Combined cerebrospinal fluid metabolomic and cytokine profiling in tuberculosis meningitis reveals robust and prolonged changes in immunometabolic networks

Much of the high mortality in tuberculosis meningitis (TBM) is attributable to excessive inflammation, making it imperative to identify targets for host-directed therapies that reduce pathologic inflammation and mortality. In this study, we investigate how cytokines and metabolites in the cerebral spinal fluid (CSF) associate with TBM at diagnosis and during TBM treatment. At diagnosis, TBM patients (n = 17) demonstrate significant increases of cytokines and chemokines that promote inflammation and cell migration including IL-17A, IL-2, TNFα, IFNγ, and IL-1β versus asymptomatic controls without known central nervous system pathology (n = 20). Inflammatory immune signaling had a strong positive correlation with immunomodulatory metabolites including kynurenine, lactic acid, and carnitine and strong negative correlations with tryptophan and itaconate. Inflammatory immunometabolic networks were only partially reversed with two months of effective TBM treatment and remained significantly different compared to CSF from controls. Together, these data highlight a critical role for host metabolism in regulating the inflammatory response to TBM and indicate the timeline for restoration of immune homeostasis in the CSF is prolonged.

Pharmacometabolomics in TB Meningitis - understanding the pharmacokinetic, metabolic, and immune factors associated with anti-TB drug concentrations in cerebrospinal fluid

Poor penetration of many anti-tuberculosis (TB) antibiotics into the central nervous system (CNS) is thought to be a major driver of morbidity and mortality in TB meningitis (TBM). While the amount of a particular drug that crosses into the cerebrospinal fluid (CSF) varies from person to person, little is known about the host factors associated with interindividual differences in CSF concentrations of anti-TB drugs. In patients diagnosed with TBM from the country of Georgia (n=17), we investigate the association between CSF concentrations of anti-TB antibiotics and multiple host factors including serum drug concentrations and CSF concentrations of metabolites and cytokines. We found >2-fold differences in CSF concentrations of anti-TB antibiotics from person to person for all drugs tested including cycloserine, ethambutol, imipenem, isoniazid, levofloxacin, linezolid, moxifloxacin pyrazinamide, and rifampin. While serum drug concentrations explained over 40% of the variation in CSF drug concentrations for cycloserine, isoniazid, linezolid, and pyrazinamide (adjusted R 2 >0.4, p<0.001 for all), there was no evidence of an association between serum concentrations of imipenem and ethambutol and their respective CSF concentrations. CSF concentrations of carnitines were significantly associated with concentrations of ethambutol and imipenem (q<0.05), and imipenem was the only antibiotic significantly associated with CSF cytokine concentrations. These results indicate that there is high interindividual variability in CSF drug concentrations in patients treated for TBM, which is only partially explained by differences in serum drug concentrations and not associated with concentrations of cytokines and chemokines in the CSF.

Combined cerebrospinal fluid metabolomic and cytokine profiling in tuberculosis meningitis reveals robust and prolonged changes in immunometabolic networks

Much of the high mortality in tuberculosis meningitis (TBM) is attributable to excessive inflammation, making it imperative to identify targets for host-directed therapies that reduce pathologic inflammation and mortality. In this study, we investigate how cytokines and metabolites in the cerebral spinal fluid (CSF) associate with TBM at diagnosis and during TBM treatment. At diagnosis, TBM patients demonstrate significant increases versus controls of cytokines and chemokines that promote inflammation and cell migration including IL-17A, IL-2, TNFα, IFNγ, and IL-1β. Inflammatory immune signaling was strongly correlated with immunomodulatory metabolites including kynurenine, lactic acid, carnitine, tryptophan, and itaconate. Inflammatory immunometabolic networks were only partially reversed with two months of effective TBM treatment and remained significantly different versus control CSF. Together, these data highlight a critical role for host metabolism in regulating the inflammatory response to TBM and indicate the timeline for restoration of immune homeostasis in the CSF is prolonged.

Cerebrospinal fluid concentrations of fluoroquinolones and carbapenems in tuberculosis meningitis

Background: Tuberculosis meningitis (TBM) is the most lethal form of TB. It is difficult to treat in part due to poor or uncertain drug penetration into the central nervous system (CNS). To help fill this knowledge gap, we evaluated the cerebrospinal fluid (CSF) concentrations of fluoroquinolones and carbapenems in patients being treated for TBM. Methods: Serial serum and CSF samples were collected from hospitalized patients being treated for TBM. CSF was collected from routine lumbar punctures between alternating timepoints of 2 and 6 h after drug administration to capture early and late CSF penetration. Rich serum sampling was collected after drug administration on day 28 for non-compartmental analysis. Results: Among 22 patients treated for TBM (8 with confirmed disease), there was high use of fluoroquinolones (levofloxacin, 21; moxifloxacin, 10; ofloxacin, 6) and carbapenems (imipenem, 11; meropenem, 6). Median CSF total concentrations of levofloxacin at 2 and 6 h were 1.34 mg/L and 3.36 mg/L with adjusted CSF/serum ratios of 0.41 and 0.63, respectively. For moxifloxacin, the median CSF total concentrations at 2 and 6 h were 0.78 mg/L and 1.02 mg/L with adjusted CSF/serum ratios of 0.44 and 0.62. Serum and CSF concentrations of moxifloxacin were not affected by rifampin use. Among the 76 CSF samples measured for carbapenem concentrations, 79% were undetectable or below the limit of detection. Conclusion: Fluoroquinolones demonstrated high CSF penetration indicating their potential usefulness for the treatment of TBM. Carbapenems had lower than expected CSF concentrations.

Cycloserine and Linezolid for Tuberculosis Meningitis: Pharmacokinetic Evidence of Potential Usefulness

BACKGROUND

The ability of antituberculosis drugs to cross the blood-brain barrier and reach the central nervous system is critical to their effectiveness in treating tuberculosis meningitis (TBM). We sought to fill a critical knowledge gap by providing data on the ability of new and repurposed antituberculosis drugs to penetrate into the cerebrospinal fluid (CSF).

METHODS

We conducted a clinical pharmacology study among patients treated for TBM in Tbilisi, Georgia, from January 2019 until January 2020. Serial serum and CSF samples were collected while patients were hospitalized. CSF was collected from routine lumbar punctures with the timing of the lumbar puncture alternating between 2 and 6 hours to capture early and late CSF penetration.

RESULTS

A total of 17 patients treated for TBM (8 with confirmed disease) were included; all received linezolid, with a subset receiving cycloserine (5), clofazimine (5), delamanid (4), and bedaquiline (2). All CSF measurements of bedaquiline (12), clofazimine (24), and delamanid (19) were below the limit of detection. The median CSF concentrations of cycloserine at 2 and 6 hours were 15.90 and 15.10 µg/mL with adjusted CSF/serum ratios of 0.52 and 0.66. CSF concentrations of linezolid were 0.90 and 3.14 µg/mL at 2 and 6 hours, with adjusted CSF/serum ratios of 0.25 and 0.59, respectively. CSF serum linezolid concentrations were not affected by rifampin coadministration.

CONCLUSIONS

Based on moderate to high CSF penetration, linezolid and cycloserine may be effective drugs for TBM treatment, whereas the utility of bedaquiline, delamanid, and clofazimine is uncertain given their low CSF penetration.

Long term outcomes of patients with tuberculous meningitis: The impact of drug resistance

BACKGROUND

Little is known about the impact of drug-resistance on clinical outcomes among patients with tuberculosis meningitis (TBM).

METHODS

A retrospective cohort study among patients treated for TBM in Tbilisi, Georgia. We performed medical chart abstraction to collect patient data. Long-term vital status was assessed using the Georgia National Death Registry. We utilized a Cox proportional-hazards model to evaluate the association of drug-resistance and mortality.

RESULTS

Among 343 TBM suspects, 237 had a presentation consistent with TBM. Drug resistance was suspected (n = 5) or confirmed (n = 31) in 36 patients including 30 with multidrug- or rifampin-resistance and 6 with isoniazid-resistance. Thirty-four patients had HIV. The median follow-up time was 1331 days (IQR, 852-1767). Overall, 73 of 237 (30%) people died with 50 deaths occurring during and 23 after treatment. The proportion of death was higher among patients with drug-resistant vs. drug-susceptible disease (67% vs. 24%, p<0.001) and with HIV versus no HIV (59% vs 27%, p<0.001). Mortality was significantly higher in patients with drug-resistant TBM after 90 days of treatment (aHR = 7.2, CI95% [3.6-14.3], p < 0.001).

CONCLUSIONS

Mortality was high among patients with drug-resistant TBM with many deaths occurring post treatment. More effective treatment options are urgently needed for drug-resistant TBM.

Clinical outcomes among patients with tuberculous meningitis receiving intensified treatment regimens

SETTING: National Center for Tuberculosis and Lung Diseases (NCTLD), Tbilisi, Georgia.OBJECTIVE: To determine clinical outcomes of patients with tuberculous meningitis (TBM) treated with an intensified regimen including a fluoroquinolone (FQ) and an injectable agent.DESIGN: Prospective cohort of patients aged ≥16 years initiating treatment for TBM at the NCTLD from January 2018 to December 2019. Treatment outcomes and neurologic disability at 1, 6 and 12 months after treatment initiation were assessed.RESULTS: Among 77 patients with median follow-up time of 363 days (IQR 269-374), 97% received a FQ, 62% an injectable agent, 44% linezolid and 39% a carbapenem. Fifty-seven patients (74%) successfully completed treatment, 2 (2.6%) had treatment failure, 6 (7.8%) died, and the remainder (12%) were lost to follow up. Among 11 patients treated for multidrug-resistant TBM, the median follow-up time was 467 days and one patient (8%) died. Regarding neurologic outcomes, 14/76 (18%) patients had Modified Rankin Scores of 0 at baseline, improving to 85% (56/66) and 94% (47/50) at 6 and 12 months, respectively.CONCLUSION: Intensified multidrug treatment regimens including a FQ and an injectable agent in all patients and newly implemented drugs in patients with multidrug-resistant TBM resulted in low mortality and favorable neurologic outcomes.