The clinical context motivating the work is sobering. Patients whose disease progresses after treatment with androgen pathway modulators, chemotherapy, and PSMA-directed radioligand therapy face limited options and typically fatal outcomes. Antibody–drug conjugates have transformed treatment in several other malignancies by coupling monoclonal antibodies to potent payloads through chemical linkers, allowing cytotoxic drugs to be delivered preferentially to antigen-expressing tumor cells. However, as the Duke team points out, ADCs evaluated so far in metastatic androgen pathway modulation resistant prostate cancer have been designed and assessed almost exclusively on their ability to kill tumor cells directly, without asking whether immune mechanisms might underpin the most durable responses seen in the clinic. Whether immune activation contributes to ADC efficacy in prostate cancer at all had remained an open and unexplored question, and it is precisely this question the study set out to answer.

The design strategy was deliberately immune-centric from the start. STEAP1 was chosen as the target because of its favorable biology: it sits on the cell surface of prostate cancer cells, its expression is driven by androgen receptor signaling, and its distribution outside the prostate lineage is minimal, reducing the risk of off-tumor toxicity that has plagued some ADC programs. The investigators generated a panel of vandortuzumab-based ADCs by conjugating the antibody to payloads across multiple clinically validated linker–payload platforms, allowing head-to-head comparison of how different payload chemistries affect not only direct cytotoxicity but also the immunological consequences of tumor cell death. This comparative approach is important because growing evidence in other cancers suggests that certain classes of payloads, particularly topoisomerase I inhibitors, can promote immunogenic cell death, a form of tumor cell killing that releases antigen and danger signals capable of priming adaptive immune responses rather than the quiet, immunologically silent apoptosis produced by many conventional cytotoxics.

The experimental pipeline combined sophisticated in vitro assays with two distinct in vivo model systems. In cell culture, the team measured Fcγ receptor engagement, the process by which the antibody’s Fc region binds to activating Fc receptors on myeloid cells such as macrophages and dendritic cells, and they tracked how this engagement influenced antigen presentation and downstream T-cell activation. The results revealed a mechanistic hierarchy that had not previously been appreciated in prostate cancer ADC development. Fcγ receptor engagement proved to be required for optimal antigen presentation by myeloid cells and for the downstream activation of T cells, meaning the antibody backbone itself, through its immune-recruiting Fc domain, plays an active role in turning drug-induced tumor death into an immune stimulus. In other words, the ADC is not just a delivery vehicle but a participant in the immune response it ultimately generates.

Among the payloads tested, exatecan-based conjugates demonstrated the strongest immunostimulatory activity. Exatecan is a potent topoisomerase I inhibitor that has attracted intense interest in recent years as a payload for next-generation ADCs in breast, lung, and gastrointestinal cancers, partly because of its membrane-permeable properties and its association with immunogenic forms of cell death. The Duke team’s data extend that rationale into prostate cancer with an added layer of mechanistic specificity: the immunostimulatory effect of vandortuzumab–exatecan depends on the Fc-mediated engagement of myeloid cells, which then present tumor antigens to T cells. This coupling of payload chemistry to innate immune engagement and adaptive immune priming offers a template for rational ADC design in which immune activation is treated as an engineered feature rather than an incidental byproduct.

The in vivo findings provided the most dramatic evidence for the platform’s potential. In a bone-metastatic model, which recapitulates the clinically devastating skeletal spread characteristic of advanced prostate cancer, vandortuzumab–exatecan mediated durable tumor control. More remarkably, in a syngeneic immunocompetent model of androgen pathway modulation resistant disease, a setting in which the animals retain fully functional immune systems, treated animals developed adaptive immune memory capable of preventing tumor rechallenge. This means that animals whose tumors had been cleared by the ADC were protected when researchers attempted to reintroduce the same cancer, a hallmark of a genuine vaccine-like immune response generated by the treatment itself. Such rechallenge protection is rarely achieved by conventional cytotoxic therapies and suggests the conjugate does more than shrink tumors; it educates the immune system to recognize and reject prostate cancer cells on subsequent encounters.

The implications for patients with advanced prostate cancer are substantial. Current PSMA-directed radioligand therapy, while effective at extending survival, delivers targeted radiation without intentionally leveraging immune activation, and resistance ultimately develops in most patients. Checkpoint immunotherapy has largely failed in metastatic castration-resistant prostate cancer outside the small subset of tumors with mismatch repair deficiency, largely because these tumors are immunologically cold, poorly infiltrated, and present few antigens to primed T cells. An ADC that kills tumor cells while simultaneously converting them into an in situ vaccine, releasing STEAP1 and other tumor antigens for uptake by Fcγ receptor–engaged macrophages and dendritic cells, could in principle warm up the tumor microenvironment and sensitize disease to checkpoint blockade or other immunotherapies. The Duke platform provides the mechanistic justification for testing such combinations.

The study also carries lessons for how ADCs across oncology are evaluated. Clinical development of ADCs in prostate cancer and beyond has traditionally relied on endpoints of radiographic response and survival, with immune endpoints largely absent from early-phase testing. By demonstrating that Fcγ receptor engagement and antigen presentation are required for optimal T-cell activation, and that different payloads sharing the same antibody target diverge sharply in their immunostimulatory capacity, the work argues that immune profiling should be incorporated into ADC design and preclinical testing from the earliest stages. Constructs that look comparable in standard cytotoxicity assays may differ profoundly in their ability to generate adaptive immune memory, and those differences could translate into the durability of response, the depth of remission, and the prevention of relapse that ultimately matter most to patients.

The research was led by John S. Wang and Zachary C. Hartman of Duke University, with contributions from a multidisciplinary team spanning Duke’s departments of Surgery, Medicine, and Pharmacology and Cancer Biology, the Duke Cancer Institute, and the Department of Chemistry at Xavier University of Louisiana, where Qiang Zhang and Guangdi Wang contributed synthetic and conjugation chemistry expertise. The work was supported by the National Institutes of Health, the Department of Defense, and the RCMI Center for Cancer and Health Disparities Research, and it was conducted under institutional animal care approval with no human participant data. The authors note that the study used the PT-09 cell line generously provided by Dr. Brian Ruffell of the Moffitt Cancer Center and relied on Duke University core facilities for flow cytometry and animal studies.

As with all preclinical research, the path from bone-metastatic mouse models to human trials involves substantial uncertainty, including questions about the therapeutic index of exatecan-based conjugates, the translatability of syngeneic model immunology to elderly, heavily pretreated patients, and the optimal sequencing with existing androgen receptor–pathway inhibitors and radioligands. Nevertheless, the study establishes proof of principle that immune activation is a determinant of ADC efficacy in androgen pathway modulation resistant prostate cancer and that vandortuzumab–exatecan functions as an immune-engaging therapeutic capable of inducing durable anti-tumor responses. If these findings hold in clinical testing, they could reshape not only how prostate cancer ADCs are built but also how the field conceives of targeted drug delivery itself, as a strategy that kills tumors and teaches the immune system to keep them dead.

Metastatic castration-resistant prostate cancer develops when prostate tumors continue to grow and spread despite therapies that suppress male hormones, which normally fuel prostate cancer progression. Although newer hormonal medicines, chemotherapy drugs and radiopharmaceuticals have improved care, the disease remains incurable once it reaches this stage. Antibody-drug conjugates, or ADCs, have revolutionized treatment for some breast, bladder and blood cancers, yet their effects in advanced prostate cancer have generally been modest and short-lived. UCLA scientists set out to determine whether existing ADC strategies could be made more effective without having to create an entirely new class of therapy.

An ADC is a molecular delivery system that links three components: an antibody, a chemical linker and a highly potent drug payload. The antibody recognizes a protein displayed on the surface of cancer cells, allowing the conjugate to bind to the tumor. After the cancer cell internalizes the ADC, the linker is broken down or chemically cleaved, releasing the payload inside the cell. This design concentrates a powerful cytotoxic drug where it is needed while potentially reducing exposure to healthy tissues. However, the approach can fail when tumors do not express enough of the target, rapidly repair the damage caused by the payload or activate survival mechanisms that prevent cell death.

The UCLA-led team first examined tumor samples from patients with advanced prostate cancer to understand whether multiple targets could be exploited at the same time. Their analysis showed that B7-H3, PSMA and STEAP1—three proteins already being investigated as targets for ADCs—were frequently present on the same cancer cells. This pattern is important because it suggests that different ADCs might be used in combination, or engineered to recognize multiple tumor-associated proteins, increasing the likelihood that cancer cells will be reached. At the same time, targeting proteins that are more abundant on tumor cells than on normal tissue could help maintain the precision that makes ADCs attractive.

The researchers then tested dozens of combinations involving payloads commonly used in ADC development. They were looking for drug pairs that produced synergy, meaning the combined effect was greater than would be expected from simply adding the activity of each drug individually. One combination consistently stood out: a DNA-damaging payload paired with a compound that inhibits BCL-XL. DNA-damaging agents can create breaks or lesions in the genetic material of cancer cells, but those cells may survive by activating molecular repair and stress-response pathways. BCL-XL acts as one of the proteins that helps prevent programmed cell death, or apoptosis, allowing damaged cells to remain alive.

Blocking BCL-XL appears to remove an important escape route. When prostate cancer cells were exposed to DNA damage while this survival protein was inhibited, they were less able to withstand the treatment and more likely to undergo apoptosis. In cell cultures, the combination caused significantly greater cancer cell death than either the DNA-damaging drug or the BCL-XL inhibitor alone. The same pattern emerged in mice implanted with advanced prostate tumors: combination treatment reduced tumor growth more strongly than single-agent therapy, supporting the idea that the two mechanisms reinforce one another inside the cancer cell.

The study also identified a possible genetic clue that could help determine which patients might benefit most. Tumors retaining an intact TP53 tumor suppressor gene responded particularly well to the treatment strategy. TP53 encodes the p53 protein, a central regulator of cellular stress responses that can halt cell division or promote apoptosis when DNA becomes severely damaged. Many cancers disable this protective system through TP53 mutations, potentially changing how they respond to DNA-damaging therapies. The researchers’ findings suggest that TP53 status may eventually become part of a biomarker strategy for selecting patients for ADC combinations, although this possibility must be tested prospectively in clinical trials.

The results also point to a broader principle in ADC design: the payload may be just as important as the target. Rather than treating an ADC as a fixed package consisting of one antibody and one chemotherapy drug, researchers can potentially match different payloads to the vulnerabilities of particular tumor types. In prostate cancer, pairing a DNA-damaging agent with a BCL-XL inhibitor may be especially effective because it attacks both the tumor’s genetic material and its ability to survive the resulting stress. Combining ADCs that recognize B7-H3, PSMA or STEAP1 could add another layer of pressure by increasing tumor-cell coverage.

The findings remain preclinical, and the treatment has not yet been shown to benefit patients. BCL-XL is also involved in the survival of some normal cells, meaning that safety, dosing and the management of possible side effects will be critical as the strategy moves toward human testing. The UCLA team is now engineering next-generation ADCs that incorporate the most promising payload combinations and evaluating additional ways to target prostate cancer cells. If these experiments are validated in clinical studies, the approach could transform ADCs from short-lived treatments into more durable, biologically tailored therapies for men with metastatic castration-resistant prostate cancer.

Subject of Research: Antibody-drug conjugate combinations targeting metastatic castration-resistant prostate cancer.

Web References: Journal of Clinical Investigation study; DOI link; UCLA Health Jonsson Comprehensive Cancer Center.

References: Semenova G. et al., Journal of Clinical Investigation, DOI: 10.1172/JCI200438.

Keywords: prostate cancer, metastatic castration-resistant prostate cancer, antibody-drug conjugates, ADCs, BCL-XL, PSMA, B7-H3, STEAP1, TP53, targeted cancer therapy, combination therapy, precision oncology.