Age as modifier of early proteomic remodeling after androgen deprivation therapy (ADT) in prostate cancer: Influence of innate immune/complement–ECM pathways.

246 Background: ADT is standard for advanced prostate cancer (PC) but increases cardiometabolic risk. The proteomic mechanisms, and whether they differ by age, remain poorly defined. We profiled longitudinal plasma proteomes around ADT initiation to test for age-modified effects. Methods: Men with PC starting ADT (n = 14; baseline, 1, 3 months) underwent Olink Explore profiling (2,874 proteins). Standardized NPX changes were modeled with linear mixed-effects (random intercepts; fixed effects for visit and age <70 vs ≥70) and FDR control. Age×time interactions, baseline traits, and pathway enrichments were analyzed. Results: Median age 65.5 years (IQR 64–72.8); 43% African American; BMI 30.4; 50% diabetes; 79% hypertension; 50% metastatic; median Gleason 8.5. Nine proteins changed over 3 months (FDR q<0.05): Upregulated: ITIH4 (acute-phase/complement), MDGA1. Downregulated: KLK3/PSA (on-target), INSL3 (Leydig), EDDM3B, galanin, PI3 (elafin), SPINT3, ENDOU. Most trajectories were consistent across ages except PI3, which declined only in <70 (β −0.71, FDR <10⁻⁴) and was unchanged in ≥70 (β −0.003, FDR 0.99; interaction q = 0.025). Galanin and ENDOU were significant only in younger men. ITIH4, INSL3, KLK3, EDDM3B, SPINT3, and MDGA1 changed in both (≥70 q ≤ 0.05). Baseline clustering revealed two proteomic phenotypes without differential response. Two deaths occurred (non-significant). Conclusions: ADT induces rapid, multisystem proteomic remodeling with suppression of androgen-linked and activation of innate immune/complement–ECM pathways. Age modifies anti-inflammatory protease regulation (PI3), suggesting divergent inflammatory and matrix-remodeling responses after ADT in younger vs older men. These findings link ADT to early complement–ECM biology underlying cardiometabolic risk and highlight age and select proteins as candidates for biomarker-guided monitoring. Larger, outcome-linked cohorts are needed for validation. Age stratified 3 month effects (standardized NPX β) and FDR q-values. Protein (biology) Direction β (<70y) q (<70y) β (≥70y) q (≥70y) Age×time q ITIH4 (acute-phase/complement-linked) Up +0.42 0.0016 +0.56 0.0218 0.65 MDGA1 (neuronal glycoprotein) Up +0.41 0.0009 +0.56 0.0501 0.65 KLK3/PSA (on-target) Down −4.20 2.9×10⁻⁵ −5.10 0.0167 0.53 INSL3 (Leydig function) Down −5.89 7×10⁻⁶ −5.09 0.0115 0.65 EDDM3B (reproductive) Down −1.55 7×10⁻⁶ −1.87 0.0167 0.65 Galanin (neuroendocrine) Down −1.35 0.00029 −0.76 0.168 0.65 PI3 / Elafin (anti-inflammatory) Down −0.71 9×10⁻⁶ −0.003 0.987 0.025 SPINT3 (serine protease inhibitor) Down −4.48 2.9×10⁻⁵ −4.65 0.0115 0.65 ENDOU (RNA processing) Down −0.72 0.00029 −0.48 0.123 0.65 β = standardized effect (baseline→3 mo); FDR = Benja