Data from A CpG Methylation Classifier to Predict Relapse in Adults with T-Cell Lymphoblastic Lymphoma

• DOI: 10.1158/1078-0432.c.6528984.v1
• Publication Date: 2023-04-01T06:21:13Z
• Authors: Xiao-Peng Tian; Ning Su; Liang Wang; Wei-Juan Huang; Yan-Hui Liu; Xi Zhang; Hui-Qiang Huang; Tong-Yu Lin; Shu-Yun Ma; Hui-Lan Rao; Mei Li; Fang Liu; Fen Zhang; Li-Ye Zhong; Li Liang; Xiao-Liang Lan; Juan Li; Bing Liao; Zhi-Hua Li; Qiong-Lan Tang; Qiong Liang; Chun-Kui Shao; Qiong-Li Zhai; Run-Fen Cheng; Qi Sun; Kun Ru; Xia Gu; Xi-Na Lin; Kun Yi; Yue-Rong Shuang; Xiao-Dong Chen; Wei Dong; Cai Sun; Wei Sang; Hui Liu; Zhi-Gang Zhu; Jun Rao; Qiao-Nan Guo; Ying Zhou; Xiang-Ling Meng; Yong Zhu; Chang-Lu Hu; Yi-Rong Jiang; Ying Zhang; Hong-Yi Gao; Wen-Jun He; Zhong-Jun Xia; Xue-Yi Pan; Lan Hai; Guo-Wei Li; Li-Yan Song; Tie-Bang Kang; Dan Xie; Qing-Qing Cai
• Keywords: CpG site

<div>AbstractPurpose:<p>Adults with T-cell lymphoblastic lymphoma (T-LBL) generally benefit from treatment acute leukemia (ALL)-like regimens, but approximately 40% will relapse after such treatment. We evaluated the value of CpG methylation in predicting for adults T-LBL treated ALL-like regimens.</p>Experimental Design:<p>A total 549 27 medical centers were included analysis. Using Illumina Methylation 850K Beadchip, 44 relapse-related CpGs identified 49 samples by two algorithms: least absolute shrinkage and selector operation (LASSO) support vector machine–recursive feature elimination (SVM-RFE). built a four-CpG classifier using LASSO Cox regression based on association between level relapse-free survival training cohort (<i>n</i> = 160). The was validated internal testing 68) independent validation 321).</p>Results:<p>The four-CpG–based discriminated patients at high risk those low (<i>P</i> < 0.001). This also showed good predictive 0.001) A nomogram incorporating five prognostic factors including CpG-based classifier, lactate dehydrogenase levels, Eastern Cooperative Oncology Group performance status, central nervous system involvement, <i>NOTCH1</i>/<i>FBXW7</i> status significantly higher accuracy than each single variable. Stratification into different subgroups helped identify subset who most benefited more intensive chemotherapy and/or sequential hematopoietic stem cell transplantation.</p>Conclusions:<p>Our could predict disease T-LBL, be used to guide decision.</p></div>