Although tuberculosis is now considered a treatable disease, central nervous system tuberculosis (CNS-TB) when managed with the current standard-of-care (SOC), still has mortality rates ranging from 30-50% even in tertiary hospital centers. At present, the SOC for the management of CNS-TB is anti-tuberculous therapy with adjunctive corticosteroids. In CNS-TB, the activity of pathogenic host matrix metalloproteinases (MMPs) is unopposed to tissue inhibitors of metalloproteinases (TIMPs), resulting in a matrix-degrading phenotype which may drive worse outcomes in CNS-TB. In a prior established CNS-TB murine model, the investigators have demonstrated that adjunctive MMP inhibition using doxycycline, a widely available and cheap drug, in addition to standard TB treatment, compared with standard TB treatment alone, improved murine survival (Manuscript in preparation). The investigators previously showed that in humans with pulmonary TB, doxycycline with anti-TB treatment is safe, accelerates the resolution of inflammation, and suppresses systemic and respiratory MMPs. Hence, the investigators are now ideally positioned to determine if adjunctive doxycycline in patients with CNS-TB can improve clinical outcomes. The investigators will perform a Phase 2 double-blind randomized-controlled trial (RCT) of adjunctive doxycycline versus placebo with standard TB treatment and steroids for 8 weeks, with the primary outcome of 8-week mortality or severe neurological deficits.

The investigators hypothesize that in patients with CNS-TB, the addition of doxycycline to SOC improves clinical outcomes.

The specific aims are to determine:

1. Whether the primary outcome of mortality rates or severe neurological deficits in CNS-TB patients in the doxycycline arm improves at 8 weeks. Neurological deficits will be defined using a modified Rankin scale score of 3 or higher to measure the degree of disability or dependence in the daily activities of people who have suffered from neurological disability.
2. Whether MMPs and inflammatory gene transcriptomics in the doxycycline arm are reduced. Plasma and cerebrospinal fluid (CSF) MMPs and TIMPs will be measured by luminex bead array technology. The functional activity of CSF MMPs will be assessed using Dye-quenched (DQ) Gelatin degradation. Inflammatory gene transcripts will be measured using whole-blood bulk RNA sequencing and single-cell RNA sequencing. CSF will also be profiled using single-cell RNA sequencing.
3. Whether concurrent SOC treatment and doxycycline for CNS-TB are safe. The investigators will monitor patients for side effects including liver function tests to evaluate any significant change in the safety profile of the patients after administration of adjunctive doxycycline, in comparison to the placebo arm. Standard measures such as grade 3 or 4 adverse events or serious adverse events will be determined.

These specific aims will determine if doxycycline will improve clinical outcomes in the management of CNS-TB, reduce long-term neurological deficits whilst demonstrating a comparable safety profile. These aims are critical to form the basis of a large-scale Phase 3 randomised-controlled trial of CNS-TB which will positively impact clinical practice and international guidelines.

Background and Clinical Significance

The global burden of tuberculosis and TB sequelae

Despite being declared an emergency by the World Health Organisation (WHO) in 1993, tuberculosis (TB) remains a global health epidemic. A significant proportion of the global population is infected with TB, particularly in lower- to middle-income countries in resource-limited settings. Despite strategies directed towards infection control and TB management, worldwide annual TB deaths are estimated to be 1.3 million. Furthermore, the coronavirus disease 2019 (COVID-19) pandemic has also impacted TB, and TB incidence is expected to rise substantially. In Singapore, TB incidence has stagnated after some significant improvements with an annual incidence of approximately 39 TB infections per 100,000 people, which translates to under 3,000 TB patients per year. Additionally, there are rising multi-drug-resistant (MDR) and extensively drug-resistant (XDR) TB regionally and in the world, which pose clinical challenges. MDR and XDR TB often require prolonged treatment courses with medications that are often unavailable in resource-limited settings. Singapore is not exempted from the scourge of MDR and XDR TB cases, which have been described in prisons, gaming arcades and even housing estates. Global and local epidemiology indicates that TB remains an infectious disease threat.

One of the most severe forms of TB is CNS-TB which can lead to significant morbidity and mortality. In CNS-TB, tissue destruction can result in long-term neurological sequelae with permanent neurological deficits. Even with appropriate treatment for CNS-TB and microbiological cure, mortality rates for CNS-TB remain high at up to 30-50%; CNS-TB may still leave behind residual neurological deficits. Patients can become bed-bound and dependent on activities of daily living. Even with good standard-of-care, there still is cognitive impairment in up to 55% of patients and motor deficits in up to 40% of patients, with significant residual neurological deficits. Adjunctive steroids significantly reduce mortality (41.3% placebo vs 31.8% in the steroid arm), but death rates remain unacceptably high. Critically, steroids did not show a significant improvement in severe disability from neurological deficits. TB sequelae cause substantial morbidity long after TB itself is treated, as currently there is no treatment to prevent severe neurological disability.

Host immunopathology and TB tissue destruction

Prior published work has shown that the cause of tissue destruction in TB resulting in sequelae is excessive host inflammatory and adaptive immune response. Although the immunopathology in TB is not fully understood, it has been shown that pro-inflammatory cytokines such as tumour necrosis factor-alpha (TNF-α) may play a key role. Proteases result in tissue destruction in TB patients, with MMPs having a unique ability to collectively degrade all extracellular matrix fibrils at a neutral power of hydrogen (pH), of which the secretion of MMPs in TB is excessive. MMP activity is dysregulated in TB, which is the key in the pathophysiology. The secretion of multiple MMPs are up-regulated, arising from human macrophages, epithelial cells and neutrophils which we showed. The upregulation of MMPs with tissue destruction has been shown in several TB animal models, which provides further evidence on the role of pathogenic MMPs in vivo.

Doxycycline is the only MMP inhibitor licensed by the U.S. Food and Drug Administration which has both anti-bacterial and immunomodulatory properties. It is an antibiotic that is used for decades for a range of clinical indications, including periodontal disease (gum disease), acting on periodontal clefts by reducing the concentration of collagenases. Doxycycline has also been shown to be bacteriostatic to Mycobacterium tuberculosis (Mtb). It has an excellent safety profile, even when given for prolonged durations. It may also be administered by nasogastric tube for individuals with swallowing disability. Furthermore, it is a drug that is widely available and cheap, making it highly accessible even in resource-limited settings. The addition of doxycycline adjunctive therapy to corticosteroids and standard TB therapy may decrease host MMP expression and inflammatory gene responses in human CNS-TB, hence improving clinical outcomes and neurological sequelae.

Doxycycline modulates TB tissue destruction

The first doxycycline pulmonary TB trial worldwide was conducted by us. In a Phase 2 double-blind randomised-controlled trial of 30 pulmonary TB patients, investigating doxycycline host-directed therapy funded by the National Medical Research Council, the investigators showed that 2 weeks of doxycycline improved pulmonary cavity resolution as indicated by the patients exhibiting significantly smaller cavities. In addition, there was also a beneficial host immunological effect that persisted 6 weeks post-discontinuation of the doxycycline, accelerating a return to healthy gene expression of the host transcriptome. MMPs were also down-regulated with suppression of plasma MMP-1 and further reduction of collagenase and gelatinase MMPs in the sputum. However, the effects of doxycycline on mortality or neurological outcomes in CNS-TB are unknown. This Phase 2 randomised-controlled-trial determines if adjunctive doxycycline treatment can be extended to CNS-TB to improve clinical outcomes and neurological sequelae. Reducing neurological disability would alleviate burden on healthcare facilities and maximise an individual's economically productive life. If positive, these findings will provide preliminary data for a Phase 3 RCT. Findings will be rapidly conveyed to the WHO with whom the investigators closely collaborate, to impact international guidelines and clinical practice.

Specific Aims and Hypothesis

The investigators hypothesize that in patients with CNS-TB, the addition of doxycycline to SOC improves clinical outcomes.

The specific aims are to determine:

1. Whether the primary outcome of mortality rates or severe neurological deficits in CNS-TB patients in the doxycycline arm improves at 8 weeks. Neurological deficits will be defined using a modified Rankin scale score of 3 or higher to measure the degree of disability or dependence in the daily activities of people who have suffered from neurological disability.
2. Whether MMPs and inflammatory gene transcriptomics in the doxycycline arm are reduced. Plasma and cerebrospinal fluid (CSF) MMPs and TIMPs will be measured by luminex bead array technology. The functional activity of CSF MMPs will be assessed using Dye-quenched (DQ) Gelatin degradation. Inflammatory gene transcripts will be measured using whole-blood bulk RNA sequencing and single-cell RNA sequencing. CSF will also be profiled using single-cell RNA sequencing.
3. Whether concurrent SOC treatment and doxycycline for CNS-TB are safe. The investigators will monitor patients for side effects including liver function tests to evaluate any significant change in the safety profile of the patients after administration of adjunctive doxycycline, in comparison to the placebo arm. Standard measures such as grade 3 or 4 adverse events or serious adverse events will be determined.

These specific aims will determine if doxycycline will improve clinical outcomes in the management of CNS-TB, reduce long-term neurological deficits whilst demonstrating a comparable safety profile. These aims are critical to form the basis of a large-scale Phase 3 randomised-controlled trial of CNS-TB which will positively impact clinical practice and international guidelines.

Inclusion Criteria:

1. Aged 21 years and above.

2. Patients receiving ≤ 7 days of TB treatment or about to start combination TB treatment, including injectable agents, where required.

3. Patients with clinical evidence of TB meningitis, as per established diagnostic criteria, defined as either definite, probable or possible CNS-TB:

   1. "Definite" CNS-TB would be defined if acid-fast bacilli (AFB) or a positive nucleic acid amplification test for M. tuberculosis in the cerebrospinal fluid of patients.
   2. "Probable" CNS-TB would be defined if the patient exhibit one or more of the following: suspected pulmonary tuberculosis on chest radiography, acid-fast bacilli found in any specimen other than the cerebrospinal fluid or clinical evidence of other extrapulmonary tuberculosis.
   3. "Possible" CNS-TB would be defined if the patients exhibit at least four of the following: a history of tuberculosis, predominance of lymphocytes in the cerebrospinal fluid, a duration of illness of more than five days, a ratio of cerebrospinal fluid glucose to plasma glucose of less than 0.5, altered consciousness, yellow cerebrospinal fluid, or focal neurologic signs.

4. Alanine aminotransferase (ALT) level < 3 times the upper limit of normal.

5. Able to provide informed consent. If the patient has no mental capacity to give consent, then consent may be provided for by the patient's next of kin.

6. Lumbar puncture and brain imaging (either computed tomography or magnetic resonance imaging, with or without contrast) is required at baseline for enrolment

Exclusion Criteria:

1. Active Cancer
2. Pregnant or breastfeeding
3. Allergies to tetracyclines
4. Patients on retinoic acid, neuromuscular blocking agents or pimozide which may increase the risk of drug toxicity.
5. Autoimmune disease and/or on systemic immunosuppressants.
6. Use of any investigational or non-registered drug, vaccine or medical device other than the study drug within 182 days preceding dosing of the study drug or planned use during the study period.
7. Enrolment in any other clinical trial involving a systemic drug or intervention involving the CNS.
8. Contraindications to the use of steroids.
9. Investigators' assessment of lack of willingness to participate and comply with all requirements including follow-up of the protocol or identification of any factor presumed to significantly increase the participant's risk of suffering an adverse outcome.

Data may be made available on reasonable request

Double-blinded