(© hin255 - stock.adobe.com)
Lab Study Finds Hidden Brake on Common Breast Cancer Type
In A Nutshell
- Cutting about 200 DNA letters erased a hormone’s growth-slowing effect on breast cancer cells in the lab.
- A gene long credited with that effect barely mattered when silenced or boosted.
- Edited cells grew faster, in both breast and prostate cancer cell lines.
- All results come from lab-grown cells, not patients.
Cutting out a sliver of DNA about 200 letters long erased a hormone’s ability to slow breast cancer cells in a lab dish, and the edited cells grew faster than normal ones. Meanwhile, the gene that sliver sits inside, long suspected of doing the work, barely mattered when scientists silenced it or pushed cells to make extra of its protein.
Estrogen-fueled tumors make up roughly 70 to 80 percent of breast cancers. Researchers already knew that the androgen receptor, a protein that responds to male hormones such as testosterone, can put the brakes on those tumors. How it does so has stayed murky. A gene called ZBTB16 was the leading suspect, so researchers at Adelaide University in Australia put it to the test. Their results, published in npj Breast Cancer, point to something hiding inside that gene instead: an enhancer, a stretch of DNA that works like a dimmer switch for genes.
Patients have a stake in the answer. A phase 2 trial of enobosarm, a drug that activates the androgen receptor, reported clinical benefit at 24 weeks in 32 percent of patients with a certain type of advanced breast cancer. Dr. Amy Dwyer, the study’s corresponding author at Adelaide University, said the work could lead to “better ways to identify and treat patients” whose advanced disease has stopped responding to hormone therapies.
Breast Cancer Tissue Holds Far Less of the Suspect Gene
Researchers started with public databases that pool genetic data from tumor and healthy tissue. ZBTB16 activity ran about threefold lower in breast tumors overall than in healthy breast tissue, and the drop showed up in all four major subtypes, a pattern typical of a tumor suppressor, a gene that normally holds cell growth in check. Among roughly 720 patients with estrogen-fueled tumors, higher ZBTB16 activity went along with longer survival. That link alone does not prove the gene protects anyone, so the team moved to the lab.
Experiments on three breast cancer cell lines, all of which stop growing when the androgen receptor is switched on, showed that a potent form of testosterone raised ZBTB16 protein levels in two of them. Scanning the DNA inside the gene, the team then found a spot where the androgen receptor latches on. It turned up in cell lines, patient-derived tumors grown in mice, and published data from breast tumor biopsies, and it carried the chemical hallmarks of an enhancer. The researchers named it HRRINT3. Most experiments in the study were repeated three times, a standard guard against flukes.
Deleting the DNA Switch Erased a Hormone’s Effect on Breast Cancer Cells
To test whether the spot mattered, scientists used CRISPR, a gene-editing tool, to snip about 200 letters out of HRRINT3 in a widely used breast cancer cell line, and confirmed the edit worked when the androgen receptor no longer attached at that spot. Without the switch, the hormone could no longer turn up ZBTB16 properly. Growth told the bigger story. In unedited cells, the hormone trimmed estrogen-fueled growth by 38 percent over a week. Across six independently engineered versions of the edited cells, that slowdown was no longer detectable.
A related hormone, progesterone, still slowed growth in the edited cells. The edit therefore did not shut down hormone signaling across the board, and the switch appears tied specifically to the androgen receptor’s brake.
Silencing the Gene Did Not Change the Hormone’s Effect
If the gene were the real tumor suppressor, removing it should blunt the hormone’s effect, as removing the switch did. It did not. When researchers silenced ZBTB16 in two breast cancer cell lines, using short strands of genetic material that turn a gene down, the hormone still slowed estrogen-driven growth, by about 30 percent in one line with or without the gene. Cells engineered to make extra ZBTB16 in two lines showed no added slowdown either, though one line already responded so strongly that a small gain would have been hard to spot.
Cells Missing the Switch Grew Faster in Breast and Prostate Lines
Even with no hormone added, cells missing the switch behaved differently. More than 7,000 genes changed activity, including many that drive cell division. That is a big ripple from such a small cut, and the authors say the effects likely reach beyond the androgen receptor’s own target genes. The same edit also sped up growth in three more cell lines: another estrogen-fueled breast line, a breast cancer line without the estrogen receptor, and a prostate cancer line.
Edited cells still made the androgen receptor, and the hormone still sent it into the nucleus, the cell’s control center. Yet some of its usual target genes responded weakly, including genes tied to better outcomes in estrogen-fueled breast cancer. The authors propose the switch may serve as a gathering spot for helper proteins, an idea that remains to be tested.
A gene with a strong reputation barely affected the result, while a 200-letter stretch of DNA inside it proved critical to the hormone’s growth-slowing effect in these cells. Treatments that harness the androgen receptor against breast cancer may need to account for that switch, a possibility that so far rests only on lab-grown cells.
Disclaimer: This summary of a published study is for general information, not medical advice. Lab results may not apply to people. Anyone with health concerns should consult a doctor.
Paper Notes
Limitations
All experiments used lab-grown cell lines, five in total (four breast cancer and one prostate cancer), so none of the results show what the switch does in patients’ tumors, and the study does not demonstrate that manipulating it would slow breast cancer in people. The CAMA-1 growth tests used two replicates instead of the usual three. The edited CAMA-1, MFM-223, and LNCaP groups were mixed groups of cells rather than lines grown from a single edited cell. In the extra-gene experiments, the authors note that a “ceiling effect” could not be ruled out, since the unmodified ZR-75-1 cells already responded strongly to the hormone treatment. The patient-database findings show associations, not cause and effect, and the paper does not report demographic details for those patients. The authors also say the exact way the switch works, including whether it shapes the local DNA environment or acts through another route, remains unresolved and will need further study, such as methods that map how DNA is packed and folded.
Funding and Disclosures
This work was supported by the National Health and Medical Research Council of Australia, Cancer Australia, and the National Breast Cancer Foundation, with facility and sequencing support from Australian research infrastructure programs. The authors declare no competing interests.
Publication Details
Titled “A functional role for an intronic enhancer of ZBTB16 in AR-mediated tumour suppression in ER+ breast cancer,” the study was written by Maliha Wajahat, Dawn Gek Lian Thean, Richard Iggo, Anzu Okada, Zeyad D. Nassar, Wayne D. Tilley, Theresa E. Hickey, and Amy R. Dwyer, who is the corresponding author. It was published online August 25, 2026, in npj Breast Cancer, a journal published in partnership with the Breast Cancer Research Foundation, after being received April 29, 2026, and accepted August 14, 2026. The version online is an early “Article in Press” release that the final edited version will replace. DOI: https://doi.org/10.1038/s41523-026-01037-5. The authors are affiliated with the Dame Roma Mitchell Cancer Research Laboratories at Adelaide University; INSERM and the University of Bordeaux in France; and the South Australian Health and Medical Research Institute and the South Australian Immunogenomics Cancer Institute in Adelaide. RNA sequencing data are available through the Gene Expression Omnibus under accession GSE322909.
