stem-like cell population within those tumors. Patient genomic data from two large cohorts confirmed that high expression of iron-related and CD44-related genes predicts worse outcomes. These findings remain pre-clinical, but they add a compelling chapter to a rapidly maturing field — and give researchers a concrete mechanistic target for one of the most challenging forms of prostate cancer.

Key Terms Explained

CRPC (Castration-Resistant Prostate Cancer)

Prostate cancer that continues to grow despite hormone therapy (ADT) that reduces testosterone to very low levels.

CRPC-SCL (Stem Cell-Like Subtype)

One of four molecularly defined CRPC subtypes, characterized by stem cell features and poor responsiveness to standard hormonal therapy. Comprises roughly 25% of CRPC cases.

CD44

A cell-surface protein (glycoprotein) that serves as a marker of cancer stem cells in prostate cancer. Cells with high CD44 ("CD44hi") are more aggressive and drug-resistant.

Ferroptosis

A form of programmed cell death driven by iron-catalyzed damage (oxidation) of lipids in the cell membrane. Different from apoptosis (the more commonly known form of cell death). The name comes from "ferrum," Latin for iron.

NRF2 (Nuclear Factor Erythroid 2-Related Factor 2)

A master transcription factor that controls the cell's antioxidant defenses and iron storage. In cancer cells, NRF2 is often overactive, protecting cancer cells from oxidative stress and ferroptosis.

Brusatol

A natural compound extracted from the plant Brucea sumatrana that inhibits NRF2. Experimental only — not an approved drug.

KDM3A

A histone demethylase (enzyme that removes silencing marks from genes), activated by iron. In CRPC-SCL cells, it helps maintain CD44 expression.

H3K9me2

A chemical "silencing tag" on histone proteins (the protein spools around which DNA is wrapped). When this mark is present at the CD44 gene, CD44 expression is suppressed. Iron-activated KDM3A removes this mark, allowing CD44 to be expressed.

Labile Iron Pool (LIP)

The pool of "free" reactive iron within the cell — not safely stored in ferritin. High LIP drives lipid peroxidation and ferroptosis.

Ferritin

A protein complex that safely stores iron inside cells. NRF2 helps maintain ferritin levels; when NRF2 is inhibited, ferritin falls and free iron rises.

FerroScore

A research scoring tool developed by the study authors, combining expression levels of six genes (CD44, FTL, FTH1, SLC11A2, KDM3A, NRF2/NFE2L2) to predict clinical outcomes in prostate cancer patients.

Verified Sources and Formal Citations

1. Primary Study (February 2026 — BioRxiv Preprint):
   Cheng W, Brunello A, Bonollo F, Marti TM, Chouvardas P, Labbé DP, De Menna M, Thalmann G, Karkampouna S, Kruithof-de Julio M. "Iron Metabolism as a Therapeutic Vulnerability in Stem Cell-Like Castration-Resistant Prostate Cancer." bioRxiv, posted February 25, 2026. doi: https://doi.org/10.64898/2026.02.24.707686
2. CRPC Molecular Subtypes (Foundational Classification — 2022):
   Tang F, Xu D, Wang S, Wong CK, et al. "Chromatin profiles classify castration-resistant prostate cancers suggesting therapeutic targets." Science. 2022 May. doi: 10.1126/science.abf8591. https://www.science.org/doi/10.1126/science.abf8591
3. Ferroptosis in CRPC — 2026 Review:
   Chen H, Lyu F, Gao X. "Advances in ferroptosis for castration-resistant prostate cancer treatment: novel drug targets and combination therapy strategies." Prostate Cancer and Prostatic Diseases 29, 36–46 (2026). doi: 10.1038/s41391-024-00933-w. https://www.nature.com/articles/s41391-024-00933-w
4. ZDHHC2-ACSL4 Ferroptosis Axis in CRPC Drug Resistance — January 2026:
   Shao S, Li W, Hong Y, et al. "ZDHHC2-Dependent Palmitoylation Dictates Ferroptosis and Castration Sensitivity in Prostate Cancer via Controlling ACSL4 Degradation and Lipid Peroxidation." Advanced Science 13(1):e14077 (2026). doi: 10.1002/advs.202514077. https://pubmed.ncbi.nlm.nih.gov/41126755/
5. RORC/GPX4 Ferroptosis and Enzalutamide Resistance — January 2026:
   "Inhibition of the RORC/GPX4 mediated ferroptosis regulatory axis suppresses tumor growth and alleviates enzalutamide resistance in prostate cancer." Cellular & Molecular Biology Letters. January 24, 2026. doi: 10.1186/s11658-025-00846-z. https://link.springer.com/article/10.1186/s11658-025-00846-z
6. PSMA-Targeted Ferroptosis Nanoplatform — 2025:
   Zhang L, Sun Q, Zheng D, et al. "PSMA-targeted theranostic nanoplatform achieves spatiotemporally precise therapy and triggers ferroptosis in prostate cancer treatment." Journal of Experimental & Clinical Cancer Research 44, 272 (2025). doi: 10.1186/s13046-025-03530-4. https://link.springer.com/article/10.1186/s13046-025-03530-4
7. Ferroptosis in Prostate Cancer — Overview Review:
   Wang Y, Ma Y, Jiang K. "The role of ferroptosis in prostate cancer: a novel therapeutic strategy." Prostate Cancer and Prostatic Diseases 26, 25–29 (2023). doi: 10.1038/s41391-022-00583-w. https://www.nature.com/articles/s41391-022-00583-w
8. Ferroptosis in Advanced Prostate Cancer — Comprehensive Overview:
   "Unlocking ferroptosis in prostate cancer — the road to novel therapies and imaging markers." PMC / Nature Reviews Urology (2025). https://pmc.ncbi.nlm.nih.gov/articles/PMC12067944/
9. Ferroptosis in Prostate Cancer — Mechanisms and Multidisciplinary Strategies:
   "Targeting ferroptosis in prostate cancer management: molecular mechanisms, multidisciplinary strategies and translational perspectives." Journal of Translational Medicine. February 7, 2025. doi: 10.1186/s12967-025-06180-4. https://translational-medicine.biomedcentral.com/articles/10.1186/s12967-025-06180-4
10. Ferroptosis in CRPC — JCI Commentary:
    "Ferroptosis: the vulnerability within a cancer monster." Journal of Clinical Investigation. 2023. https://www.jci.org/articles/view/170027
11. Brusatol as NRF2 Inhibitor — Overview:
    Cai SJ, Liu Y, Han S, et al. "Brusatol, an NRF2 inhibitor for future cancer therapeutic." Cell & Bioscience 9, 45 (2019). doi: 10.1186/s13578-019-0309-8. https://cellandbioscience.biomedcentral.com/articles/10.1186/s13578-019-0309-8 | PMC: https://pmc.ncbi.nlm.nih.gov/articles/PMC6554866/
12. NRF2 Caution — Brusatol and Osteosarcoma:
    Stephens E, Greenhough A, Mansell JP. "The Nrf2 Inhibitor Brusatol Promotes Human Osteosarcoma (MG63) Growth and Blocks EB1089-Induced Differentiation." International Journal of Molecular Sciences (MDPI). Published October 3, 2025. doi: 10.3390/ijms26199675. https://www.mdpi.com/1422-0067/26/19/9675
13. NRF2 — Thirty Years of Advances:
    Zhang DD. "Thirty years of NRF2: advances and therapeutic challenges." Nature Reviews Drug Discovery 24, 421–444 (2025). doi: 10.1038/s41573-025-01106-z. https://www.nature.com/articles/s41573-025-01106-z
14. CD44 Mediates Iron Endocytosis — Key Mechanistic Paper:
    Müller S, Sindikubwabo F, Cañeque T, et al. "CD44 regulates epigenetic plasticity by mediating iron endocytosis." Nature Chemistry 12, 929–938 (2020). doi: 10.1038/s41557-020-0513-5. https://www.nature.com/articles/s41557-020-0513-5
15. NRF2 Controls Iron and Ferroptosis via HERC2/VAMP8:
    Anandhan A, et al. "NRF2 controls iron homeostasis and ferroptosis through HERC2 and VAMP8." Science Advances 9, eade9585 (2023). doi: 10.1126/sciadv.ade9585. https://www.science.org/doi/10.1126/sciadv.ade9585
16. CD44 in Cancer — Role and Therapeutic Implications:
    Chen C, Zhao S, Karnad A, Freeman JW. "The biology and role of CD44 in cancer progression: therapeutic implications." Journal of Hematology & Oncology 11(1):64 (2018). doi: 10.1186/s13045-018-0605-5. https://jhoonline.biomedcentral.com/articles/10.1186/s13045-018-0605-5
17. Cancer Stem Cells in CRPC and NEPC — 2025 Review:
    "Mechanisms of neuroendocrine differentiation in castration-resistant prostate cancer." Annals of Medicine. Published online October 1, 2025. doi: 10.1080/07853890.2025.2548975. https://www.tandfonline.com/doi/full/10.1080/07853890.2025.2548975
18. Prostate Cancer Stem Cells — Targeted Therapies Review:
    "Prostate cancer stem cells and their targeted therapies." Frontiers in Cell and Developmental Biology. Published December 2025. doi: 10.3389/fcell.2024.1410102. https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2024.1410102/full
19. Natural NRF2 Inhibitors for Cancer:
    "Natural Nrf2 Inhibitors: A Review of Their Potential for Cancer Treatment." PMC / Molecules (2023). https://pmc.ncbi.nlm.nih.gov/articles/PMC10321279/
20. Ferroptosis Inducers as Therapeutics for Advanced Prostate Cancer:
    Ghoochani A, Hsu EC, Aslan M, et al. "Ferroptosis inducers are a novel therapeutic approach for advanced prostate cancer." Cancer Research 81, 1583–1594 (2021). doi: 10.1158/0008-5472.CAN-20-1868. https://pubmed.ncbi.nlm.nih.gov/33234511/