Department of Hematologic Oncology, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, Guangdong, P.R. China.
Department of VIP Region, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, Guangdong, P.R. China.
Clin Cancer Res. 2021 Jan 1;27(1):255-266. doi: 10.1158/1078-0432.CCR-20-3417. Epub 2020 Dec 1.
Prediction models for acute myeloid leukemia (AML) are useful, but have considerable inaccuracy and imprecision. No current model includes covariates related to immune cells in the AML microenvironment. Here, an immune risk score was explored to predict the survival of patients with AML.
We evaluated the predictive accuracy of several algorithms for immune composition in AML based on a reference of multi-parameter flow cytometry. CIBERSORTx was chosen to enumerate immune cells from public datasets and develop an immune risk score for survival in a training cohort using least absolute shrinkage and selection operator Cox regression model.
Six flow cytometry-validated immune cell features were informative. The model had high predictive accuracy in the training and four external validation cohorts. Subjects in the training cohort with low scores had prolonged survival compared with subjects with high scores, with 5-year survival rates of 46% versus 19% ( < 0.001). Parallel survival rates in validation cohorts-1, -2, -3, and -4 were 46% versus 6% ( < 0.001), 44% versus 18% ( = 0.041), 44% versus 24% ( = 0.004), and 62% versus 32% ( < 0.001). Gene set enrichment analysis indicated significant enrichment of immune relation pathways in the low-score cohort. In multivariable analyses, high-risk score independently predicted shorter survival with HRs of 1.45 ( = 0.005), 2.12 ( = 0.004), 2.02 ( = 0.034), 1.66 ( = 0.019), and 1.59 ( = 0.001) in the training and validation cohorts, respectively.
Our immune risk score complements current AML prediction models.
急性髓系白血病（AML）的预测模型虽然有用，但准确性和精密度仍存在较大不足。目前尚无模型包含与 AML 微环境中免疫细胞相关的协变量。本研究旨在探索免疫风险评分，以预测 AML 患者的生存情况。
我们评估了几种基于多参数流式细胞术参考资料的 AML 免疫成分预测算法的准确性。选择 CIBERSORTx 从公共数据集估算免疫细胞，并使用最小绝对收缩和选择算子 Cox 回归模型，在训练队列中开发用于生存的免疫风险评分。
有 6 种流式细胞术验证的免疫细胞特征具有信息性。该模型在训练队列和 4 个外部验证队列中具有较高的预测准确性。与高评分者相比，训练队列中低评分者的生存时间更长，5 年生存率分别为 46%和 19%（<0.001）。验证队列 1、2、3 和 4 的平行生存率分别为 46%和 6%（<0.001）、44%和 18%（=0.041）、44%和 24%（=0.004）、62%和 32%（<0.001）。基因集富集分析表明，低评分队列中存在显著富集的免疫相关通路。多变量分析显示，高风险评分独立预测生存时间更短，HR 分别为 1.45（=0.005）、2.12（=0.004）、2.02（=0.034）、1.66（=0.019）和 1.59（=0.001），在训练和验证队列中分别为。
我们的免疫风险评分补充了当前的 AML 预测模型。
