Journal of Infection (2015) xx, 1e9
www.elsevierhealth.com/journals/jinf
Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis Jiwon Jung a,c, Hyo-Lim Hong b,c, Sang-Oh Lee a, Sang-Ho Choi a, Yang Soo Kim a, Jun Hee Woo a, Sung-Han Kim a,* a
Department of Infectious Diseases, Asan Medical Center, University of Ulsan College of Medicine, University of Ulsan, Seoul, Republic of Korea b Division of Infectious Diseases, Department of Internal Medicine, Daegu Catholic University Medical Center, Daegu, Republic of Korea Accepted 18 December 2014 Available online - - -
KEYWORDS Invasive pulmonary aspergillosis; Neutropenia; Immune reconstitution inflammatory syndrome
Summary Objectives: Clinical and radiologic deterioration is sometimes observed during neutrophil recovery in patients with invasive pulmonary aspergillosis (IPA). This deterioration can be caused by immune reconstitution inflammatory syndrome (IRIS) as well as by progression of the IPA. However, there is limited data on IRIS in neutropenic patients. Methods: Over a 6-year period, adult patients with neutropenia who met the criteria for probable or proven IPA by the revised EORTC/MSG definition were retrospectively enrolled. IRIS was defined as de novo appearance or worsening of radiologic pulmonary findings temporally related to neutrophil recovery, with evidence of a decrease of 50% in serum galactomannan level. Results: Of 153 patients, 36 (24%, 95% CI 18%e31%) developed IRIS during neutrophil recovery. More of these patients received voriconazole than did those with non-IRIS (42% vs. 25%, P Z 0.05). Thirty- and ninety-day mortalities were lower in the patients with IRIS than in those with non-IRIS (11% vs. 33%, P Z 0.01, and 33% vs. 58%, P Z 0.01, respectively). Conclusion: IRIS is relatively common among neutropenic patients with IPA, occurring in about one quarter of such patients. It is associated with voriconazole use and has a good prognosis. ª 2015 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Department of Infectious Diseases, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnapdong, Songpa-gu, Seoul, 138-736, Republic of Korea. Tel.: þ82 2 3010 3305; fax: þ82 2 3010 6970. E-mail address: [email protected] (S.-H. Kim). c These authors contributed equally to this manuscript. http://dx.doi.org/10.1016/j.jinf.2014.12.020 0163-4453/ª 2015 The British Infection Association. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
2
Introduction Invasive pulmonary aspergillosis (IPA) is associated with high morbidity and mortality in patients with hematologic malignancies who are undergoing myelosuppressive chemotherapy or receiving hematopoietic stem cell transplants.1 Since alveolar macrophages and neutrophils play a dominant role in host defenses against aspergillosis,2 neutropenia from chemotherapy or underlying hematologic disease is an important predisposing host factor for IPA. Sometimes clinical and radiologic deterioration is observed during neutrophil recovery in patients with IPA, and this deterioration can be associated with immune reconstitution inflammatory syndrome (IRIS) or with the progression of the IPA. Caillot et al. showed that the median volume of lesions in neutropenic patients with IPA, as measured serially by CT, increased four-fold from day 0 to day 7 despite antifungal therapy.3 However, many physicians consider that worsening of the radiologic findings of IPA during neutrophil recovery is associated with progression of the infection and they modify the antifungal therapy with or without surgery.4 However, if there is no microbiologic documentation, it is unclear whether this radiologic deterioration is associated with IPA progression or with IRIS. Recently, the GM assay has been proposed as a microbiologic marker of response.5,6 If so, it might be possible to decide whether the radiologic worsening of IPA in neutropenic patients is due to IPA progression or immune reconstitution inflammatory syndrome (IRIS) by assessing clinical and microbiologic parameter as well as immunologic parameters such as neutrophil count. This distinction helps clinicians to guide antifungal therapy in this difficult clinical situation. In addition, this procedure could detect refractory IPA more objectively, and could advance clinical study of salvage therapy in refractory IPA without dilution of the results of studies by the inclusion of patients with IRIS. The concept of fungal IRIS was originally proposed in patients starting antiretroviral therapy for human immunodeficiency virus (HIV) infection and was clearly demonstrated in those with cryptococcal infection, histoplasmosis, and candidiasis.7 In non-HIV patients, cases of fungal IRIS have been reported in transplant recipients,8e10 neutropenic patients,11,12 and recipients of TNFa receptor inhibitors.13,14 Recently, Singh et al. reported that about 5% of solid organ transplant recipients with cryptococcal infections had IRIS-like illnesses, and these appeared to be associated with a certain immunosuppressive regimen.9 In addition, Legrand, et al. showed that some neutropenic cancer patients with chronic disseminated candidiasis who were unresponsive to antifungal therapy and improved dramatically after corticosteroid therapy appeared to suffer from what they suggested to be fungal IRIS.15 Miceli et al. proposed a similar concept of IRIS in patients with IPA; this was defined as the onset or worsening of clinical and radiological findings together with neutrophil recovery and a 50% decrease in serum GM titers that could not be explained by newly acquired infection, failure of treatment, or side effects of treatment.4 They described 19 neutropenic non-HIV patients conforming to this definition of fungal IRIS.4 However, data are limited on the incidence, clinical characteristics, and
J. Jung et al. outcomes of IRIS associated with IPA in neutropenic cancer patients. We therefore evaluated the incidence, clinical characteristics, and outcomes of IRIS in neutropenic patients with IPA.
Patients and methods Study population and data collection This study was performed at the Asan Medical Center, a 2700-bed tertiary care teaching hospital in Seoul, South Korea. The medical records of adult patients ( 16 years old) with neutropenia who met the criteria for proven or probable IPA from January 2008 to December 2013 were retrospectively reviewed.
Definitions Patients were assigned a proven or probable diagnosis of IPA based on a modification of the consensus definition of EORTC/MSG and our previous studies.1,16e20 Proven IPA was defined by histologic evidence of tissue invasion including septated, acutely branching filamentous fungi and positive culture or positive GM assay. Probable IPA was defined as the presence of host factors, together with one or more clinical indication such as dense, well circumscribed lesions with or without a halo sign, and an air-crescent sign, or cavity on CT; and mycologic evidence of fungal infection (by culture or cytologic analysis of bronchoalveolar lavage fluid for Aspergillus species, or GM assay of serum or bronchoalveolar lavage fluid). Possible IPA was defined as the presence of a host factor and clinical criteria without mycologic evidence for IPA. Cases classified as possible IPA were not included in this analysis. Neutropenia was defined as an absolute neutrophil count 3 weeks.1 Immunosuppressant use was defined as treatment with a T-cell immunosuppressant such as cyclosporin, TNF-a blocker, specific monoclonal antibody (such as alemtuzumab), or purine analogs during the previous 90 days.1 Serum GM antigen levels were measured as described previously (Platelia Aspergillus EIA; BioRad, Redmond, WA).18,19 They were considered positive if the GM index was 0.5 serum GM antigen level. IRIS was defined as new or worsening radiologic pulmonary findings temporally related to neutrophil recovery, with evidence of a 50% decrease in a single serum GM index titer combined without evidence of persistent positive fungal culture and absence of new extrapulmonary lesions of aspergillosis (e.g., new skin lesions) and other problem, such as newly acquired infection, failure of treatment of a known infection, or side effect of medication.4 Clinical response to antifungal therapy was classified as success (complete or partial response) or failure (stable response, progression of disease, or death) as described previously22 and assessed 10 days after neutrophil recovery (if a patient died before neutrophil recovery, failure was assigned to the day of death), 1 month, and 3 months after diagnosis,
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
Immune reconstitution syndrome in IPA respectively. Radiologic criteria used in the classification of clinical response were as follows22; complete response was defined as resolution of radiological lesion(s), partial response as at least a 25% reduction in diameter of the radiological lesion(s), stable response as 0%e25% reduction in the diameter of the lesion(s), and progression of disease as new sites of disease or worsening of preexisting lesions(s). Antifungal prophylaxis was defined as the use of an antifungal agent with anti-Aspergillus activity (e.g. voriconazole, posaconazole, itraconazole, caspofungin, or amphotericin B) for > 5 continuous days without any signs or symptoms of invasive mold infection.23
Statistical analysis All statistical analyses were performed with SPSS version 20.0 (SPSS, Chicago, IL). Categorical variables were compared using the c2or Fisher’s exact test, and continuous variables were compared using the ManneWhitney U test, as appropriate. A P value 0.10 from the equation.
Results Patient characteristics During the study period, a total 153 neutropenic patients with IPA (10 proven and 143 probable) were analyzed. The most frequent underlying hematologic disease was acute myeloid leukemia (76% [116/153]). Of the 153 patients, 51 (33%) received hematopoietic stem cell transplantations (HCT) including 48 allogeneic HCT and 3 autologous HCT. The median neutrophil count of the 153 patients at diagnosis of IPA was 3/mL (interquartile range [IQR] 0e61). Neutrophil recovery occurred in 103 of the 153 patients. The median length of time between neutropenia and the diagnosis of IPA was 23 days (IQR 15e44). The baseline clinical characteristics of these patients are shown in Table 1.
IRIS versus non-IRIS in neutropenic cancer patients with IPA Of the 153 patients with IPA, 36 (24%, 95% CI 18%e31%) developed IRIS during neutrophil recovery (Fig. 1). The proportions of patients who underwent HCT (IRIS 25% vs. nonIRIS 36%; P Z 0.23) were not significantly different in the two groups. Although the effect was not statistically significant, fewer of the patients with IRIS received corticosteroids (IRIS, 19% vs. non-IRIS 34%; P Z 0.09), more of the patients with IRIS received colony stimulating factor after diagnosis of IPA (IRIS 72% vs. non-IRIS 55%; P Z 0.06), and more of the patients with IRIS had the underlying disease
3 controlled within 3 months (31% vs. 17%, P Z 0.08). The median length of neutropenic periods in the patients with IRIS was shorter than in the patients with non-IRIS (29 days [IQR 19-49] vs. 40 [IQR 23-87], P Z 0.03). Among the patients with IRIS, clinical deterioration such as hypoxia, tachycardia and fever was observed in 19 (53%) patients. The median days from initiation of antifungal therapy to IRIS was 16 days (IQR 6e18). Fourteen (39%) patients received changed antifungal therapy upon development of IRIS. Interestingly, patients with IRIS received voriconazole more frequently on the day of diagnosis of IPA than those with non-IRIS (42% vs. 25%, P Z 0.05). The 30- and 90-day mortalities were lower in patients with IRIS than in those with non-IRIS (11% vs. 33%, P Z 0.01, and 33% vs. 58%, P Z 0.01, respectively).
IRIS versus non-IRIS in 54 neutropenic cancer patients with IPA who underwent neutrophil recovery and radiologically progressive disease Of the 153 patients, 50 (33%) who died without neutrophil recovery were excluded and the remaining 103 (67%) who underwent neutrophil recovery were further analyzed. Of these 103 patients, 54 (52%) had radiologically progressive disease during the neutrophil recovery while 49 (48%) had radiologically stable disease or partial/complete responses during the neutrophil recovery (Fig. 1). In a subgroup analysis of the 54 patients who had radiologically progressive disease during neutrophil recovery, 36 (67%, 95% CI 53%e 78%) were assigned to the IRIS group and the remaining 18 to the non-IRIS group (Table 2). The 30- and 90-day mortalities were higher in the patients with non-IRIS who had progressive disease during neutrophil recovery than in those with IRIS (11% vs. 39%, P Z 0.03 and 33% vs. 72%, P Z 0.007, respectively).
Outcomes and prognostic factors according to host status with fungal burden and radiologic response Fig. 1 shows the detailed classification of the 153 patients with IPA according to host status with fungal burden and radiologic progression. The baseline clinical characteristics and outcomes depending on host status with fungal burden and radiologic progression are shown in Supplemental Table 1. We analyzed risk factors for 30-day mortality in the overall patient population (Table 3). Multivariate analysis indicated that radiologic stable or partial/complete response during neutrophil recovery (OR 0.07, P < 0.001) and IRIS during neutrophil recovery (OR 0.26, P < 0.001) were independently associated with 30-day mortality. The survival curves stratified according to host status with fungal burden and radiologic progression are presented in Fig. 2.
Discussion Here we found that IRIS occurred in about one quarter of all neutropenic patients with IPA, and in two-thirds of patients with IPA who underwent radiologically progressive disease during neutrophil recovery. We found that IRIS was
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
4
J. Jung et al. Table 1 Clinical characteristics and outcomes of neutropenic patients with and without immune reconstitution inflammatory syndrome (IRIS).
Age, median years (IQR) Male gender Underlying disease and/or conditions Acute leukemia Aplastic anemia Lymphoma Receipt of HCT Allogeneic HCT Autologous HCT Diabetes mellitus Prior corticosteroid usea Prior immune suppressant useb Colony stimulating factor use after diagnosis of IPA State of underlying disease Controlled Relapsed or refractory Mechanical ventilation Length of neutropenic period, median days (IQR) Length of time between neutropenia and diagnosis of IPA, median days (IQR) IPA diagnosis category Proven Probable Positive Aspergillus culture Sputum Bronchioalveolar lavage Positive Aspergillus culture after neutrophil recovery Peak galactomannan level, median value (IQR) Antifungal therapy on diagnosis day of IPA Amphotericin B Voriconazole Itraconazole Liposomal amphotericin B Caspofungin Received prophylactic antifungal agents Itraconazole Amphotericin B Micafungin Voriconazole Posaconazole Antifungal therapy during IRIS (Liposomal) Amphotericin-B to voriconazole Voriconazole to echinocandin or combination of echinocandin Voriconazole to amphotericin-B Corticosteroid use during IRIS Length of time from initiation of antifungal therapy to IRIS, median days (IQR) Length of time between IPA diagnosis and voriconazole initiation, median days (IQR) Response to treatment (10 days after neutrophil recovery) Complete response Partial response Stable response Progression of disease Response to treatment (1 Month) Complete response
IRIS (n Z 36)
Non-IRIS (n Z 117)
P value
55 (46e64) 24 (67)
52 (42e62) 71 (61)
0.54 0.52
28 (78) 1 (3) 2 (6) 9 (25) 8 (22) 1 (3) 6 (17) 7 (19) 10 (29) 26 (72)
88 (75) 12 (10) 8 (7) 42 (36) 40 (34) 2 (2) 13 (11) 40 (34) 44 (38) 64 (55)
0.75 0.30 >0.99 0.23 0.18 0.56 0.39 0.09 0.28 0.06
11 (31) 25 (69) 7 (19) 29 (19e49) 21 (14e40)
20 97 30 40 23
0.08
3 (8) 33 (92) 5 (14) 4 (11) 1 (3) 0 (0) 1.48 (0.73e3.59)
7 (6) 110 (94) 21 (18) 20 (17) 1 (1) 7 (6) 1.53 (0.65e3.54)
0.57 0.39 0.42 0.20 0.66
18 (50) 15 (42) 1 (3) 1 (3) 2 (6) 10 (28) 4 (11) 3 (8) 3 (8) 0 (0) 0 (0) 14 (39) 8 5 1 1 (3) 16 (6e18)
69 (59) 29 (25) 2 (2) 3 (3) 12 (10) 29 (25) 15 (13) 6 (5) 5 (4) 2 (2) 1 (1) NA NA NA NA NA NA
0.34 0.050 0.56 >0.99 0.52 0.72 0.72 0.67 0.40 >0.99 >0.99 NA NA NA NA NA NA
2 (1e6)
2 (1e8)
0.53
0 0 0 36 (100)
5 (4) 28 (24) 27 (23) 57 (49)
0.60 0.001 0.001 0.99 0.01
4 (11) 12 (33)
38 (33) 68 (58)
0.01 0.01
Note. Data are no. (%) of patients, unless otherwise indicated. IRIS: Immune reconstitution inflammatory syndrome, IPA: Invasive pulmonary aspergillosis, IQR: Interquartile range, HCT: Hematopoietic stem cell transplantation. a Corticosteroid users are defined as use of corticosteroids at a mean minimum dose of 0.3 mg/kg/day of prednisolone equivalent for > 3 weeks. b Treatment with other recognized T cell immunosuppressants, such as cyclosporine, TNF-a blockers, specific monoclonal antibodies (such as alemtuzumab), or purine analogs during the past 90 days.
associated with voriconazole use and with a good prognosis. In addition, we showed that the prognosis in patients with IRIS appear to be as good as that in patients with neutrophil recovery and favorable radiologic response (Fig. 2). To our knowledge, this is the largest study to systemically evaluate fungal IRIS in neutropenic patients. In a previous study it was reported that a majority of neutropenic patients (73% [19/26]) with IPA were improved
Figure 1
or cured by antifungal treatment combined with surgery despite an increase in the extent of aspergillosis on CT in the first week.3 Although that study showed in directly that radiologically progressive disease was not associated with poor outcomes, it could not be determined from this study that the apparent failure of medical treatment followed by surgery affected the outcome in these patients. Furthermore, the recent advances in antifungal agents have
Proportions of IRIS and non-IRIS patients with different responses (n Z 153).
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
6
J. Jung et al. Table 2 Clinical characteristics and outcomes of neutropenic patients with and without immune reconstitution inflammatory syndrome (IRIS) who experienced neutrophil recovery and progressive disease.
Age, median years (IQR) Male gender Underlying disease and/or conditions Acute leukemia Aplastic anemia Lymphoma Receipt of HCT Allogeneic HCT Autologous HCT Diabetes mellitus Prior corticosteroid use Prior immune suppressant use Colony stimulating factor use after diagnosis of IPA State of underlying disease Controlled Relapsed or refractory Mechanical ventilation Length of neutropenic period, median days (IQR) Length of time between neutropenia and diagnosis of IPA, median days (IQR) IPA diagnosis category Proven Probable Positive Aspergillus culture Sputum Bronchioalveolar lavage Positive Aspergillus culture after neutrophil recovery Peak galactomannan level, median value (IQR) Antifungal therapy on diagnosis day of IPA Amphotericin B Voriconazole Itraconazole Liposomal amphotericin B Caspofungin Received prophylactic antifungal agents Itraconazole Amphotericin B Micafungin Voriconazole Posaconazole Antifungal therapy during IRIS (Liposomal) Amphotericin-B to voriconazole Voriconazole to echinocandin or combination of echinocandin Voriconazole to amphotericin-B Corticosteroid use during IRIS Length of time from initiation of antifungal therapy to IRIS, median days (IQR) Length of time between IPA diagnosis and voriconazole initiation, median days (IQR) Response to treatment (3 Months) Complete response Partial response Stable response Progression of disease Outcome after diagnosis of IPA
IRIS (n Z 36)
Non- IRIS (n Z 18)
P value
55 (44e64) 24 (67)
58 (54e71) 11 (61)
0.11 0.69
28 (78) 1 (3) 2 (6) 9 (25) 8 (22) 1 (3) 6 (17) 7 (19) 10 (29) 26 (72)
15 (83) 1 (6) 1 (6) 7 (39) 7 (39) 0 (0) 5 (28) 6 (33) 9 (50) 9 (50)
0.73 >0.99 >0.99 0.29 0.20 >0.99 0.48 0.32 0.10 0.11
11 (31) 25 (69) 7 (19) 29 (19e49) 21 (14e40)
3 (17) 15 (83) 7 (39) 31 (16e69) 20 (11e50)
0.34
3 (8) 33 (92) 5 (14) 4 (11) 1 (3) 0 (0) 1.48 (0.73e3.59)
0 (0) 18 (100) 6 (33) 6 (33) 0 (0) 3 (17) 1.17 (0.59e6.76)
0.54 0.15 0.07 >0.99 0.03 0.59
18 (50) 15 (42) 1 (3) 1 (3) 2 (6) 10 (28) 4 (11) 3 (8) 3 (8) 0 (0) 0 (0) 14 (39) 8 5 1 1 (3) 16 (6e18)
12 (67) 5 (28) 0 (0) 0 (0) 1 (6) 7 (39) 3 (17) 3 (17) 1 (6) 0 (0) 0 (0) NA NA NA NA NA NA
0.25 0.32 >0.99 >0.99 >0.99 0.41 >0.99 0.64 0.60 NA NA NA NA NA NA NA NA
2 (1e6)
1 (1e6)
0.91
7 (19) 18 (50) 1 (3) 10 (28)
0 (0) 6 (33) 0 (0) 12 (67)
0.08 0.25 >0.99 0.006
0.19 0.98 0.88
(continued on next page)
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
Immune reconstitution syndrome in IPA
7
Table 2 (continued )
30 day mortality 90 day mortality
IRIS (n Z 36)
Non- IRIS (n Z 18)
P value
4 (11) 12 (33)
7 (39) 13 (72)
0.03 0.007
Note. Data are no. (%) of patients, unless otherwise indicated. IRIS: Immune reconstitution inflammatory syndrome, IPA: Invasive pulmonary aspergillosis, IQR: Interquartile range, HCT: Hematopoietic stem cell transplantation.
Table 3
Univariate and multivariate analyses for 30 day mortality in 153 neutropenic patients with invasive aspergillosis. Univariate analysis OR (95% CI)
Old age (‡ 60 years) Uncontrolled underlying disease Host status with fungal burden and radiologic response Radiologic favorable response with neutrophil recovery IRIS with neutrophil recovery Radiologic progression with neutrophil recovery No neutrophil recovery Renal replacement therapy Voriconazole use on diagnosis day of IPA Amphotericin use on diagnosis day of IPA
1.66 (0.78e3.52) 0.26 (1.74-0.66-4.61) 0.07 0.26 1.8 9.5 3.5 0.84 1.33
increased the tendency for physicians to change or combine antifungal agents in patients with deteriorating radiologic findings of IPA during neutrophil recovery. We found that about two-thirds of such patients had IRIS and this was associated with decreased mortality and the prognosis was as good as patients with favorable radiologic response. We therefore believe that close monitoring without modification of antifungal treatment is warranted in these patients. The definition of fungal IRIS in non-HIV patients is not established. So, it is to be settled that clinical symptoms or findings should be incorporated into clinical criteria of the
Figure 2 The survival curves stratified according to host status with fungal burden and radiologic progression among the 153 neutropenic cancer patients with IPA.
(0.02e0.30) (0.09e0.79) (0.65e5.06) (4.26e21.46) (1.10e11.12) (0.38e1.87) (0.64e2.75)
Multivariate analysis P value
Adjusted OR (95% CI)
P value
0.04 (0.01e0.17) 0.11 (0.04e0.35)
www.elsevierhealth.com/journals/jinf
Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis Jiwon Jung a,c, Hyo-Lim Hong b,c, Sang-Oh Lee a, Sang-Ho Choi a, Yang Soo Kim a, Jun Hee Woo a, Sung-Han Kim a,* a
Department of Infectious Diseases, Asan Medical Center, University of Ulsan College of Medicine, University of Ulsan, Seoul, Republic of Korea b Division of Infectious Diseases, Department of Internal Medicine, Daegu Catholic University Medical Center, Daegu, Republic of Korea Accepted 18 December 2014 Available online - - -
KEYWORDS Invasive pulmonary aspergillosis; Neutropenia; Immune reconstitution inflammatory syndrome
Summary Objectives: Clinical and radiologic deterioration is sometimes observed during neutrophil recovery in patients with invasive pulmonary aspergillosis (IPA). This deterioration can be caused by immune reconstitution inflammatory syndrome (IRIS) as well as by progression of the IPA. However, there is limited data on IRIS in neutropenic patients. Methods: Over a 6-year period, adult patients with neutropenia who met the criteria for probable or proven IPA by the revised EORTC/MSG definition were retrospectively enrolled. IRIS was defined as de novo appearance or worsening of radiologic pulmonary findings temporally related to neutrophil recovery, with evidence of a decrease of 50% in serum galactomannan level. Results: Of 153 patients, 36 (24%, 95% CI 18%e31%) developed IRIS during neutrophil recovery. More of these patients received voriconazole than did those with non-IRIS (42% vs. 25%, P Z 0.05). Thirty- and ninety-day mortalities were lower in the patients with IRIS than in those with non-IRIS (11% vs. 33%, P Z 0.01, and 33% vs. 58%, P Z 0.01, respectively). Conclusion: IRIS is relatively common among neutropenic patients with IPA, occurring in about one quarter of such patients. It is associated with voriconazole use and has a good prognosis. ª 2015 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Department of Infectious Diseases, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnapdong, Songpa-gu, Seoul, 138-736, Republic of Korea. Tel.: þ82 2 3010 3305; fax: þ82 2 3010 6970. E-mail address: [email protected] (S.-H. Kim). c These authors contributed equally to this manuscript. http://dx.doi.org/10.1016/j.jinf.2014.12.020 0163-4453/ª 2015 The British Infection Association. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
2
Introduction Invasive pulmonary aspergillosis (IPA) is associated with high morbidity and mortality in patients with hematologic malignancies who are undergoing myelosuppressive chemotherapy or receiving hematopoietic stem cell transplants.1 Since alveolar macrophages and neutrophils play a dominant role in host defenses against aspergillosis,2 neutropenia from chemotherapy or underlying hematologic disease is an important predisposing host factor for IPA. Sometimes clinical and radiologic deterioration is observed during neutrophil recovery in patients with IPA, and this deterioration can be associated with immune reconstitution inflammatory syndrome (IRIS) or with the progression of the IPA. Caillot et al. showed that the median volume of lesions in neutropenic patients with IPA, as measured serially by CT, increased four-fold from day 0 to day 7 despite antifungal therapy.3 However, many physicians consider that worsening of the radiologic findings of IPA during neutrophil recovery is associated with progression of the infection and they modify the antifungal therapy with or without surgery.4 However, if there is no microbiologic documentation, it is unclear whether this radiologic deterioration is associated with IPA progression or with IRIS. Recently, the GM assay has been proposed as a microbiologic marker of response.5,6 If so, it might be possible to decide whether the radiologic worsening of IPA in neutropenic patients is due to IPA progression or immune reconstitution inflammatory syndrome (IRIS) by assessing clinical and microbiologic parameter as well as immunologic parameters such as neutrophil count. This distinction helps clinicians to guide antifungal therapy in this difficult clinical situation. In addition, this procedure could detect refractory IPA more objectively, and could advance clinical study of salvage therapy in refractory IPA without dilution of the results of studies by the inclusion of patients with IRIS. The concept of fungal IRIS was originally proposed in patients starting antiretroviral therapy for human immunodeficiency virus (HIV) infection and was clearly demonstrated in those with cryptococcal infection, histoplasmosis, and candidiasis.7 In non-HIV patients, cases of fungal IRIS have been reported in transplant recipients,8e10 neutropenic patients,11,12 and recipients of TNFa receptor inhibitors.13,14 Recently, Singh et al. reported that about 5% of solid organ transplant recipients with cryptococcal infections had IRIS-like illnesses, and these appeared to be associated with a certain immunosuppressive regimen.9 In addition, Legrand, et al. showed that some neutropenic cancer patients with chronic disseminated candidiasis who were unresponsive to antifungal therapy and improved dramatically after corticosteroid therapy appeared to suffer from what they suggested to be fungal IRIS.15 Miceli et al. proposed a similar concept of IRIS in patients with IPA; this was defined as the onset or worsening of clinical and radiological findings together with neutrophil recovery and a 50% decrease in serum GM titers that could not be explained by newly acquired infection, failure of treatment, or side effects of treatment.4 They described 19 neutropenic non-HIV patients conforming to this definition of fungal IRIS.4 However, data are limited on the incidence, clinical characteristics, and
J. Jung et al. outcomes of IRIS associated with IPA in neutropenic cancer patients. We therefore evaluated the incidence, clinical characteristics, and outcomes of IRIS in neutropenic patients with IPA.
Patients and methods Study population and data collection This study was performed at the Asan Medical Center, a 2700-bed tertiary care teaching hospital in Seoul, South Korea. The medical records of adult patients ( 16 years old) with neutropenia who met the criteria for proven or probable IPA from January 2008 to December 2013 were retrospectively reviewed.
Definitions Patients were assigned a proven or probable diagnosis of IPA based on a modification of the consensus definition of EORTC/MSG and our previous studies.1,16e20 Proven IPA was defined by histologic evidence of tissue invasion including septated, acutely branching filamentous fungi and positive culture or positive GM assay. Probable IPA was defined as the presence of host factors, together with one or more clinical indication such as dense, well circumscribed lesions with or without a halo sign, and an air-crescent sign, or cavity on CT; and mycologic evidence of fungal infection (by culture or cytologic analysis of bronchoalveolar lavage fluid for Aspergillus species, or GM assay of serum or bronchoalveolar lavage fluid). Possible IPA was defined as the presence of a host factor and clinical criteria without mycologic evidence for IPA. Cases classified as possible IPA were not included in this analysis. Neutropenia was defined as an absolute neutrophil count 3 weeks.1 Immunosuppressant use was defined as treatment with a T-cell immunosuppressant such as cyclosporin, TNF-a blocker, specific monoclonal antibody (such as alemtuzumab), or purine analogs during the previous 90 days.1 Serum GM antigen levels were measured as described previously (Platelia Aspergillus EIA; BioRad, Redmond, WA).18,19 They were considered positive if the GM index was 0.5 serum GM antigen level. IRIS was defined as new or worsening radiologic pulmonary findings temporally related to neutrophil recovery, with evidence of a 50% decrease in a single serum GM index titer combined without evidence of persistent positive fungal culture and absence of new extrapulmonary lesions of aspergillosis (e.g., new skin lesions) and other problem, such as newly acquired infection, failure of treatment of a known infection, or side effect of medication.4 Clinical response to antifungal therapy was classified as success (complete or partial response) or failure (stable response, progression of disease, or death) as described previously22 and assessed 10 days after neutrophil recovery (if a patient died before neutrophil recovery, failure was assigned to the day of death), 1 month, and 3 months after diagnosis,
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
Immune reconstitution syndrome in IPA respectively. Radiologic criteria used in the classification of clinical response were as follows22; complete response was defined as resolution of radiological lesion(s), partial response as at least a 25% reduction in diameter of the radiological lesion(s), stable response as 0%e25% reduction in the diameter of the lesion(s), and progression of disease as new sites of disease or worsening of preexisting lesions(s). Antifungal prophylaxis was defined as the use of an antifungal agent with anti-Aspergillus activity (e.g. voriconazole, posaconazole, itraconazole, caspofungin, or amphotericin B) for > 5 continuous days without any signs or symptoms of invasive mold infection.23
Statistical analysis All statistical analyses were performed with SPSS version 20.0 (SPSS, Chicago, IL). Categorical variables were compared using the c2or Fisher’s exact test, and continuous variables were compared using the ManneWhitney U test, as appropriate. A P value 0.10 from the equation.
Results Patient characteristics During the study period, a total 153 neutropenic patients with IPA (10 proven and 143 probable) were analyzed. The most frequent underlying hematologic disease was acute myeloid leukemia (76% [116/153]). Of the 153 patients, 51 (33%) received hematopoietic stem cell transplantations (HCT) including 48 allogeneic HCT and 3 autologous HCT. The median neutrophil count of the 153 patients at diagnosis of IPA was 3/mL (interquartile range [IQR] 0e61). Neutrophil recovery occurred in 103 of the 153 patients. The median length of time between neutropenia and the diagnosis of IPA was 23 days (IQR 15e44). The baseline clinical characteristics of these patients are shown in Table 1.
IRIS versus non-IRIS in neutropenic cancer patients with IPA Of the 153 patients with IPA, 36 (24%, 95% CI 18%e31%) developed IRIS during neutrophil recovery (Fig. 1). The proportions of patients who underwent HCT (IRIS 25% vs. nonIRIS 36%; P Z 0.23) were not significantly different in the two groups. Although the effect was not statistically significant, fewer of the patients with IRIS received corticosteroids (IRIS, 19% vs. non-IRIS 34%; P Z 0.09), more of the patients with IRIS received colony stimulating factor after diagnosis of IPA (IRIS 72% vs. non-IRIS 55%; P Z 0.06), and more of the patients with IRIS had the underlying disease
3 controlled within 3 months (31% vs. 17%, P Z 0.08). The median length of neutropenic periods in the patients with IRIS was shorter than in the patients with non-IRIS (29 days [IQR 19-49] vs. 40 [IQR 23-87], P Z 0.03). Among the patients with IRIS, clinical deterioration such as hypoxia, tachycardia and fever was observed in 19 (53%) patients. The median days from initiation of antifungal therapy to IRIS was 16 days (IQR 6e18). Fourteen (39%) patients received changed antifungal therapy upon development of IRIS. Interestingly, patients with IRIS received voriconazole more frequently on the day of diagnosis of IPA than those with non-IRIS (42% vs. 25%, P Z 0.05). The 30- and 90-day mortalities were lower in patients with IRIS than in those with non-IRIS (11% vs. 33%, P Z 0.01, and 33% vs. 58%, P Z 0.01, respectively).
IRIS versus non-IRIS in 54 neutropenic cancer patients with IPA who underwent neutrophil recovery and radiologically progressive disease Of the 153 patients, 50 (33%) who died without neutrophil recovery were excluded and the remaining 103 (67%) who underwent neutrophil recovery were further analyzed. Of these 103 patients, 54 (52%) had radiologically progressive disease during the neutrophil recovery while 49 (48%) had radiologically stable disease or partial/complete responses during the neutrophil recovery (Fig. 1). In a subgroup analysis of the 54 patients who had radiologically progressive disease during neutrophil recovery, 36 (67%, 95% CI 53%e 78%) were assigned to the IRIS group and the remaining 18 to the non-IRIS group (Table 2). The 30- and 90-day mortalities were higher in the patients with non-IRIS who had progressive disease during neutrophil recovery than in those with IRIS (11% vs. 39%, P Z 0.03 and 33% vs. 72%, P Z 0.007, respectively).
Outcomes and prognostic factors according to host status with fungal burden and radiologic response Fig. 1 shows the detailed classification of the 153 patients with IPA according to host status with fungal burden and radiologic progression. The baseline clinical characteristics and outcomes depending on host status with fungal burden and radiologic progression are shown in Supplemental Table 1. We analyzed risk factors for 30-day mortality in the overall patient population (Table 3). Multivariate analysis indicated that radiologic stable or partial/complete response during neutrophil recovery (OR 0.07, P < 0.001) and IRIS during neutrophil recovery (OR 0.26, P < 0.001) were independently associated with 30-day mortality. The survival curves stratified according to host status with fungal burden and radiologic progression are presented in Fig. 2.
Discussion Here we found that IRIS occurred in about one quarter of all neutropenic patients with IPA, and in two-thirds of patients with IPA who underwent radiologically progressive disease during neutrophil recovery. We found that IRIS was
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
4
J. Jung et al. Table 1 Clinical characteristics and outcomes of neutropenic patients with and without immune reconstitution inflammatory syndrome (IRIS).
Age, median years (IQR) Male gender Underlying disease and/or conditions Acute leukemia Aplastic anemia Lymphoma Receipt of HCT Allogeneic HCT Autologous HCT Diabetes mellitus Prior corticosteroid usea Prior immune suppressant useb Colony stimulating factor use after diagnosis of IPA State of underlying disease Controlled Relapsed or refractory Mechanical ventilation Length of neutropenic period, median days (IQR) Length of time between neutropenia and diagnosis of IPA, median days (IQR) IPA diagnosis category Proven Probable Positive Aspergillus culture Sputum Bronchioalveolar lavage Positive Aspergillus culture after neutrophil recovery Peak galactomannan level, median value (IQR) Antifungal therapy on diagnosis day of IPA Amphotericin B Voriconazole Itraconazole Liposomal amphotericin B Caspofungin Received prophylactic antifungal agents Itraconazole Amphotericin B Micafungin Voriconazole Posaconazole Antifungal therapy during IRIS (Liposomal) Amphotericin-B to voriconazole Voriconazole to echinocandin or combination of echinocandin Voriconazole to amphotericin-B Corticosteroid use during IRIS Length of time from initiation of antifungal therapy to IRIS, median days (IQR) Length of time between IPA diagnosis and voriconazole initiation, median days (IQR) Response to treatment (10 days after neutrophil recovery) Complete response Partial response Stable response Progression of disease Response to treatment (1 Month) Complete response
IRIS (n Z 36)
Non-IRIS (n Z 117)
P value
55 (46e64) 24 (67)
52 (42e62) 71 (61)
0.54 0.52
28 (78) 1 (3) 2 (6) 9 (25) 8 (22) 1 (3) 6 (17) 7 (19) 10 (29) 26 (72)
88 (75) 12 (10) 8 (7) 42 (36) 40 (34) 2 (2) 13 (11) 40 (34) 44 (38) 64 (55)
0.75 0.30 >0.99 0.23 0.18 0.56 0.39 0.09 0.28 0.06
11 (31) 25 (69) 7 (19) 29 (19e49) 21 (14e40)
20 97 30 40 23
0.08
3 (8) 33 (92) 5 (14) 4 (11) 1 (3) 0 (0) 1.48 (0.73e3.59)
7 (6) 110 (94) 21 (18) 20 (17) 1 (1) 7 (6) 1.53 (0.65e3.54)
0.57 0.39 0.42 0.20 0.66
18 (50) 15 (42) 1 (3) 1 (3) 2 (6) 10 (28) 4 (11) 3 (8) 3 (8) 0 (0) 0 (0) 14 (39) 8 5 1 1 (3) 16 (6e18)
69 (59) 29 (25) 2 (2) 3 (3) 12 (10) 29 (25) 15 (13) 6 (5) 5 (4) 2 (2) 1 (1) NA NA NA NA NA NA
0.34 0.050 0.56 >0.99 0.52 0.72 0.72 0.67 0.40 >0.99 >0.99 NA NA NA NA NA NA
2 (1e6)
2 (1e8)
0.53
0 0 0 36 (100)
5 (4) 28 (24) 27 (23) 57 (49)
0.60 0.001 0.001 0.99 0.01
4 (11) 12 (33)
38 (33) 68 (58)
0.01 0.01
Note. Data are no. (%) of patients, unless otherwise indicated. IRIS: Immune reconstitution inflammatory syndrome, IPA: Invasive pulmonary aspergillosis, IQR: Interquartile range, HCT: Hematopoietic stem cell transplantation. a Corticosteroid users are defined as use of corticosteroids at a mean minimum dose of 0.3 mg/kg/day of prednisolone equivalent for > 3 weeks. b Treatment with other recognized T cell immunosuppressants, such as cyclosporine, TNF-a blockers, specific monoclonal antibodies (such as alemtuzumab), or purine analogs during the past 90 days.
associated with voriconazole use and with a good prognosis. In addition, we showed that the prognosis in patients with IRIS appear to be as good as that in patients with neutrophil recovery and favorable radiologic response (Fig. 2). To our knowledge, this is the largest study to systemically evaluate fungal IRIS in neutropenic patients. In a previous study it was reported that a majority of neutropenic patients (73% [19/26]) with IPA were improved
Figure 1
or cured by antifungal treatment combined with surgery despite an increase in the extent of aspergillosis on CT in the first week.3 Although that study showed in directly that radiologically progressive disease was not associated with poor outcomes, it could not be determined from this study that the apparent failure of medical treatment followed by surgery affected the outcome in these patients. Furthermore, the recent advances in antifungal agents have
Proportions of IRIS and non-IRIS patients with different responses (n Z 153).
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
6
J. Jung et al. Table 2 Clinical characteristics and outcomes of neutropenic patients with and without immune reconstitution inflammatory syndrome (IRIS) who experienced neutrophil recovery and progressive disease.
Age, median years (IQR) Male gender Underlying disease and/or conditions Acute leukemia Aplastic anemia Lymphoma Receipt of HCT Allogeneic HCT Autologous HCT Diabetes mellitus Prior corticosteroid use Prior immune suppressant use Colony stimulating factor use after diagnosis of IPA State of underlying disease Controlled Relapsed or refractory Mechanical ventilation Length of neutropenic period, median days (IQR) Length of time between neutropenia and diagnosis of IPA, median days (IQR) IPA diagnosis category Proven Probable Positive Aspergillus culture Sputum Bronchioalveolar lavage Positive Aspergillus culture after neutrophil recovery Peak galactomannan level, median value (IQR) Antifungal therapy on diagnosis day of IPA Amphotericin B Voriconazole Itraconazole Liposomal amphotericin B Caspofungin Received prophylactic antifungal agents Itraconazole Amphotericin B Micafungin Voriconazole Posaconazole Antifungal therapy during IRIS (Liposomal) Amphotericin-B to voriconazole Voriconazole to echinocandin or combination of echinocandin Voriconazole to amphotericin-B Corticosteroid use during IRIS Length of time from initiation of antifungal therapy to IRIS, median days (IQR) Length of time between IPA diagnosis and voriconazole initiation, median days (IQR) Response to treatment (3 Months) Complete response Partial response Stable response Progression of disease Outcome after diagnosis of IPA
IRIS (n Z 36)
Non- IRIS (n Z 18)
P value
55 (44e64) 24 (67)
58 (54e71) 11 (61)
0.11 0.69
28 (78) 1 (3) 2 (6) 9 (25) 8 (22) 1 (3) 6 (17) 7 (19) 10 (29) 26 (72)
15 (83) 1 (6) 1 (6) 7 (39) 7 (39) 0 (0) 5 (28) 6 (33) 9 (50) 9 (50)
0.73 >0.99 >0.99 0.29 0.20 >0.99 0.48 0.32 0.10 0.11
11 (31) 25 (69) 7 (19) 29 (19e49) 21 (14e40)
3 (17) 15 (83) 7 (39) 31 (16e69) 20 (11e50)
0.34
3 (8) 33 (92) 5 (14) 4 (11) 1 (3) 0 (0) 1.48 (0.73e3.59)
0 (0) 18 (100) 6 (33) 6 (33) 0 (0) 3 (17) 1.17 (0.59e6.76)
0.54 0.15 0.07 >0.99 0.03 0.59
18 (50) 15 (42) 1 (3) 1 (3) 2 (6) 10 (28) 4 (11) 3 (8) 3 (8) 0 (0) 0 (0) 14 (39) 8 5 1 1 (3) 16 (6e18)
12 (67) 5 (28) 0 (0) 0 (0) 1 (6) 7 (39) 3 (17) 3 (17) 1 (6) 0 (0) 0 (0) NA NA NA NA NA NA
0.25 0.32 >0.99 >0.99 >0.99 0.41 >0.99 0.64 0.60 NA NA NA NA NA NA NA NA
2 (1e6)
1 (1e6)
0.91
7 (19) 18 (50) 1 (3) 10 (28)
0 (0) 6 (33) 0 (0) 12 (67)
0.08 0.25 >0.99 0.006
0.19 0.98 0.88
(continued on next page)
Please cite this article in press as: Jung J, et al., Immune reconstitution inflammatory syndrome in neutropenic patients with invasive pulmonary aspergillosis, J Infect (2015), http://dx.doi.org/10.1016/j.jinf.2014.12.020
Immune reconstitution syndrome in IPA
7
Table 2 (continued )
30 day mortality 90 day mortality
IRIS (n Z 36)
Non- IRIS (n Z 18)
P value
4 (11) 12 (33)
7 (39) 13 (72)
0.03 0.007
Note. Data are no. (%) of patients, unless otherwise indicated. IRIS: Immune reconstitution inflammatory syndrome, IPA: Invasive pulmonary aspergillosis, IQR: Interquartile range, HCT: Hematopoietic stem cell transplantation.
Table 3
Univariate and multivariate analyses for 30 day mortality in 153 neutropenic patients with invasive aspergillosis. Univariate analysis OR (95% CI)
Old age (‡ 60 years) Uncontrolled underlying disease Host status with fungal burden and radiologic response Radiologic favorable response with neutrophil recovery IRIS with neutrophil recovery Radiologic progression with neutrophil recovery No neutrophil recovery Renal replacement therapy Voriconazole use on diagnosis day of IPA Amphotericin use on diagnosis day of IPA
1.66 (0.78e3.52) 0.26 (1.74-0.66-4.61) 0.07 0.26 1.8 9.5 3.5 0.84 1.33
increased the tendency for physicians to change or combine antifungal agents in patients with deteriorating radiologic findings of IPA during neutrophil recovery. We found that about two-thirds of such patients had IRIS and this was associated with decreased mortality and the prognosis was as good as patients with favorable radiologic response. We therefore believe that close monitoring without modification of antifungal treatment is warranted in these patients. The definition of fungal IRIS in non-HIV patients is not established. So, it is to be settled that clinical symptoms or findings should be incorporated into clinical criteria of the
Figure 2 The survival curves stratified according to host status with fungal burden and radiologic progression among the 153 neutropenic cancer patients with IPA.
(0.02e0.30) (0.09e0.79) (0.65e5.06) (4.26e21.46) (1.10e11.12) (0.38e1.87) (0.64e2.75)
Multivariate analysis P value
Adjusted OR (95% CI)
P value
0.04 (0.01e0.17) 0.11 (0.04e0.35)
