Abstract
Isoniazid (INH) is a prodrug activated by the mycobacterial enzyme KatG, a multifunctional catalase peroxidase. KatG converts INH to reactive antimycobacterial species. For decades, an association between INH and drug-induced lupus erythematosus has been recognized. We present the case of a patient with primary progressive multiple sclerosis whose disease commenced weeks after initiating INH therapy for prevention of tuberculosis. Possible mechanisms by which INH may trigger autoimmunity in humans are discussed.
Case report
This is a case of a 39-year-old man with primary progressive multiple sclerosis (PPMS) who was originally diagnosed in 2008. His neurological symptoms started after isoniazid (INH) therapy was initiated for prevention of tuberculosis (TB) after he had a positive purified protein derivative (PPD) tuberculin skin test. Within weeks, the patient developed fatigue, a tremor of his neck, arms, and legs, and heat sensitivity. At an outside hospital, he was initially diagnosed with drug-induced lupus erythematosus (DILE). However, his progressive accumulation of neurological disability without disease relapses or periods of remission led to further diagnostic testing, which included magnetic resonance imaging (MRI) of his brain and spinal cord, as well as cerebrospinal fluid (CSF) analyses. Eventually, the patient met diagnostic criteria for PPMS [Poser et al. 1983]: 1 year of disease progression, determined retrospectively; multiple T2-weighted lesions on MRI in the periventricular cerebral white matter (Figure 1); multiple T2-weighted MRI lesions in the cervical spinal cord (Figure 2); and oligoclonal bands in the CSF. Mimics of multiple sclerosis (MS) were considered during his evaluation, and the following laboratory tests were negative or within normal limits: anti aquaporin-4 antibody, antinuclear antibodies, hemoglobin A1C, human immunodeficiency virus 1 and 2, rheumatoid factor, thyroid stimulating hormone, and a vitamin B12 level. A chest X-ray was read as normal.

Magnetic resonance imaging (MRI) scans of the brain are consistent with a diagnosis of multiple sclerosis (MS). (A) Axial T2-weighted brain MRI scan shows periventricular ovoid lesions, classic for MS (arrows). (B) Sagittal T1-weighted brain MRI scan shows a persistent black hole, a radiographic sign suggestive of axonal degeneration.

Magnetic resonance imaging (MRI) scan of the cervical spinal cord is consistent with a diagnosis of multiple sclerosis. Sagittal T2-weighted MRI scan of the cervical spinal cord shows extensive abnormal cord signal and atrophy.
At the time of his first assessment in our clinic, his Expanded Disability Status Scale (EDSS) score was 8.0 [Kurtzke, 1983]. The patient was unable to ambulate, and he reported urinary incontinence and erectile dysfunction. He reported intermittent dysphagia, and intermittent muscle spasms in his legs, as well as a resting tremor in his arms and neck. On neurological examination, there was a bilateral intermittent gaze-associated pendular nystagmus, one-fifth strength in the lower extremities, a 4–5-beat clonus over both ankles, a coarse intermittent irregular holding tremor of the hands, and a bilateral intention tremor. The patient sat in a motorized wheelchair.
Isoniazid and autoimmunity
INH, more than half a century after its discovery, is still the centerpiece of treatment for TB. INH is a prodrug activated by the mycobacterial enzyme KatG, a multifunctional catalase peroxidase. KatG converts INH to reactive antimycobacterial species [Timmins and Deretic, 2006]
INH has been associated with DILE. Approximately one-fifth of the patients treated with INH for more than 6 months develop antinuclear antibodies. Arthralgia, anemia, fever, pleuritis and pericarditis are common manifestations of DILE [Vasoo, 2006]. An antibody against histone (H2A-H2B)–DNA complex is a specific marker of INH-induced lupus erythematosus [Vázquez-Del Mercado et al. 1995].
Putative contribution of INH to the pathogenesis of systemic lupus erythematosus and MS
The mechanism by which this medication induces specific clinical and serological manifestations is unclear. Initially, haptenization, binding of a small molecule to a macromolecule and thus rendering it immunogenic, was postulated as a common mechanism of drug-induced autoimmunity [Chang and Gershwin, 2011]. However, there is accumulating evidence that different mechanisms lead to loss of central or peripheral tolerance to self-antigens, and environmental exposure and host characteristics modulate these events [Chang and Gershwin, 2011]. In the case of INH, it was demonstrated that it is oxidized by activated leukocytes to isonicotinic acid [Timmins and Deretic, 2006]. It is speculated that reactive intermediates produced by activated leukocytes might have implications in the pathogenesis of INH-induced lupus. Kucharz and Sierakowski studied the immunomodulatory properties of INH more than 20 years ago [Kucharz and Sierakowski, 1990]. INH at different concentrations has opposing effects on the proliferation and viability of CD3+ mononuclear cells. Specifically, interleukin-2 (IL-2) production by T cells and phorbol myristate induced proliferation of T cells appear to be dose dependent. It was also shown that INH at low concentrations increased the proliferation of IL-2 receptor-negative cells in vitro.
T lymphocytes very likely play a critical role in many forms of drug-induced lupus. For example, it was demonstrated that hydralazine can inhibit ERK pathway signaling and subsequent DNA hypomethylation, which then leads to lymphocyte function associated antigen 1 overexpression and resultant T-cell autoreactivity [Deng et al. 2003]. Anti-TNFα antibodies (e.g. etanercept and infliximab) have been reported to induce central nervous system (CNS) demyelination or worsen the preexisting demyelinating disease [Kaltsonoudis et al. 2014]. These monoclonal antibodies can induce a drug-induced lupus syndrome with clinical and laboratory features different from traditional drug-induced lupus, including a higher incidence of rash, anti-DNA antibodies, leukopenia and thrombocytopenia, and a lower incidence of antihistone antibodies [Chang and Gershwin, 2011].
Interestingly, an association between INH and MS in the literature goes back to the reports of using this medication in the symptomatic therapy of MS, particularly cerebellar tremors [Sabra et al. 1982]. While later studies refuted any beneficial claims [Koller, 1984], there was no mention of worsening of the disease by the INH. However, given that MS is a relentlessly progressing disease in many patients, it would be difficult to ascertain an additive effect of INH. To the best of our knowledge, there is no report in the literature about INH inducing MS.
MS is considered a T-cell-driven inflammatory disease of the CNS [Frohman et al. 2006]. The strongest genetic association of MS is with HLA-DRB1*15:01 [Haines et al. 1998]. It is conceivable that INH, in a genetically susceptible individual, may activate CNS autoantigen-reactive CD4+ T cells and ultimately initiate a relapsing demyelinating disease. As stated above, our patient was diagnosed with PPMS almost immediately after INH therapy was initiated. An estimated 10–15% of patients with MS display this clinical phenotype [Lublin and Reingold, 1996]. The clinical progression in PPMS is considered to be more rapid than in other MS subgroups [Cottrell et al. 1999]. While it is thought that systemic inflammatory events are less relevant than in relapsing forms of MS, recent histological studies showed that meningeal and perivascular inflammatory leukocyte infiltrates are abundantly present, and that meningeal inflammation is closely associated with cortical demyelination [Choi et al. 2012].
A potential association of INH and MS disease activity may be more relevant in geographical areas in which the prevalence of TB is high, and in which patients with MS and concomitant TB may be exposed to INH therapy. It would be prudent to monitor MS disease activity in such patients very closely.
Footnotes
Funding
Olaf Stuve is an associate editor of JAMA Neurology, and he serves on the editorial boards of the Multiple Sclerosis Journal, Clinical and Experimental Immunology, and Therapeutic Advances in Neurological Disorders. He has participated in data and safety monitoring committees for Pfizer and Sanofi. Dr. Stuve has received grant support from Teva Pharmaceuticals.
Conflict of interest statement
The authors declare that there is no conflict of interest.
