Purpose <p>Hematologic adverse events (HAEs) are common in patients with locally advanced rectal cancer (LARC) undergoing neoadjuvant chemoradiotherapy (nCRT), potentially leading to treatment interruptions and compromised efficacy. This study aimed to develop a hematologic adverse event prediction model using volumetric dose parameters.</p> Methods and materials <p>All doses were converted to the equivalent dose in 2&#xa0;Gy fractions (EQD2) using an α/β ratio of 10&#xa0;Gy for bone marrow. Pelvic bone marrow (PBM) was contoured, and dosimetric parameters (Dmean_EQD2, V3_EQD2-V40_EQD2) were analysed. The association between ≥ 2 grade HAE (HAE2+) and dosimetric/clinical parameters was evaluated through logistic regression. The normal tissue complication probability (NTCP) model was constructed through logistic regression analysis of HAE2+. The PBM dosimetric threshold was determined through receiver operating characteristic (ROC) analysis. The predictive performance of the model was internally validated using bootstrap resampling.</p> Results <p>Of the 141 patients, 49 (34.8%) developed HAE2+. In multivariate analysis, Dmean_EQD2 and V40_EQD2 were associated with HAE2+ (all <i>P</i> &lt; 0.05). In the NTCP analysis, the dose levels corresponding to a 50% probability of HAE2 + were 24.2&#xa0;Gy for DmeanEQD2 and 13.3% for V40EQD2. ROC analysis identified optimal thresholds of Dmean_EQD2 ≤ 24.09&#xa0;Gy (AUC = 0.78, 95% CI: 0.70–0.86) and V40_EQD2 ≤ 13.00% (AUC = 0.83, 95% CI: 0.76–0.90) for predicting HAE2+.</p> Conclusion <p>In LARC patients receiving nCRT, the dose limitation of PBM Dmean_EQD2 ≤ 24.09&#xa0;Gy and V40_EQD2 ≤ 13.00% may reduce the risk of HAE2+. These findings provide quantitative dosimetric references for pelvic bone marrow protection and may help reduce the risk of moderate-to-severe hematologic adverse events.</p> Clinical trial number <p>Not applicable.</p>