Department of Medical Laboratory Medicine, The First Affiliated Hospital of Zhejiang Chinese Medical University (Zhejiang Provincial Hospital of Chinese Medicine), Hangzhou, Zhejiang, China.
Clinical Research Institute, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou Medical College, Hangzhou, Zhejiang, China.
Front Immunol. 2024 May 21;15:1402724. doi: 10.3389/fimmu.2024.1402724. eCollection 2024.
Acute ischemic stroke (AIS) is a leading cause of mortality, severe neurological and long-term disability world-wide. Blood-based indicators may provide valuable information on identified prognostic factors. However, currently, there is still a lack of peripheral blood indicators for the prognosis of AIS. We aimed to identify the most promising prognostic indicators and establish prognostic models for AIS.
484 subjects enrolled from four centers were analyzed immunophenotypic indicators of peripheral blood by flow cytometry. Least absolute shrinkage and selection operator (LASSO) regression was applied to minimize the potential collinearity and over-fitting of variables measured from the same subject and over-fitting of variables. Univariate and multivariable Cox survival analysis of differences between and within cohorts was performed by log-rank test. The areas under the receiving operating characteristic (ROC) curves were used to evaluate the selection accuracy of immunophenotypic indicators in identifying AIS subjects with survival risk. The prognostic model was constructed using a multivariate Cox model, consisting of 402 subjects as a training cohort and 82 subjects as a testing cohort.
In the prospective study, 7 immunophenotypic indicators of distinct significance were screened out of 72 peripheral blood immunophenotypic indicators by LASSO. In multivariate cox regression, CTL (%) [HR: 1.18, 95% CI: 1.03-1.33], monocytes/μl [HR: 1.13, 95% CI: 1.05-1.21], non-classical monocytes/μl [HR: 1.09, 95% CI: 1.02-1.16] and CD56 NK cells/μl [HR: 1.13, 95% CI: 1.05-1.21] were detected to decrease the survival probability of AIS, while Tregs/μl [HR:0.97, 95% CI: 0.95-0.99, p=0.004], B/μl [HR:0.90, 95% CI: 0.85-0.95, p=0.023] and CD16NK cells/μl [HR:0.93, 95% CI: 0.88-0.98, p=0.034] may have the protective effect. As for indicators' discriminative ability, the AUC for CD56NK cells/μl attained the highest of 0.912. In stratification analysis, the survival probability for AIS subjects with a higher level of Tregs/μl, B/μl, CD16NK cells/μl, or lower levels of CD56NK cells/μl, CTL (%), non-classical monocytes/μl, Monocytes/μl were more likely to survive after AIS. The multivariate Cox model showed an area under the curve (AUC) of 0.805, 0.781 and 0.819 and 0.961, 0.924 and 0.982 in the training and testing cohort, respectively.
Our study identified 7 immunophenotypic indicators in peripheral blood may have great clinical significance in monitoring the prognosis of AIS and provide a convenient and valuable predictive model for AIS.
急性缺血性脑卒中（AIS）是全球范围内导致死亡、严重神经功能障碍和长期残疾的主要原因。基于血液的指标可能为已确定的预后因素提供有价值的信息。然而，目前仍然缺乏用于 AIS 预后的外周血指标。我们旨在确定最有前途的预后指标，并为 AIS 建立预后模型。
通过流式细胞术对来自四个中心的 484 名受试者的外周血免疫表型指标进行分析。最小绝对收缩和选择算子（LASSO）回归用于最小化从同一受试者测量的变量的潜在共线性和过拟合以及变量的过度拟合。对数秩检验用于对队列之间和队列内的差异进行单变量和多变量 Cox 生存分析。通过接收者操作特征（ROC）曲线下面积评估免疫表型指标在识别具有生存风险的 AIS 受试者中的选择准确性。使用包含 402 名受试者的训练队列和 82 名受试者的测试队列的多变量 Cox 模型构建预后模型。
在前瞻性研究中，通过 LASSO 从 72 个外周血免疫表型指标中筛选出 7 个具有显著意义的免疫表型指标。在多变量 Cox 回归中，CTL（%）[HR：1.18，95%CI：1.03-1.33]、单核细胞/μl[HR：1.13，95%CI：1.05-1.21]、非经典单核细胞/μl[HR：1.09，95%CI：1.02-1.16]和 CD56NK 细胞/μl[HR：1.13，95%CI：1.05-1.21]被检测到降低 AIS 的生存概率，而 Tregs/μl[HR：0.97，95%CI：0.95-0.99，p=0.004]、B/μl[HR：0.90，95%CI：0.85-0.95，p=0.023]和 CD16NK 细胞/μl[HR：0.93，95%CI：0.88-0.98，p=0.034]可能具有保护作用。就指标的判别能力而言，CD56NK 细胞/μl 的 AUC 最高，为 0.912。在分层分析中，Tregs/μl、B/μl、CD16NK 细胞/μl 水平较高或 CD56NK 细胞/μl、CTL（%）、非经典单核细胞/μl、单核细胞/μl 水平较低的 AIS 受试者的生存概率更有可能在 AIS 后存活。多变量 Cox 模型在训练队列和测试队列中的曲线下面积（AUC）分别为 0.805、0.781 和 0.819 以及 0.961、0.924 和 0.982。
我们的研究确定了外周血中的 7 种免疫表型指标可能在监测 AIS 预后方面具有重要的临床意义，并为 AIS 提供了一种方便且有价值的预测模型。
