Predictive value of bowel dose-volume for severe radiation-induced lymphopenia and survival in cervical cancer

Background: Radiation-induced lymphopenia (RIL) is closely related to the prognosis of cervical cancer patients and may affect the efficacy of immune checkpoint inhibitors (ICIs). However, the factors influencing RIL are not very clear. In addition to bone marrow (BM) dose-volume, animal studies indicate radiation-induced bowel injury may be a more crucial factor. Further clarification of the correlation between RIL and bowel dose-volume is important for cervical cancer treatment. Methods: Cervical cancer patients treated with postoperative radiotherapy or radical radiotherapy were eligible for this retrospective study. Clinical characteristics, dose parameters of bowel and BM, planning target volume (PTV) size, overall survival (OS) and progression-free survival (PFS) were recorded. The absolute lymphocyte count<0.5x10(9)/L at radiotherapy end was defined as severe RIL (sRIL). Hazard ratio (HR) and 95% confidence interval (Cl)were estimated using Cox regression models. Survival curve was plotted using the Kaplan-Meier method. On this basis, the receiver operating characteristics (ROC) curve was used to calculate the area under the curve (AUC) for radiation parameters with sRIL as the state variable. Result: A total of 118 cervical cancer patients were included in this study, with a median follow-up time of 57.6 months. In multivariable Cox regression analysis, international Federation of Gynecology and obstetrics (FIGO) stage (HR, 11.806; 95% CI, 3.256-42.809; p<0.001), concurrent chemotherapy (HR, 0.200; 95% CI, 0.054-0.748; p=0.017), sRIL after radiotherapy (HR, 6.009; 95% CI, 1.361-26.539; p=0.018), and pathological type (HR, 2.261; 95% CI, 1.043-4.901; p=0.039) were significantly correlated with OS. Patients with sRIL had significantly decreased OS (79.1% vs 94.1%; HR, 3.81; 95%CI, 1.46-9.92; p=0.023). In binary logistic regression analysis, sRIL was significantly correlated with bowel V45 (Odds radio (OR), 1.025; 95%CI, 1.007-1.044; p=0.007), BM V10 (OR, 0.987; 95%CI, 0.978-0.997; p=0.011), BM V20 (OR, 1.017; 95%CI, 1.002-1.031, p=0.027), and PTV size (OR, 0.998; 95%CI, 0.996-1.000; p=0.026). The ROC curve showed, bowel V45 (AUC=0.787, p<0.001) was the best indicator for predicting sRIL. Conclusion: SRIL after radiotherapy could significantly predict decreased OS. In addition, sRIL is associated with higher bowel, BM dose-volume, PTV size, indicating that the bowel may be an important organ leading to an increased risk of sRIL.