Amusan Precious, Power Christopher, Gill M John, Gomez Daniela, Johnson Erika, Rubin Leah H, Fujiwara Esther
Department of Psychiatry, University of Alberta, Edmonton, AB, Canada.
Department of Medicine, University of Alberta, Edmonton, AB, Canada.
J Neurovirol. 2020 Oct;26(5):743-753. doi: 10.1007/s13365-020-00870-z. Epub 2020 Jul 27.
Despite the availability of modern antiretroviral therapy (ART), neurocognitive impairment persists among some persons with HIV (PWH). We investigated the role of exposure to four major classes of ARTs in neurocognitive impairment in PWH. A single-site cohort of 343 PWH was recruited. Lifetime ART medication history was obtained from medical health records. We evaluated the role of ART exposure as a predictor of neurocognitive impairment using univariate analyses and machine learning, while accounting for potential effects of demographic, clinical, and comorbidity-related risk factors. Out of a total of 26 tested variables, two random forest analyses identified the most important characteristics of a neurocognitively impaired group (N = 59): Compared with a neurocognitively high-performing group (N = 132; F-score = 0.79), we uncovered 13 important risk factors; compared with an intermediate-performing group (N = 152; F-score = 0.75), 16 risk factors emerged. Longer lifetime ART exposure, especially to integrase inhibitors, was one of the most important predictors of neurocognitive impairment in both analyses (rank 2 of 13 and rank 4 of 16, respectively), superseding effects of age (rank 11/13, rank 15/16) and HIV duration (rank 13/13, rank 16/16). Concerning specific integrase inhibitors, the impaired group had significantly longer dolutegravir exposure (p = 0.011) compared with the high-performing group (p = 0.012; trend compared with the intermediate group p = 0.063). A longer duration to integrase inhibitor intake was negatively related to cognition in this cohort. Our findings suggest that possible cognitive complications of long-term exposure to integrase inhibitors, in particular dolutegravir, should be closely monitored in PWH.
尽管有现代抗逆转录病毒疗法(ART),但部分感染艾滋病毒者(PWH)仍存在神经认知障碍。我们调查了接触四类主要抗逆转录病毒药物对PWH神经认知障碍的影响。招募了一个由343名PWH组成的单中心队列。从医疗健康记录中获取终身抗逆转录病毒药物治疗史。我们在考虑人口统计学、临床和合并症相关风险因素的潜在影响的同时,使用单变量分析和机器学习评估抗逆转录病毒药物暴露作为神经认知障碍预测指标的作用。在总共26个测试变量中,两项随机森林分析确定了神经认知障碍组(N = 59)的最重要特征:与神经认知功能良好组(N = 132;F分数 = 0.79)相比,我们发现了13个重要风险因素;与中等表现组(N = 152;F分数 = 0.75)相比,出现了16个风险因素。在两项分析中,终身抗逆转录病毒药物暴露时间越长,尤其是接触整合酶抑制剂,是神经认知障碍最重要的预测指标之一(分别在13个因素中排第2,在16个因素中排第4),超过了年龄(分别排第11/13,第15/16)和艾滋病毒感染持续时间(分别排第13/13, 第16/16)的影响。关于特定的整合酶抑制剂,与表现良好组相比,障碍组的多替拉韦暴露时间显著更长(p = 0.011)(表现良好组p = 0.012;与中等组相比的趋势p = 0.063)。在该队列中,整合酶抑制剂摄入时间越长与认知呈负相关。我们的研究结果表明,在PWH中应密切监测长期接触整合酶抑制剂,特别是多替拉韦可能出现的认知并发症。