KM Research Hub
← 返回论文库
已完成 入排:排除 图源:可得 入排置信度 0.98

Genotype-stratified adjunctive dexamethasone for tuberculous meningitis in HIV-negative adults: a randomized controlled phase 3 trial.

Joseph Donovan, Nguyen Duc Bang, Huu Khanh Trinh Dong, Dang Trung Nghia Ho, Thi Anh Thu Nguyen, Thi Thu Hiep Nguyen, Hong Bao Ngoc Lam, Vo Khac Nguyen Phung, Truc Thanh Nguyen, Ho Hong Hanh Nguyen, Kieu Nguyet Oanh Pham, Dang Anh Thu Do, Thi Mai Trang Nguyen, Thi Minh Ha Dang, Huu Lan Nguyen, Van Vinh Chau Nguyen, Thanh Hai Hoang, Dinh Dinh Tran, Khanh Lam Phung, Lalita Ramakrishnan, Thanh Hoang Nhat Le, Thuy Thuong Thuong Nguyen, Marcel Wolbers, Evelyne Kestelyn, Ronald B Geskus, Hoan Phu Nguyen, Guy E Thwaites

Nat Med 32(3) 849-858

PMID41540106
PMCIDPMC13004695
DOI10.1038/s41591-025-04138-z
发表日期2026-01-15
入排依据:未发现恶性肿瘤人群证据。
摘要

摘要

Adjunctive corticosteroids such as dexamethasone are recommended in tuberculous meningitis treatment, despite modest and heterogeneous survival benefit. Leukotriene A4 hydrolase (LTA4H) genotypes associate with distinct intracerebral inflammatory phenotypes and may determine corticosteroid response in tuberculous meningitis, with benefit observed in hyperinflammatory TT genotype but uncertain benefit in lower inflammation CC and CT genotypes. Here, in a phase 3, placebo-controlled trial of human immunodeficiency virus-negative Vietnamese adults with tuberculous meningitis, we randomized 613 LTA4H CC- and CT-genotype participants to 6-8 weeks of dexamethasone or placebo, aiming to show noninferiority of placebo (hazard ratio margin of 0.75) or its superiority. Given the significant survival benefit of dexamethasone previously seen in LTA4H TT-genotype individuals, TT-genotype participants all received open-label dexamethasone and were not randomized. A total of 89 TT-genotype participants received open-label dexamethasone. In CC- and CT-genotype participants, the primary endpoint of all-cause death or new neurological event over 12 months from randomization occurred in 108/305 (35.4%) given dexamethasone and 110/308 (35.7%) given placebo (hazard ratio of 0.99, 96% confidence interval (adjusted for multiple testing) 0.748-1.31). The number of observed primary endpoints (n = 218) exceeded the prespecified number (n = 184) used to calculate the trial's sample size and power. Placebo noninferiority was not established in the CC and CT population or in individual genotype subpopulations. Benefit or heterogeneity of effect was not observed by any prespecified subgroup. In TT-genotype participants, the primary endpoint occurred in 28/89 (31.5%) participants, similar to CC and CT participants. Outcomes were not significantly better in TT-genotype participants versus CC- or CT-genotype participants. In CC- and CT-genotype participants, serious adverse events occurred in 161/305 (52.8%) dexamethasone-treated participants and 160/308 (51.9%) placebo-treated participants. In conclusion, neither noninferiority nor superiority of placebo was established in human immunodeficiency virus-negative LTA4H CC- and CT-genotype adults with tuberculous meningitis, and dexamethasone was safe. The modest and heterogeneous benefit of dexamethasone indicates that greater understanding of tuberculous meningitis pathophysiology is needed, alongside better targeted, more effective anti-inflammatory agents than corticosteroids (ClinicalTrials.gov NCT03100786 ).

KM 提取结果

2 个 Figure · 9 个 KM 面板

进入结果中心
KM Figure3 个面板

Fig. 3

Fig. 3. Death or new neurological events during the first 12 months after randomization for the intention-to-treat population. Kaplan–Meier curves are shown for the primary endpoint. The solid lines represent the estimated probability of event-free survival in each stratum. The colored shading represents the 95% CIs. A forest plot is shown displaying the hazard ratio (HR) and CIs, with corresponding confidence levels at 96% for CC and CT combined and 98.3% for CC. The points represent the estimated HRs in the primary endpoint between two treatment arms. The black lines represent the 95% CIs. The red vertical line demonstrates the noninferiority threshold at HR of 0.75. The dashed vertical line demonstrates the superiority threshold at HR of 1. For analysis, a Cox proportional hazard model was used, with MRC grade and LTA4H genotype as strata. Analyses are performed for the following populations: for CC and CT genotype combined, 613 participants; for CT genotype, 322 participants; and for CC genotype, 291 participants.

Number_at_risk_in_CC_and_CT_combined

event-free survival

Fig. 3 · Panel Number_at_risk_in_CC_and_CT_combined

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel Number_at_risk_in_CC_and_CT_combined 是同一 Fig. 3 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CC

event-free survival

Fig. 3 · Panel CC

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CC 是同一 Fig. 3 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CT

event-free survival

Fig. 3 · Panel CT

quality_warning 待 EXE 对齐 预期 2 条 已提取 1 条

Panel CT 是同一 Fig. 3 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示曲线缺失:预期 2 条,实际恢复 1 条曲线 1:extreme upward reversal曲线 2:off origin start曲线 2:extreme upward reversal
查看结果
KM Figure6 个面板

Fig. 4

Fig. 4. Death or new neurological events for all genotypes. A comparison of TT-genotype individuals receiving dexamethasone with CC, CC and CT, and CT-genotype individuals receiving dexamethasone. This is followed by comparison of TT-genotype individuals receiving dexamethasone with CC, CC and CT, and CT-genotype individuals receiving placebo. The solid lines represent the estimated probability of event-free survival in each stratum. The colored shading represents the 95% CIs. Analyses are performed for the following populations: TT-genotype participants receiving dexamethasone, 89; CC-genotype participants receiving dexamethasone, 146; CT-genotype participants receiving dexamethasone, 159; CC-genotype participants receiving placebo, 145; and CT-genotype participants receiving placebo, 163.

CC

event-free survival

Fig. 4 · Panel CC

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CC 是同一 Fig. 4 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CC_and_CT

event-free survival

Fig. 4 · Panel CC_and_CT

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CC_and_CT 是同一 Fig. 4 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CT

event-free survival

Fig. 4 · Panel CT

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CT 是同一 Fig. 4 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CC_4

event-free survival

Fig. 4 · Panel CC_4

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CC_4 是同一 Fig. 4 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CC_and_CT_5

event-free survival

Fig. 4 · Panel CC_and_CT_5

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CC_and_CT_5 是同一 Fig. 4 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
CT_6

event-free survival

Fig. 4 · Panel CT_6

complete 待 EXE 对齐 预期 2 条 已提取 2 条

Panel CT_6 是同一 Fig. 4 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

查看结果
其他 Figure(2,默认折叠)

这些图片未产生 KM 曲线结果,不在主页面重复展示。

Fig. 1Fig. 1. Screening, enrollment and randomization. ITT, intention-to-treat; PP, per-protocol; TB, tuberculosis.
非 KM Figure
Fig. 2Fig. 2. CSF cytokine concentrations at baseline according to LTA4H genotype, changes in CSF cytokine concentrations over time in CC- and CT-genotype participants randomized to dexamethasone or placebo and changes in cytokine concentrations over time in dexamethasone-treated CC-, CT- and TT-genotype participants. a , The distribution of CSF cytokine levels (normalized protein expression (NPX), log 2 normalized expression) across different genotypic groups (CC, CT and TT). Each subplot represents a distinct cytokine: TNF, IL-1β, IL-6, IL-12β, IFN-γ and IL-10. The violin plots illustrate the data distribution for each genotype, with individual data points shown as dots. Each red central point indicates the median cytokine level within each group, and the horizontal black bars indicate the interquartile range. Pairwise comparisons between genotype groups were performed using the Wilcoxon rank-sum test (two-sided), with the corresponding P values displayed. The cytokine data are shown for a total of 638 participants (262 of CC genotype, 295 of CT genotype and 81 of TT genotype), for each cytokine subpanel. Eight participants (of 646 individuals who underwent proteomic profiling) did not have proteomic profiling data at baseline. b , CSF cytokine changes over 30 days, by placebo and dexamethasone arms. The solid blue lines show the fit mean cytokine levels at each timepoint, with shaded 95% Bayesian credible intervals obtained via Markov Chain Monte Carlo (MCMC) estimation from the longitudinal submodel of the Bayesian joint model. Rates of change of CSF cytokines for CC- and CT-genotype participants were as follows: for TNF, placebo −0.023, dexamethasone −0.049; for IL-1β, placebo −0.038, dexamethasone −0.072; for IL-6, placebo −0.078, dexamethasone −0.098; for IL-12β, placebo 0.041, dexamethasone −0.002; for IFN-γ, placebo −0.044, dexamethasone −0.070; for IL-10, placebo −0.021, dexamethasone −0.035. The cytokine data are shown for 281 participants allocated to dexamethasone (data shown for 278 participants at day 0 and 175 participants at day 30) and for 283 participants allocated to placebo (data shown for 279 participants at day 0 and 166 participants at day 30) for each cytokine subpanel. c , CSF cytokine changes over 30 days in dexamethasone-treated participants, by LTA4H genotype. The solid blue lines show the fit mean cytokine levels at each timepoint, with shaded 95% Bayesian credible intervals obtained via MCMC estimation from the longitudinal submodel of the Bayesian joint model. The cytokine data are shown for 363 dexamethasone-treated participants: for 136 participants of CC genotype (134 participants at day 0 and 83 participants at day 30), for 145 participants of CT genotype (144 participants at day 0 and 92 participants at day 30) and for 82 participants of TT genotype (81 participants at day 0 and 50 participants at day 30).
非 KM Figure