Effect of Pelvic Bone Marrow Dose-Volume Parameters on Acute Hematologic Toxicity in Neoadjuvant Pelvic Intensity-Modulated Radiotherapy for Local Advance Rectal Cancer

Abstract

<h3>Purpose/Objective(s)</h3> Neoadjuvant concurrent chemoradiotherapy effectively improves survival and local control of locally advanced rectal cancer (LARC). However, because the pelvic bone marrow is inevitably irradiated, pelvic radiotherapy often results in varying degrees of acute hematologic toxicity (HT). To effectively control the risk of acute HT, this study analyzed the correlation between the volume of irradiated pelvic bone marrow and acute HT based on clinical cases, and identified the corresponding risk threshold. <h3>Materials/Methods</h3> From October 2017 to May 2019, 41 LARC patients who received neoadjuvant CRT were retrospectively reviewed in our center. All patients were treated with 5-field intensity-modulated radiotherapy (IMRT), and the prescription dose delivered to PTV was 45∼50.4 Gy in 25∼28 fractions. Capecitabine or 5-fluorouracil were administered daily 5 days a week during radiotherapy. Different HT were recorded according to National Cancer Institute Common Toxicity Criteria Version 3.0 (NCI‐CTC.V3.0) based on laboratory tests. Multivariable logistic regression was performed to evaluate the association between the volume of irradiated pelvic bone marrow and different HT. Generalized additive model (GAM) and piecewise linear regression were used to further analyze the possible non-linear relationship and threshold effect between them. <h3>Results</h3> Multivariate Logistic regression analysis showed that low-dose (5,10,15,20 Gy) of irradiated total pelvic bone marrow volume (TVx) and coxal bone marrow volume (CVx) were significantly correlated with ≥I grade hemoglobin decreased, ≥I grade neutropenia and ≥II grade leukopenia. There was a significant negative correlation between the irradiated sacrum bone marrow volume (SVx) and ≥I grade hemoglobin decreased, ≥II grade leukopenia. Threshold effect has been observed between CV₅, CV₁₀, CV₁₅ and ≥I grade neutropenia by Generalized additive model (GAM) and piecewise linear regression, and the thresholds were 597ml, 575ml, 518ml, OR (95%CI) were 1.86 (1.04, 3.33), 2.36 (1.13, 4.94), 1.57 (1.08, 2.29), respectively. The thresholds between CV₅, CV₁₀ and ≥II grade leukopenia was 595ml, 575ml, OR (95%CI) was 1.86 (1.06, 3.25), 1.85(1.08, 3.16), respectively. <h3>Conclusion</h3> In neoadjuvant IMRT of rectal cancer, CV is a better predictor of acute HT induced by concurrent chemoradiotherapy than TV. The irradiated volume of CV associated with acute HT was mainly low dose levels (CV₅, CV₁₀, CV₁₅, CV₂₀). In order to effectively reduce the risk of acute HT, keeping CV₁₀ within 575ml may be a good choice.