Drug targets in prostate cancer: an appetite for KLK2-mediated destruction

Summary:

Human Kallikrein 2 (KLK2) is a prostate cancer tissue specific protein that is regulated by androgen receptor (AR) signaling. KLK2 was not previously recognized as a therapeutic target as it is secreted. It has now been demonstrated that KLK2 is expressed on the cell surface and is targetable by various methodologies.

In this issue of Clinical Cancer Research, Shen and colleagues demonstrate that the novel prostate cancer antigen KLK2 has highly prostate cancer specific expression, localizes to the cell surface, and is an intriguing therapeutic vulnerability (1). KLK2 is a trypsin-like serine protease that shares 80% amino acid homology with Human Kallikrein 3 (KLK3), also known as prostate-specific antigen (PSA), and is located 12 kb from KLK3 on chromosome 19. In normal development, KLK2 is expressed in prostate epithelial cells and is secreted into the prostatic fluid where it plays a role in seminal liquefaction. In prostate cancer, KLK2 is regulated by AR signaling and the authors show it is highly expressed in tumors at different stages of disease including metastatic castration-resistant prostate cancer (mCRPC), where it has more homogenous expression than previously investigated antigens. Shen et al. (1) establish that KLK2 is targetable in pre-clinical models by multiple strategies including T-cell engagers, CAR T-cell and radioligand therapy (RLT; Figure 1).

Figure 1.

KLK2 expressing prostate cancer is targetable by multiple methodologies including T-cell engagers, CAR T-cell therapy and radioligand therapy (RLT). KLK2 is also secreted into the peripheral circulation, potentially acting as a therapeutic sink. (Created in BioRender. Blinka, S. [2025] https://BioRender.com/r27iaxg.)

Targeting surface antigens has revolutionized cancer treatment from the development of antibody drug conjugates and RLTs to a range of immuno-oncology (IO) approaches including CAR T-cell therapy. While immunotherapy has transformed treatment practice for many solid tumors, overcoming the “cold” tumor microenvironment in prostate cancer remains elusive. Checkpoint inhibition has failed to show clinical efficacy in prostate cancer with only a 5% overall response rate in unselected and heavily pre-treated patient populations (2). Sipuleucel-T, an autologous dendritic cell vaccine targeting prostatic acid phosphatase (PAP), remains the only FDA approved prostate cancer specific immunotherapy with an improved overall survival of approximately 4 months. T-cell engagers are an intriguing addition to the IO arsenal, especially as T-cells have been shown to be restricted to the prostate cancer stroma. However, several early phase clinical trials, many of which targeted prostate-specific membrane antigen (PSMA), were limited by toxicity and/or modest efficacy (3). Clinical trials targeting STEAP1 by different methodologies are currently under clinical investigation, however, have they have yet to report safe, robust and durable responses. Thus, there remains a significant unmet need to develop tissue specific as well as immune-harnessing therapies for prostate cancer.

In this study, Shen et al. (1) demonstrate high levels of KLK2 expression across the spectrum of prostate cancer. While they observed increased KLK2 heterogeneity in heavily pre-treated mCRPC, 75% of these samples retain higher expression (3+) than PSMA. Specifically, KLK2 expression was higher than PSMA in lymph node and bone with comparable expression in soft tissues. KLK2 was not previously recognized as a therapeutic target as it was shown to be a secreted serine protease with no evidence for localization to the cell surface. To address this unknown, the authors confirmed that KLK2 is present on the surface of the AR+ VCaP cell line and dissociated bone metastatic tissue from patients. Finally, they validate KLK2 surface expression by immunofluorescence with co-localization of EpCAM.

Along with actionable mutations, tumor specific surface antigens are “holy grail” oncologic vulnerabilities and targetable by multimodal therapies. To evaluate KLK2 as a pre-clinical target, Shen et al. (1) show a KLK2-CD3 bispecific T-cell engager resulted in effective T-cell mediated cytotoxicity and pro-inflammatory cytokine release in a dose dependent manner in VCaP cells. Importantly, they observed dose-dependent tumor growth inhibition (TGI) with increased CD8+ and CD4+ T-cell infiltration. Second, to determine if KLK2 is amenable to additional surface targeting therapies, such as RLT, the authors demonstrated that KLK2-mediated internalization occurs via a KLK2-immunofluorescent antibody. They subsequently show a single intravenous injection with a ²²⁵Ac-KLK2 alpha-emitter resulted in anti-tumor activity with 110% TGI. Finally, the authors evaluated a KLK2 CAR T-cell therapy targeting VCaP xenografts. They observed robust responses to a single intravenous treatment of CAR T-cells with 10/10 complete tumor regression following infusions of 10×10⁶ CAR T-cells. Tumor reduction was accompanied by increased CD4+ and CD8+ T-cell infiltration, cytokine release, and CAR T-cell expansion.

Limitations of the study include a single pre-clinical model tested for tumor regression, which was restricted to the hormone sensitive and AR-V7 expressing VCaP cell line. While the authors demonstrate that KLK2 is localized to the cell membrane, they do not test if secreted KLK2 functions as a therapeutic sink in the peripheral circulation. It remains an open question in the field if inter-individual heterogeneity of secreted KLK2 results in differential treatment response.

A recently reported phase I trial with pasritamig, a first-in-human, first-in-class bispecific T-cell engager targeting CD3-KLK2 demonstrated a favorable safety profile with very low rates of cytokine release syndrome (CRS) using intravenous dosing at RP2D (ref 4; NCT04898634). All incidences of CRS were grade 1 (fever only) and occurred in only 4 patients (8.9%) and did not require tocilizumab, dramatically differentiating pasritamig from prior PSMA and STEAP1 targeting T-cell engagers (3,5). Importantly, this permitted outpatient dosing at 300mg every 6 weeks which will potentially expand future access to this therapy to the community. Excitingly, the authors reported a median radiographic progression-free survival of 7.85 months with 42.4% of patients achieving a PSA50 response in a heavily pretreated population with a median of 4 prior lines of therapy. Results from this trial are highly encouraging and the safety and convenience of the RP2D regimen opens the door for combination therapy approaches. At present, two clinical trials are enrolling to determine the RP2D of KLK2 targeting Actinium-225 labeled antibodies (NCT04644770) as well as combinatorial T-cell engager studies targeting PSMA-CD28 and KLK2-CD3 (NCT06095089).

This study by Shen et al. (1) provides critical evidence as to why KLK2 may be a superior prostate cancer surface antigen, though several clinically relevant questions remain. Results from the first-in-human pasritamig study in mCRPC are promising, however, predictive biomarker studies of tissue expression have yet to be evaluated. In addition, although mCRPC samples have higher KLK2 expression compared to PSMA, KLK2 expression is more heterogenous in heavily treated disease, thus it may be a better target earlier in the disease course. Further, it remains unclear if the remarkably low rate of CRS is due to high specificity of the antigen for prostate cancer or high steroid dosing during the step-up phase of treatment. To enhance efficacy, it would be reasonable to increase the dose of pasritamig or pre-medicate with lower steroid dosing. Finally, given the increased tissue specificity of KLK2 for prostate cancer compared to STEAP1 and PSMA, there is potential to expand KLK2-directed clinical indications to diagnostic imaging with positron emission tomography as well as antibody drug conjugates given clear evidence of internalization with RLT. The clinical utility of KLK2-mediated therapies is indeed promising, and we eagerly await first-in-human studies testing additional mechanisms of action, such KLK2-Actinium-225.