Preclinical evidence for the use of anti‐Trop‐2 antibody‐drug conjugate Sacituzumab govitecan in cerebral metastasized castration‐resistant prostate cancer

Abstract Purpose Improved survival rates have been observed in castration‐resistant prostate cancer (CRPC) due to advancements in treatment options. However, individuals with brain metastases still have limited therapeutic options and an unfavorable prognosis. Therefore, there is an urgent need to explore new therapeutic avenues, such as antibody‐drug conjugates (ADCs), which have demonstrated significant clinical activity against active brain metastases in solid tumors. Our objective was to determine the expression levels of the ADC targets Trop‐2 and NECTIN‐4 in cerebral metastasized CRPC (mCRPC). Methods Immunohistochemical staining of Trop‐2 and NECTIN‐4 with evaluation of H‐score was performed in CRPC brain metastases (n = 31). Additionally, we examined Trop‐2 protein expression in prostate cancer cell lines and studied their responsiveness to the anti‐Trop‐2 ADC Sacituzumab govitecan (SG) in vitro. Results Our analysis revealed that most patients exhibited moderate to strong Trop‐2 expression [n = 27/31 with H‐score ≥100, median H‐score 220 (IQR 180–280)], while NECTIN‐4 was absent in all cerebral metastases. Mechanistically, we demonstrated that the efficacy of SG depends on Trop‐2 expression levels in vitro. Overexpression of Trop‐2 in Trop‐2‐negative PC‐3 cells led to sensitization to SG, whereas CRISPR‐Cas9‐mediated knockdown of Trop‐2 in Trop‐2‐expressing DU‐145 cells conferred resistance to SG. Conclusion The substantial expression of Trop‐2 in cerebral metastases, along with our preclinical in vitro results, supports the efficacy of SG in treating cerebral mCRPC. Thus, our results extend the understanding of the potential of ADCs in prostate cancer treatment and provide an additional treatment strategy for the challenging subset of patients with cerebral metastases.


| INTRODUCTION
The incidence of brain metastases in patients with castration-resistant prostate cancer (CRPC) is gradually increasing, primarily attributed to improved systemic therapies.Persisting as a rare event, with an incidence below 1%. 1,2However, these patients are confronted with limited therapeutic options and a significantly compromised prognosis. 3Despite the utilization of new antiandrogen therapies and chemotherapies, treatment remains challenging in this subset and there is a high clinical need to explore new therapeutic strategies. 4,5dditionally, there is limited evidence for the effectiveness of approved drugs in cerebral mCRPC, as these patients are often excluded from contemporary clinical trials. 4,5n this context, antibody-drug conjugates (ADCs) have emerged as a promising new class of drugs with high efficacy in the treatment of cerebral metastasized solid tumors, for example, Her-2 targeting ADC for cerebral metastasized breast cancer (Phase II TUXEDO trial). 6revious investigations have demonstrated an upregulation of Trophoblast cell surface antigen-2 (Trop-2) across various human solid tumors, including both primary and metastasized prostate cancer, which is associated with an increased risk of disease progression and unfavorable clinical outcomes. 7Moreover, Trop-2 has been identified as a potential therapeutic target for mCRPC treatment. 7,8In the IMMU-132-01 phase I/ II trial (NCT01631552) the efficacy of Sacituzumab govitecan (SG), an anti-Trop-2 ADC, which is already approved for treating patients with metastatic urothelial carcinoma (mUC) or triple-negative breast cancer (TNBC), 9 has been investigated in patients with advanced epithelial cancers, including 11 patients with mCRPC.SG exhibited a clinical benefit rate, defined as an objective response or stable disease ≥6 months, of 27.3% (n = 3/11) in CRPC patients. 7,10imilar to TROP-2, NECTIN-4 overexpression has been reported in various tumor types, 11,12 and is linked with several facets of tumor progression, like proliferation, recurrence, and metastasis. 11,12Hence, alongside Trop-2, NECTIN-4 provides a promising therapeutic target across different cancer types.Furthermore, the effectiveness of Enfortumab vedotin, an anti-NECTIN-4 ADC, which has already received approval for treating patients with metastatic urothelial cancer, 13 is presently under assessment in the ENCORE phase II trial (NCT0475191) for mCRPC.
In this study, we aimed to investigate the expression levels of Trop-2 and NECTIN-4 in cerebral metastases of patients with CRPC to assess the potential use of ADC in this hard-to-treat population.

| Patient cohort
We conducted a retrospective analysis of two cohorts comprising in total n = 31 patients with cerebral metastasized CRPC treated at the University Medical Center Bonn (n = 10) and University Medical Center Göttingen (n = 21).Clinicopathologic data of the discovery cohort #1 from the University Hospital Bonn (UKB) are summarized in

Conclusion:
The substantial expression of Trop-2 in cerebral metastases, along with our preclinical in vitro results, supports the efficacy of SG in treating cerebral mCRPC.Thus, our results extend the understanding of the potential of ADCs in prostate cancer treatment and provide an additional treatment strategy for the challenging subset of patients with cerebral metastases.
Two independent pathologists (M.Toma, and M. Eckstein) evaluated H-score for membranous NECTIN-4 and Trop-2 staining.The study was approved by the ethical review board of the Medical Faculty of the University of Bonn (approval number: 372/21) and conducted in accordance with the Declaration of Helsinki.All patients provided written informed consent.

| Retroviral transduction of Trop-2 expression plasmid
For retroviral transduction of Trop-2 overexpressing cultures (PC-3 OV) a pRP-Trop-2 vector was utilized.Retroviral particles were produced in HEK 293 cells by transfecting 1 μg of the retroviral pCL-Eco and 5 μg of pRP-Trop-2.The next day, the medium was replaced by fresh medium.On Day 2, the supernatant was harvested, sterile flirted, and applied on the target cells at different concentrations (50, 75, and 100% virus media) followed by selection with puromycin.

| Measurement of cell viability
To measure the cell viability after treatment with the ADC Sacituzumab govitecan (SG) we used a crystal violet assay (0.05% crystal violet staining, 0.1% acetic acid).The assay involves exposing the cells (1 × 10 4 ) to different concentrations of SG (0-10 μg/mL) for 48 h, fixing them with 37% formaldehyde, and staining them with crystal violet.The absorbance of the stained cells is then measured using an ultraviolet-visible spectrometer [570 nm, Safire Reader (Tecan)], and the relative viability of the cells is calculated based on the absorbance values.

| Analysis of the SG payload, SN-38 on PCa cell lines
To evaluate the cytotoxic effect of the payload of SG (SN-38, a topoisomerase-1 inhibitor) on well-established PCa cell lines in-silico analysis of DepMap portal (https:// depmap.org/ portal/ ) was used.Analysis was performed for PCa cell lines 22Rv1, PC-3, LNCaP, and DU-145.The curves illustrate a dose-dependent inhibition of SN-38 in DU-145 and PC-3, as depicted in Figure S1.

| Statistical analysis
Statistical analysis was performed using SPSS (Version 28.0.1.1)and GraphPad Prism (Version 9.4.0).Comparisons between two groups were statistically tested with a non-parametric Mann-Whitney test and a parametric t-test.All p-values were calculated two-sided, and a p < 0.05 was considered statistically significant.

| Robust membranous Trop-2 expression in CRPC brain metastases
First, we evaluated the membranous expression of the two ADC targets Trop-2 and NECTIN-4 in a discovery cohort consisting of n = 10 patients diagnosed with cerebral mCRPC [discovery cohort #1 from the University Hospital Bonn (UKB)].Subsequently, we validated our findings by analyzing an independent validation cohort comprising n = 21 cerebral CRPC specimens [validation cohort #2 from the University Hospital Göttingen (UMG)].To assess the membranous expression of the ADC targets, we performed immunohistochemical staining with subsequent evaluation of the H-score. 15The representative images of Trop-2 immunohistochemical staining on CRPC brain metastasis demonstrate robust membranous expression of Trop-2 in the tumor cells, while the surrounding brain tissue consistently shows no Trop-2 expression (Figure 1A).In the discovery cohort #1, 100% (n = 10/10) of the cerebral mCRPC samples exhibited moderate to strong membranous Trop-2 expression, defined as H-score ≥100 as previously described by us and others, 14,15 with a median H-score of 215 (IQR 195-282.5),whereas NECTIN-4 was absent in all samples (Figure 1B).In validation cohort #2, moderate to strong membranous Trop-2 expression was observed in 80.9% (n = 17/21) of cerebral mCRPC samples, with a median H-score of 220 (IQR 130-270), while only 1 (=4,8%) sample displayed the weak membranous expression of Trop-2 (H-score 90).Of note, only three samples (=14.3%) of cohort #2 showed no Trop-2 expression (Figure 1B).Histopathological analysis revealed that those three tumor samples lacking Trop-2 expression were all neuroendocrine dedifferentiated.There was no statistically significant difference between cohort #1 and #2 (p = 0.5) regarding Trop-2 expression.

| On-target efficacy of Sacituzumab govitecan in PCa cell lines
To assess Trop-2 expression levels, Western blotting, and flow cytometry were performed on a panel of prostate cancer (PCa) cell lines, including 22Rv1, C4-2B, PC-3, DU-145, and LNCaP.Among these cell lines, only DU-145 exhibited strong Trop-2 protein expression (Figure 2A,B).Furthermore, the efficacy of SG on PCa cell growth in vitro was evaluated by treating DU-145 and PC-3 cells with varying concentrations of SG.The results showed that SG significantly inhibited cell growth in Trop-2expressing cancer cells (DU-145), while Trop-2 negative PC-3 cells exhibited resistance to SG treatment (p < 0.001) (Figure 2C).

| DISCUSSION
In this report, we demonstrated that Trop-2 is highly expressed in cerebral CRPC metastases, and provided a compelling preclinical rationale for investigating SG treatment for patients with cerebral mCRPC.In addition, the current findings align with previous research that Trop-2 is overexpressed in CRPC, and targeting Trop-2 induced dose-dependent inhibition of cell growth in Trop-2 expressing cancer cells. 17he strong expression of Trop-2 on tumor cells, coupled with the absence of Trop-2 expression in adjacent brain tissue and the ability of the SG payload SN-38 to cross the blood-brain barrier, 18,19 may lead to a significant therapeutic opportunity.In a subgroup analysis of TNBC patients with brain metastases from the phase 3 ASCENT trial, SG demonstrated a clinical benefit without regarding the intracranial efficacy of SG in breast cancer (BC) brain metastases. 20Based on preclinical evidence showing intracranial accumulation of SG and survival benefit in murine xenograft models of BC, 21 the central nervous system (CNS) activity of SG is being evaluated in a trial involving patients with BC brain metastases or glioblastoma. 22Initial findings revealed therapeutic levels of the payload SN-38 in brain tumor samples, indicating potential accumulation of SG, and intracranial responses were reported in the first patients. 22These findings suggest a relevant therapeutic window for treatment in patients with cerebral metastases of CRPC with no or low on-target toxicity on adjacent brain tissue. 23However, it is crucial to determine whether Trop-2 expression can serve as a predictive biomarker for predicting the therapeutic response to SG.In addition, clinical studies are necessary to thoroughly evaluate the effectiveness and safety of SG in treating cerebral mCRPC.Careful patient selection is necessary before administering anti-Trop-2 ADCs, although the expression of Trop-2 exceeds 80% in cerebral metastases.To address this issue, complementary biomarkers, for example, the analysis of circulating tumor cells (CTCs) or molecular imaging may be a promising approach for assessing ADC targets in cases of cerebral metastasis. 7,24However, further investigations are needed to establish the correlation between the Trop-2 status in (cerebral) metastases and CTCs.Valuable insights on this matter are expected from an ongoing phase II clinical trial involving SG (NCT03725761), which aims to explore the correlation between CTC status and tissue-based status in patients with mCRPC.In addition, we envision the use of Trop-2 PET/CT imaging to noninvasively capture the expression of Trop-2 in tumors before initiating Trop-2-directed therapy. 25Molecular imaging holds great potential for providing valuable information on the ADC target expression status, as has been shown for NECTIN-4 in urothelial cancer, allowing for more targeted and personalized treatment decisions in the sense of ADC precision oncology. 26,27urthermore, diverse observations have been noted regarding Trop-2 expression in neuroendocrine prostate cancer, making it an interesting avenue for future research. 7,17resently, conclusive inferences regarding intrinsic resistance to Trop-2-targeted ADCs are not feasible due to the limited number of cases.Further investigation involving larger cohorts of these rare subtypes is necessary.
Limitations of our study include the complexity of translating in vitro findings into clinical practice, particularly with regard to ADC payload stability, as spontaneous deconjugation at higher concentrations may also lead to non-target-specific cytotoxicity in vitro, and retrospective data collection, which may lead to potential bias, underscoring the need for cautious generalization of results to broader populations.In addition, the sample size is only n = 31, but for the specific subset of CRPC with cerebral metastases, this cohort size might be considered appropriate given the rarity of such cases.
In conclusion, the strong expression of Trop-2 in cerebral mCRPC highlights the potential of SG as a promising therapeutic option for the hard-to-treat population of cerebral mCRPC.