|
| |||||||||||||||||||||||||||||||
Background Studies conducted in vitro and in animals suggest that cytokine signals to monocytes or macrophages by interferon gamma are important in the containment and clearance of disseminated nontuberculous mycobacterial infections.
Methods We studied seven patients with refractory disseminated nontuberculous mycobacterial infections who were not infected with the human immunodeficiency virus. Three patients were from a family predisposed to the development of Mycobacterium avium complex infections; four patients had idiopathic CD4+ T-lymphocytopenia. Their infections were culture- or biopsy-proved, involved at least two organ systems, and had been treated with the maximal tolerated medical therapy. Cellular proliferation, cytokine production, and phagocyte function were assessed in peripheral-blood cells. Interferon gamma was administered subcutaneously two or three times weekly in a dose of 25 to 50 µg per square meter of body-surface area in addition to antimycobacterial medications. Clinical effects were monitored by cultures, biopsies, radiographs, and in one patient a change in the need for paracentesis.
Results In response to phytohemagglutinin, the production of interferon gamma by mononuclear cells from the patients was lower than in normal subjects (P<0.001), whereas stimulation with ionomycin and phorbol myristate acetate led to normal production of interferon gamma in the patients. Within eight weeks of the start of interferon gamma therapy, all seven patients had marked clinical improvement, with abatement of fever, clearing of many lesions and quiescence of others, radiographic improvement, and a reduction in the need for paracentesis.
Conclusions Interferon gamma in combination with conventional therapy may be effective for some cases of refractory disseminated nontuberculous mycobacterial infection.
Methods
Subjects
Seven patients with culture-proved, nontuberculous disseminated mycobacterial infection that was refractory to chemotherapy were evaluated at the National Institutes of Health according to approved protocol 92-I-0255. Three of the seven patients (Patients 4, 5, and 6) had been given a diagnosis of idiopathic CD4+ T-lymphocytopenia. The extent and severity of mycobacterial disease were determined on the basis of a physical examination, nutritional status, radiographs, biopsies, cultures, and in one patient the need for paracentesis. Infection with the human immunodeficiency virus (HIV) was ruled out. In all patients except Patient 3 interferon gamma was added to antimycobacterial chemotherapeutic regimens only if there was no improvement in the patient's condition after three months or if there was progression of disease for one month while the maximal tolerated therapy was being given. Disease progression was defined as the appearance of new lesions or the exacerbation of old ones. Sixteen healthy volunteer blood-bank donors served as controls for the in vitro studies. Four patients with culture-proved pulmonary Mycobacterium tuberculosis who had been receiving chemotherapy for 1 to 11 months, were clinically responding to therapy, and were not infected with HIV were included in the in vitro studies as a means of controlling for mycobacterial infection itself and for the effects of some antimycobacterial drugs.
Microbiologic Techniques
Blood samples were obtained from all patients for mycobacterial culture, and specimens for culture were also obtained from medically indicated sites or fluids (e.g., lymph nodes, sputum, and skin lesions). The AccuProbe system (Gen-Probe, San Diego, Calif.) was used for the definitive identification of M. avium complex excluding the X cluster (referred to hereafter as M. avium complex), M. avium complex including the X cluster, and M. kansasii17,18. The M. avium X cluster consists of organisms that are biochemically and phenotypically M. avium complex but do not react with either the M. avium or M. intracellulare probes. These organisms do react with a probe specific for the M. avium X cluster.
Assessment of Immune Function
The superoxide production and staphylocidal activity of granulocytes from selected patients were assessed as described previously19. Peripheral-blood mononuclear cells were obtained from apheresis packs or whole blood. The cells were cultured in RPMI medium at 106 cells per milliliter in microtiter plates and stimulated for 48 hours with phytohemagglutinin or ionomycin and phorbol myristate acetate. The cultures were assayed after 48 hours to determine the rate of proliferation and the level of cytokine production. The cytokines were assayed as described elsewhere, except that interleukin-2 was measured by enzyme-linked immunosorbent assay (R&D Systems, Minneapolis)20. Cells from Patients 1 through 5 and the mother of Patient 1 were assayed together with those from the healthy controls and the patients infected with M. tuberculosis; specimens from Patient 6 were assayed separately, and the results were not included in the proliferation or cytokine-production figures. No samples from Patient 7 were assayed. Lymphocyte immunophenotyping was performed with the use of a standard whole-blood lysis technique. Samples from the mother and maternal grandmother of Patient 1 (the latter of whom was also the mother of Patients 2 and 3) were also subjected to lymphocyte phenotyping.
Treatment
Interferon gamma (Actimmune) was supplied by Genentech (South San Francisco, Calif.) and administered subcutaneously three times weekly in a dose of 50 µg per square meter of body-surface area. Toxicity was graded according to hematologic, hepatic, renal, gastrointestinal, neurologic, and functional criteria as previously reported14. Efficacy was reflected by changes in the physical examination, nutritional status, radiographs, biopsies, cultures, and need for paracentesis. If after receiving interferon gamma, patients experienced malaise, fatigue, or headache that was unresponsive to acetaminophen and seriously impeded daily activities, the dose of the drug was halved. Interferon gamma was discontinued in the event of adverse clinical changes not due to another identified source and was reinstituted with careful monitoring after the symptoms resolved. Interferon gamma therapy was to be given for one year, but was prolonged if the patient was continuing to improve at that point.
Statistical Analysis
The proliferation and cytokine data were expressed as the average of the determinations for a given subject. We compared the log of the values using Student's t-test (StatWorks for Macintosh, Cricket Software, Philadelphia). The values were expressed as the means ±SE.
Case Reports
Members of a Family Predisposed to M. avium Complex Infection
Several members of a family that is apparently predisposed to disseminated M. avium complex infection constituted the core of this study. The disease has so far occurred only in two generations of male family members.
Patient 1
Patient 1 was a five-year-old boy born to unrelated parents. He had agenesis of the corpus callosum and the right kidney and had had frequent otitis media and respiratory infections in early childhood. At the age of five, he had night sweats, fever, sinusitis, cervical adenitis, diarrhea, hepatosplenomegaly, and failure to thrive. Cultures of sinus, lymph node, blood, marrow, and gastric aspirate grew M. avium complex. Despite having elevated immunoglobulin levels, the patient had no detectable antibodies to polysaccharide antigens; antibodies to viral antigens were present. Fever and M. avium complex bacteremia persisted despite treatment with clarithromycin, rifabutin, ethambutol, ciprofloxacin, clofazimine, amikacin, cycloserine, and intravenous immune globulin.
Patient 2
Patient 2 was a 29-year-old chemical worker (a maternal uncle of Patient 1) who has been described previously21. At the age of 26 he had erosive lesions on the face and arm and M. avium complex osteomyelitis of the distal left second metacarpal phalangeal joint that required debridement and synovectomy. He was treated with rifampin, ethambutol, clofazimine, isoniazid, and several months of streptomycin. After 2.5 years of therapy, smears and cultures of skin from his face and arm were positive for M. avium complex X cluster. The skin lesions were unchanged by the addition of clarithromycin to the treatment regimen, and smears and cultures of biopsy specimens remained positive.
Patient 3
Patient 3 was a 46-year-old man (a maternal uncle of Patient 1 and a brother of Patient 2) with bronchiectasis. At the age of 28 he had miliary pulmonary infection with culture-proved M. tuberculosis that was treated conventionally without recurrence. At the age of 44 he had disseminated M. avium complex treated with several months of streptomycin and two years of rifampin, ethambutol, clofazimine, and isoniazid21. Eleven months after the discontinuation of therapy, a sputum sample was again positive for M. avium complex and there was mild, diffuse erythema of the face. Three months later, during which no therapy was given, multiple sputum samples, a skin-biopsy specimen, and blood samples were all positive for M. avium complex. Treatment with clarithromycin, ethambutol, rifampin, and clofazimine was begun, with clearing of the skin and negative blood cultures at one month. In view of the systemic disease, relapse following definitive therapy, and persistent facial erythema, treatment with interferon gamma was begun.
Additional Family Members
The mother of Patient 1 (a sister of Patients 2 and 3) was born to unrelated parents and had no unusual infections during childhood or adulthood. She underwent splenectomy and partial pancreatectomy for trauma at the age of 18 and had a mild persistent leukocytosis and mild glucose intolerance without evidence of M. avium complex infection. The maternal grandmother of Patient 1 (the mother of Patients 2 and 3) had pulmonary tuberculosis at the age of 31 that responded to standard therapy without recurrence. Patient 1 had two apparently unaffected brothers and one unaffected maternal uncle.
Unrelated Patients with Mycobacterial Infection
Patient 4
Patient 4 was a 42-year-old man born to unrelated parents. At the age of 28 he had the first of multiple episodes of recurrent pneumonia, predominantly due to encapsulated organisms. An extensive immunologic evaluation revealed no abnormalities. At the age of 40, the patient had M. avium complex bacteremia and hepatitis. The CD4+ T-lymphocyte count was 166 per cubic millimeter, but an extensive search for HIV was negative. Despite therapy with clarithromycin, rifampin, ethambutol, ciprofloxacin, clofazimine, and amikacin, M. avium complex peritonitis developed, which required drainage of 4 to 6 liters of ascitic fluid three times monthly. Liver biopsies showed acid-fast organisms, but no cultures from any site were positive for M. avium complex beginning seven months after the diagnosis of the infection.
Patient 5
Patient 5 was a 41-year-old woman whose mother had disseminated M. tuberculosis in the setting of HIV-negative CD4+ T-lymphocytopenia and subsequently had a fatal myeloproliferative disorder. An open-lung biopsy performed when the patient was 40 showed bronchiolitis obliterans organizing pneumonia, which was treated with corticosteroids. After three months of therapy she noted multiple skin nodules, fever, and night sweats. Cultures of skin-biopsy specimens, blood, marrow, pleural fluid, stool, and urine grew M. avium complex. The CD4+ T-lymphocyte count was 143 per cubic millimeter, but an extensive search for HIV was negative. Steroids were discontinued, and antimycobacterial therapy initiated. Multiple drug reactions and intolerances resulted in the use of a regimen of amikacin, with marked ototoxicity, and clofazimine. Cultures were negative for M. avium complex beginning three months after the diagnosis of infection, although smears and histopathological analyses of various specimens were positive. Fifteen months after the initial diagnosis and five months before the initiation of interferon gamma therapy, diplopia and headache due to a posterior orbital mass developed. After nine months of interferon gamma treatment, this mass was operatively diagnosed as a schwannoma. Throughout the patient's illness her skin lesions were intermittently tender and draining, and she was intermittently febrile. Nausea and malabsorption led to endoscopy, which revealed the "cobblestone" appearance of the duodenum typical of M. avium complex involvement; hyperalimentation was required.
Patient 6
Patient 6 was a 49-year-old woman with an anticardiolipin-antibody syndrome (Patient 6 in the report by Smith et al.3). Nodular lesions had developed on the left thigh 1.3 years previously that contained numerous acid-fast bacilli and were positive for M. avium complex. The CD4+ T-lymphocyte count was low, but an extensive search for HIV was negative. After 11 months of therapy with clarithromycin, rifampin, ciprofloxacin, isoniazid, and amikacin, the cutaneous lesions were unchanged and remained histopathologically and culture positive for M. avium complex. The skin lesions ultimately progressed and a left fibular lytic lesion was shown to be due to M. avium complex osteomyelitis. The isolate was subsequently confirmed to be M. avium complex X cluster.
Patient 7
Patient 7 was a 53-year-old woman who received corticosteroids for a prolonged fever after a cholecystectomy at the age of 51. Pleural, pericardial, and mediastinal infection with M. kansasii subsequently developed. Therapy with isoniazid, rifampin, ethambutol, and pyrazinamide led to some improvement, but two months after the initiation of therapy multiple, painful, culture-positive cutaneous lesions developed that required surgical drainage. Nine months and 13 months after the initiation of therapy mediastinal pus spontaneously drained through the supraclavicular area and the sternal notch, respectively. The addition of clarithromycin to the regimen had no clinical effect. After 15 months of therapy, biopsy of a tender subcutaneous nodule showed frank pus without mycobacteria. Magnetic resonance imaging and computed tomography of the chest showed mediastinal inflammation extending from the aortic arch to the sternal notch.
Results
Characteristics of the Patients
The patients' characteristics are listed in Table 1. Analysis with the AccuProbe system showed that at least two types of isolates of M. avium complex were recovered from these patients. The strain of M. kansasii isolated from Patient 7 had a typical pattern of sensitivity. Staphylocidal activity was assessed in Patients 1, 4, 5, and 7 and was normal (Table 1). Superoxide-generating capacity was assessed in Patients 1, 4, and 5 and was normal (data not shown). None of the patients had serologic evidence of HIV infection; analyses involving the polymerase chain reaction, p24 antigen assay, and viral culture were also negative for HIV in the patients with CD4+ T-lymphocytopenia (Patients 4, 5, and 6).
|
|
Cellular proliferation induced by phytohemagglutinin and by ionomycin and phorbol myristate acetate was normal among the patients who were related and among the patients with tuberculosis but significantly reduced in the patients with CD4+ T-lymphocytopenia (Figure 1). Cells from the patients who were related and cells from the patients with CD4+ T-lymphocytopenia produced significantly less interferon gamma than did cells from normal subjects after phytohemagglutinin stimulation (P<0.001 for both) (Figure 1). However, stimulation by ionomycin and phorbol myristate acetate of cells from both patients who were related and patients with CD4+ T-lymphocytopenia resulted in normal production of interferon gamma (Figure 1). The production of interferon gamma by phytohemagglutinin-stimulated mononuclear cells was not altered by the inclusion of interleukin-2 in the culture medium, excluding the possibility of an interleukin-2-dependent abnormality in interferon gamma production (data not shown). Cells from these two groups of patients also produced significantly less interleukin-2 than did cells from normal subjects after stimulation by phytohemagglutinin (P<0.01 and P<0.001, respectively) (Figure 1).
|
There was no dramatic change in CD4+ T-cell counts or cell-activation markers in our patients during treatment with interferon gamma (Table 2). Anergy in the patients who were related was not reversed by interferon gamma therapy; Patient 4 acquired delayed hypersensitivity during treatment (Table 3).
|
|
Two patients required a reduction in the dose of interferon gamma because of fatigue, malaise, and myalgias. Patient 5 had an episode of delirium that cleared in association with the discontinuation of interferon gamma but did not recur with its reinstitution. Pneumocystis carinii pneumonia also developed during therapy at a time when her CD4+ T-lymphocyte count was 147 per cubic millimeter and she had received no prophylaxis. She subsequently was found to have a myeloproliferative syndrome similar to that of her mother.
Patient 6 had a CD4+ T-lymphocyte count below 50 per cubic millimeter at the time of her initial evaluation, and the count fell below 15 per cubic millimeter before treatment with interferon gamma was begun. After 14 weeks of therapy, a period during which her M. avium complex X cluster infection improved markedly, she was found to have a monoclonal, non-Burkitt's large-cell lymphoma that was positive for Epstein-Barr virus and was consistent with her immunodeficiency23. Interferon gamma was discontinued, and she was treated for lymphoma.
Discussion
The seven patients described here had persistent, well-documented disseminated nontuberculous mycobacterial infections that were refractory to aggressive therapy with conventional antimycobacterial agents. All seven patients had dramatic subjective and objective clinical responses when interferon gamma was added to their regimens, and this therapy was not associated with serious side effects. We therefore conclude that interferon gamma in combination with conventional therapy may be beneficial in the treatment of some cases of refractory disseminated nontuberculous mycobacterial infection.
The patients with familial disseminated mycobacterial infection merit separate consideration. The three patients in this cohort (Patients 1, 2, and 3) were infected with at least two genetically distinct organisms, M. avium complex and M. avium complex including the X cluster, which excludes the possibility of a common-source outbreak of a virulent strain of M. avium complex. Previous attempts to identify familial clusters of infection with leishmaniasis or leprosy have been difficult, since these are pathogenic organisms that occur predominantly in areas of endemic disease24,25. Very few families have been described in which genetic susceptibility to mycobacterial infection has been convincingly documented26. In the few families with disseminated M. avium complex or bacille Calmette-Guerin infection that have been described, studies on both the organisms and the infected hosts were limited, making it hard to distinguish disease due to common-source outbreaks with virulent strains from that due to true host-defense defects27,28. We believe that the family we describe is genetically predisposed to disseminated M. avium complex infection. The type of cell specifically affected in this family is under investigation. The abnormalities could be explained by defects in the lymphocyte, the monocyte, or the macrophage.
Our other patients with disseminated infection had CD4+ T-lymphocytopenia3,4,5,6. These patients may be similar to previously described patients with disseminated nontuberculous mycobacterial infection without HIV infection9,10,11,12,29,30. Although they are an immunologically heterogeneous group, they may share a common defect resulting in altered interferon gamma production and an inability to kill mycobacteria.
Interferon gamma therapy for chronic granulomatous disease is effective clinically, although it has been shown to correct the underlying phagocytic-cell defect in only some patients with this disease14,31,32. Previous trials of interferon gamma in the treatment of lepromatous leprosy used an intradermal route of administration and resulted in accentuated formation of local granulomas, increased numbers of keratinocyte activation markers, and decreased bacterial burdens15,33. However, the treatment of leprosy with interferon gamma has been complicated by the frequent "upgrading" of immune responses and occurrence of erythema nodosum leprosum34. The toxic effects encountered in patients with lepromatous leprosy treated with interferon gamma were not encountered in a large cohort of patients with chronic granulomatous disease,14 in patients with visceral leishmaniasis,16 or in our patients. The toxic effects seen with interferon gamma therapy for lepromatous leprosy are not specific to this type of treatment, but are more common with it34. Interferon gamma has also been used in the treatment of visceral and diffuse cutaneous leishmaniasis, occasionally at doses considerably higher than those given to our patients without serious toxic effects35. The fact that substantial toxicity has not been seen with the doses used in patients with chronic granulomatous disease, leishmaniasis, or disseminated nontuberculous mycobacterial infections suggests that the toxic effects reported in lepromatous leprosy may be particular to that disease.
The severity and extent of disease in our patients with refractory disseminated nontuberculous mycobacterial infection before therapy and the rapidity and extent of their improvement with therapy suggest that interferon gamma may be a powerful adjunct to the treatment of these infections, probably through the stimulation of cellular elements of the inflammatory and immune responses. Interferon gamma may be useful in the treatment of other mycobacterial infections, including M. tuberculosis.
We are indebted to Ms. Judi T. Miller and Ms. Ellen S. DeCarlo for ongoing excellent clinical care of the patients; to Dr. David Alling, National Institute of Allergy and Infectious Diseases, for statistical advice; to Dr. John Curd, Genentech, for assistance and advice; to Ms. Janet Andrews, Ms. Patricia Conville, and Dr. Frank Witebsky for microbiologic assistance; to Ms. Margaret Brown, Clinical Center, National Institutes of Health, for fluorescence-activated cell sorting; and to the physicians who referred and treated the patients: Patient 1 -- Dr. Raymond Caron, Orlando, Fla., and Dr. Blaese Congeni, Akron, Ohio; Patient 2 -- Dr. William Gardner, Akron, Ohio; Patient 3 -- Dr. J. Walton Tomford, Cleveland; Patient 4 -- Dr. David Mushatt, New Orleans; Patient 5 -- Dr. Steven Davis, Dallas; Patient 6 -- Dr. Louis Hammerman, Newark, N.J., and Ms. Doreen Chaitt, Clinical Center, National Institutes of Health; Patient 7 -- Dr. Benjamin Bridges and Dr. Richard J. Wallace, Jr., Tyler, Tex.; and the patients with tuberculosis -- Dr. Thomas Walsh, Montgomery County, Md., Department of Health.
Source Information
From the Laboratories of Host Defenses (S.M.H., J.I.G.) and Clinical Investigation (E.M.E., W.S.), National Institute of Allergy and Infectious Diseases; the Dermatology Branch, National Cancer Institute (M.L.T.); and Warren Grant Magnuson Clinical Center (T.A.F.) -- all at the National Institutes of Health, Bethesda, Md.; and PRI/DynCorp, Inc., Frederick, Md. (D.B.K.).
Address reprint requests to Dr. Holland at the Laboratory of Host Defenses, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bldg. 10, Rm. 11N103, 9000 Rockville Pike, Bethesda, MD 20892.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |