Abstract
PURPOSE: Patients with cervical cancer who are at high risk for para-aortic lymphatic involvement may receive extended-field chemoradiation (EF-CRT), with inclusion of the para-aortic region. Increased radiation to bone marrow (BM) may heighten hematologic toxicity (HT) and affect timely delivery of chemoradiation. Factors associated with HT in this setting have not been well studied. METHODS AND MATERIALS: This study was a retrospective analysis of women treated with EF-CRT from 2012 to 2018 with platinum-based chemotherapy. Factors including age, body mass index (BMI), race, Charlson Comorbidity Index (CCI), and nadirs for white blood cell count, absolute neutrophil count, hemoglobin, and platelet count were collected. The BM metrics included V5Gy, V10Gy, V15Gy, V20Gy, V25Gy, V30Gy, V35Gy, V40Gy and V45Gy (VxGy was defined as the percentage of BM volume receiving x Gy). Hematologic toxicity was defined as grade ≥2 (Cooperative Group Common Toxicity Criteria) leukopenia, anemia, neutropenia, or thrombocytopenia. Univariate analysis (UVA) and multivariate analysis (MVA) were performed using the χ(2) test, the Fisher exact test, and logistic regression. Previously published dosimetric BM constraints were examined as detailed in each respective study. RESULTS: Fifty-two women underwent EF-CRT with cisplatin. UVA showed no association between HT and age, BMI, or CCI. When accounting for race, V5Gy ≥98% was associated with grade ≥2 leukopenia (P = .02) and grade ≥2 HT (P = .05). Most previously described radiation metrics were not reproduced in our cohort, but a similar constraint, V20Gy <70%, was associated with reduced leukopenia of grade ≥2 on UVA (P = .02) and MVA (P < .05). CONCLUSIONS: Acute HT in patients receiving EF-CRT was associated with large volumes of low-dose radiation to the BM and was also associated with race. Restricting the BM V20Gy to less than 70% to 75% may be beneficial in reducing HT, but other pelvic radiation BM constraints may not be applicable to this population.