Supplementary Material for: Exclusion of Acute Myeloid Leukemia Patients with Central Nervous System Involvement from Clinical Trials: An Analysis of the National Institutes of Health Clinical Trials Registry from 2012-2022

Introduction: Central nervous system (CNS) involvement in acute myeloid leukemia (AML) can be successfully treated with intrathecal chemotherapy and carries debatable prognostic impact. However, patients with CNS involvement are commonly excluded from clinical trials at an unknown rate. We systematically evaluated exclusion criteria of AML clinical trials based on CNS involvement and determined associations with clinical trial characteristics. Methods: The National Institutes of Health Clinical Trials Registry was searched for interventional adult AML trials between 2012-2022 that were phase I, II, or III and relevant trial characteristics were extracted. Results: 1270 trials were included in the analysis with 790 trials (62.1%) explicitly excluding CNS involvement. There was no significant change in rates of CNS exclusion over the past decade. CNS exclusion was higher in trials that included the non-transplant population compared to trials exclusive to the transplant population (66.9% vs. 43.8%, p<0.01). Non-transplant trials were also more likely to exclude patients with a history of or ambiguous timing of CNS involvement (p<0.01). Phase III trials were associated with more liberal definitions of CNS exclusion (history or ambiguous timing) as compared to phase I and II trials that had higher rates of excluding patients with only active CNS involvement (P<0.01). Conclusion: A majority of AML clinical trials, particularly in the non-transplant setting, exclude patients with CNS involvement. Many of these trials, most notably phase 3 trials, exclude patients not only with active, but any history of CNS involvement. Further research is needed to determine optimal management of these patients in order to increase representation in large clinical trials.

引言：急性髓系白血病（Acute Myeloid Leukemia, AML）合并中枢神经系统（Central Nervous System, CNS）受累可通过鞘内化疗成功治疗，但其对预后的影响尚存争议。然而，合并CNS受累的患者常以未知比例被排除在临床试验之外。本研究系统评估了基于CNS受累情况的AML临床试验排除标准，并明确了其与临床试验特征的关联。

方法：检索美国国立卫生研究院（National Institutes of Health, NIH）临床试验注册库，筛选2012年至2022年间开展的针对成人AML的I、II、III期干预性临床试验，并提取相关试验特征数据。

结果：本分析共纳入1270项临床试验，其中790项（62.1%）明确排除CNS受累患者。近十年间，CNS受累患者的排除率未出现显著变化。纳入非移植人群的试验中，CNS受累患者排除率高于仅纳入移植人群的试验（66.9% vs. 43.8%，p<0.01）。非移植类试验也更倾向于排除有CNS受累病史或发病时间不明的患者（p<0.01）。与仅排除活动性CNS受累患者的I、II期试验相比，III期试验对CNS受累排除的定义更为宽松（涵盖病史或发病时间不明的情况，P<0.01）。

结论：多数AML临床试验，尤其是非移植场景下的试验，均排除合并CNS受累的患者。其中许多试验（尤以III期试验为甚）不仅排除活动性CNS受累患者，还排除任何有CNS受累病史的患者。未来需开展进一步研究以明确此类患者的最优管理方案，从而提升其在大型临床试验中的入组比例。