Dynamic Neutrophil-to-Lymphocyte Ratio at 3-4 weeks of Intensity-Modulated Radiotherapy Combined with Normal Liver Volume Predicts Radiation-Induced Hepatic Toxicity in Hepatocellular Carcinoma: A Retrospective Study.

Purpose: This study aimed to investigate whether early dynamic variation in circulating neutrophil-to-lymphocyte ratio(NLR) during intensity-modulated radiotherapy (IMRT) can predict radiation-induced hepatic toxicity (RIHT) in patients with hepatocellular carcinoma(HCC).

Patients And Methods: Neutrophil and lymphocyte counts of 103 HCC patients were measured every 2 weeks before, during and after completion of IMRT.Generalized estimating equations analyses was used to analyze the dynamic changes of neutrophil and lymphocyte counts. The prognostic significant factors were assessed through logistic regression analyses. Statistical power was assessed using power analysis, and the model was adjusted for multiple comparisons via the Bonferroni correction. Receiver operating characteristic (ROC) curves and calibration curves were used to evaluate the prediction accuracy.The predictive model was internally validated using 5-fold cross-validation.

Results: Radiation-induced liver disease(RILD) is a type of RIHT and is a relatively severe phenomenon of hepatic toxicity.Overall, 23 patients (22%) developed RILD. In RILD group, NLR were significantly changed in the first 3 to 4 weeks during IMRT (p=0.006).In multivariate analysis, NLR in first 3-4 weeks(NLR 4), the ratio of NLR in first 3-4 weeks to 1-2 weeks(NLR 4/2),and normal liver volume(NLV) were independent predictive factors for RILD. The area under the curve(AUC) was 0.860 (95% CI, 0.783-0.937).Larger NLV correlated with a lower likelihood of developing RILD(p = 0.041).The mean AUC values was 0.86 in the training set and 0.81 in the test sets across 5-fold cross-validation (p=0.41).

Conclusion: Circulating NLR in first 3 to 4 weeks and its relatively change during IMRT were significantly associated with RIHT.The model based on early dynamic variation of NLR and dosimetric factors NLV can predict RIHT with high accuracy in HCC patients.It can timely assist clinician to take preventive measures and adjusting treatment plans.