BACKGROUND. Assessing the outcome of patients with invasive pulmonary aspergillosis

Transcription

1 112 Immune Reconstitution Inflammatory Syndrome in Cancer Patients With Pulmonary Aspergillosis Recovering From Neutropenia: Proof of Principle, Description, and Clinical and Research Implications Marisa H. Miceli, MD 1 Johan Maertens, MD 2 Kristel Buvé, MD 2 Monica Grazziutti, MD 1 Gail Woods, MD 3 Mazhar Rahman, MD 1 Bart Barlogie, MD, PhD 1 Elias J. Anaissie, MD 1 1 Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, Little Rock, Arkansas. 2 Department of Hematology, University Hospital Gasthuisberg, Leuven, Belgium. 3 Department of Pathology, University of Arkansas for Medical Sciences, Little Rock, Arkansas. BACKGROUND. Assessing the outcome of patients with invasive pulmonary aspergillosis by using conventional criteria is difficult, particularly when clinical and radiologic worsening coincides with neutrophil recovery. Usually, it is assumed that this deterioration is related to progressive aspergillosis, prompting changes in patient management. However, its temporal relation with neutrophil recovery suggests that it may be caused by an immune reconstitution syndrome (IRIS). Galactomannan is an Aspergillus-specific polysaccharide that is released during aspergillosis and is detected by the serum galactomannan test, which has been approved by the United States Food and Drug Administration for the diagnosis of invasive aspergillosis. In this study, the authors used sequential galactomannan testing to distinguish IRIS responses from progressive aspergillosis. METHODS. From April 2001 to December 2006, patients with hematologic malignancies underwent galactomannan screening during periods when they were at risk. The clinical and laboratory findings from patients who had 2 consecutive positive galactomannan assays (optical density, 0.5) were reviewed. RESULTS. Nineteen neutropenic patients with aspergillosis developed clinical and radiologic pulmonary deterioration during neutrophil recovery. Deterioration coincided with microbiologic response, as documented by rapid normalization of serum galactomannan, and, in 16 patients, was followed by complete clinical response and survival at 3 months, although there were no changes in antifungal therapy. The 3 patients who died during the first month had no evidence of aspergillosis at autopsy examination. CONCLUSIONS. The authors propose that IRIS was responsible for the current findings and provide a definition for the syndrome. They also recommend serial galactomannan testing to guide aspergillosis management. Declining galactomannan values imply IRIS with an aspergillus response and obviate the need for invasive procedures and alternative antifungal therapies, whereas persistent galactomannan elevation indicates progressive aspergillosis and requires prompt treatment modification. Cancer 2007;110: Ó 2007 American Cancer Society. Address for reprints: Elias J. Anaissie, MD, Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, 4301 West Markham, Slot 816, Little Rock, AR 72205; Fax: (501) ; anaissieeliasj@ uams.edu Received February 14, 2007; revision received February 28, 2007; accepted March 1, KEYWORDS: galactomannan, hematologic malignancies, immune reconstitution syndrome, invasive aspergillosis. Invasive pulmonary aspergillosis is a life-threatening infection in patients with hematologic malignancies who are undergoing myelosuppressive chemotherapy. 1 Phagocytic cells (neutrophils and the macrophage/monocyte system) are the key immune defenses against aspergillosis, as demonstrated previously by experimental studies, 2 which showed the increased susceptibility of neutropenic patients to ª 2007 American Cancer Society DOI /cncr Published online 24 May 2007 in Wiley InterScience (

2 Aspergillosis and IRIS/Miceli et al. 113 aspergillosis and a strong relation between aspergillosis outcome and the resolution of neutropenia. 3 Two studies reported on 11 patients who had worsening clinical and radiologic pulmonary findings after neutrophil recovery. 4,5 Typically, this deterioration is considered progressive aspergillosis 6 and prompts invasive diagnostic procedures and treatment modifications, such as the use of cytokines, granulocyte transfusions and/or surgery, and the addition of new antifungal agents. 6,7 For example, Caillot et al. described a 4-fold increase in infiltrate size coinciding with neutrophil recovery among 25 patients who had invasive pulmonary aspergillosis. On the assumption that they had refractory aspergillosis, patients underwent modification of antifungal therapy and surgery. 7 Similar patients usually are enrolled in salvage trials of antifungal therapy in which refractory aspergillosis is defined as progression of clinical or radiographic signs despite 7 days of standard antifungal therapy. 6 Pivotal trials continue to use this enrollment strategy 8 despite its major pitfalls, as reported by Almyroudis et al. 6 We propose that the temporal association of pulmonary deterioration with neutrophil recovery, in fact, may be caused by the restored ability to mount an inflammatory response, in a manner reminiscent of the immune reconstitution inflammatory syndrome (IRIS) in patients with the human immunodeficiency virus (HIV) who respond to highly active antiretroviral therapy (HAART). 9,10 In the latter population, the diagnosis of IRIS is supported by new symptoms consistent with an infectious/inflammatory condition that is related temporally to the initiation of HAART and a robust decrease in the HIV-RNA level, which is a pathogen-specific, reproducible, and quantitative test. 9,10 In contrast, the assessment of aspergillosis response still relies on unvalidated, complex, composite endpoints (clinical, radiologic, and rarely histologic or microbiologic), which mostly are nonpathogenspecific, qualitative, and observer-dependent. 11 The United States Food and Drug Administrationapproved diagnostic test for aspergillosis is based on galactomannan, an Aspergillus-specific polysaccharide that is released during invasive infection. The test is reported as a galactomannan index, is quantitative and reproducible, 12 and appears to correlate with aspergillosis burden and outcome. 13,14 Herein, we report the development of IRIS in cancer patients with invasive pulmonary aspergillosis who were recovering from neutropenia and offer a galactomannan-based strategy to distinguish responding pulmonary aspergillosis with IRIS from progressive aspergillosis. We also describe the incidence, clinical and radiologic findings, and favorable outcome of patients with aspergillosis-related IRIS. MATERIALS AND METHODS This study was conducted at the University of Arkansas for Medical Sciences from November 2003 to September 2006 and was approved by the Institutional Review Board. The medical records of 67 patients with 2 consecutive positive galactomannan indices (optical density 0.5) were reviewed; 22 patients were excluded, including 14 patients who were receiving piperacillin-tazobactam and 8 patients who did not undergo optimal radiologic testing and/ or follow-up. In addition, 13 cancer patients with pulmonary aspergillosis who were cared for at the University Hospital Gasthuisberg (Leuven, Belgium) from April 2001 to December 2006 and who met our inclusion criteria were identified, and their medical records were reviewed. Galactomannan testing was performed and reported according to the manufacturer s instructions (Platelia Aspergillus EIA; Bio-Rad, Redmond, WA). Sera with an index 0.5 were retested the next day and were considered positive if the galactomannan index was 0.5. Neutropenia was defined as an absolute neutrophil count (ANC) 0 cells/ll. Invasive pulmonary aspergillosis was defined according to European Organization for Research and Treatment of Cancer/ Mycoses Study Group criteria. 15 Successful outcome was defined as a persistently negative galactomannan index with survival at 3 months in the absence of new extrapulmonary lesions of aspergillosis (eg, a culture-positive skin lesion and/or a skin lesion with hyphal tissue invasion consistent with aspergillosis) or a negative autopsy for aspergillosis in the event of death. Failure was defined as a persistently positive galactomannan index and/or death, unless autopsy examination failed to reveal aspergillosis. Pulmonary IRIS was defined as a new onset of or worsening clinical and radiologic pulmonary findings consistent with an infectious/inflammatory pulmonary condition temporally related to neutrophil recovery with evidence of microbiologic response (a decrease 50% in serum galactomannan index titers on 2 consecutive tests within 4 days of each other) despite no change in antifungal therapy. The definition also required absence of new extrapulmonary lesions of aspergillosis (eg, new skin lesions as described above) and/or other processes, such as newly acquired infection, failure of treatment of a known infection, or medication side effects. RESULTS Nineteen neutropenic patients with proven/probable invasive pulmonary aspergillosis fulfilled the IRIS criteria, including 11 patients with acute leukemia (58%), 6

3 114 CANCER July 1, 2007 / Volume 110 / Number 1 patients with myeloma (32%), and 1 patient each with non-hodgkin lymphoma and chronic myelogenous leukemia. The median patient age was 56 years (range, years), and there were 12 men and 7 women (Table 1). Aspergillosis developed after stem cell transplantation (autologous in 4 patients and allogeneic in 3 patients) or after conventional chemotherapy (12 patients). The mean days to clinical and radiologic findings of IRIS after an ANC >100/lL andananc >500/lL was 3.5 days (median, 3 days; range, from 7 days to 18 days) and 2 days (median, 1 day; range, from 8 daysto115 days), respectively. Clinical deterioration included worsening or new onset of hypoxia requiring oxygen therapy (12 patients), including mechanical ventilation in 4 patients, cough (8 patients), chest pain (6 patients), dyspnea (5 patients), and hemoptysis (3 patients). Radiologic findings consisted of increasing and/or new onset pulmonary infiltrates in 14 patients, pleural effusions in 10 patients, nodular lesions in 9 patients, intrathoracic lymphadenopathy in 3 patients, and cavitation in 1 patient. At diagnosis, all 19 patients had 2 consecutive positive serum galactomannan tests, prompting the initiation of antimould therapy. Declining titers of sequential serum galactomannan were observed during clinical and radiologic deterioration. Three patients died during the first month of follow-up (Patients 12, 14, and 16); autopsies on those 3 patients failed to demonstrate invasive aspergillosis. At 1-month follow-up, the remaining 16 patients improved without changes in antifungal therapy or additional therapies, such as cytokines or surgery; at 3 months, all 16 patients were alive without clinical, radiologic, or microbiologic evidence of aspergillosis. The serum Aspergillus galactomannan index was also negative. It is noteworthy that 2 patients with myeloma (Patients 1 and 5) received intravenous methylprednisone for impending respiratory failure. DISCUSSION Several new findings emerge from this study. First, we provide a proof of principle for pulmonary IRIS among patients with hematologic cancer and invasive pulmonary aspergillosis who are recovering from neutropenia. The immune-related nature of the pulmonary deterioration is supported by 1) the temporal association with neutrophil recovery, 2) the robust microbiologic response (decrease in galactomannan index titers), 3) the absence of new lesions from aspergillosis or other processes, 4) the subsequent resolution of aspergillosis without treatment modifications (confirming that the deterioration was not related to progressive aspergillosis), and 5) the improved outcome after methylprednisone in 2 patients with impending respiratory failure (providing further support for the immune-related nature of this syndrome). Worsening of pulmonary findings of aspergillosis during neutrophil recovery has been reported among patients with hematologic cancer 4,7 butalmostalways was considered evidence of progressive infection. However, because of the absence of microbiologic documentation in those reports, it remains unclear whether patients had resolving aspergillosis with IRIS or progressive infection. In another report, 3 patients with pulmonary aspergillosis developed respiratory failure during neutrophil recovery; 2 of those patients responded to methylprednisone without changes in antifungal therapy, suggesting that deterioration was related to the resolution of neutropenia. 4 Our data support the latter findings and now provide evidence, for the first time, that pulmonary deterioration can be immune-related with infection response and not refractory aspergillosis. 11 Indeed, patients who have worsening clinical and radiologic findings always are considered to have progressive infection, prompting changes in their management. Then, the subsequent clinical and radiologic improvements are considered proof of efficacy of this agent (often leading to regulatory approval), without any consideration for the possibility that deterioration leading to enrollment, in fact, may be aspergillosis-related IRIS with response occurring prior to investigational therapy. The incidence of HIV-associated IRIS is from 10% to 25%, 9 and pulmonary IRIS has been associated with various pathogens, including Aspergillus spp. 16 HIVassociated IRIS appears to develop later (mean, 33 days; range, days after HAART) 9 than among our patients (mean, 2 days; range, from 8 days to 115 days after the resolution of neutropenia). It remains to be determined whether IRIS in our patients was caused by neutrophil recovery only or by the combined effect of recovery and antifungal therapy. Second, we provide a definition of pulmonary IRIS among cancer patients that was adapted from the definition proposed for patients with HIV 10 (Table 2). Similar to the definition of HIV-IRIS, our definition is clinical, includes a microbiologic marker of response (serum galactomannan index), but also requires measuring the key immune parameter in aspergillosis among neutropenic patients, ie, the neutrophil count. Third, we describe the clinical and radiologic spectrum of IRIS in 19 cancer patients with pulmonary aspergillosis who were recovering from neutropenia andreportthatirismaynotbeuncommonamong such patients. Fourth, our data suggest that cancer patients with pulmonary aspergillosis and IRIS may have a good prognosis, similar to that for HIV-infected

4 Aspergillosis and IRIS/Miceli et al. 115 TABLE 1 Cancer Patients With Invasive Pulmonary Aspergillosis and Immune Reconstitution Inflammatory Syndrome After Neutrophil Recovery* Pulmonary IRIS features y Patient sex Age, y Underlying disease and status Therapy Findings at diagnosis Date of IRIS Clinical deterioration Worsening radiology Temporal relation to neutrophil recovery 1 W 61 Recurrent myeloma 2 W 60 Recurrent myeloma 3 M 67 Recurrent myeloma 4 W 73 Myeloma in remission 5 M 72 Myeloma in remission 6 M 58 Recurrent myeloma 3/27/04: 4/1-24/04: ANC 7/8/04: ASCT; 7/13-19/04: ANC 12/20/04: 12/28/04-1/3/05: 3/17/05: ASCT; 3/21-27/05: ANC 8/29/05: ASCT; 9/4-8/05: ANC 12/5/05: ASCT; 12/14-24/05: ANC 4/12-22/04: Positive (peak, 4.14; 4 tests), cough, crackles; 4/5/04: CT shows apical scarring, Voriconazole 400 mg/d (started 4/13/04) 7/14-31/04: Positive (peak, 2.64; 7 tests); 7/14-19/04: Fever and hypoxia, CT shows infiltrates and effusions; Voriconazole 600 mg/d (7/19-23/04), Ambisome 150 mg/d (started 7/23/04) Minimal crackles; 12/30/05: CT shows minimal bilateral interstitial infiltrates; 12/30/04-1/5/05: Positive (peak, 6.5; 5 tests), Ambisome 200 mg/d (started 1/3/05) 3/25/05: RML rales and hypoxia, CT shows bilateral upper lobe infiltrates; 3/25/05-4/3/05: Positive (peak, 3.3; 6 tests), Voriconazole 600 mg/d (started 3/29/05) 9/2/05: ANC 400, 5.8, fever; 9/2-7/05: Positive (peak, 6; 3 tests); 9/9/05: ANC >1000, fever, CT shows scarring LLL; Voriconazole 400 mg/d (9/9-11/05), Ambisome 300 mg/d (started 9/12/05) 12/20-24/05: Fever, cough, and crackles; CT shows bilateral lower lobe consolidations; 12/ 20/05-1/2/06: Positive (peak, 6.3; 8 tests); Voriconazole 400 mg/d (started on 12/22/05) 5/3-17/04 Cough, crackles, wheezing, hypoxia requiring O 2, and IV methylprednisone 8/5-19/04 Elevated fever, worsening hypoxia 1/12-18/05 Fever, cough, crackles, wheezing, hypoxia 3/31/ 05-4/7/05 Cough, bilateral crackles, wheezing, hypoxia 9/11-15/05 Elevated fever, wheezing, and worsening hypoxia requiring ICU transfer and IV methylprednisone 12/25/05-1/7/06 Mechanical ventilation for respiratory failure 5/3/04: CT shows extensive bilateral effusions, infiltrates, nodules; 5/17/04: CT shows new mediastinal and precarinal lymph nodes 8/20/04: CT shows worsening infiltrates 1/12/05: CT shows bilateral alveolar opacities and effusions, new right lung nodule 4/7/05: CT shows stable infiltrates; new pretracheal, precarinal and aortopulmonary window lymph nodes 9/11/05: CT shows new, diffuse, bilateral infiltrates and nodules 12/26/05: CT shows worsening bilateral infiltrates, new effusions 5/1/04: ANC >500; 5/5/04: ANC 7/20/04: ANC >500; 7/24/04: ANC 1/5/05: ANC >500; 1/8/05: ANC 3/28/05: ANC >500; 3/31/05: ANC 9/9/05: ANC >500; 9/10/05: ANC 12/25/05: ANC >500; 12/29/05: ANC Resolving antigenemia 4/19/04: decreased >50% decrease; 4/22/05: Negative 7/29/04: 8/3/04: Negative ; 1/13/05: Negative 4/1/05: 4/4/05: Negative 9/12/05: Negative 1/2/06: 2/26/06: Negative (continued)

5 116 CANCER July 1, 2007 / Volume 110 / Number 1 TABLE 1 (Continued) Pulmonary IRIS features y Patient Sex Age, y Underlying disease and status Therapy Findings at diagnosis Date of IRIS Clinical deterioration Worsening radiology Temporal relation to neutrophil recovery 7 M 53 Recurrent AML 10/22/06: 10/31/06-11/18/ 06: 8 W 52 NHL in remission 7/22/06: Allo-SCT; 7/21/06-12/1/06: 9 W 36 Recurrent ALL 5/16/06: 5/21/06-6/21/06: 10 W 25 Recurrent ALL 1/8/03: 1/15/03-2/13/03: 11 M 59 ALL in remission 5/7/03: 5/19/03-6/22/03: 10/31/06: Fever, dyspnea, and cough; CT shows bilateral lower lobe consolidations; 10/31/06-11/18/06: Positive (peak, 8.2; 21 tests); Ambisome 5 mg/ kg/d (started 11/3/06) 9/16/06: Fever, dry cough, and dyspnea; CT shows 2 nodules with halo sign in RUL; 9/16/06-12/3/06: Positive (peak, 6.8; 83 tests); Voriconazole 400 mg/d (started on 9/16/06) 6/6/06: Confusion, fever, dyspnea, cough; CT shows nodule in RUL with halo sign; 6/10-21/06: Positive (peak, 3.1; 12 tests); Caspofungin mg/d (started on 6/10/06) 1/30/03: Fever and cough, CT shows small nodular infiltrates in the RUL; 1/30/03-2/20/03: Positive (peak, 2.2; 18 tests); Ambisome 5 mg/k/d (started on 1/30/03); 2/6/03: CT shows significant increase of the initial lesion and pleuritic fluid 6/13/03: Fever, dyspnea, cough; CT not available; chest x-ray shows bilateral patchy infiltrates; 6/13/03-7/29/03: Positive (peak, 2.7; 25 tests); Voriconazole 400 mg/d (started on 6/14/03) 11/18/06-12/5/06 Mechanical ventilation for respiratory failure 12/4-15/06 Pleuritic chest pain, hypoxia requiring O 2, worsening productive cough 6/16-22/06 Increased dry cough, persistent fever, pleuritic pain 2/17-28/03 Hypoxia requiring O 2, bilateral crackles 6/24/03-7/3/03 Mechanical ventilation for respiratory failure CT not feasible; daily chest x-ray; progression of pulmonary infiltrates 12/6/06: CT shows increased volume of previous lesions, new nodular lesion in LLL 6/16/06 and 6/23/06: CT shows new infiltrates with halo in RLL, subpleural and lingula; 6/30/06: CT shows air crescent formation in 2 lesions 2/17/03: CT shows further increase of the original lesion and new apical infiltrate in right lower mediastinal and axillary lymph nodes, pleural fluid X-ray shows bilateral, patchy infiltrates while in mechanical ventilation; pneumothorax on 7/2/03 11/20/06: ANC >500; 11/24/06: ANC 12/3/06: ANC >500; 12/8/06: ANC 6/23/06: ANC >500; 6/29/06: ANC 2/14/03: ANC >500; 2/21/03: ANC 6/25/03: ANC >500; 7/4/03: ANC Resolving antigenemia 11/22/06: 11/28/06: Negative 12/1/06: 12/4/06: Negative 6/20/06: 6/22/06: Negative 2/17/03: 2/21/03: Negative 6/30/03: 7/30/03: Negative (continued)

6 Aspergillosis and IRIS/Miceli et al. 117 TABLE 1 (Continued) Pulmonary IRIS features y Patient Sex Age, y Underlying disease and status Therapy Findings at diagnosis Date of IRIS Clinical deterioration Worsening radiology Temporal relation to neutrophil recovery 12 M 71 Recurrence AML 9/12/03: Allo-SCT; 9/8-21/03: ANC 13 M 18 Recurrent ALL 12/28/03: 12/31/03-1/12/04: 14 M 68 Newly diagnosed AML 7/13/02: 7/21/02-8/6/02: 15 W 54 Recurrent AML 4/6/01: 4/15/01-5/15/01: 16 M 67 Recurrent AML 30/10/06: 11/6-12/9/06: 17 M 60 Chronic-phase CML 10/6/04: Allo-SCT; 6/1-28/05: ANC 9/11/03: Fever and dyspnea; 9/12/03: CT shows nodule in RML and lingula with halo sign; 9/11/03-10/9/03: Positive (peak, 6.1; 24 tests); Ambisome 5 mg/k/d (started on 9/12/03) 1/1/04: Fever, cough, and hemoptysis; 1/5/04: CT shows bilateral nodular lesions surrounded by halo;1/1-12/04: Positive (peak, 2.1; 10 tests); Ambisome 5 mg/k/d (started on 1/3/04) 8/2/02: Fever, dyspnea; 8/4/02: CT shows bilateral lower lobe consolidations; 8/2-22/02: Positive (peak, 5.2; 19 tests); Anidulafungin 200 mg/d and Ambisome 5 mg/k/d (started on 8/4/02) 4/30/01: Fever, cough, dyspnea; CT shows consolidation in LUL; 4/30/01-5/19/01: Positive (peak, 4.1; 20 tests); Ambisome 5 mg/kg/d (started on 8/4/01) 12/5/06: Fever, dry cough; CT shows consolidation in left suprahilar region with halo; 12/ 6-12/06: Positive (peak, 1.8; 7 tests); Caspofungin 70/50 mg/d (started on 12/5/06) 6/13/05: Fever, cough; CT shows multiple nodules in the right lung with halo; 6/13-22/05: Positive (peak, 1.4; 5 tests); Caspofungin 70/50 mg/d (started on 6/13/05) 9/23-28/03 Mechanical ventilation for respiratory failure 1/14-/26/04 Increased cough, hypoxia requiring O2, pleuritic chest pain requiring morphine, crackles 8/12-17/02 Worsening dyspnea, hemoptysis, wheezing 5/18-24/01 Worsening cough, hypoxia requiring O 2, pleuritic chest pain requiring morphine, transfer to ICU 12/13-22/06 Worsening cough, dyspnea, hypoxia requiring O2, pleuritic chest pain requiring morphine 6/29/05-7/5/05 Worsening dyspnea, hemoptysis Chest x-rays show progressive increase in radiologic abnormalities; 10/15/03: CT shows cavitation of previously seen nodule, new nodules and infiltrates in both lungs, large bilateral pleural fluid 1/14/04: CT shows new lesions, pleural fluid, increased volume of initial lesions 8/12/02: CT shows increased volume of initial lesions; new pleural fluid 5/18/01: CT shows new lesion RUL, pleural fluid, increased volume of initial infiltrate in LUL; 6/5/01: CT shows further increase of volume lesions in both lungs 12/13/06: CT shows increase of initial consolidation, new nodule in RLL, ground-glass opacities, bilateral pleural fluid 6/29/05: CT shows doubling volume of initial lesions, pleural fluid, multiple mediastinal lymph nodes 9/22/03: ANC >500; 9/23/03: ANC 1/14/04: ANC > 500; 1/19/04: ANC 8/8/02: ANC >500; 8/12/02: ANC 5/17/01: ANC >500; 5/23/01: ANC 12/11/06: ANC >500; 12/17/06: ANC 6/30/05: ANC >500; 7/1/05: ANC Resolving antigenemia 9/28/03: 10/10/03: Negative 1/9/04: 1/13/04: Negative 8/9/02: 8/23/02: Negative 5/15/01: 5/20/01: Negative 12/11/06: 12/13/06: Negative 6/21/05: 6/23/05: Negative (continued)

7 118 CANCER July 1, 2007 / Volume 110 / Number 1 TABLE 1 (Continued) Pulmonary IRIS features y Patient Sex Age, y Underlying disease and status Therapy Findings at diagnosis Date of IRIS Clinical deterioration Worsening radiology Temporal relation to neutrophil recovery Resolving antigenemia 18 M 57 Newly diagnosed AML 12/12/05: 12/25/05-1/18/06: 19 M 57 Recurrent AML 8/7/06: 8/14-29/06: ANC 1/11/06: Fever, productive cough, mild chest pain; CT shows nodule in the RUL with halo; 1/11-22/06: Positive (peak, 3.9; 12 tests); Caspofungin 70/ 50 mg/d (started on 1/14/06) 8/25/06: Fever, dyspnea, cough; 8/28/06: Bilateral lower lobe consolidations surrounded by halo; 8/25/06-9/5/06: Positive (peak, 2.8; 12 tests); Caspofungin 70/50 mg/d (started on 8/28/06) 1/20-25/06 Worsening chest pain, 1/20/06: CT shows increased hypoxia requiring O 2 volume of initial nodule, progression of infiltrates 9/5-14/06 Hemoptysis, hypoxia requiring O2 and BiPAP 9/8/06: CT shows worsening of initial lesions, new apical infiltrate, bilateral pleural fluid 1/21/06: ANC >500; 1/27/06: ANC 8/31/06: ANC >500; 9/6/06: ANC 1/18/06: 1/23/06: Negative 9/3/06: 9/6/06: Negative IRIS indicates immune reconstitution inflammatory syndrome; W, woman; ANC, absolute neutrophil count (cells/ll); CT, chest computed tomography scan; IV, intravenous;, serum galactomannan index; ASCT, autologous stem cell transplantation; M, man; RML, right middle lobe; LLL, left lower lobe; ICU, intensive care unit; AML, acute myeloid leukemia; NHL, non-hodgkin lymphoma; Allo-SCT, allogenic stem cell transplantation; RUL, right upper lobe; ALL, acute lymphoid lympyhoma; LUL, left upper lobe; RLL, right lower lobe; BiPaP, bilevel positive air pressure. * Patients 12, 14, and 16 died during the first month of follow-up; in those 3 patients, autopsies failed to demonstrate invasive aspergillosis. At 1 month of follow-up, the remaining 16 patients improved without changes in antifungal therapy; at 3 months, all patients were alive without clinical, radiologic, or microbiologic evidence of aspergillosis. The serum Aspergillus galactomannan index also was negative. y Pulmonary IRIS was defined as new onset or worsening of clinical and radiologic pulmonary findings consistent with an infectious/inflammatory pulmonary condition temporally related to neutrophil recovery and a decrease 50% in serum titers (2 consecutive tests within 4 days) in the absence of new extrapulmonary lesions of aspergillosis (eg, the aforementioned new skin lesions) and after excluding other causes, such as newly acquired infection, failure of treatment of a known infection, or medication side effects.

8 Aspergillosis and IRIS/Miceli et al. 119 TABLE 2 Definition of Immune Reconstitution Inflammatory Syndrome (IRIS): HIV Patients After HAART vs. Cancer Patients Following Neutrophil Recovery Criteria IRIS in HIV patients Pulmonary IRIS in cancer patients with aspergillosis Required Clinical Worsening symptoms of inflammation/infection New onset of or worsening clinical and radiological pulmonary findings consistent with an infectious/inflammatory pulmonary condition Temporal relationship with starting antiretroviral treatment Temporal relationship with neutrophil recovery Symptoms not explained by newly acquired infection or disease or the usual course of a previously acquired disease Absence of new extrapulmonary lesions of aspergillosis (eg, new skin lesions above-described) and after exclusion of other causes, such as newly acquired infection, failure of treatment of a known infection, or medication side effects Laboratory 1 Log 10 decrease in plasma HIV load 50% decrease in serum titers without treatment modifications Supportive Clinical Subsequent resolution of aspergillosis without treatment modifications Laboratory Increase in CD4 1 cell count of 25 cells/mm 3 To be determined but likely to include: Immune function studies such as immunophenotype (CD4, CD8, others), cytokines, others Biopsy demonstrating well-formed granulomatous inflammation or unusually exuberant inflammatory response IRIS indicates immune reconstitution inflammatory syndrome;, serum galactomannan index. Biopsy demonstrating well-formed granulomatous inflammation or unusually exuberant inflammatory response patients in whom IRIS was associated with successful immune reconstitution, decreasing HIV viral load, and a trend toward increased survival. 17 Additional experience will be needed to confirm our findings, because it is likely that some patients with invasive pulmonary aspergillosis and IRIS may not survive in face of overwhelming respiratory deterioration. Our findings imply that pulmonary deterioration among cancer patients with invasive pulmonary aspergillosis who are recovering from neutropenia does not necessarily indicate progressive aspergillosis, particularly when such deterioration coincides with a robust decrease in aspergillus antigenemia. Such patients should be continued on the same therapy with consideration for a 3- to 7-day course of corticosteroids (eg, 2 mg/kg methylprednisone per day) for the occasional patient with impending respiratory failure (Table 1). Corticosteroids have been used successfully in HIV-positive patients who had IRIS associated with various opportunistic infections. 17,18 In contrast, persistently elevated galactomannan titers suggest progressive aspergillosis and should prompt treatment modification. Hence, we recommend serial serum galactomannan testing in patients with pulmonary deterioration. Work-up to exclude other infections may be needed, although breakthrough infections are least likely during neutrophil recovery. The current findings also imply that clinical trials of progressive aspergillosis should rely on serial galactomannan testing and should exclude patients who have resolving antigenemia and no other evidence of progressive aspergillosis. Our study should be confirmed by others and broadened to include other populations at risk. 1,3 A prospective validation of our definition also is needed. We conclude that pulmonary IRIS may develop among patients with invasive aspergillosis and hematologic cancers who are recovering from neutropenia. Serial galactomannan testing is critical to distinguish this syndrome from progressive aspergillosis. This strategy is likely to improve patient outcomes and to enhance the precision of trials that investigate treatment strategies for aspergillosis. REFERENCES 1. Perfect JR, Cox GM, Lee JY, et al. The impact of culture isolation of Aspergillus species: a hospital-based survey of aspergillosis. Clin Infect Dis. 2001;33: Robertson MD, Seaton A, Raeburn JA. Phagocytic cell responses to Aspergillus fumigatus. FEMS Microbiol Immunol. 1989;1: Segal BH, Walsh TJ. Current approaches to diagnosis and treatment of invasive aspergillosis. Am J Respir Crit Care Med. 2006;173: Takuma T, Okada K, Uchida Y, Yamagata A, Sawae Y. Invasive pulmonary aspergillosis resulting in respiratory failure during neutrophil recovery from postchemotherapy neutropenia in three patients with acute leukaemia. J Intern Med. 2002;252: Todeschini G, Murari C, Bonesi R, et al. Invasive aspergillosis in neutropenic patients: rapid neutrophil recovery is a risk factor for severe pulmonary complications. Eur J Clin Invest. 1999;29:

9 120 CANCER July 1, 2007 / Volume 110 / Number 1 6. Almyroudis NG, Kontoyiannis DP, Sepkowitz KA, DePauw BE, Walsh TJ, Segal BH. Issues related to the design and interpretation of clinical trials of salvage therapy for invasive mold infection. Clin Infect Dis. 2006;43: Caillot D, Couaillier JF, Bernard A, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. J Clin Oncol. 2001;19: Walsh TJ, Raad I, Patterson TF, et al. Treatment of invasive aspergillosis with posaconazole in patients who are refractory to or intolerant of conventional therapy: an externally controlled trial. Clin Infect Dis. 2007;44: Shelburne SA 3rd, Hamill RJ, Rodriguez-Barradas MC, et al. Immune reconstitution inflammatory syndrome: emergence of a unique syndrome during highly active antiretroviral therapy. Medicine (Baltimore). 2002;81: Robertson J, Meier M, Wall J, Ying J, Fichtenbaum CJ. Immune reconstitution syndrome in HIV: validating a case definition and identifying clinical predictors in persons initiating antiretroviral therapy. Clin Infect Dis. 2006;42: Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N Engl J Med. 2002;347: Wheat LJ. Rapid diagnosis of invasive aspergillosis by antigen detection. Transpl Infect Dis. 2003;5: Maertens J, Verhaegen J, Demuynck H, et al. Autopsycontrolled prospective evaluation of serial screening for circulating galactomannan by a sandwich enzyme-linked immunosorbent assay for hematological patients at risk for invasive Aspergillosis. J Clin Microbiol. 1999;37: Woods G, Miceli M, Grazziutti M, et al. Validation of serum aspergillus galactomannan index as a surrogate endpoint for outcome of invasive aspergillosis: clinical and research implications. Paper presented at: 48th Annual Meeting of the American Society of Hematology, December 11, 2006; Orlando, Fla. Abstract Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis. 2002;34: Sambatakou H, Denning DW. Invasive pulmonary aspergillosis transformed into fatal mucous impaction by immune reconstitution in an AIDS patient. Eur J Clin Microbiol Infect Dis. 2005;24: Shelburne SA, Visnegarwala F, Darcourt J, et al. Incidence and risk factors for immune reconstitution inflammatory syndrome during highly active antiretroviral therapy. AIDS. 2005;19: Montaner JS, Lawson LM, Levitt N, Belzberg A, Schechter MT, Ruedy J. Corticosteroids prevent early deterioration in patients with moderately severe Pneumocystis carinii pneumonia and the acquired immunodeficiency syndrome (AIDS). Ann Intern Med. 1990;113:14 20.