Digestive Diseases and Sciences

, Volume 52, Issue 6, pp 1481–1484

Pneumocystis jiroveci (carinii) Pneumonia After Infliximab Therapy: A Review of 84 Cases

Authors

  • Nirmal Kaur
    • Department of Internal MedicineUniversity of Michigan
    • Division of GastroenterologyDepartment of Medicine, VA Upstate New York, State University of New York at Buffalo School of Medicine
Original Paper

DOI: 10.1007/s10620-006-9250-x

Cite this article as:
Kaur, N. & Mahl, T.C. Dig Dis Sci (2007) 52: 1481. doi:10.1007/s10620-006-9250-x

Abstract

Anti-tumor necrosis factor-α therapy, infliximab, has become an established effective therapy for Crohn's disease and rheumatoid arthritis. However, infliximab has been associated with various opportunistic pathogens such as tuberculosis, histoplasmosis, listeriosis, aspergillosis, and Pneumocystis jiroveci (carinii) pneumonia. We reviewed the FDA Adverse Event Reporting System for cases of Pneumocystis associated with infliximab use from January 1998 through December 2003. The database revealed 84 cases of PCP following infliximab therapy. Concomitant immunosuppressive medications included methotrexate, prednisone, azathioprine, 6-mercaptopurine, and cyclosporine. Mean time between infliximab infusion and onset of symptoms of pneumonia, when reported, was 21 days (±18 days; n=40). Twenty-three of the 84 (27%) patients died. The use of infliximab is associated with PCP infection. Further, the mortality rate for Pneumocystis following the use of infliximab is significant. The potential for severe disease, mortality, and often subtle presentation of these infections warrant close follow-up and careful monitoring after therapy.

Keywords

InfliximabRemicadePneumocystisInfectionPneumonia

Introduction

Tumor necrosis factor-α (TNF-α) is a proinflammatory cytokine that has been found to play an important role in autoimmune diseases including rheumatoid arthritis and Crohn's disease. In patients with Crohn's disease, mucosal biopsies have shown increased levels of TNF-α protein and messenger RNA [1]. The blood and feces of patients with Crohn's disease have also demonstrated increased levels of this cytokine [2, 3]. In rheumatoid arthritis, TNF-α has been shown to stimulate the production of matrix metalloproteinase enzymes that degrade components of the extracellular matrix and lead to joint destruction [4].

Specifically in Crohn's disease, TNF-α binds to lamina propria mucosal cells and activates the CD2 pathway, which then increases production of T helper-1 cytokines such as interferon-γ [5]. Additionally, TNF-α has been shown to promote the expression of adhesion molecules which encourage an influx of cells into sites of inflammation and subsequently induce the release of proteolytic enzymes which lead to tissue destruction [6, 7]. The role of TNF-α receptors p55 and p75 in Crohn's disease has also been further developed: both take part in apoptosis, and the p55 subunit plays a role in lymphocyte proliferation [8].

Anti-TNF-α, infliximab (Remicade; Centocor, Malvern, PA), is a chimeric monoclonal antibody comprised of 75% human and 25% mouse protein. Infliximab binds to TNF-α and inhibits its ability to bind to receptors on lamina propria mucosal cells. This drug has become an established effective therapy for Crohn's disease as well as rheumatoid arthritis. In each disease, this medication is usually reserved for patients who fail to respond to conventional therapy [9]. To date, infliximab has been used to treat more than 700,000 patients [10]. The use of infliximab as a therapy for other autoimmune diseases including sarcoidosis [11], psoriatic arthritis [12], ankylosing spondylitis [13], and Wegener's granulomatosis [14] has been reported.

Despite its role in chronic inflammatory conditions, evidence shows that TNF-α is also a critical factor in host defense. TNF-α initiates the inflammatory cascade by stimulating the production of other pro-inflammatory cytokines and influencing the maturation of cytokines that provide defense against pathogens. Although little is known about its role in host defense against Pneumocystis, TNF-α has been shown to be critical in defending against many pathogens. In TNF knockout mice, IgG and IgE antbody reactions are altered, antibody responses are impaired, and antigen presentation is inconsistent [15].

While anti-TNF therapy is helpful in many who suffer from inflammatory disease, it is not without hazard. Infliximab has been associated with several infections such as tuberculosis [16], invasive aspergillosis [17], histoplasmosis [18], cytomegalovirus [19], listeriosis [20], cryptococcosis [21], salmonella [22], varicella [23], and nocardiosis [24]. There have been reports of demyelinating disease secondary to infliximab therapy [25]. Four case reports of Pneumocystis jiroveci (carinii) pneumonia (PCP) after infliximab therapy have been published thus far [2629]. Here we report all cases of PCP associated with infliximab therapy that were reported to the Adverse Event Reporting System (AERS) of the United States Food and Drug Administration between January 1998 and December 2003.

Methods

The AERS is a passive reporting system that documents adverse reactions to medications used in the United States. Drug manufacturers are required to submit reports of adverse events to this database, while health care professionals voluntarily send reports. We reviewed the AERS database for cases of PCP associated with infliximab use. The database was accessed on October 15, 2004. At that time, the database included all cases reported between January 1998 and December 2003. AERS reports of PCP were imported into a Microsoft Excel file. No patients were excluded. When multiple reports were found to be associated with the same patient, the most recent report was selected. The temporal relationship between infliximab therapy and development of infection was indexed using the date of the drug infusion and the date of the reported adverse event. These dates were not included in all AERS reports. The database was also examined for demographic information, indication for infliximab, and use of concomitant immunosuppressive medications.

Results

The database contained 84 cases of PCP following infliximab therapy (Table 1). Demographics, reported in 74 patients, included 47 women and 27 men, with a mean age of 55 years. Indications for infliximab were rheumatoid arthritis (n=49), Crohn's disease (n=14), ulcerative colitis (n=2), ankylosing spondylitis (n=2), Wegener's granulomatosis (n=2), psoriatic arthritis (n=1), dermatomyositis (n=1), polymyositis (n=1), and Still's disease (n=1). Concomitant immunosuppressive medications included methotrexate (n=38), prednisone (n=37), azathioprine (n=6), leflunomide (n=6), 6-mercaptopurine (n=5), and cyclosporine (n=4). Six of the 84 patients had chronic lung disease including asthma, pulmonary fibrosis, and bronchitis. Mean time between infliximab infusion and onset of symptoms of pneumonia, when reported, was 21 days ± 18 days; (n=40). Patients had received an average of 2.1 ± 1.3 infusions before onset of symptoms (Table 2). Twenty-three of the 84 (27%) cases resulted in death.
Table 1

Demographics, indications for infliximab therapy, concomitant immunosuppressants, and comorbid diseases

 

Number of patients

I. Demographics

 

 Men

27

 Women

47

 Mean age = 55 ± 15 years

 

II. Indications for infliximab

 

 Rheumatoid arthritis

49

 Crohn's disease

14

 Ulcerative colitis

 2

 Ankylosing spondylitis

 2

 Wegener's granulomatosis

 2

 Psoriatic arthritis

 1

 Dermatomyositis

 1

 Polymyositis

 1

 Still's disease

 1

III. Concomitant immunosuppressants

 

 Methotrexate

38

 Prednisone

37

 Azathioprine

 6

 Leflunomide

 6

 6-Mercaptopurine

 5

 Cyclosporine

 4

IV. Comorbid diseases

 

 Anemia

 3

 Asthma

 3

 Pulmonary fibrosis

 2

 Bronchitis

 2

Table 2

Temporal relationship

Mean time between infliximab infusion and onset of symptoms of pneumonia

21 ± 18 days (n=40)

Number of infusions before onset of symptoms

2.1 ± 1.3 (n=76)

Discussion

The use of infliximab is associated with PCP infection. This review is the largest study to date that demonstrates the relationship between PCP and infliximab. Although PCP infection after infliximab therapy is certainly not common, the incidence is probably higher than clinicians and patients suspect. Further, the mortality rate for PCP is quite significant: the patients we reviewed had a 27% mortality rate. Mortality is 39% for non-HIV PCP cases and 59% for intubated patients [30]. The potential for severe disease, mortality, and often subtle presentation of these infections warrant close follow-up and careful monitoring after therapy.

The spontaneous nature of adverse event reporting by health care professionals limits our knowledge of the true incidence of PCP, as well as other infections, in relation to TNF-α antagonists. Analyses of the nature of adverse event reporting in the United States conclude that the actual incidence of adverse events is higher than what is reported [31]. The true incidence of PCP after infliximab is likely higher, to an unknown extent, than what we report here. Additionally, our study is limited by the fact that the AERS reports were not always entirely complete. When the data were obtained by a phone call from a health care professional, details such as number of infliximab infusions, date of infusion, and indication for therapy were sometimes omitted. Subsequent processing of the AERS report, in accordance with the Health Insurance Portability and Accountability Act of 1996 (HIPAA), required that the patient's identifying information be deleted; thus, tracking of the aforementioned details is impossible.

Several of these patients received multiple immunosuppressive medications. While PCP has been reported after methotrexate [32], it does not appear to be a common clinical event. Evidence exists that chronic administration of glucocorticoids leads to immunosuppression; however, corticosteroid-induced immunosuppression does not typically lead to opportunistic infection [33].

We previously reported a case of PCP after infliximab and also submitted the adverse event data to the AERS [26]. Our patient, at the time of diagnosis, had a CD4 count of 100. Infliximab has been shown to decrease CD4 lymphocyte counts; Baert et al. demonstrated a 50% reduction in CD4 lymphocyte counts in both ileal and colonic lamina propria mucosal cells of Crohn's patients after receiving an infliximab infusion [34]. In addition, Van den Brande et al. demonstrated that infliximab induces apoptosis of activated lamina propria T lymphocytes; it acts as an immunosuppressant [35]. Further, agranulocytosis after infliximab therapy has been reported [36]. The CD4 counts of the other 83 patients reported in this study are unknown. To our knowledge, there have been no reports of infliximab's effect on peripheral blood CD4 counts. Surveillance of this parameter during therapy may be warranted.

The studies which led to FDA approval of infliximab for Crohn's disease did not report PCP or other opportunistic infection [37]. Of note, patients receiving concomitant cyclosporine were excluded from that study. Thirty-four percent of patients receiving infliximab experienced infections requiring antibiotics, and 3% of patients in the treatment group had "serious infections" which were not further described. Centocor, the manufacturer of infliximab, added pneumocystis to its black box warning on August 8, 2001, after reports of PCP were received from health professionals [10]. A more recent surveillance study of 500 Mayo Clinic patients receiving infliximab for Crohn's disease reported 8 cases of pneumonia, 2 of which were fatal, and 9 upper respiratory tract infections [38]. It is possible that these pneumonias may have been caused by opportunistic pathogens. Of these 17 patients, infectious agents were identified in only 2. The possibility that PCP was present in some of these patients, especially the two that died, cannot be excluded.

We conclude that the use of infliximab is associated with the development of PCP pneumonia. We recommend consideration of prophylaxis from opportunistic infection prior to and after infliximab therapy. Further investigation regarding duration and types of prophylaxis is warranted. Continued long-term monitoring of patients during and after anti-TNF-α therapy will be valuable in determining the appropriate level of surveillance and intervention that these patients require. As additional immunomodulators enter clinical use, and the use of currently available agents expands, clinicians must be vigilant for unexpected infectious complications.

Copyright information

© Springer Science + Business Media, Inc. 2006