Factors Predicting Acute Hematologic Toxicity In Cervical Cancer Treated With Postoperative Intensity-Modulated Radiotherapy And Concurrent Nedaplatin

Purpose:To identify predictors associated with acute hematologic toxicity (HT) in cervical cancer treated with postoperative intensity-modulated radiotherapy (IMRT) and concurrent nedaplatin.Methods and Materials:From August 2012 to October 2014,40 patients with cervical cancer after radical surgery were treated at our institution with IMRT and concurrent nedaplatin. The pelvic bone marrow (PBM) was delineated retrospectively on the initial planning CT scan for each patient and divided into three subsites:ilium, lower pelvis, and lumbosacral spine. The volume of each region receiving 10,20,30,40, and≥45Gy(V10, V20, V30, V40, and V45, respectively) was recalculated and a dose-volume histogram was generated based on the original treatment plan. Hematologic toxicity was defined by using World Health Organization (WHO) criteria. Variables predicting for HT included age, body mass index (BMI), performance status, clinical stage, duration of chemotherapy and radiotherapy, cycles of chemotherapy, total dose and dose per fraction of radiotherapy, total dose and weekly dose of chemotherapy, and bone marrow volumes irradiated. Endpoints included the white blood cell count, absolute neutrophil count, hemoglobin, and platelet count nadirs. Multivariate regression models were used to test associations between dosimetric parameters and HT.Results:Age, BMI, performance status, clinical stage, fractionated dose, duration of concurrent chemoradiotherapy, cycles of chemotherapy and weekly dose of chemotherapy were not significantly correlated with HT (P>0.05). On multiple regression analysis, the volume of pelvic bone marrow receiving 10 Gy (PBM-V10) and 20 Gy (PBM-V20) were risk factors associated with Grade 2 or worse HT (HT2+) (odds ratio [OR],1.838; 95% confidence interval [CI], 1.198-2.821; P=0.005; and OR,1.351; 95% CI,1.085-1.681; P=0.007, respectively). If PBM-V10 exceeded 90%, the risk of developing^ Grade 2 leucopenia and neutropenia increased by a factor (odds ratio) of 6 (OR,5.952 and 6.0, respectively) (p< 0.05). Patients with PBM-V20>75%had higher rates of≥Grade 2 leucopenia and neutropenia than those with PBM-V20<75% (27.5% vs.52.5%, p= 0.08; and15.0% vs.47.5%, p<0.001). No correlations were observed between HT and V30/V40/V45. Dosimetric parameters for the lower pelvis and lumbosacral spine had stronger associations with HT than those for the ilium. Cycles of chemotherapy and chemotherapy total dose were associated with WBC nadirs and HT. Weekly dose of chemotherapy mainly affected hemoglobin content. The content of hemoglobin and platelet count nadirs decreased with the increase of the irradiated volume, but no significant correlations were observed between the two hematological parameter and the dosimetric parameters for each portion of the pelvis bone marrow. There was a moderate negative correlation between platelet count nadirs and the total radiation dose (β=-0.402, P=0.01).Conclusions:The volume of pelvic BM receiving low-dose radiation is associated with HT for cervical cancer patients undergoing concurrent chemoradiotherapy. Cycles of chemotherapy, total dose and weekly dose of nedaplatin affect hemogram in varying degrees.