2496 - Radiation-Induced Immune Reconstitution Characterization To Assess Recurrence Risk in Oligometastatic Prostate Cancer
Presenter(s)
S. C. Kamran1, J. J. Smith2,3, R. Dzeng2, Y. Otani1, G. E. Cerrato4, K. Kim5, J. M. Heather6, I. Pompa1, R. Pittie1, E. Chung1, K. Otani1, C. L. Wu7, S. Wu7, J. Y. Wo1, A. L. Zietman1, H. Willers8, P. M. Reeves9, E. M. Van Allen10, J. A. Efstathiou2, and D. T. Miyamoto2; 1Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, 2Massachusetts General Hospital and Harvard Medical School, Boston, MA, 3Department of Infectious Disease, Imperial College London, London, United Kingdom, 4Department of Radiation Oncology, Massachusetts General Hospital, Boston, MA, 5Massachusetts General Hospital Cancer Center, Boston, MA, 6Department of Medicine, Harvard Medical School, Boston, MA, 7Department of Pathology, Massachusetts General Hospital, Boston, MA, 8Department of Radiation Oncology, Mass General Brigham/ Massachusetts General Hospital, Boston, MA, 9Department of Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, MA, 10Dana Farber Cancer Institute, Boston, MA
Purpose/Objective(s):
Radiation therapy (RT) can induce systemic remission in oligometastatic prostate cancer (OMPC), but responses are highly variable. Immune-mediated mechanisms may contribute to tumor control, yet predictors of recurrence remain unclear. We hypothesized that baseline immune composition and post-RT immune reconstitution could differentiate patients by recurrence status. Understanding these dynamics may guide risk stratification and treatment optimization.
Materials/Methods:
Peripheral blood, serum, and tissue were collected from patients (pts) with de novo or recurrent OMPC at baseline, immediately post-RT, and90 and 180 days post– RT. Multi-omic profiling included 39-marker CyTOF of PBMCs, serum proteomics (Olink), and TCR sequencing of PBMCs and tumor tissue. TCR diversity metrics were calculated. Logistic regression, generalized linear mixed models, and unpaired t tests identified immune features associated with recurrence.
Results:
14 pts met inclusion criteria (median age 62.5 years). 8/14 experienced PSA-defined recurrence (median follow-up 34 months). Baseline immune features poorly predicted recurrence (51% by elastic net regression), though elevated NK cells and terminally differentiated CD8+ EMRA T cells (p= 0.028) expressing CD57 (p=0.013) or TIGIT (p=0.023) were enriched in recurrent pts. Baseline TCR diversity and clonality did not differ by recurrence status.
RT induced significant lymphopenia (mean ALC decrease 67%), non-recurrent pts exhibited significantly lower ALC nadir compared to recurrent (p=0.002). Immune trajectories diverged immediately following RT, with elastic net regression achieving 83% accuracy in distinguishing recurrence status. Recurrent pts were characterised by increased terminally differentiated CD8+ (EMRA) T cell abundance (p=0.029) and limited TCR reorganization following RT. In contrast, non-recurrent pts exhibited enrichment of NKT cells, myeloid DCs along with increased T-cell expression of activation/ costim markers CD38, CD27 and CCR7 (all p<0.05). Further, these pts exhibited significant TCR repertoire reorganization (p= 0.004). In serum proteomics, increased CCL17 (p=0.02) and Il7 (p=0.05) along with reduced CXCL13 (p=0.03) were strong predictors of non-recurrence by logistic regression which achieved 78% accuracy across 92 features.
Conclusion:
Multi-omic profiling identified immune features that distinguish pts by recurrence following RT. Baseline immune state was minimally predictive, whereas immediate post-RT immune dynamics, including T cell differentiation and TCR repertoire reorganization, correlated with outcome. Future work will expand the cohort (N=66), incorporate tissue-based imaging, and explore antigen specificity through TCR clustering and expanded proteomics. These findings highlight the potential of immune monitoring to inform risk stratification in OMPC.