The PI3K/AKT/mTOR pathway sits at the heart of cellular growth control. When functioning normally, it relays signals from growth factor receptors on the cell surface to internal machinery that drives metabolism, protein synthesis, and proliferation. In many cancers, this circuit becomes permanently switched on, fueling uncontrolled division and survival. Pharmaceutical companies have developed a series of PI3K inhibitors to shut down this pathway, but clinical experience has repeatedly shown that tumors often possess intrinsic resistance mechanisms that blunt the drugs’ effectiveness from the very first dose. Understanding these mechanisms has become a priority for oncology research.

Compound 8 emerged from the team’s medicinal chemistry efforts as a novel fluorinated derivative of isoflavone, a class of natural plant compounds. In laboratory tests, the molecule showed robust anti-tumor activity against triple-negative breast cancer cells that express low levels of the androgen receptor, specifically the MDA-MB-231 cell line. However, when the researchers tested it against MDA-MB-453 cells, a triple-negative breast cancer line belonging to the luminal androgen receptor subtype that overexpresses the androgen receptor, the compound’s potency dropped dramatically. This disparity provided the investigators with a natural experiment: two cancer cell types, one vulnerable and one resistant, differing primarily in their androgen receptor abundance.

The androgen receptor, best known for its role in prostate cancer, is a nuclear transcription factor that binds testosterone and other androgens before migrating to DNA to activate specific gene programs. The luminal androgen receptor subtype of triple-negative breast cancer is characterized by high levels of this receptor, which has made anti-androgen therapy an attractive therapeutic concept for these tumors. The new study reveals that the receptor does far more than simply mark a tumor subtype. It actively orchestrates resistance to PI3K inhibition by blocking two of the key processes that compound 8 relies upon to kill cancer cells: autophagy-associated cell death and cell cycle arrest.

Autophagy is a cellular recycling system in which the cell digests its own components to survive stress. In cancer therapy, excessive or dysregulated autophagy can tip over into a form of programmed cell death, and many anti-cancer drugs work partly by pushing cells across this threshold. Using transcriptomic analysis, the researchers discovered that high androgen receptor abundance upregulates a gene called UBE2C, which encodes an enzyme involved in tagging proteins for destruction during cell division. Elevated UBE2C levels performed a double duty that protected resistant cells. First, UBE2C sustained mitotic progression, allowing cancer cells to keep dividing even under drug pressure. Second, and perhaps more surprisingly, UBE2C maintained autophagy in an inhibited state, preventing compound 8 from unleashing the autophagy-associated cytotoxicity that would normally kill the cells.

To confirm that this mechanism was not an artifact of a single cell line, the team validated their findings in prostate cancer models. They used LNCaP cells, which express high levels of the androgen receptor, and PC-3 cells, which express low levels. The same pattern emerged: androgen receptor-high cells resisted compound 8, while androgen receptor-low cells remained sensitive. This consistency across two hormonally distinct cancer types suggests that the AR-UBE2C-PI3K resistance axis is a conserved feature rather than a peculiarity of breast cancer biology, broadening the potential clinical relevance of the discovery considerably.

The therapeutic implications became clear when the researchers combined compound 8 with strategies to suppress the androgen receptor. Genetic silencing of the receptor, or pharmacological inhibition using enzalutamide, a clinically approved anti-androgen drug, sensitized the resistant cells to compound 8. The combination restored the drug’s ability to induce cell cycle arrest and promoted the autophagy-associated cytotoxicity that the androgen receptor had been suppressing. In effect, blocking the androgen receptor removed the molecular brake that UBE2C had placed on the cancer cells’ self-destruction program, allowing the PI3K inhibitor to work as intended.

The team then moved from cell culture to living organisms. In mouse models of cancer, the combination therapy synergistically suppressed tumor growth and metastasis, outperforming either agent alone. All animal experiments were conducted under the approval of the Institutional Animal Care and Use Committee at West China Hospital, Sichuan University. These in vivo results provide compelling preclinical evidence that the drug combination strategy could translate into a meaningful clinical approach for patients whose tumors overexpress the androgen receptor.

The study was supported by funding from the National Natural Science Foundation of China, the Fundamental Research Funds for the Central Universities, the China Postdoctoral Science Foundation, the Natural Science Foundation of Sichuan Province, and several institutional programs. The corresponding authors are Wei Liu, Lan Zhang, and Xiaojun Man, whose teams brought together expertise in biomimetic drug synthesis, cancer biotherapy, and urology. The article is published open access under a Creative Commons Attribution 4.0 license, making the full dataset and methods available to researchers worldwide.

For patients with androgen receptor-overexpressing breast or prostate cancers, the findings offer a molecular explanation for why PI3K inhibitors have sometimes fallen short and a concrete rationale for combining them with anti-androgen therapy. Rather than abandoning a promising drug class in the face of resistance, oncologists may soon be able to disarm the resistance mechanism directly, converting a futile single-agent approach into a synergistic two-pronged attack. As clinical trials test this combination strategy, the AR-UBE2C axis may become a biomarker that identifies which patients stand to benefit most from this newly illuminated therapeutic partnership.

Subject of Research: Targeting the AR-UBE2C axis unleashes autophagy-associated cell death to overcome intrinsic resistance to PI3K Inhibition in AR-overexpressed breast and prostate cancers

Article Title: Targeting the AR-UBE2C axis unleashes autophagy-associated cell death to overcome intrinsic resistance to PI3K Inhibition in AR-overexpressed breast and prostate cancers

Article References: Fu, L., Li, H., Li, Z., Liu, W., Zhang, L., & Man, X. (2026). Targeting the AR-UBE2C axis unleashes autophagy-associated cell death to overcome intrinsic resistance to PI3K Inhibition in AR-overexpressed breast and prostate cancers. Cell Death Discovery. https://doi.org/10.1038/s41420-026-03389-0

Image Credits: AI Generated

DOI: 10.1038/s41420-026-03389-0

Keywords: Targeting, AR-UBE2C, axis, unleashes, autophagy-associated, cell, death, overcome, intrinsic, resistance, PI3K, Inhibition

The drug targets the phosphatidylinositol-4,5-bisphosphate 3-kinase, AKT, mechanistic target of rapamycin signaling cascade, a central growth and survival pathway that is frequently dysregulated in both canine and human cancers. When this pathway runs amok, it drives neoplastic cells to proliferate relentlessly, evade apoptosis, crank up metabolism, and, in the case of lymphoid malignancies, stimulate the production of vascular endothelial growth factor, a molecule that supports tumor neovascularization. Because single-node inhibitors of the pathway often deliver only brief or incomplete responses, likely due to well-known feedback loops within the signaling network, the researchers turned to VDC597, a compound that blocks PI3K along with both mTOR complexes, mTORC1 and mTORC2, thereby hitting the pathway at multiple points simultaneously.

In their experiments, the team tested VDC597 against two canine cell lines: 1771, derived from a spontaneously occurring B-cell lymphoma, and CLL1390, derived from chronic lymphocytic leukemia. Both lines had previously been documented to carry mutations in or constitutively express activated proteins of the PI3K-AKT-mTOR pathway, making them appropriate models for pathway-targeted therapy. Western blot analysis showed that a 24-hour treatment with the drug produced a clear, dose-dependent reduction in phosphorylated AKT at the serine 473 residue, the molecular signature of pathway activation. The leukemia line proved more sensitive, showing significant suppression at concentrations as low as 0.2 micromolar, while the lymphoma line required a full micromolar dose to achieve statistically significant inhibition.

Timing experiments revealed that the drug’s effects on signal transduction are both rapid and durable. Following a single 1 micromolar dose, phosphorylation of AKT dropped quickly and remained at roughly half of control levels for a full 24 hours. Downstream phosphorylation of 4EBP1, a translation-regulating target of mTORC1, fell even more sharply, with statistically significant suppression measurable from half an hour after treatment through the two-hour mark. The researchers verified these biochemical findings by immunohistochemistry on fixed cell pellets, observing reduced perinuclear and cytoplasmic staining for both phosphoproteins after drug exposure, an encouraging confirmation that the compound shuts down signaling across the entire pathway rather than just a single branch.

The functional consequences were equally striking. After 72 hours of exposure, VDC597 halved the viable cell population at concentrations within or near what is pharmacologically achievable, with mean half-maximal inhibitory concentrations of 1.41 micromolar for CLL1390 and 2.56 micromolar for 1771, values informed by an earlier phase I trial of the drug in tumor-bearing dogs. Live-cell imaging, tracking fluorescently labeled nuclei and dead-cell stains over 48 hours, showed that the drug did not merely slow proliferation: it actively killed the lymphoma cells, with cell death increasing in a dose-dependent manner and becoming significantly elevated at concentrations of 0.25 micromolar and above. Enzyme-linked immunosorbent assays further demonstrated that VDC597 suppressed vascular endothelial growth factor production in both cell lines, again in a dose-dependent fashion, suggesting the drug may undermine the angiogenic support system that lymphoid tumors rely upon.

Because combination therapy, not single agents, ultimately determines how well targeted drugs translate into clinical benefit, the team paired VDC597 with the three chemotherapy drugs most commonly used in canine lymphoma protocols: dexamethasone, doxorubicin, and vincristine. The results were revealing. Combining VDC597 with dexamethasone produced genuinely synergistic growth inhibition, quantified by combination indices below 1 using the Chou-Talalay method, with the effect strongest at lower drug concentrations. Pairing the drug with doxorubicin yielded additive to mildly synergistic effects at lower doses, though the highest doxorubicin concentration produced antagonism, possibly because PI3K inhibition dampens the reactive oxygen species generation through which doxorubicin promotes apoptosis. Vincristine, whose mechanism overlaps with the pathway’s role in microtubule function during mitosis, showed only additive to mildly antagonistic interactions.

The strong synergy with dexamethasone is mechanistically compelling. Glucocorticoids kill lymphoid cells through receptors whose nuclear translocation and downstream signaling are potentiated by PI3K-AKT-mTOR inhibition, a finding previously documented in human B-cell acute lymphoblastic leukemia. The Colorado State results thus support the hypothesis that adding a dual PI3K/mTOR inhibitor to standard chemotherapy could deepen responses or sensitize resistant tumors, particularly in lymphoma, where the clinical reservoir of effective options is thin.

Importantly, the study went beyond cell culture to examine pathway activation in actual patients. Using tissue microarrays built from 118 lymph node samples collected from 84 dogs with spontaneous B-cell lymphoma, the researchers scored phosphorylated AKT and phosphorylated 4EBP1 by immunohistochemistry, employing both traditional manual Allred scoring and artificial intelligence-assisted whole-slide analysis with Visiopharm software to generate finer-grained H-scores. The two approaches correlated well, validating the use of machine learning tools for quantitative pathology in veterinary specimens. The phosphorylation signals proved highly variable from tumor to tumor, underscoring that not every canine B-cell lymphoma is driven by an equally active PI3K-AKT-mTOR pathway.

Two clinically meaningful correlations emerged. First, tumors with lower expression of major histocompatibility complex class II, a known adverse prognostic marker measured by flow cytometry, showed significantly higher levels of phosphorylated AKT, linking aggressive immune phenotype to stronger pathway activity. Second, samples taken after relapse following CHOP chemotherapy showed higher phosphorylated 4EBP1 than matched pre-treatment samples, a result that retained significance in paired manual scoring. However, no correlation was found between either phosphoprotein and progression-free interval, overall survival, WHO stage or substage, or flow cytometric immunophenotypic markers, and phosphorylated AKT did not rise in parallel with phosphorylated 4EBP1 in the relapse samples. The authors caution against overinterpreting these limited associations and suggest several possible explanations, including AKT phosphorylation at sites not detected by the assay, alternative 4EBP1 kinases such as PIM2 or RAS/MAPK signaling, or mutations downstream of AKT.

The study has clear limitations that its authors acknowledge. Only one cell line each was tested for lymphoma and leukemia, so the full heterogeneity of canine lymphoid malignancies, including T-cell lymphoma and acute lymphoblastic leukemia, remains unexplored. No washout experiments were performed to determine how long signaling stays suppressed after the drug is withdrawn, information that would be essential for designing oral dosing schedules. And the finding of elevated phosphorylated 4EBP1 without corresponding phosphorylated AKT in relapsed tumors awaits mechanistic confirmation. Nevertheless, the core message is clear: VDC597 collapses PI3K-AKT-mTOR signaling, kills lymphoid cancer cells in vitro at achievable concentrations, synergizes with dexamethasone, and may sensitize tumors to standard chemotherapy. As dogs with spontaneous diffuse large B-cell lymphoma are increasingly recognized as valuable models for human non-Hodgkin lymphoma, these results position dual PI3K/mTOR inhibition as a strategy worth advancing toward larger cell line panels, animal studies, and ultimately clinical trials in both species.

Subject of Research: Evaluation of the dual PI3K/mTOR inhibitor VDC597 against canine B-cell lymphoma and chronic lymphocytic leukemia cells in vitro, with immunohistochemical assessment of PI3K-AKT-mTOR pathway activation in spontaneous canine B-cell lymphoma.

Article Title: In vitro effects of the PI3K/mTOR dual-inhibitor, VDC597, in canine lymphoma and leukemia

Article References: Meuten, T., Brill, S. A., Brady, R. V., Farrell, K. B., Rose, B. J., Schlein, L. J., & Thamm, D. H. (2026). In vitro effects of the PI3K/mTOR dual-inhibitor, VDC597, in canine lymphoma and leukemia. Veterinary Oncology, 3(1), Article 13. https://doi.org/10.1186/s44356-026-00065-4

Image Credits: AI Generated

DOI: 10.1186/s44356-026-00065-4

Keywords: canine lymphoma, canine leukemia, VDC597, PI3K, mTOR, AKT, 4EBP1, signal transduction, chemotherapy, immunohistochemistry, veterinary oncology, drug combination