Two Sides of a Shape-Shifting Cancer:

Scientists find a new weakness in treatment-resistant prostate cancer | ScienceDaily

IPCSG  Informed Prostate Cancer Support Group

Research Update - Treatment Resistance · Precision Medicine

A New Drug-Pairing Strategy for Treatment-Resistant Prostate Cancer

When prostate cancer changes its identity to dodge hormone therapy, one drug may not be enough. A University of Michigan team makes the case for hitting the disguise from both directions at once.

Prepared for the IPCSG Newsletter  •  September 2026

Bottom Line Up Front

Some advanced prostate cancers escape hormone-blocking drugs by physically changing what kind of cell they are — a process called transdifferentiation or lineage plasticity. Researchers at the University of Michigan Rogel Cancer Center report that this escape has two moving parts, and that pairing two classes of “epigenetic” drugs — a BET bromodomain inhibitor and a DNMT inhibitor — blocked both parts and slowed tumor growth far better than either drug alone in laboratory cells and in mice. The work is promising but still preclinical: no patients were treated in this study, and a human trial of the specific pairing has not yet begun. It arrives, however, inside a fast-moving field where related drugs are already in clinical testing, giving patients real reason for measured optimism.

Why resistance is the central problem

About one in eight men will be diagnosed with prostate cancer in his lifetime, and in the United States the disease remains the second-leading cause of cancer death among men. Most prostate tumors begin life looking and behaving like the gland they came from, and their growth depends on male hormones (androgens) such as testosterone. That dependence is the reason hormone-blocking drugs — the androgen-receptor inhibitors abiraterone, enzalutamide, apalutamide and darolutamide — are a mainstay of treatment for advanced disease.

These drugs often work well at first. The hard truth is that nearly every patient eventually develops resistance. One of the ways cancer resists is not by mutating a single gene, but by reinventing itself: the tumor stops depending on the androgen receptor and adopts a different cellular identity altogether. In its most aggressive form this is called neuroendocrine prostate cancer (NEPC), a small-cell-type disease that no longer makes much PSA, spreads quickly to organs, and responds poorly to standard therapy.

Transdifferentiation / lineage plasticity

A cancer cell changing its identity — shedding its original “glandular” program and switching on a different one — to survive treatment.

Neuroendocrine prostate cancer (NEPC)

An aggressive, hormone-independent form the cancer can transform into. Rare at first diagnosis (under ~2% of cases), it emerges in an estimated 10–20% of men whose disease becomes castration-resistant.

Epigenetics

The system of chemical “switches” that turn genes on or off without changing the underlying DNA code. Because switches can be flipped back, they are attractive drug targets.

What the Michigan team discovered

Earlier work had tied the transformation to the loss of two well-known tumor-suppressor genes, TP53 and RB1. What had stayed murky was why losing them causes such a dramatic change in a cell’s identity. To find out, the group led by Joshi Alumkal, MD, examined prostate cancer cell lines and mapped which biological programs shifted when TP53 and RB1 were switched off.

The key insight is that the escape has two sides. As Dr. Alumkal put it, the cells lose their glandular gene programs while simultaneously turning on stem-cell-like identity programs. In other words, the tumor both erases its old self and activates a new, more primitive and adaptable one. A therapy that addresses only one side leaves the other side free to keep driving the cancer.

Two drug classes for two problems

Alumkal’s lab had previously shown that BET bromodomain inhibitors can interfere with the pathways that let prostate cancer switch on those alternate identity programs. In the new experiments the BET inhibitor again slowed the growth of cancer cells — but it did not kill them. Blocking the “new self” alone was not enough.

So the team added a second, complementary drug class: DNMT inhibitors (DNA methyltransferase inhibitors). These drugs can un-silence genes that a cancer has switched off. The team was especially interested in restoring the glandular genes that get shut down as the cell changes identity — in effect, coaxing the “old self” back on. Notably, DNMT inhibitors are not exotic: two of them (azacitidine and decitabine) have been FDA-approved for years to treat blood cancers such as myelodysplastic syndromes and acute myeloid leukemia.

When the researchers combined the two drugs, the pairing suppressed growth of transdifferentiated prostate cancer cell lines more effectively than either agent used by itself, and the same benefit held up in human tumors implanted in mice. The combination reversed a meaningful share of the gene-expression changes that accompany the transformation. Encouragingly, the effect appeared even at doses well below the usual recommended levels, and the combination was well tolerated by the animals — an early hint that a tolerable human dose might be achievable. The findings were published in JCI Insight in August 2026.

An honest caveat

This is a laboratory and animal study. It shows a mechanism and a rational strategy — not a proven treatment. Results in cell lines and mice frequently fail to translate to people, and the specific BET-plus-DNMT combination has not yet been tested in a prostate cancer trial. The authors’ own next steps make this clear: they want to find biomarkers that identify which patients would benefit, understand which genes drive the effect, and ideally learn to prevent the transformation before it happens rather than treat it afterward. In Dr. Alumkal’s framing, heading off the transformation early could be key to survival.

How this fits the wider research landscape

What makes this study worth a patient’s attention is that it is not an isolated idea. Lineage plasticity has become one of the most active fronts in prostate cancer research, and several related approaches are already further along.

The BET inhibitor is already in the clinic

The BET inhibitor most associated with this line of work, ZEN-3694 (from Zenith Epigenetics), has been studied in men with metastatic castration-resistant prostate cancer for several years. An early Phase 1b/2a trial pairing ZEN-3694 with enzalutamide in 75 patients found the combination reasonably tolerable, with a median radiographic progression-free survival of about nine months; importantly, men whose tumors already had low androgen-receptor activity — a signature of the plastic, resistance-prone state — tended to benefit longer. A later Phase 2 study (NCT04471974) explicitly built in a cohort of men with clinically or genomically defined transdifferentiated disease, testing ZEN-3694 alongside enzalutamide and the immunotherapy pembrolizumab, and a larger randomized Phase 2b trial has been undertaken with Zenith’s partner. In 2025 the FDA granted ZEN-3694 Fast Track and Orphan Drug designations — though, importantly, those designations were for a different rare cancer (NUT carcinoma), not prostate cancer. The prostate program continues as investigational.

Other ways to attack a shape-shifting tumor

Researchers are pursuing the same problem from several angles. One promising route bypasses the identity question entirely and simply targets a protein called DLL3 that sits on the surface of many neuroendocrine tumor cells. Two “T-cell engager” drugs that harness the immune system against DLL3 — tarlatamab and an investigational agent known as MK-6070 (gocatamig) — have shown early signs of activity in neuroendocrine prostate cancer, with response rates that have drawn genuine interest at recent oncology meetings. Others are probing epigenetic regulators such as EZH2 and transcription factors like ASCL1, SOX2 and FOXA2 that help enforce the neuroendocrine state.

A cautionary footnote from the literature

The value of independent replication was underscored this year. In late 2025 a high-profile study in Nature reported that inhibiting a different epigenetic enzyme, NSD2, could reverse lineage plasticity and re-sensitize prostate tumors to hormone therapy in the lab — an exciting result widely covered at the time. That paper was retracted at the authors’ request in June 2026. The underlying biological idea — that plasticity may be reversible — remains actively studied, but the episode is a useful reminder that even prominent findings must be confirmed before patients should count on them.

What this means for you

• It reinforces the case for genomic testing. The transformation is tied to loss of TP53 and RB1 (often together with PTEN). Knowing your tumor’s genetic profile — through tissue biopsy or, increasingly, blood-based tests — can help you and your oncologist anticipate aggressive resistance and consider a repeat biopsy if the disease behaves unexpectedly (for example, rapid progression with a low or falling PSA).
• “Low-PSA progression” deserves attention. Because neuroendocrine transformation can advance while PSA stays low, worsening scans or new symptoms in the face of a reassuring PSA are worth raising with your care team.
• Trials are where these ideas are tested. If you have castration-resistant disease that has stopped responding to androgen-receptor inhibitors, ask whether clinical trials targeting lineage plasticity, NEPC, or DLL3 are appropriate for you. ClinicalTrials.gov and your treatment center’s trial office are good starting points.
• Temper timelines, keep the optimism. A preclinical study like this one typically takes years to reach — if ever — a proven human therapy. But the convergence of multiple independent groups on the same problem, using drugs that in some cases already exist, is exactly the pattern that precedes real clinical progress.

Verified sources

1. Storck WK, Flores D, Kumaraswamy A, et al. Combined BET bromodomain and DNMT inhibition targets critical survival pathways in transdifferentiated prostate cancer. JCI Insight. 2026 Aug 11 (online ahead of print). doi:10.1172/jci.insight.207543. PMID 42579341. — https://pubmed.ncbi.nlm.nih.gov/42579341/  |  https://doi.org/10.1172/jci.insight.207543
2. Michigan Medicine – University of Michigan. New drug combination may help treat advanced prostate cancer. Michigan Health Lab, 2026. — https://www.michiganmedicine.org/health-lab/new-drug-combination-may-help-treat-advanced-prostate-cancer
3. Michigan Medicine – University of Michigan. Scientists find a new weakness in treatment-resistant prostate cancer. ScienceDaily, 8 September 2026. — https://www.sciencedaily.com/releases/2026/09/260904000324.htm
4. Kregel S, Chen JL, Tom W, ... Alumkal JJ, et al. BET Bromodomain Inhibition Blocks an AR-Repressed, E2F1-Activated Treatment-Emergent Neuroendocrine Prostate Cancer Lineage Plasticity Program. Clin Cancer Res. 2021;27(17):4923–4936. doi:10.1158/1078-0432.CCR-20-4968. PMID 34145028. — https://pubmed.ncbi.nlm.nih.gov/34145028/
5. Aggarwal RR, Schweizer MT, Nanus DM, et al. A Phase Ib/IIa Study of the Pan-BET Inhibitor ZEN-3694 in Combination with Enzalutamide in Patients with Metastatic Castration-resistant Prostate Cancer. Clin Cancer Res. 2020;26(20):5338–5347. doi:10.1158/1078-0432.CCR-20-1707. PMID 32694156. — https://pubmed.ncbi.nlm.nih.gov/32694156/
6. U.S. National Library of Medicine. ZEN-3694, Enzalutamide, and Pembrolizumab for Metastatic Castration-Resistant Prostate Cancer (NCT04471974). ClinicalTrials.gov. — https://clinicaltrials.gov/study/NCT04471974
7. Zenith Epigenetics Ltd. FDA grants Zenith’s ZEN-3694 Fast Track status (14 July 2025) and ZEN-3694 receives Orphan Drug designation from FDA (27 October 2025). Company news releases. — https://www.biospace.com/press-releases/fda-grants-zeniths-zen-3694-fast-track-status  |  https://www.zenithepigenetics.com/programs/pipeline
8. Li JJ, Vasciaveo A, Karagiannis D, et al. NSD2 targeting reverses plasticity and drug resistance in prostate cancer. Nature. 2026;649(8095):216–226. doi:10.1038/s41586-025-09727-z. Retracted: Nature. 2026;656(8128):780. doi:10.1038/s41586-026-10816-w. — https://www.nature.com/articles/s41586-026-10816-w (retraction note)
9. UroToday. APCCC 2026: Neuroendocrine Prostate Cancer — Treatment Options in 2026 and Hope for New Treatments? (DLL3 T-cell engagers tarlatamab and MK-6070/gocatamig; emerging targets). — https://www.urotoday.com/conference-highlights/apccc-2026/168671-...
10. Niu Y, Zhang S, Zhang S, Tao J, Zhu Y. Mechanistic insights and molecular therapy for neuroendocrine prostate cancer. Front Oncol. 2026;16:1806742. doi:10.3389/fonc.2026.1806742. PMID 42100391. — https://pubmed.ncbi.nlm.nih.gov/42100391/
11. Cai M, Zheng F, Ren YZ, et al. New Insights into Potential Therapeutic Targets for Neuroendocrine Prostate Cancer: From Bench to Clinic. Research (Wash DC). 2025;8:0791. doi:10.34133/research.0791. PMID 40746825. — https://pubmed.ncbi.nlm.nih.gov/40746825/
12. Gjyrezi A, et al. Epidemiology and overall survival of de novo and transformed neuroendocrine prostate cancer: a two-nation population-based study of 1,465 patients (SEER and NCRAS). 2026. PMID 42573114. — https://pubmed.ncbi.nlm.nih.gov/42573114/
13. Gnanapragasam V, et al. (review). DNA methyltransferase inhibitors combination therapy for the treatment of solid tumors: mechanism and clinical application. Clin Epigenetics. 2021;13:166. doi:10.1186/s13148-021-01154-x. — https://clinicalepigeneticsjournal.biomedcentral.com/articles/10.1186/s13148-021-01154-x

This article is provided for education and patient advocacy. It is not medical advice and does not describe an available treatment. Discuss your individual situation, testing options, and any clinical trials with your own oncology team. Author names on the primary study are abbreviated in some citations; full author lists appear at the linked sources.