Main Session
Sep
28
PQA 03 - Digital Health Innovation and Informatics, Patient Safety & Quality, and Radiation and Cancer Biology
2590 - Circulating Biomarkers Predictive of Response to Lu-PSMA-617 in Metastatic Castrate Resistant Prostate Cancer Patients
Presenter(s)
Daniel Rosen, MD, PhD - Brigham and Women's Hospital/Dana Farber Cancer Institute, Boston, MA
D. B. Rosen1, J. Brady2,3, M. Lee3, H. Jacene4, A. K. Tewari5, and P. Ravi2; 1Department of Radiation Oncology, Brigham and Women's / Dana-Farber Cancer Center, Harvard Medical School, Boston, MA, 2Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, 3Arthur and Linda Gelb Center for Translational Research, Dana-Farber Cancer Institute, Boston, MA, 4Department of Radiology, Brigham and Women’s Hospital / Department of Imaging, Dana-Farber Cancer Institute, Boston, MA, 5Dana-Farber Cancer Institute, Boston, MA
Purpose/Objective(s):
Lu-PSMA-617 is now a standard therapy for patients with metastatic castration-resistant prostate cancer (mCRPC). However, only about half of treated patients achieve a systemic response (PSA50), and all eventually progress. The tumor-intrinsic and host mechanisms that drive treatment response remain poorly understood.Materials/Methods:
Patients with mCRPC treated with Lu-PSMA-617 at out institution were enrolled on a biospecimen protocol (DF/HCC #21-629). Plasma samples were collected before treatment (prior to cycle 1, C1) and after 2 cycles of treatment (prior to cycle 3, C3). Clinical demographics, laboratory values, and treatment decisions were recorded prospectively. Patients who received C1 between 06/20/2022 and 02/05/2025 were evaluated. PSA change from C1 to C3 was used to categorize response. A =90% PSA decline over two cycles (PSA90) defined an excellent response. Stable or increased PSA (PSA0) defined non-response. Categorical variables were compared using Fisher’s exact test, and continuous variables using unpaired t-tests. Paired pre-C1 and pre-C3 plasma samples were analyzed by Nomic Bio using the Omni 1000 proteomic panel. Within-patient changes were assessed using paired t-tests; between-group differences were assessed using unpaired t-tests. Pathway overrepresentation analysis was performed using Reactome.org with false discovery rate (FDR) correction by the Benjamini–Hochberg method.Results:
Twenty-four excellent responders and 24 non-responders underwent proteomic profiling. Technical validity was supported by expected biomarker changes. In excellent responders, KLK3 (PSA) decreased from a mean of 75 to 17 ng/mL (p = 0.00016), whereas non-responders showed a nonsignificant increase from 100 to 121 ng/mL (p = 0.31). Acid phosphatase (ACP3), a recently reported prostate cancer biomarker, decreased from 97 to 16 ng/mL in excellent responders (p = 0.016). Among excellent responders, 44 proteins changed significantly after treatment (p < 0.05). Among non-responders, 40 proteins changed significantly. Five treatment-related marker changes overlapped between groups: FAS-L, IL-36a, and CD244 decreased; Flt-3 ligand increased; and ST14 increased in non-responders but decreased in excellent responders. Twenty-two proteins differed significantly between responders and non-responders at baseline. Reactome.org analysis showed significant overrepresentation in immune system pathways (13 proteins, q = 0.009) and cytokine signaling pathways (8 proteins, q = 0.009).Conclusion:
High-plexity proteomic profiling of patients with mCRPC receiving Lu-PSMA-617 demonstrates treatment-associated immune changes and baseline pathway differences between excellent responders and non-responders. These preliminary findings require validation in larger cohorts. The results also support investigation of combination strategies incorporating radiopharmaceutical therapy and immune checkpoint inhibition.