Correlation Between Radiation Dose to 18F-FDG-PET Defined Active Bone Marrow Subregions and Acute Hematologic Toxicity in Cervical Cancer Patients Treated With Chemoradiotherapy

Purpose: To test the hypothesis that radiation dose to F-18-fluorodeoxyglucose positron emission tomography (F-18-FDG-PET)-defined active bone marrow (BMACT) subregions is correlated with hematologic toxicity in cervical cancer patients treated with chemoradiotherapy. Methods and Materials: The conditions of 26 women with cervical cancer who underwent F-18-FDG-PET before treatment with concurrent cisplatin and intensity-modulated radiation therapy were analyzed. BMACT was defined as the subregion of total bone marrow (BMTOT) with a standardized uptake value (SUV) equal to or above the mean for that individual. Inactive bone marrow (BMINACT) was defined as BMTOT - BMACT. Generalized linear modeling was used to test the correlation between BMACT and BMINACT dose-volume metrics and hematologic nadirs, particularly white blood cell count (WBC) and absolute neutrophil count (ANC). Results: Increased BMACT mean dose was significantly associated with decreased log(WBC) nadir (beta = -0.04; 95% CI, -0.07to -0.01; p = 0.009), decreased log(ANC) nadir (beta = -0.05; 95% CI, -0.08 to -0.02; p = 0.006), decreased hemoglobin nadir (beta = -0.16; 95% CI, 0.27 to 0.05; p = 0.010), and decreased platelet nadir (beta = 6.16; 95% CI, 9.37 to -2.96; p < 0.001). By contrast, there was no association between BMINACT mean dose and log(WBC) nadir (beta = -0.01; 95% CI, -0.06 to 0.05; p = 0.84), log(ANC) nadir (beta = -0.03; 95% CI, -0.10 to 0.04; p = 0.40), hemoglobin nadir (beta = -0.09; 95% CI, -0.31 to 0.14; p = 0.452), or platelet nadir (beta = -3.47; 95% CI, -10.44 to 3.50; p = 0.339). Conclusions: Irradiation of BM subregions with higher F-18-FDG-PET activity was associated with hematologic toxicity, supporting the hypothesis that reducing dose to BMACT subregions could mitigate hematologic toxicity. Future investigation should seek to confirm these findings and to identify optimal SUV thresholds to define BMACT. (C) 2012 Elsevier Inc.