**4. Nontuberculous mycobacterial infections in the setting of TNF-α inhibitors**

Nontuberculous mycobacteria (NTM) include species of mycobacteria other than those belonging to the *Mycobacterium tuberculosis* complex and *M. leprae*. They are a large group of ubiquitous environmental organisms that can cause pulmonary and extrapulmonary infections. Pulmonary NTM disease is often associated with underlying structural lung disease, including chronic obstructive pulmonary disease and bronchiectasis. In many other cases, there is no obvious underlying lung disease or overt immune incompetence. The relative proportions of NTM lung disease patients with and without underlying lung disease probably varies by population, but it appears that the majority of cases occur without demonstrable predisposing factors. NTM organisms may be isolated from the sputum in the absence of clinically relevant disease, and thus the diagnosis of pulmonary NTM disease rests on the presence of multiple positive cultures, and clinical (symptoms and radiology) data (Griffith et al. 2007). Extrapulmonary NTM disease is less common than lung disease, and may manifest as a localized infection of lymph nodes, skin, soft tissue, or bone, or may be disseminated. It is diagnosed when biopsy specimens of the involved organ(s) culture the causative organism, or, in the case of disseminated disease, with positive blood cultures (Griffith et al. 2007).

#### **4.1 The risk of NTM disease associated with TNF-α inhibitors**

Given the known role of tumor necrosis factor (TNF) in granuloma formation and maintenance, it is likely that TNF-α inhibitors increase the risk of all granulomatous

Mycobacterial Infections Associated with TNFαnhibitors 85

rheumatoid arthritis (95%CI 1.5–8.8) (Winthrop et al. 2009). There are several possible reasons for this. For one, rheumatoid lung disease, which can include bronchiolitis and bronchiectasis, occurs in about 10% of people with rheumatoid arthritis, and can predispose to NTM disease (Winthrop et al. 2009). Also, rheumatoid arthritis and NTM lung disease have similar epidemiologic risk profiles, as both occur more commonly in elderly women (Winthrop et al. 2009). Additionally, rheumatoid arthritis is more common in the elderly, who may have comorbidities predisposing to NTM lung disease, such as chronic obstructive

NTM disease is associated with a high level of morbidity and mortality when it develops on TNF-α inhibitor therapy. In the report from Winthrop et al., 61% of patients with NTM

A broad range of different NTM species have been described in association with TNF-α inhibitors, including those of high and low pathogenicity (van Ingen et al. 2008). In the US Medwatch study, *M*. *Avium* was the most common etiologic organism reported (49%), followed by rapidly growing mycobacteria (19%), and *M*. *marinum* (8%) (Winthrop et al.

NTM disease seems to occur after many months of TNF-α inhibitor therapy. The report of the US FDA Medwatch data showed that the median time between TNF-α inhibitor start date and infection diagnosis was 43 weeks for infliximab (range 2-200 weeks), 35 weeks for etanercept (range 0-288 weeks), and 18 weeks for adalimumab (range 4–94 weeks) (Winthrop et al. 2009). This group therefore surmised that most cases represent newly acquired infection. However, given the natural course of pulmonary NTM disease, which typically is insidious in onset and slowly progressive, the possibility exists that some patients had undiagnosed pulmonary NTM disease before starting TNF-antagonist therapy. The experience of treating NTM disease in the setting of TNF-α inhibitor therapy is limited. Anti-TNF-α therapy should likely be held for an unknown duration. However, one case report described a paradoxical worsening of NTM disease after withdrawal of infliximab (Salvana et al. 2007), similar to the paradoxical reaction sometimes seen with tuberculosis; the clinician should be aware of this complication. Treatment of NTM disease is complicated because different regimens exist for the different NTM species. Futhermore, prolonged antimicrobial therapy is required and results are often disappointing; expert consultation

The best approach to screening and prevention of NTM disease prior to initiation of TNF-α inhibitor therapy is unknown. Unlike tuberculosis, there is no evidence of a latent phase in NTM disease. Additionally, screening is complicated by the possibility of NTM colonization without active disease, and the ongoing environmental inoculation that is likely present. However, given the insidious nature of NTM disease and its slow progression, unrecognized NTM disease may be present in some patients prior to starting TNFantagonist therapy. Screening for such patients should be considered. Screening should include chest radiography, which must be done for all patients prior to starting TNF-α inhibitors to screen for tuberculosis. However, chest radiographs are not sensitive for detecting bronchiectasis or other parenchymal abnormalities associated with pulmonary NTM disease. Computerized tomography (CT) should therefore be considered in patients suspected of predisposing pulmonary diseases, including those with chronic unexplained cough. If chest CT is suggestive of possible NTM disease, sputum or bronchoscopy specimens should be cultured to rule out active NTM disease prior to initiation of TNF-

infections were hospitalized, and 9% died (Winthrop et al. 2009).

antagonist therapy (van Ingen et al. 2008; Winthrop et al. 2009).

pulmonary disease.

should always be sought.

2009).

infections, including NTM infection. However, in comparion to tuberculosis, relatively little has been published on this association. NTM infection is more difficult to study than tuberculosis, since the diagnosis is more complex as it relies on clinical data in addition to positive cultures. Additionally, in many jurisdictions, NTM isolation and NTM disease are not reportable to public health authorities.

Most literature associating NTM disease with TNF-α inhibitors has been in the form of case reports; a broad range of different NTM species infecting different body sites have been described in association with infliximab, adalimumab, and etanercept therapy (van Ingen et al. 2008). However, incidence studies are scarce. In 2004, Wallis *et al*. reported rates of granulomatous infections in persons treated with infliximab and etanercept. The cases were voluntarily reported to the United States Food and Drug Administration (FDA) Adverse Events Reporting System (AERS) from January 1998 to September 2002. A correction was published soon after (Wallis et al. 2004), which removed erroneously included cases from Europe. This study identified 29 cases of unspecified NTM infections, which translates to a rate of 17 per 100,000 treated persons (Wallis et al. 2004). This is much higher than the background incidence of 4 cases per 100,000 persons, reported in the United States in 1996 (Centers for Disease et al. 1996).

An updated study of the same Medwatch database, extending the time period to 2007, reported 105 confirmed or probable cases of NTM infection associated with TNF-α inhibitors (Winthrop et al. 2009). These cases were most frequently associated with infliximab (n = 73, 69%), followed by etanercept (n = 25, 24%), and then adalimumab (n = 7, 7%). Unfortunately, they did not have information regarding drug exposure, and so were unable to calculate rates of infection.

Interestingly, the original report of the US FDA Medwatch data found that the incidence of NTM infection was significantly lower than the incidence of tuberculosis in patients on TNF-α inhibitors (Wallis et al. 2004). However, a more recent report, based on a survey of infectious disease physicians in the United States, found the opposite; there were more cases of NTM infection than tuberculosis infection associated with TNF-α inhibitors in the United States (32 vs. 17 cases) (Winthrop et al. 2008). This finding is not unexpected, given the low prevalence of tuberculosis in the United States, and the widespread belief that rates of NTM disease are increasing (Griffith et al. 2007), but it highlights the fact that NTM disease is an underrecognized but important complication of TNF-antagonist therapy.

#### **4.2 The presentation, treatment, and prevention of NTM disease associated with TNFα inhibitors**

Similar to tuberculosis, in the setting of TNF-α inhibitor therapy, extra-pulmonary and disseminated NTM disease appear to be more common. In the 2009 report of the US FDA Medwatch data, 56% of the confirmed or probable NTM cases were pulmonary, and 44% were extrapulmonary; 26% involved skin or soft tissue, 9% bone or joint, and 8% were disseminated (Winthrop et al. 2009). In contrast, in the United States during the period from 1993 through 1996, the NTM isolates reported by state public health laboratories were divided as follows; 75% were pulmonary, 5% were from blood, 2% from skin/soft tissue, and 0.4% from lymph node isolates (Centers for Disease et al. 1996).

Patients with rheumatoid arthritis, however, may be more likely than those with other indications for TNF-antagonist therapy to have pulmonary NTM disease. The US FDA Medwatch study showed that compared with patients with extrapulmonary NTM disease, patients with pulmonary NTM disease were 3.6 times more likely to have underlying

infections, including NTM infection. However, in comparion to tuberculosis, relatively little has been published on this association. NTM infection is more difficult to study than tuberculosis, since the diagnosis is more complex as it relies on clinical data in addition to positive cultures. Additionally, in many jurisdictions, NTM isolation and NTM disease are

Most literature associating NTM disease with TNF-α inhibitors has been in the form of case reports; a broad range of different NTM species infecting different body sites have been described in association with infliximab, adalimumab, and etanercept therapy (van Ingen et al. 2008). However, incidence studies are scarce. In 2004, Wallis *et al*. reported rates of granulomatous infections in persons treated with infliximab and etanercept. The cases were voluntarily reported to the United States Food and Drug Administration (FDA) Adverse Events Reporting System (AERS) from January 1998 to September 2002. A correction was published soon after (Wallis et al. 2004), which removed erroneously included cases from Europe. This study identified 29 cases of unspecified NTM infections, which translates to a rate of 17 per 100,000 treated persons (Wallis et al. 2004). This is much higher than the background incidence of 4 cases per 100,000 persons, reported in the United States in 1996

An updated study of the same Medwatch database, extending the time period to 2007, reported 105 confirmed or probable cases of NTM infection associated with TNF-α inhibitors (Winthrop et al. 2009). These cases were most frequently associated with infliximab (n = 73, 69%), followed by etanercept (n = 25, 24%), and then adalimumab (n = 7, 7%). Unfortunately, they did not have information regarding drug exposure, and so were

Interestingly, the original report of the US FDA Medwatch data found that the incidence of NTM infection was significantly lower than the incidence of tuberculosis in patients on TNF-α inhibitors (Wallis et al. 2004). However, a more recent report, based on a survey of infectious disease physicians in the United States, found the opposite; there were more cases of NTM infection than tuberculosis infection associated with TNF-α inhibitors in the United States (32 vs. 17 cases) (Winthrop et al. 2008). This finding is not unexpected, given the low prevalence of tuberculosis in the United States, and the widespread belief that rates of NTM disease are increasing (Griffith et al. 2007), but it highlights the fact that NTM disease is an

**4.2 The presentation, treatment, and prevention of NTM disease associated with TNF-**

Similar to tuberculosis, in the setting of TNF-α inhibitor therapy, extra-pulmonary and disseminated NTM disease appear to be more common. In the 2009 report of the US FDA Medwatch data, 56% of the confirmed or probable NTM cases were pulmonary, and 44% were extrapulmonary; 26% involved skin or soft tissue, 9% bone or joint, and 8% were disseminated (Winthrop et al. 2009). In contrast, in the United States during the period from 1993 through 1996, the NTM isolates reported by state public health laboratories were divided as follows; 75% were pulmonary, 5% were from blood, 2% from skin/soft tissue,

Patients with rheumatoid arthritis, however, may be more likely than those with other indications for TNF-antagonist therapy to have pulmonary NTM disease. The US FDA Medwatch study showed that compared with patients with extrapulmonary NTM disease, patients with pulmonary NTM disease were 3.6 times more likely to have underlying

underrecognized but important complication of TNF-antagonist therapy.

and 0.4% from lymph node isolates (Centers for Disease et al. 1996).

not reportable to public health authorities.

(Centers for Disease et al. 1996).

unable to calculate rates of infection.

**α inhibitors** 

rheumatoid arthritis (95%CI 1.5–8.8) (Winthrop et al. 2009). There are several possible reasons for this. For one, rheumatoid lung disease, which can include bronchiolitis and bronchiectasis, occurs in about 10% of people with rheumatoid arthritis, and can predispose to NTM disease (Winthrop et al. 2009). Also, rheumatoid arthritis and NTM lung disease have similar epidemiologic risk profiles, as both occur more commonly in elderly women (Winthrop et al. 2009). Additionally, rheumatoid arthritis is more common in the elderly, who may have comorbidities predisposing to NTM lung disease, such as chronic obstructive pulmonary disease.

NTM disease is associated with a high level of morbidity and mortality when it develops on TNF-α inhibitor therapy. In the report from Winthrop et al., 61% of patients with NTM infections were hospitalized, and 9% died (Winthrop et al. 2009).

A broad range of different NTM species have been described in association with TNF-α inhibitors, including those of high and low pathogenicity (van Ingen et al. 2008). In the US Medwatch study, *M*. *Avium* was the most common etiologic organism reported (49%), followed by rapidly growing mycobacteria (19%), and *M*. *marinum* (8%) (Winthrop et al. 2009).

NTM disease seems to occur after many months of TNF-α inhibitor therapy. The report of the US FDA Medwatch data showed that the median time between TNF-α inhibitor start date and infection diagnosis was 43 weeks for infliximab (range 2-200 weeks), 35 weeks for etanercept (range 0-288 weeks), and 18 weeks for adalimumab (range 4–94 weeks) (Winthrop et al. 2009). This group therefore surmised that most cases represent newly acquired infection. However, given the natural course of pulmonary NTM disease, which typically is insidious in onset and slowly progressive, the possibility exists that some patients had undiagnosed pulmonary NTM disease before starting TNF-antagonist therapy.

The experience of treating NTM disease in the setting of TNF-α inhibitor therapy is limited. Anti-TNF-α therapy should likely be held for an unknown duration. However, one case report described a paradoxical worsening of NTM disease after withdrawal of infliximab (Salvana et al. 2007), similar to the paradoxical reaction sometimes seen with tuberculosis; the clinician should be aware of this complication. Treatment of NTM disease is complicated because different regimens exist for the different NTM species. Futhermore, prolonged antimicrobial therapy is required and results are often disappointing; expert consultation should always be sought.

The best approach to screening and prevention of NTM disease prior to initiation of TNF-α inhibitor therapy is unknown. Unlike tuberculosis, there is no evidence of a latent phase in NTM disease. Additionally, screening is complicated by the possibility of NTM colonization without active disease, and the ongoing environmental inoculation that is likely present. However, given the insidious nature of NTM disease and its slow progression, unrecognized NTM disease may be present in some patients prior to starting TNFantagonist therapy. Screening for such patients should be considered. Screening should include chest radiography, which must be done for all patients prior to starting TNF-α inhibitors to screen for tuberculosis. However, chest radiographs are not sensitive for detecting bronchiectasis or other parenchymal abnormalities associated with pulmonary NTM disease. Computerized tomography (CT) should therefore be considered in patients suspected of predisposing pulmonary diseases, including those with chronic unexplained cough. If chest CT is suggestive of possible NTM disease, sputum or bronchoscopy specimens should be cultured to rule out active NTM disease prior to initiation of TNFantagonist therapy (van Ingen et al. 2008; Winthrop et al. 2009).

Mycobacterial Infections Associated with TNFαnhibitors 87

preventive strategies. Prospective studies estimating the positive predictive value of IGRA responses compared with tuberculin skin test responses for the developent of tuberculosis in patients treated with TNF-α inhibitors are also needed. Reports on treatment and outcomes of tuberculosis and NTM disease in the setting of TNF-α inhibitor therapy are also necessary.

Anonymous (1982). Efficacy of various durations of isoniazid preventive therapy for

Arend, S. M., Leyten, E. M. S., Franken, W. P. J., Huisman, E. M. & van Dissel, J. T. (2007). A

Askling, J., Fored, C. M., Brandt, L., Baecklund, E., Bertilsson, L., Coster, L., Geborek, P.,

alpha therapy. *Swiss Medical Weekly,* Vol. 137, No. 43-44, (Nov 3), pp. 620-2. Bekker, L. G., Freeman, S., Murray, P. J., Ryffel, B. & Kaplan, G. (2001). TNF-alpha controls

Belknap, R., Reves, R. & Burman, W. (2005). Immune reconstitution to Mycobacterium

Blackmore, T. K., Manning, L., Taylor, W. J. & Wallis, R. S. (2008). Therapeutic use of

lymph nodes. *Clinical Infectious Diseases,* Vol. 47, No. 10, (Nov 15), pp. e83-5. Brassard, P., Kezouh, A. & Suissa, S. (2006). Antirheumatic drugs and the risk of tuberculosis. *Clinical Infectious Diseases,* Vol. 43, No. 6, (Sep 15), pp. 717-22. Brassard, P., Lowe, A.-M., Bernatsky, S., Kezouh, A. & Suissa, S. (2009). Rheumatoid

British Thoracic Society Standards of Care, C. (2005). BTS recommendations for assessing

Carmona, L., Gomez-Reino, J. J., Rodriguez-Valverde, V., Montero, D., Pascual-Gomez, E.,

*Infectious Diseases,* Vol. 45, No. 11, (Dec 1), pp. 1470-5.

*Immunology,* Vol. 166, No. 11, (Jun 1), pp. 6728-34.

*Lung Disease,* Vol. 9, No. 9, (Sep), pp. 1057-8.

*Rheumatism,* Vol. 61, No. 3, (Mar 15), pp. 300-4.

800-5.

1766-72.

tuberculosis: five years of follow-up in the IUAT trial. International Union Against Tuberculosis Committee on Prophylaxis. *Bulletin of the World Health Organization,*

patient with de novo tuberculosis during anti-tumor necrosis factor-alpha therapy illustrating diagnostic pitfalls and paradoxical response to treatment. *Clinical* 

Jacobsson, L. T., Lindblad, S., Lysholm, J., Rantapaa-Dahlqvist, S., Saxne, T., Romanus, V., Klareskog, L. & Feltelius, N. (2005). Risk and case characteristics of tuberculosis in rheumatoid arthritis associated with tumor necrosis factor antagonists in Sweden. *Arthritis & Rheumatism,* Vol. 52, No. 7, (Jul), pp. 1986-92. Beglinger, C., Dudler, J., Mottet, C., Nicod, L., Seibold, F., Villiger, P. M. & Zellweger, J.-P.

(2007). Screening for tuberculosis infection before the initiation of an anti-TNF-

intracellular mycobacterial growth by both inducible nitric oxide synthasedependent and inducible nitric oxide synthase-independent pathways. *Journal of* 

tuberculosis after discontinuing infliximab. *International Journal of Tuberculosis &* 

infliximab in tuberculosis to control severe paradoxical reaction of the brain and

arthritis, its treatments, and the risk of tuberculosis in Quebec, Canada. *Arthritis &* 

risk and for managing Mycobacterium tuberculosis infection and disease in patients due to start anti-TNF-alpha treatment. *Thorax,* Vol. 60, No. 10, (Oct), pp.

Mola, E. M., Carreno, L. & Figueroa, M. (2005). Effectiveness of recommendations to prevent reactivation of latent tuberculosis infection in patients treated with tumor necrosis factor antagonists. *Arthritis & Rheumatism,* Vol. 52, No. 6, (Jun), pp.

**6. References** 

Vol. 60, No. 4, pp. 555-64.

During therapy with TNF-α inhibitors, patients should be regularly assessed to rule out active infections. With respect to NTM disease, repeated sputum cultures during therapy should be considered in the setting of chest symptoms or co-morbid pulmonary disease, as well as chest radiography or CT scans. Extrapulmonary disease should be thoroughly investigated, and biopsy specimens should be stained for acid-fast bacilli and cultured for mycobacteria (van Ingen et al. 2008).
