Blood Radon Activity and Model-Derived Uranium Estimates in Gastrointestinal Cancer Patients from Southern Iraq

Uranium is a naturally occurring trace element with established radiological and chemical toxicity, yet its blood burden in patients with cancer remains poorly characterized. This exploratory cross-sectional study used radon^-222 as a tracer to derive model-based uranium estimates in patients with gastrointestinal (GI) cancer and healthy controls from southern Iraq, a region with documented elevated environmental radioactivity. Blood samples were collected from 78 patients with histologically confirmed GI cancer (39 females and 39 males) and 58 healthy controls (29 females and 29 males) at the National Oncology and Hematology Hospital, Najaf. CR-39 poly(allyl diglycol carbonate) solid-state nuclear track detectors were sealed with blood samples for 90 days. Track densities were quantified using TASLIMAGE automated analysis, and radon concentrations were calculated using an experimentally established calibration factor (0.048 tracks cm^-2 Bq^-1 m³ d). Model-based uranium estimates were derived assuming secular equilibrium between uranium-238 and radon^-222. Daily alpha-particle flux, annual effective dose, and excess lifetime cancer risk were calculated according to ICRP Publications 103 and 119. Group differences were evaluated using Levene-corrected independent-samples t-tests, and effect sizes were quantified using Cohen's d. Mean estimated blood uranium concentrations were higher in male patients than male controls (15.48 ± 5.16 vs. 10.17 ± 3.39 parts per billion (ppb)) and in female patients than female controls. Correspondingly, mean blood radon concentrations were higher in patients (males: 3.99 ± 1.50 vs. 2.62 ± 0.80 Bq m^-3; females: 3.21 ± 1.07 vs. 2.31 ± 0.77 Bq m^-3). Excess lifetime cancer risk remained well below the ICRP reference level in all groups (1.10 × 10^-7 for male patients and 8.82 × 10^-8 for female patients). Although between-group differences were not statistically significant (males: p = 0.275; females: p = 0.187), the observed effect sizes were large (Cohen's d = 1.10 for male radon), suggesting that the study may have been underpowered to detect statistically significant differences. These findings indicate consistently higher model-based uranium estimates in patients with GI cancer than in healthy controls. However, because uranium concentrations were estimated indirectly using a model-based approach and the study employed a cross-sectional design, no causal relationship between uranium exposure and GI cancer can be inferred. Further studies with larger cohorts and direct uranium measurements using ICP-MS are warranted.