Cytokine Profiling for the Prediction of Lethality and High-Dose Exposure in a Murine Partial Body Irradiation Model

Accurate radiation biodosimetry is urgently needed for medical management after large-scale radiation exposure. Partial-body irradiation with 5% bone marrow sparing (PBI/BM5) provides a realistic radiation model. The current study specifically focused on the high-dose lethality window (12–16 Gy), where survival transitioned from 100% to 0%, representing a clinically distinct and underserved scenario requiring dedicated biodosimetry tools. We defined the survival profile of male C57BL/6 mice after PBI/BM5 and found that doses of 13.5–14.0 Gy were nonlethal within 12 days, whereas 15.0–15.5 Gy caused 100% mortality within 12 days, with a calculated LD50/12 of 14.68 Gy. A separate cohort of 14.0 Gy showed 100% survival up to 90 days post-radiation. To develop serum cytokine-based biodosimetry in high-dose radiation exposure, mice were exposed to 12.0–16.0 Gy PBI/BM5, and serum was collected on days 1, 3, and 7. A multiplex cytokine assay was used to quantify 70 total cytokines/chemokines. After the exclusion of 4 targets outside detection limits, 66 markers were utilized for downstream analysis. PCA, clustering and heatmaps, and LASSO classification revealed that cytokine signatures can classify radiation groups/status/doses. A 4-cytokine panel (IL-7, GDF-15, IL-16 and FLT3L) could distinguish naïve vs. irradiated mice on all study days. A 24-cytokine signature panel distinguished radiation survivors vs. non-survivors, and another 34-cytokine panel separated radiation doses (12–16 Gy); the prediction was better on day 7 compared to earlier time points. This exploratory study was specifically designed to define the systemic inflammatory response in a high-dose window where survival transitions from 100% to 0% (the ‘lethality threshold’) in a clinically relevant partial-body irradiation model. These findings show that serum cytokines have strong potential for high-dose triage, survival prediction, and dose discrimination within the near-lethal exposure range in a clinically relevant PBI/BM5 model. Extension to lower dose ranges is an important direction for future work.