Clinical Rheumatology

, Volume 29, Issue 10, pp 1201–1204

Cerebral tuberculoma in a patient receiving anti-TNF alpha (adalimumab) treatment

Authors

    • Department of NeuropathologyBeaumont Hospital
    • Department of Neurology Tufts UniversitySt Elizabeth’s Medical Center
  • Michael Farrell
    • Department of NeuropathologyBeaumont Hospital
Case Report

DOI: 10.1007/s10067-010-1466-7

Cite this article as:
Lynch, K. & Farrell, M. Clin Rheumatol (2010) 29: 1201. doi:10.1007/s10067-010-1466-7

Abstract

We report a case of a cerebral tuberculoma in a 60-year-old woman with rheumatoid arthritis while receiving the anti-tumor necrosis factor alpha monoclonal antibody, adalimumab (Humira), for active disease. MR brain imaging for dyspraxia revealed a left parietal ring-enhancing lesion, which on resection was shown to be a necrotizing granuloma. There were no associated pulmonary lesions, and the patient was systemically well. Sputum and urine cultures were negative for tuberculosis. The patient was treated with anti-tuberculous medications and made an excellent recovery. We consider this to be the first documented case of tuberculosis involving the central nervous system occurring in the setting of adalimumab treatment.

Keywords

AdalimumabAnti-TNFCerebralRheumatoid arthritis.Tuberculoma

Abbreviations

CDC

Centers for Disease Control

TNF

Tumor necrosis factor

ZN

Ziehl Nielson

GFAP

Glial fibrillary acidic protein

PPD

Purified protein derivative

RA

Rheumatoid arthritis

DMARDs

Disease-modifying anti-rheumatic drugs

LTBI

Latent tuberculous bacilli infection

IGRA

Interferon-gamma release assay.

Introduction

Rheumatoid arthritis (RA) is a chronic, progressive, systemic inflammatory disease of undetermined etiology involving primarily the synovial membranes and articular structures of joints. Clinically, it may result in significant pain, stiffness, and swelling of joints with subsequent deformity and ankylosis in later stages of the disease. Early administration of disease-modifying anti-rheumatic drugs (DMARDs), the use of non-steroidal anti-inflammatory drugs, and glucocorticoids are all essential in the medical management of RA. With the discovery of the role of cytokine tumor necrosis factor (TNF) alpha in the mechanism of inflammation in RA, anti-TNF alpha drugs, a new biological class of DMARDs was introduced. Three of these medications have been commercialized since 1999 for use in RA patients: infliximab, etanercept, and adalimumab. Since their introduction, there have been numerous studies and meta-analysis comparing this new class of drug from efficacy to safety profile [1]. In recent years, the emphasis has been on the risk of malignancy and serious infection (tuberculosis, histoplasmosis, listeriosis, candidiasis, and aspergillosis), with a number of comparative studies being conducted in the US and Europe [1]. Data from these studies shows clearly that reactivation of latent TB infection (LTBI) has been the most commonly reported adverse effect of anti-TNF inhibitors. This susceptibility to mycobacterial infections, particularly reactivation of LTBI, is due to the role that TNF alpha plays in containing and ensuring the preservation of the granuloma structure [1, 5].

Among the TNF alpha inhibitors used in RA, there are differences in the risk of TB reactivation. Infliximab has been shown to have a higher rate of reactivation compared to etancercept, and data so far on adalimumab suggest that its risk is similar to that of infliximab [9]. The increased risk associated with infliximab and adalimumab may be due to the combined biological effect of these agents on both CD4 TB-responsive cells and their known suppression of interferon-gamma production. Etanercept on the other hand has no great effect on these parameters.

We describe a patient with RA, who presented with central neurological symptoms while receiving adalimumab treatment and on investigation was found to have a cerebral tuberculoma. There have been an unusually large number of cases of tuberculosis, predominantly in the miliary form, reported in patients on anti-TNF alpha drugs, particularly infliximab, but we present the first documented case of a cerebral tuberculoma without systemic involvement following adalimumab treatment.

Case report

A 60-year-old woman with a 19-year history of rheumatoid arthritis was investigated for a 2-week history of left neck pain and right hand dyspraxia. She also had an associated sensation of pressure behind the left eye, with minimal intermittent headaches. Medications included methotrexate 15 mg/week for 8 years and adalimumab (Humira) 40 ml taken fortnightly for 2 years. Her arthritis activity had been quiescent for over 2 years, and apart from raised serum inflammatory markers, she was systemically well on presentation. Magnetic resonance imaging (MRI) of the brain showed a left parietal ring-enhancing lesion, with surrounding vasogenic edema (Fig. 1) and the patient proceeded to resection.
https://static-content.springer.com/image/art%3A10.1007%2Fs10067-010-1466-7/MediaObjects/10067_2010_1466_Fig1_HTML.jpg
Fig. 1

Coronal T1-weighted MRI brain post-gadolinium demonstrating ring-enhancing left parietal lesion

Microscopic examination of the resection specimen revealed a large central area of necrosis surrounded by fibrocollagenous connective tissue, lymphocytes, and multinucleate giant cells with reactive gliosis (Fig. 2). Immunocytochemistry for CD3 confirmed T-lymphocytes at the periphery of the lesion, and ZN stain was negative for tuberculous bacilli. Appearances were consistent with a cerebral tuberculoma. Further ZN stain and culture of sputum and urine as well as polymerase chain reaction (PCR) and acid fast bacilli (AFB) staining of CSF were negative for Mycobacterium tuberculosis. Prior to commencement of adalimumab, the patient had a negative screening PPD test and chest X-ray for latent TB.
https://static-content.springer.com/image/art%3A10.1007%2Fs10067-010-1466-7/MediaObjects/10067_2010_1466_Fig2_HTML.jpg
Fig. 2

Cerebral granuloma comprising of epithelioid cells, reactive astrocytes, and fibrous gliosis surrounding a central core of caseous necrosis

The patient was started on four antituberculosis medications (rifampicin, isoniazid, ethambutol, and pyrazinamide) and was discharged 1 week after admission. At a follow-up visit 6 weeks later, she had resolution of her initial neurological symptoms with significant reduction in size of the tuberculoma on MRI. She continued treatment for 12 months, and her rheumatoid arthritis is currently stable.

Discussion

Adalimumab (Humira) is the first and only human monoclonal anti-TNF alpha antibody approved for use in patients with active rheumatoid arthritis that have not responded adequately to standard anti-rheumatic therapies such as low dose corticosteroids, NSAIDs, and disease-modifying anti-rheumatic drugs. The immune-modulating effects of the three current TNF antagonists used clinically (etanercept, infliximab, and adalimumab) have been associated with various adverse effects including opportunistic infections such as TB, demyelinating disorders, systemic lupus erythematosus-like syndrome, and lymphoproliferative disorders [1].

There have been well-documented reports of TB in the setting of patients receiving anti-TNF alpha medications, with a higher likelihood of occurring in extrapulmonary sites or being in the disseminated/miliary forms [2]. The Adverse Event Reporting System (AERS), a database established by the Food and Drug Administration (FDA) for post-marketing adverse drug event surveillance, found that over the period of 1998–2002, rates of TB among US patients treated with infliximab and etanercept were 54 and 28 cases/100,000, respectively. The overall rate of TB in the US over the same period of time was 5.2–6.8 cases/100,000. Comparing data worldwide over the same period of time, there were a total of 70 cases of TB following treatment with infliximab, 64 of which were in countries with a low background incidence of TB. These cases occurred on average 4 months following infusion, and over half [3] had extrapulmonary disease, with one meningeal case described [4].

In Europe, several studies conducted in Spain and Portugal, where there is a high incidence of latent TB compared to the US, showed substantially higher rates of TB in patients receiving anti-TNF alpha inhibitors. The calculated incidence of TB with infliximab use was 1,900/100,000 patients in 2000, compared to the baseline incidence of TB in Spain of 21/100,000 inhabitants [6]. Susceptibility to reactivation of LTBI cited in a later AERS database [7] include co-administration of other immunosuppressive drugs and history of latent or active TB, being born in or spent time in a TB endemic region. It is also worth noting that RA alone can be associated with the development of TB [4]; however, TNF alpha inhibitors would appear to increase the risk beyond what would be expected to be related to RA alone.

The above studies have limitations in assessing the impact of these medications on TB risk because of overall decreasing overall rates of TB worldwide, varying TB notification rates by country, and a prescribing bias of anti-TNF alpha inhibitors. The absence of a gold standard test for latent TB infection also complicates assessing risk of reactivation of latent TB. We know that false positives of tuberculin skin testing (TST) and whole blood interferon-gamma release assay cannot distinguish successful immune system eradication or persistence of infection, and false negatives in patients with active RA or whilst receiving other immunosuppressive treatments complicate the assessment of the true impact of TNF alpha inhibitors on TB risk.

Among the TNF alpha inhibitors, there are differences in the risk of TB reactivation, with infliximab showing a higher risk compared to etancercept, and data so far on adalimumab suggest that its risk is similar to that of infliximab [9]. After the approval of adalimumab in 2003 by the European Medicines Agency and the initiation of routine TB screening prior to commencement of treatment, a total of 23 TB reactivation cases were reported (0.33/100 PYs) over a 3-year period in the drugs’ post marketing surveillance data [5, 10]. Other than the case we present here, there have been to date no documented cases of isolated cerebral tuberculoma with adalimumab and just two reports of a cerebral and cerebellar tuberculoma following infliximab [1, 8].

CNS tuberculosis overall accounts for almost 1% of TB cases worldwide, manifesting as tuberculous meningitis, and less commonly as tuberculous encephalitis, tuberculoma, and tuberculous brain abscess. Groups susceptible include children and HIV-infected patients, and risk factors include alcoholism, malignancy, and the concomitant use of immunosuppressive agents. Even with treatment, the mortality rate of CNS tuberculosis can vary from 20% to 80% depending on the time of presentation [12]. Deaths from CNS tuberculoma alone are more difficult to quantify probably due to the relative rarity of cases worldwide and lack of registries in the underdeveloped or developing countries where the majority of the cases occur.

The relative rarity of cases worldwide has meant that CNS TB remains a difficult diagnostic challenge. Traditional CSF analysis identifying AFB on smear and culture is the most widely used method for diagnosis; however, sensitivity is highly variable, in part due to the typical scarcity of AFB in CNS tuberculosis, their relative absence in tuberculoma cases, the volume of CSF collected, and time taken examining the sample. Molecular techniques used such as nucleic acid amplification (NAA) and other PCR-based methods, which accurately and rapidly diagnose TB in respiratory specimens, are being evaluated now for their applicability in the diagnosis of CNS tuberculosis. The typical lack of bacilli present in CSF in CNS tuberculosis and tuberculomas in particular, as well as the presence of amplification inhibitors in CSF, means that NAA and PCR methods have been shown to have a poor sensitivity of 56% and are not ideal for ruling out CNS tuberculosis [13]. One study confirmed this by comparing PCR assays in the diagnosis of tuberculomas. They demonstrated that immunohistochemical techniques outperformed PCR on surgically resected tuberculomas [14].

Clinical judgment remains paramount in the diagnosis of CNS tuberculosis even with negative diagnostic tests. Molecular techniques may not the best method in the diagnosis, but may be useful as an adjunct to traditional histology and culture or in monitoring response to treatment. For tuberculomas, MRI can further improve the diagnostic accuracy. Typical findings include solitary or multiple, round “nodules,” often located in the frontal or parietal lobes, with irregular walls and showing ring enhancement after contrast. Standard treatment of CNS tuberculosis is 2 months of isoniazid, rifampin, pyrazinamide, and ethambutol followed by 7–10 months of isoniazid and rifampin. The use of dexamethasone in addition is relatively controversial with some mortality benefit among children and non-HIV-infected adults [15]

It is assumed that most cases of TB in patients on anti-TNF inhibitors or other immunosuppressives are due to reactivation of latent infection and would therefore be preventable with treatment for LTBI. However, patients living or born in TB endemic countries could be also at risk for newly acquired infection. At present, there are limited case reports on the incidence of new TB due to TNF alpha inhibitors, and recent data from the AERS database would suggest that as more patients undergo TST and are treated for LTBI, it is possible that the difference in rates between infliximab and etanercept will be reduced in regions with high TB prevalence and that time to TB onset after treatment initiation will increase [7].

Screening for latent TB prior to anti-TNF treatment and the initiation of prophylaxis (isoniazid for 9 months) where indicated is paramount in reducing the overall numbers of reactivation cases. Currently, the CDC recommends that an induration of ≥5 mm diameter be considered a positive PPD result in patients with RA. Together with a careful history and chest X-ray, the effectiveness of pre-treatment screening can be maximized and prophylaxis can be initiated for latent TB cases where appropriate. Benefits in screening for LTBI were demonstrated in the Spanish registry BIOBADASER [11], after recommendations were implemented in 2002. They showed a 74% reduction in TB cases in patients with RA on infliximab, but still demonstrated an overall increased risk compared to the general population. As our case highlights, despite a negative screening test, it is always necessary to consider TB in patients being treated with TNF alpha inhibitors as screening may be complicated by a number of factors. These include the lack of a gold standard test for latent TB, false negatives in patients with active TB or the immunocompromised RA patient, and the difficulty in distinguishing between reactivation and progression of a newly acquired TB infection. As we continue to see more widespread use of TNF alpha inhibitors in the treatment of RA, further work will be needed on improving the accuracy of diagnostic screening techniques for LTBI and prevent the potential devastating manifestation of CNS tuberculosis.

Disclosures

The following case report is authored by Karen Lynch MD and Michael Farrell MD. It adheres to ethical guidelines and standards, and there was no financial or other conflict of interests associated with this case or the authors.

Copyright information

© Clinical Rheumatology 2010