Eugene, Kwon

Abstract OBJECTIVE To demonstrate that Prostate-specific membrane antigen (PSMA) PET imaging, though primarily used in prostate cancer surveillance, can reveal PSMA uptake in non-prostatic tumors due to neovascular PSMA expression. BACKGROUND PSMA PET imaging is increasingly used for staging and monitoring prostate cancer. However, PSMA uptake is not exclusive to prostate carcinoma and may be observed in non-prostatic malignancies, including glioblastoma, due to neovascular expression. CASE REPORT We present a 67-year-old gentleman with metastatic prostate carcinoma, diagnosed in 2014 and treated with neoadjuvant Lupron, prostatectomy, and salvage radiation. In 2019, he developed nodal recurrence in retroperitoneal and bilateral iliac chain nodes, treated with lymphadenopathy, six cycles of Taxotere-based chemotherapy, and consolidative proton beam radiation (56.25 Gy/25 fractions). He later developed progression to the right proximal humerus and left coracoid process, treated with stereotactic body radiation therapy (25Gy/5 fractions). Follow-up PSMA PET/CT revealed no systemic abnormalities but showed a new PSMA-avid lesion in the right medial temporal lobe. Brain MRI confirmed a 3.5 x 1.7 x 2.5 cm ring-enhancing, intra-axial lesion with edema. The patient was asymptomatic. PSA was undetectable on abiraterone/prednisone and leuprolide. He underwent right frontotemporal craniotomy and resection. Pathology confirmed Glioblastoma, IDH-wildtype, CNS WHO Grade 4 with focal epithelioid features and positive MGMT promoter methylation. He received standard-of-care treatment with radiation, temozolomide, and TTF device. Surveillance PSMA PET and MRI showed no residual brain avidity or progression. Discussion: PSMA is a type II transmembrane glycoprotein expressed by prostatic and various non-prostatic tissues. In healthy brain parenchyma, PSMA expression is typically absent but is overexpressed in neovasculature of high-grade gliomas, including glioblastoma, IDH-wildtype, allowing for radiotracer uptake, as seen in this case. CONCLUSION This case illustrates the potential role of PSMA as a diagnostic tool, as it allows for easy identification of glioblastoma neovasculature, as demonstrated in our case report.

Abstract NRG-1 signals through epidermal growth factor receptor (EGFR) family members. Despite promising preclinical data, HER2 blockade in prostate cancer has failed to improve outcomes in clinical trials.1-4 We have discovered a novel complex systemic role of NRG-1 that drives sensitivity and resistance to androgen deprivation therapy (ADT) in castration-sensitive (CSPC) and castration resistant prostate cancer (CRPC), respectively. Significantly, sensitivity to ADT can be restored in CRPC by manipulating NRG-1 signaling. Our data posit NRG-1 as a systemic “switch” in prostate cancer. Plasma NRG-1 is elevated in our cohort of patients with metastatic CRPC (mean 4.9ng/ml) versus healthy controls (mean 3.2ng/ml, p<0.0001) or metastatic CSPC (mean 3.4ng/ml, p<0.0001). Paradoxically, high systemic NRG-1 predicted superior OS and PFS in CSPC but inferior clinical outcomes in CRPC. Consistently, opposite cell survival between LNCaP (CSPC) vs Du145 & PC3 (CRPC) was observed with recombinant human NRG-1 and enzalutamide (ENZ). ENZ combined with increasing NRG-1 concentration induced LNCaP cell death in an additive manner, but NRG-1 rescued Du145 & PC3 from ENZ-induced death in a dose-dependent manner. We next hypothesized that differential NRG-1 downstream signaling is mediated by engagement with distinct EGFR receptor dimers. While HER2 expression is similar, LNCaP express higher HER3 than Du145 & PC3. High ratio of HER2 to HER3 in Du145 & PC3 mirrors clinical reports of HER2 upregulation in CRPC5,6 and also predicts patient survival in the Stand Up to Cancer cohort.7 Significantly, we recapitulated the CRPC phenotype in LNCaP by inducing HER2 overexpression, HER3 suppression, and chemically-mediated HER2/HER3 dimerization. Conversely, HER2 knockdown sensitized Du145 cells to ENZ. Next, we queried composite single-cell whole human8 and CRPC9 transcriptomes for NRG-1 expression and discovered that the source of NRG-1 is the patients’ own mesenchymal stem cells (MSCs). Bone-derived MSC outgrowth cells produced copious NRG-1, which was increased substantially by coincubation with prostate cancer cell supernatants. Importantly, we were able to manipulate MSCs to reduce their NRG-1 production as a possible clinical translation. Similarly, patients treated with these compounds showed significant reduction in plasma NRG-1 levels after treatment. Additional clinical studies are pending. Together, our observations confirm that NRG-1 has differential signaling: HER3 homodimerization leads to prostate cancer cell death, whereas heterodimerization of HER3 with HER2 leads to prostate cancer cell survival via ADT resistance in a plasma NRG-1 dependent manner. Consequently, HER2 and HER3 receptor availability determine the molecular switch of NRG-1 in prostate cancer. Therefore, NRG-1 can be both a prognostic marker and therapeutic target in CRPC, and multiple clinical trials are pending. Citation Format: Jacob Orme, Roman Thaler, Dejie Wang, Siven Chinniah, Jacob Hirdler, Fernando Quevedo, Lance Pagliaro, Kwon Eugene, Alan Bryce, Brian Costello, Haojie Huang, Sean Park. Mesenchymal stem cell-derived systemic NRG-1 mediates a targetable molecular switch in metastatic prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 397.