Synthetic and Functional Biomolecules Center, Beijing National Laboratory for Molecular Sciences, Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of Education, College of Chemistry and Molecular Engineering, Innovation Center for Genomics, Peking University, Beijing, 100871, People's Republic of China.
Department of Hematology, Lymphoma Research Center, Peking University Third Hospital, Beijing, 100191, People's Republic of China.
Clin Epigenetics. 2021 Feb 11;13(1):33. doi: 10.1186/s13148-020-00973-8.
Although R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) remains the standard chemotherapy regimen for diffuse large B cell lymphoma (DLBCL) patients, not all patients are responsive to the scheme, and there is no effective method to predict treatment response.
We utilized 5hmC-Seal to generate genome-wide 5hmC profiles in plasma cell-free DNA (cfDNA) from 86 DLBCL patients before they received R-CHOP chemotherapy. To investigate the correlation between 5hmC modifications and curative effectiveness, we separated patients into training (n = 56) and validation (n = 30) cohorts and developed a 5hmC-based logistic regression model from the training cohort to predict the treatment response in the validation cohort.
In this study, we identified thirteen 5hmC markers associated with treatment response. The prediction performance of the logistic regression model, achieving 0.82 sensitivity and 0.75 specificity (AUC = 0.78), was superior to existing clinical indicators, such as LDH and stage.
Our findings suggest that the 5hmC modifications in cfDNA at the time before R-CHOP treatment are associated with treatment response and that 5hmC-Seal may potentially serve as a clinical-applicable, minimally invasive approach to predict R-CHOP treatment response for DLBCL patients.
尽管 R-CHOP（利妥昔单抗、环磷酰胺、多柔比星、长春新碱和泼尼松）仍然是弥漫性大 B 细胞淋巴瘤（DLBCL）患者的标准化疗方案，但并非所有患者对该方案都有反应，并且目前尚无有效的方法来预测治疗反应。
我们利用 5hmC-Seal 在 86 名接受 R-CHOP 化疗前的 DLBCL 患者的血浆无细胞 DNA（cfDNA）中生成全基因组 5hmC 图谱。为了研究 5hmC 修饰与疗效之间的相关性，我们将患者分为训练（n=56）和验证（n=30）队列，并从训练队列中开发了基于 5hmC 的逻辑回归模型来预测验证队列中的治疗反应。
在这项研究中，我们确定了与治疗反应相关的 13 个 5hmC 标志物。逻辑回归模型的预测性能（敏感性为 0.82，特异性为 0.75，AUC=0.78）优于现有的临床指标，如 LDH 和分期。
我们的研究结果表明，R-CHOP 治疗前 cfDNA 中的 5hmC 修饰与治疗反应相关，5hmC-Seal 可能是一种潜在的临床应用、微创方法，可用于预测 DLBCL 患者对 R-CHOP 的治疗反应。
