Sun Peng, Chen Cui, Zhao Bai-Tian, Zhang Bo-Yan, Yang Hang, Wang Yu, Xu Fei, Huang Kang-Ming, Liu Yin-Ting, Ouyang Mei-Shuo, Huang Jia-Jia, Liu Pan-Pan, Guo Ying, Li Zhi-Ming
Department of Medical Oncology, Sun Yat-sen University Cancer Center, Guangzhou, 510060, China.
State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Guangzhou, 510060, China.
Cancer Immunol Immunother. 2025 Jun 4;74(7):236. doi: 10.1007/s00262-025-04080-6.
Targeting CD3 × CD20 bispecific antibodies (BsAbs) represents a new milestone in the salvage therapy of relapsed/relapsed large B cell lymphoma and follicular lymphoma. However, cytokine release syndrome (CRS) remains one of the major concerns in clinical practice of CD3 × CD20 BsAbs. This study aimed to identify the potential predictive factors and to construct a nomogram of grade ≥ 2 CRS in CD3 × CD20 BsAbs in Chinese patients.
A total of 87 consecutive patients with B-NHL who received CD3 × CD20 BsAbs at Sun Yat-sen University Cancer Center from January 2021 to December 2023 were included and analyzed. Clinical data were collected, and various methods including machine learning algorithms were introduced.
The median age of the patients was 55 years, and the median number of previous treatment lines was 2. CRS occurred in 42 patients (48.3%), including 27 cases of grade 1, 9 cases of grade 2, 5 cases of grade 3, and 1 case of grade 5. Four variables were revealed and selected: bulky disease (> 5 cm), number of previous treatment lines (≥ 3), monocyte-to-lymphocyte ratio (MLR), and platelet count (PLT). Bootstrap resampling was introduced for both internal validation and model building. A predictive nomogram was ultimately established, with the ROC-AUC of 0.867 (95% CI 0.719-0.954).
This study suggests four potential predictive factors and provides a feasible and easy-to-use nomogram for grade ≥ 2 CRS in Chinese patients for the first time, which is important for guiding personalized management and early intervention of CRS.
靶向CD3×CD20双特异性抗体（BsAbs）代表复发/难治性大B细胞淋巴瘤和滤泡性淋巴瘤挽救治疗的一个新里程碑。然而，细胞因子释放综合征（CRS）仍然是CD3×CD20 BsAbs临床实践中的主要关注点之一。本研究旨在识别潜在的预测因素，并构建中国患者中CD3×CD20 BsAbs≥2级CRS的列线图。
纳入并分析了2021年1月至2023年12月在中山大学肿瘤防治中心接受CD3×CD20 BsAbs治疗的87例连续B-NHL患者。收集临床数据，并引入包括机器学习算法在内的各种方法。
患者的中位年龄为55岁，既往治疗线数的中位数为2。42例患者（48.3%）发生CRS，包括1级27例、2级9例、3级5例和5级1例。揭示并选择了四个变量：大包块疾病（>5cm）、既往治疗线数（≥3）、单核细胞与淋巴细胞比值（MLR）和血小板计数（PLT）。引入自助重采样进行内部验证和模型构建。最终建立了一个预测列线图，ROC-AUC为0.867（95%CI 0.719-0.954）。
本研究提示了四个潜在的预测因素，并首次为中国患者中≥2级CRS提供了一个可行且易于使用的列线图，这对于指导CRS的个性化管理和早期干预具有重要意义。
