A Phase 2A Trial of the Safety and Tolerability of Increased Dose Rifampicin and Adjunctive Linezolid, With or Without Aspirin, for Human Immunodeficiency Virus-Associated Tuberculous Meningitis: The LASER-TBM Trial

Abstract

Background: Drug regimens that include intensified antibiotics alongside effective anti-inflammatory therapies may improve outcomes in tuberculous meningitis (TBM). Safety data on their use in combination and in the context of human immunodeficiency virus (HIV) are needed to inform clinical trial design.

Methods: We conducted a phase 2, open-label, parallel-design, randomized, controlled trial to assess the safety of high-dose rifampicin, linezolid, and high-dose aspirin in HIV-associated TBM. Participants were randomized (1.4:1:1) to 3 treatment arms (1, standard of care [SOC]; 2, SOC + additional rifampicin [up to 35 mg/kg/d] + linezolid 1200 mg/d reducing after 28 days to 600 mg/d; 3, as per arm 2 + aspirin 1000 mg/d) for 56 days, when the primary outcome of adverse events of special interest (AESI) or death was assessed.

Results: A total of 52 participants with HIV-associated TBM were randomized; 59% had mild disease (British Medical Research Council (MRC) grade 1) vs 39% (grade 2) vs 2% (grade 3). AESI or death occurred in 10 of 16 (63%; arm 3) vs 4 of 14 (29%; arm 2) vs 6 of 20 (30%; arm 1; P = .083). The cumulative proportion of AESI or death (Kaplan-Meier) demonstrated worse outcomes in arm 3 vs arm 1 (P = .04); however, only 1 event in arm 3 was attributable to aspirin and was mild. There was no difference in efficacy (modified Rankin scale) between arms.

Conclusions: High-dose rifampicin and adjunctive linezolid can safely be added to the standard of care in HIV-associated TBM. Larger studies are required to determine whether potential toxicity associated with these interventions, particularly high-dose aspirin, is outweighed by mortality or morbidity benefit.

Clinical trials registration: NCT03927313.

Keywords: HIV; aspirin; linezolid; rifampicin; tuberculous meningitis.

Figure 1.

Consolidated Standards of Reporting Trials diagram describing recruitment and arm allocation. One participant was randomized but excluded prior to any study investigational product (IP) being dispensed due to emergence of an exclusion criterion (estimated glomerular filtration rate <20) on a hospital blood test performed prior to randomization. Another participant was excluded from the modified intention-to-treat analysis as they died prior to receiving any dose of study drug. Six participants discontinued the study prior to day 56, and an additional 4 participants discontinued between day 56 and day 180 (Supplementary Table 2). Reasons for screening exclusions and early study withdrawals are listed in Supplementary Tables 1 and 2. *Patient randomized but withdrawn prior to receiving study IP due to emergence of exclusion criteria. Abbreviations: IV, intravenous; PO, by mouth; Rif, rifampicin.

Figure 2.

Time to worst-grade adverse events of special interest (AESI) or death. Kaplan–Meier analysis of time to worst grade AESI or death, comparing arms 1, 2, and 3 (A), arm 2 vs arm 1 (B), arm 3 vs arm 1 (C), and arms 2 and 3 combined vs arm 1 (D). Arm 3 vs arm 1 (C) demonstrated a significant P value (P = .043); however, all other comparisons showed no significant difference between arms.

Figure 3.

Time to death. Kaplan–Meier analysis of time to death, comparing arms 1, 2, and 3 (A), arm 2 vs arm 1 (B), arm 3 vs arm 1 (C), and arms 2 and 3 combined vs arm 1 (D). Note that no analysis demonstrates a statistically significant difference between arms.

Figure 4.

Time to adverse events of special interest (AESI). Kaplan–Meier analysis of time to AESI, comparing arms 1, 2, and 3 (A) and arms 2 and 3 combined vs arm 1 (B). Note that comparison of time to AESI in arm 3 vs arm 1 demonstrated a statistically significant difference (P = .02); however, arm 2 vs arm 1 and arms 2 and 3 combined vs arm 1 were not significantly different (P values of .84 and .18, respectively).

Figure 5.

Functional outcome at day 56 as defined by modified Rankin scale (MRS). A, Comparison between good outcome (MRS, 0–3) vs bad outcome stratified by arm at day 56 with no significant difference between the 3 treatment arms (P = .616). B, Change in MRS between enrollment and day 56 across treatment arms with no statistically significant difference between the 3 treatment arms (P = .611).

Figure 6.

Change in cerebrospinal fluid (CSF) parameters over time. Spaghetti plots for CSF parameters, lymphocyte count (A), polymorphonuclear cells (B), glucose (C), and protein (D), plotted as individual values over time (faint lines), with mean values for each treatment arm represented by superimposed line (bold lines). Individual values are plotted, and the superimposed bold line represents the mean values at each time point in each treatment arm as per the color key. The t tests comparing mean and variance at each time point demonstrated no difference between arms.