A Multi-Biomarker Signature of Immune Activation and Metabolic Shift Predicts Treatment Success in Localized Prostate Cancer

AbdulQader M. AbdulQader; Muqdad Khamis Abd; Kawthar Amer Al-Shamari; Waleed Khalid Ahmed

doi:10.31557/APJCB.2026.11.4.1021

Authors

AbdulQader M. AbdulQader College of Dentistry, Al-Iraqia University, Baghdad, Iraq.
Muqdad Khamis Abd Chemistry Department, College of Education for Pure Science/Ibn Al-Haitham, University of Baghdad, Iraq.
Kawthar Amer Al-Shamari College of Medicine, University of Al-Mustansiriyah, Baghdad, Iraq.
Waleed Khalid Ahmed College of Medicine, University of Fallujah, Baghdad, Iraq.

DOI:

https://doi.org/10.31557/APJCB.2026.11.4.1021

Keywords:

Prostate Cancer, Biomarkers, SDF1 (CXCL12), Adenosine, Immunometabolic Profiling, Treatment Response, Predictive Modeling

Abstract

Background: Prostate cancer’s response to therapy varies significantly, and traditional markers like PSA offer limited prognostic insight. A thorough understanding of systemic immune and metabolic changes post-treatment is essential for developing better predictive tools.

Objective: To characterize the profiles of circulating immune and metabolic biomarkers prior to and following definitive therapy in patients with localized prostate cancer, and to assess their efficacy in distinguishing biochemical responders from non-responders.

Materials and Methods: This prospective study involved 90 patients with prostate adenocarcinoma treated at the Oncology Teaching Hospital / Baghdad Teaching Hospital, between February 2024 and February 2025. Patients underwent radical prostatectomy or radiotherapy, with or without androgen deprivation therapy. Blood samples were collected at baseline and three months post-treatment to measure six biomarkers (GFRAL, adenosine (AD), interleukin-39 (IL-39), TRAIL, lipopolysaccharide (LPS), and SDF1) using ELISA. Treatment response was defined as PSA below 6.5 ng/mL. Data were analyzed using nonparametric tests, machine learning, ROC analysis, and multivariate regression, with stratification by treatment type to evaluate intervention-specific changes in biomarkers.

Results: Following treatment, responders exhibited a distinct molecular profile characterized by significantly increased levels of SDF1 and IL-39, along with decreased concentrations of AD and LPS. SDF1 was identified as the most reliable individual predictor, with an area under the curve (AUC) of 0.89. A multivariable model including SDF1, AD, and IL-39 demonstrated improved predictive performance, achieving an AUC of 0.92, with a sensitivity of 89.1% and specificity of 85.7%. Multivariate analysis confirmed that SDF1 and AD are independent predictors of response, after adjusting for age, Gleason score, baseline PSA, and treatment modality.

Conclusion: Effective prostate cancer treatment prompts a systemic immunometabolic transformation towards a reparative and immune-active state. A multi-biomarker panel, notably including SDF1, AD, and IL-39, emerges as a promising discovery-phase signature that, following independent external validation, may provide a non-invasive method for monitoring therapeutic response, but further validation is needed before clinical use.