- Email: [email protected]
Clinical utility and prognostic value of galactomannan in neutropenic patients with invasive aspergillosis
Pathologie Biologie 60 (2012) 357–361
Available online at
www.sciencedirect.com
Original article
Clinical utility and prognostic value of galactomannan in neutropenic patients with invasive aspergillosis Inte´reˆt clinique et valeur pronostique du galactomannane chez les patients neutrope´niques atteints d’aspergillose invasive I. Hadrich a,*, F. Makni a, F. Cheikhrouhou a, S. Neji a, I. Amouri a, H. Sellami a, H. Trabelsi a, H. Bellaaj b, M. Elloumi b, A. Ayadi a a b
Fungal and Parasitic Molecular Biology Laboratory, Sfax School of Medicine, Magida Boulila street, 3029 Sfax, Tunisia Hematology wards, He´di Chaker Hospital, 3029 Sfax, Tunisia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 28 May 2011 Accepted 27 October 2011
Invasive aspergillosis (IA) is a major cause of morbidity and mortality in profoundly neutropenic patients. Delayed diagnosis and therapy may lead to poor outcomes. Aims. – The objective of this study was to assess the performance characteristics of the galactomannan (GM) assay in serum and bronchoalveolar lavage specimens for the diagnosis of IA in neutropenic patients with hematological malignancies. We also evaluated the prognostic outcome. Patients and methods. – A total of 1198 serum samples and 42 BAL from 235 neutropenic patients were tested with a GM elisa platelia test. We used Cox modeling of time to 6- and 12-week mortality for GM level at the time of diagnosis (GM0) and GM decay in the week following diagnosis in proven and probable IA patients with more than two GM values. Results. – There were three proven, 55 probable, and four possible cases of IA. The sensitivity and specificity of the GM test were 96.8% and 82.4% respectively. In BAL samples, sensitivity was 86% and the specificity 93%. BAL GM was more sensitive than microscopy (22.2%) and BAL culture (38.9%). Among patients with proven/probable IA, serum and BAL GM were in agreement for 92.8% of paired samples. The hazard ratio (HR) of GM0 and 1-week GM decay per unit increase in Aspergillus enzyme immunoassay (EIA) was 1.044 (95% CI, 0.738 to 1.476) and 0.709 (95% CI, 0.236 to 2.130) respectively. Conclusion. – We found good correlation between the GM0 and GM decay combination and outcome of IA patients. The GM is a useful tool for diagnosis and monitoring of IA. ß 2011 Elsevier Masson SAS. All rights reserved.
Keywords: Invasive aspergillosis Galactomannan Pronostic value
R E´ S U M E´
Mots cle´s: Aspergillose invasive Galactomannan Valeur pronostique
L’aspergillose invasive (AI) est une cause majeure de morbidite´ et mortalite´ chez les patients neutrope´niques. Le diagnostic de l’AI repose sur des faisceaux complexes d’arguments et souvent porte´ trop tardivement. But. – L’objectif de cette e´tude e´tait d’e´valuer les performances diagnostiques du galactomannane (GM) dans le se´rum et les lavages broncho-alve´olaire (LBA) pour le diagnostic de l’AI chez les patients neutrope´niques atteints d’he´mopathies malignes. Nous avons e´galement e´value´ la valeur pronostique du GM. Patients et me´thodes. – Un total de 1198 e´chantillons de se´rums et 42 LBA collecte´s a` partir de 235 patients neutrope´niques ont e´te´ teste´s avec Elisa Platelia GM. Nous avons utilise´ la mode´lisation Cox pour la mortalite´ a` six et 12 semaines pour le GM au moment du diagnostic (GM0) et le GM dans la semaine suivant le diagnostic chez les patients atteints d’AI prouve´es et probables avec les valeurs de GM supe´rieures a` deux. Re´sultats. – Soixante-deux patients ont de´veloppe´ une AI avec trois cas d’AI prouve´es, 55 cas d’AI probables et quatre cas d’AI possibles. La sensibilite´ et la spe´cificite´ du test GM ont e´te´ de 96,8 % et 82,4 %
* Corresponding author. E-mail address: [email protected] (I. Hadrich). 0369-8114/$ – see front matter ß 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.patbio.2011.10.011
358
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
respectivement. Concernant les LBA, la sensibilite´ e´tait de 86 % et la spe´cificite´ e´tait de 93 %. La de´tection du GM dans LBA a e´te´ plus sensible que l’examen microscopique direct (22,2 %) et la culture (38,9 %). Pour les patients atteints d’AI prouve´s ou probables, le GM dans le se´rum et le LBA e´taient en accord dans 92,8 % d’e´chantillons apparie´s. Le risque au hasard de GM0 et de 1-semaine GM par augmentation d’une unite´ a` l’Aspergillus immunoenzymatique essai (AIE) ont e´te´ de 1,044 (IC 95 %, 0,738 a` 1,476) et de 0,709 (IC 95 %, 0,236 a` 2,130) respectivement. Conclusion. – Nous avons conclu une bonne corre´lation entre la combinaison du GM0 et 1-semaine GM et le pronostic des patients atteints d’AI. Le GM est un outil utile pour le diagnostic et le suivi de l’e´volution de l’AI. ß 2011 Elsevier Masson SAS. Tous droits re´serve´s.
1. Introduction The incidence of invasive aspergillosis (IA) has significantly increased in neutropenic patients and in hematopoietic stem cell transplant (HSCT) recipients [1]. This severe opportunistic fungal infection is characterized by high mortality in these at-risk patients [2,3]. The major criteria for IA suspicion is essentially based on a set of arguments such as abnormal chest computed tomography scans, a positive fungal culture, and the detection of circulating galactomannan (GM) [4]. Tissue biopsy provides a means of making the diagnosis, but is invasive and not always feasible, especially among neutropenic patients [5,6]. Much attention has been focused on developing a non-invasive diagnostic test. Methods designed to detect serum GM antigen and 1.3-ß-D-glucan levels are useful for IA diagnosis [7]. Recently, studies proposed a binarized GM outcome as a possible surrogate outcome measure for IA, based on a strong correlation between GM outcomes and poor clinical outcomes in hematologic malignancies [8,9]. Our study evaluated the performance of the GM assay in serum and BAL specimens for IA diagnosis in a cohort of patients with hematological malignancies. We also analyzed important factors predictive of clinical outcome such as baseline GM level at diagnosis and the subsequent rate of GM decay during the first week following diagnosis. 2. Patients and methods 2.1. Study participants 2.1.1. Patients The cohort study included immunocompromised patients at high risk of IA hospitalized in the hematology wards of the Sfax university Hospital, Tunisia, for the treatment of hematological malignancies from January 2002 to December 2009. Patients were considered to be at high risk for IA if they underwent chemotherapy, neutropenic (< 500 neutrophils per mL) for at least 10 days, and presented with persistent fever unresponsive to appropriate broad-spectrum antibacterial treatments more than 96 hours. No patients received piparacillin–tazobactam or any other b-lactam antibiotics. IA was classified as proven, probable or possible using the international definitions of invasive fungal disease (IFD) published by the EORTC/MSG [10]. 2.1.2. Controls Because BAL had been performed in only a subgroup of patients, diagnostic indices on BAL specimens had to be evaluated using a nested case-control design on a subgroup of patients (IA and controls) in whom BAL was done. IA cases were matched with two non-IA controls respecting the age, sex, and follow-up duration. For the 28 controls: 16 had acute lymphocytic leukemia (ALL), 10 had acute myelocytic leukemia (AML) and two medullar aplasia. 2.2. GM assay (GMA) levels Serum and BAL GM levels were measured using the PlateliaTM Aspergillus enzyme immunoassay (EIA) kit (BioRad, France), as described previously by Maertens et al. [11]. The assay was run twice weekly. An index cutoff value of 0.5 was chosen for positivity. The GM level at the time of diagnosis (GM0) was defined as the first GM value of more or equal to 0.5. For patients with two serum GM values, a daily GM decay value was calculated by dividing the difference between GM0 and a GM measurement obtained around 1 week after GM0 by the number of days between the tests. If a day 7 GM was unavailable, the GM value soonest after day 7 was
preferentially used for the calculation, dividing by the actual number of days between tests. Daily GM decay values were multiplied by seven to yield a 1-week GM decay value in EIA units per week. 2.3. Statistical analysis The diagnostic indices of GMA in serum and BAL were calculated. Patients with possible IA were excluded from the diagnostic indices calculation. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR) and diagnostic odds ratio (DOR) were computed with their 95% confidence interval (CI). Receiver-operator characteristics (ROC) analysis was performed for serum and BAL GM results. Cox proportional hazards modeling was used to generate adjusted HRs of time to all-cause mortality at 6 and 12 weeks. Possible predictors of mortality were included in the multivariable Cox regression model if they were associated with this outcome in the univariate analysis or previously identified in the literature as significant risk factors. We assessed the impact of the inclusion of an interaction term between GM0 and 1-week GM decay. All analyses were performed using STATA version 10 (STATA Corporation, College Station, TX).
3. Results 3.1. Patients’ characteristics Two hundred and thirty-five patients participated in this prospective study. Sixty-two patients (26.4%) developed IA; their mean age was 34.6 years (range: 1–69 years), with a 1.42 sex-ratio. Three patients were classified as proven, 55 as probable and four as possible IA. Sixteen had ALL, 42 had AML and four had medullar aplasia (Table 1). A CT scan was performed for 45 patients. The halo sign was found in 62.7% and the crescent air sign in 37.3%. Twenty-eight patients who did not develop IA (two per case) were used as controls for the evaluation of BAL-based diagnosis. The mean age of this group was 37.6 years (range: 1–84 years) with a 2.5 sex-ratio. 3.1.1. Antifungal therapy Forty-three patients (93.2%) received systemic antifungal therapy: amphotericin B (74.5%), voriconazole (6.7%) and amphotericin B followed by voriconazole (8.4%). Thirty-two patients (58.1%) started antifungal therapy during the week following GM0, nine (16.3%) on the day of GM0, six (10.9%) the day before GM0 and four (7.2%) the day following GM0. The three patients who did not receive systemic antifungal therapy died 1, 6, and 18 days after GM0. 3.1.2. GM serum parameters A total of 1198 serum samples were collected from the 235 patients included in the cohort. The ROC curve and table of statistics for cutoff GM index are shown in Fig. 1 and Table 2 respectively. The ROC analysis suggested an optimal cutoff more or equal to 0.758 which yielded 90.3% sensitivity and 91.8% specificity. Using a cutoff value of more or equal to 0.5, the sensitivity and specificity were 96.8% and 82.4% respectively. The specificity was 97.6% using a cutoff of 1.5.
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
359
Table 1 Characteristics of 58 proven/probable IA patients. Patients characteristics
IA diagnosis
Number
Sex, age (years)
Primary disease
Outcome
EORTC 2008 criteria
Direct exam
Fungal culture
GM BAL
GM Blood
CT scan
Antifungal treatment
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
M, 21 F, 47 M, 3 M, 42 M, 24 M, 21 M, 19 M, 17 M, 50 M, 18 M, 20 M, 59 M, 58 F, 25 F, 21 F, 25 F, 42 F, 42 M, 55 M, 54 F, 63 M, 50 M, 43 M, 47 M, 40 F, 56 F, 1 F, 28 F, 42 F, 38 M, 19 F, 34 M, 27 M, 12 M, 32 F, 47 M, 43 M, 10 M, 51 F, 8 F, 38 F, 38 F, 35 F, 32 M, 69 M, 17 M, 5 M, 32 M, 9 F, 36 M, 44 M, 30 F, 65 M, 25 F, 54 M, 21 M, 6 M, 36
ALL AML AML AML AML AML AML AML AML AML AML AML AML AML AML ALL AML ALL AML AML AML AML AML MA MA ALL AML AML AML AML ALL AML ALL AML AML AML AML ALL AML ALL ALL MA AML ALL AML ALL AML AML ALL AML AML AML AML AML AML ALL ALL AML
Survival Survival Survival Death Death Survival Death Death Survival Death Death Death Death Death Death Death Death Survival Death Death Death Death Death Death Survival Death Death Death Death Survival Survival Survival Survival Survival Survival Death Survival Survival Death Survival Death Survival Death Death Survival Survival Survival Survival Death Survival Survival Survival Survival Survival Survival Survival Survival Survival
Provena Proven Proven Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable
Pos Pos Neg ND Neg Neg ND ND Neg Pos Neg Neg ND Pos Neg Neg ND ND ND ND Neg ND ND Neg ND ND ND ND ND ND ND Neg ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND Neg ND ND ND ND Neg ND Neg ND ND ND
A. flavus A. flavus A. flavus ND 0/1 0/1 ND ND A. flavus A. flavus 0/1 A. fumigatus ND A. flavus 0/1 0/1 ND ND ND ND 0/1 ND ND 0/1 ND ND ND ND ND ND ND 0/1 ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND 0/1 ND ND ND ND 0/1 ND 0/1 ND ND ND
1/1 ND ND ND 1/1 1/1 ND ND 1/1 1/1 1/1 1/1 ND 1/1 1/1 1/1 ND ND ND ND 1/1 ND ND 1/1 ND ND ND ND ND ND ND 0/1 ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND
4/4 11/14 2/2 7/17 6/7 2/3 4/5 4/5 7/10 9/14 3/4 4/4 2/5 10/10 4/4 8/10 2/3 5/9 4/5 2/2 1/1 2/3 4/4 11/13 6/6 16/27 2/4 5/9 2/3 5/8 4/4 3/3 6/6 7/8 4/4 2/3 3/7 3/5 2/5 2/3 3/6 3/4 4/7 2/2 2/2 4/4 2/2 3/4 1/1 5/7 1/6 3/4 1/1 3/5 4/13 3/6 1/4 4/9
Pos Pos Pos Pos Pos Pos ND Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos ND ND ND ND ND ND ND ND ND ND ND ND Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos
Ampho/Vorico Vorico Ampho/Vorico Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho No Vorico Ampho Ampho Ampho No Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Vorico No Ampho/Vorico Ampho No Ampho Ampho Ampho Ampho Ampho Ampho/Vorico Ampho Vorico Ampho Ampho Ampho Ampho Ampho Ampho Ampho/Vorico
F: female; M: male; AML: acute myelocytic leukemia; ALL: acute lymphocytic leukemia; MA: medullar aplasia; Neg: negative; ND: not done; GM: galactomannan; BAL: bronchoalveolar lavage; Direct exam Pos: presence of hyphea; CT scan Pos: positive high-resolution computerized tomography results indicate findings consistent with invasive aspergillosis. a Diagnosis of proven IA was made at biopsy; lung culture at biopsy grew A. flavus; Ampho: amphoterecin B; Vorico: voriconazole; No: not treated.
The median GM0 for the 58 proven/probable IA patients was 0.95 (range, 0.508 to 9.9). Median GM decay was 0.0001 EIA units per week (range, 1.19 to 1.07). A negative number indicates an increase in GM in the week following GM0. In patients with IA, serum-based biological diagnosis was predictive of the CT scan results in 72% of the 45 patients who
underwent a CT scan. Elevated GMA was detected on average eight days before the CT scan results. 3.1.3. GM BAL parameters GM detection was performed on 14 BAL samples from proven (1) and probable (13) IA and 28 non-IA patients considered as controls. ROC curve and table of statistics for cutoff GM index are
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
360
Fig. 1. Receiver-operator characteristic curve values for cutoff galactomannan index in blood samples for all patients. AUC: area under ROC curve; IC: confidential interval.
shown in Fig. 2 and Table 2 respectively. The diagnostic indices in BAL are listed in Table 3. Using BAL GM index more or equal to 0.5, the sensitivity and specificity of GM were 85.7% and 92.9% respectively. Microscopy and BAL culture were positive in 22.2% and 38.9% respectively. 85.7% (6/7) of positive BAL culture isolated A. flavus and 14.2% (1/7) isolated A. fumigatus.
Fig. 2. Receiver-operator characteristic curve values for cutoff galactomannan index in BAL samples for all patients. AUC: area under ROC curve; IC: confidential interval.
Table 3 Diagnostic performance in BAL samples for the diagnosis of IA in (n = 42) at-risk patients with hematological malignancies.
Sensitivity [95% CI] Specificity [95% CI] LR+ [95% CI] LR [95% CI] DOR [95% CI]
3.1.4. Correlation between BAL and serum GM Among the patients with proven/probable IA, BAL and serum GM were in agreement for 92.8% (13/14) of paired samples. BAL and serum GM indices were both positive in 12 patients, and both negative in one patient. One patient had negative BAL GM and positive serum GM. 3.1.5. Outcomes For the GM level at the time of diagnosis (GM0), the mortality rates were 51.7% and 56.87% for 6 and 12 weeks respectively. Among all proven and probable IA patients, the crude HR of GM0 for time to mortality was 1.044 (95% CI, 0.738 to 1.476) per unit increase. The HR for 1-week GM decay was 0.709 (95% CI, 0.236 to 2.130) per EIA unit decline over the week following GM0.
4. Discussion Our study analyzed the routine use of serum GM test for monitoring IA in a large population of neutropenic patients with hematological malignancies. We confirmed the high sensitivity (96.8%) and specificity (82.4%) reported in other series [11–13]. We found that GM detection in BAL increased the sensitivity of IA diagnosis from 38.9% (based on conventional mycological cultures) to 85.7%. Indeed, culture and cytology of BAL samples are known to
GMA
Direct examination
Culture
85.7 [67.4–94.3] 92.9 [83.7–97.2] 12 [4.144–33.322] 0.154 [0.058–0.389] 78.00 [10.66–571.86]
22.2 [11.8–22.2] 100 [95.5–100] +1 [2.642–Inf] 0.778 [0.778–0.923] +1 [2.862–inf]
38.9 [264–38.9] 100 [94.6–100] +1 [4.906–Inf] 0.611 [0.611–0.778] +1 [6.304–inf]
LR+: positive likelihood ratio; LR: negative likelihood ratio; DOR: diagnostic odds ratio; CI: confidence interval.
have low sensibility despite extensive fungal infection observed on pathology specimens [14–17]. Our results are in keeping with previous studies showing BAL GM sensitivity for IA diagnosis ranging from 58 to 100% and specificity from 94 to 100% [14,15,18]. We cannot conclusively define an interpretive BAL GM cutoff for diagnosis IA due to the small sample size. Our observations are in keeping with the recommendation that BAL GM more or equal to 1.0 is optimal in patients with hematological disorders [18]. Long MI et al. indicated that raising the cutoff level of positivity from 0.5 to 2 improves the specificity without decreasing sensitivity [18]. For patients with proven or probable IA, we identified one patient with negative BAL GM and positive serum GM indices. On one hand, such discordance might reflect the fact that the airway and vascular compartments express different disease stages. Using in vitro and rabbit models of IA, investigators demonstrated that
Table 2 Diagnostics performance for cutoff galactomannan index in blood and BAL samples in patients with proven and probable invasive aspergillosis as a standard. BAL
Blood GM index
Sensitivity
Specificity
PPV
NPV
Sensitivity
Specificity
LR+
LR
0.50 0.75 1 1.25 1.5
0.968 0.903 0.774 0.710 0.629
0.824 0.918 0.959 0.971 0.976
0.667 0.803 0.889 0.898 0.907
0.986 0.966 0.921 0.902 0.878
0.857 0.786 0.571 0.571 0.357
0.929 0.929 0.929 0.929 0.964
12.000 11.000 8.000 8.000 10.000
0.154 0.231 0.462 0.462 0.667
PPV: positive predictive value; NPV: negative predictive value; LR+: positive likelihood ratio; LR: negative likelihood ratio.
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
the appearance of GM in alveolar fluid correlates with airway cellular invasion of Aspergillus [19]. On the other hand, serum GM correlates with the subsequent penetration of hyphae through the endothelial cell layer [19,20]. We found that both GM0 and 1-week GM decay were predictive of time to all-cause mortality at 6 and 12 weeks. Each EIA unit increase in GM0 increased the hazard of time to all-cause mortality at 6 weeks. Similar results were found by Koo et al. who confirmed that the negative number indicates an increase in GM in the week following GM0. These authors also found that the increase in GM0 increases the hazard of time to all-cause mortality at 6 weeks by 25% [9]. Woods et al. used a kappa correlation coefficient test to assess the correlation between GM values and survival. They indicate that serum GM values correlate strongly with aspergillosis outcome in both neutropenic and non neutropenic adults with hematologic cancer receiving various antineoplastic therapies [8]. In conclusion, the GM assay was useful for monitoring IA in patients with hematological malignancy. BAL and serum GM testing are complementary for early diagnosis of IA. Our findings support a strong correlation between GM0 and GM decay and outcome in IA patients. Persistent GM positivity is associated with death and treatment failure, whereas a successful outcome is associated with GM normalization. More study is needed to better define the role of this biomarker. Further confirmation of this strong correlation will have significant implications for patient management and clinical trial design.
[5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
[16]
References
[17]
[1] Jantunen E, Ruutu P, Niskanen L, et al. Incidence and risk factors for invasive fungal infections in allogeneic BMT recipients. Bone Marrow Transplant 1997;19:801–8. [2] Maertens J, Verhaegen J, Lagrou K, Van Eldere J, Boogaerts M. Screening for circulating galactomannan as a noninvasive diagnostic tool for invasive aspergillosis in prolonged neutropenic patients and stem cell transplantation recipients: a prospective validation. Blood 2001;97:1604–10. [3] Denning DW, Stevens DA. Antifungal and surgical treatment of invasive aspergillosis: review of 2,121 published cases. Rev Infect Dis 1990;12:1147–201. [4] Caillot D, Couaillier JF, Bernard A, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic
[18]
[19]
[20]
361
computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19:253–9. Vandewoude KH, Blot SI, Depuydt P, et al. Clinical relevance of Aspergillus isolation from respiratory tract samples in critically ill patients. Crit Care 2006;10:R31. Yeo SF, Wong B. Current status of nonculture methods for diagnosis of invasive fungal infections. Clin Microbiol Rev 2002;15:465–84. Persat F, Ranque S, Derouin F, Michel-Nguyen A, Picot S, Sulahian A. Contribution of the (1–>3)-beta-D-glucan assay for diagnosis of invasive fungal infections. J Clin Microbiol 2008;46:1009–13. Woods G, Miceli MH, Grazziutti ML, Zhao W, Barlogie B, Anaissie E. Serum Aspergillus galactomannan antigen values strongly correlate with outcome of invasive aspergillosis: a study of 56 patients with hematologic cancer. Cancer 2007;110:830–4. Koo S, Bryar JM, Baden LR, Marty FM. Prognostic features of galactomannan antigenemia in galactomannan-positive invasive aspergillosis. J Clin Microbiol 2010;48:1255–60. De Pauw B, Walsh TJ, Donnelly JP, et al. Revised definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis 2008;46:1813–21. Maertens J, Theunissen K, Verbeken E, et al. Prospective clinical evaluation of lower cut-offs for galactomannan detection in adult neutropenic cancer patients and haematological stem cell transplant recipients. Br J Haematol 2004;126:852–60. Yao JF, Su D, Huang Y, et al. Circulating galactomannan screening for early diagnosis and treatment of invasive aspergillosis. Zhonghua Xue Ye Xue Za Zhi 2009;30:592–5. Sarrafzadeh SA, Hoseinpoor Rafati A, Ardalan M, Mansouri D, Tabarsi P, Pourpak Z. The accuracy of serum galactomannan assay in diagnosing invasive pulmonary aspergillosis. Iran J Allergy Asthma Immunol 2010;9: 149–55. Maertens J, Maertens V, Theunissen K, et al. Bronchoalveolar lavage fluid galactomannan for the diagnosis of invasive pulmonary aspergillosis in patients with hematologic diseases. Clin Infect Dis 2009;49:1688–93. Nguyen MH, Leather H, Clancy CJ, et al. Galactomannan testing in bronchoalveolar lavage fluid facilitates the diagnosis of invasive pulmonary aspergillosis in patients with hematologic malignancies and stem cell transplant recipients. Biol Blood Marrow Transplant 2011;17:1042–50. Park SY, Lee SO, Choi SH, et al. Aspergillus galactomannan antigen assay in bronchoalveolar lavage fluid for diagnosis of invasive pulmonary aspergillosis. J Infect 2010;61:492–8. Reichenberger F, Habicht J, Matt P, et al. Diagnostic yield of bronchoscopy in histologically proven invasive pulmonary aspergillosis. Bone Marrow Transplant 1999;24:1195–9. Luong ML, Filion C, Labbe AC, et al. Clinical utility and prognostic value of bronchoalveolar lavage galactomannan in patients with hematologic malignancies. Diagn Microbiol Infect Dis 2010;68:132–9. Hope WW, Kruhlak MJ, Lyman CA, et al. Pathogenesis of Aspergillus fumigatus and the kinetics of galactomannan in an in vitro model of early invasive pulmonary aspergillosis: implications for antifungal therapy. J Infect Dis 2007;195:455–66. Mennink-Kersten MA, Donnelly JP, Verweij PE. Detection of circulating galactomannan for the diagnosis and management of invasive aspergillosis. Lancet Infect Dis 2004;4:349–57.
Available online at
www.sciencedirect.com
Original article
Clinical utility and prognostic value of galactomannan in neutropenic patients with invasive aspergillosis Inte´reˆt clinique et valeur pronostique du galactomannane chez les patients neutrope´niques atteints d’aspergillose invasive I. Hadrich a,*, F. Makni a, F. Cheikhrouhou a, S. Neji a, I. Amouri a, H. Sellami a, H. Trabelsi a, H. Bellaaj b, M. Elloumi b, A. Ayadi a a b
Fungal and Parasitic Molecular Biology Laboratory, Sfax School of Medicine, Magida Boulila street, 3029 Sfax, Tunisia Hematology wards, He´di Chaker Hospital, 3029 Sfax, Tunisia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 28 May 2011 Accepted 27 October 2011
Invasive aspergillosis (IA) is a major cause of morbidity and mortality in profoundly neutropenic patients. Delayed diagnosis and therapy may lead to poor outcomes. Aims. – The objective of this study was to assess the performance characteristics of the galactomannan (GM) assay in serum and bronchoalveolar lavage specimens for the diagnosis of IA in neutropenic patients with hematological malignancies. We also evaluated the prognostic outcome. Patients and methods. – A total of 1198 serum samples and 42 BAL from 235 neutropenic patients were tested with a GM elisa platelia test. We used Cox modeling of time to 6- and 12-week mortality for GM level at the time of diagnosis (GM0) and GM decay in the week following diagnosis in proven and probable IA patients with more than two GM values. Results. – There were three proven, 55 probable, and four possible cases of IA. The sensitivity and specificity of the GM test were 96.8% and 82.4% respectively. In BAL samples, sensitivity was 86% and the specificity 93%. BAL GM was more sensitive than microscopy (22.2%) and BAL culture (38.9%). Among patients with proven/probable IA, serum and BAL GM were in agreement for 92.8% of paired samples. The hazard ratio (HR) of GM0 and 1-week GM decay per unit increase in Aspergillus enzyme immunoassay (EIA) was 1.044 (95% CI, 0.738 to 1.476) and 0.709 (95% CI, 0.236 to 2.130) respectively. Conclusion. – We found good correlation between the GM0 and GM decay combination and outcome of IA patients. The GM is a useful tool for diagnosis and monitoring of IA. ß 2011 Elsevier Masson SAS. All rights reserved.
Keywords: Invasive aspergillosis Galactomannan Pronostic value
R E´ S U M E´
Mots cle´s: Aspergillose invasive Galactomannan Valeur pronostique
L’aspergillose invasive (AI) est une cause majeure de morbidite´ et mortalite´ chez les patients neutrope´niques. Le diagnostic de l’AI repose sur des faisceaux complexes d’arguments et souvent porte´ trop tardivement. But. – L’objectif de cette e´tude e´tait d’e´valuer les performances diagnostiques du galactomannane (GM) dans le se´rum et les lavages broncho-alve´olaire (LBA) pour le diagnostic de l’AI chez les patients neutrope´niques atteints d’he´mopathies malignes. Nous avons e´galement e´value´ la valeur pronostique du GM. Patients et me´thodes. – Un total de 1198 e´chantillons de se´rums et 42 LBA collecte´s a` partir de 235 patients neutrope´niques ont e´te´ teste´s avec Elisa Platelia GM. Nous avons utilise´ la mode´lisation Cox pour la mortalite´ a` six et 12 semaines pour le GM au moment du diagnostic (GM0) et le GM dans la semaine suivant le diagnostic chez les patients atteints d’AI prouve´es et probables avec les valeurs de GM supe´rieures a` deux. Re´sultats. – Soixante-deux patients ont de´veloppe´ une AI avec trois cas d’AI prouve´es, 55 cas d’AI probables et quatre cas d’AI possibles. La sensibilite´ et la spe´cificite´ du test GM ont e´te´ de 96,8 % et 82,4 %
* Corresponding author. E-mail address: [email protected] (I. Hadrich). 0369-8114/$ – see front matter ß 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.patbio.2011.10.011
358
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
respectivement. Concernant les LBA, la sensibilite´ e´tait de 86 % et la spe´cificite´ e´tait de 93 %. La de´tection du GM dans LBA a e´te´ plus sensible que l’examen microscopique direct (22,2 %) et la culture (38,9 %). Pour les patients atteints d’AI prouve´s ou probables, le GM dans le se´rum et le LBA e´taient en accord dans 92,8 % d’e´chantillons apparie´s. Le risque au hasard de GM0 et de 1-semaine GM par augmentation d’une unite´ a` l’Aspergillus immunoenzymatique essai (AIE) ont e´te´ de 1,044 (IC 95 %, 0,738 a` 1,476) et de 0,709 (IC 95 %, 0,236 a` 2,130) respectivement. Conclusion. – Nous avons conclu une bonne corre´lation entre la combinaison du GM0 et 1-semaine GM et le pronostic des patients atteints d’AI. Le GM est un outil utile pour le diagnostic et le suivi de l’e´volution de l’AI. ß 2011 Elsevier Masson SAS. Tous droits re´serve´s.
1. Introduction The incidence of invasive aspergillosis (IA) has significantly increased in neutropenic patients and in hematopoietic stem cell transplant (HSCT) recipients [1]. This severe opportunistic fungal infection is characterized by high mortality in these at-risk patients [2,3]. The major criteria for IA suspicion is essentially based on a set of arguments such as abnormal chest computed tomography scans, a positive fungal culture, and the detection of circulating galactomannan (GM) [4]. Tissue biopsy provides a means of making the diagnosis, but is invasive and not always feasible, especially among neutropenic patients [5,6]. Much attention has been focused on developing a non-invasive diagnostic test. Methods designed to detect serum GM antigen and 1.3-ß-D-glucan levels are useful for IA diagnosis [7]. Recently, studies proposed a binarized GM outcome as a possible surrogate outcome measure for IA, based on a strong correlation between GM outcomes and poor clinical outcomes in hematologic malignancies [8,9]. Our study evaluated the performance of the GM assay in serum and BAL specimens for IA diagnosis in a cohort of patients with hematological malignancies. We also analyzed important factors predictive of clinical outcome such as baseline GM level at diagnosis and the subsequent rate of GM decay during the first week following diagnosis. 2. Patients and methods 2.1. Study participants 2.1.1. Patients The cohort study included immunocompromised patients at high risk of IA hospitalized in the hematology wards of the Sfax university Hospital, Tunisia, for the treatment of hematological malignancies from January 2002 to December 2009. Patients were considered to be at high risk for IA if they underwent chemotherapy, neutropenic (< 500 neutrophils per mL) for at least 10 days, and presented with persistent fever unresponsive to appropriate broad-spectrum antibacterial treatments more than 96 hours. No patients received piparacillin–tazobactam or any other b-lactam antibiotics. IA was classified as proven, probable or possible using the international definitions of invasive fungal disease (IFD) published by the EORTC/MSG [10]. 2.1.2. Controls Because BAL had been performed in only a subgroup of patients, diagnostic indices on BAL specimens had to be evaluated using a nested case-control design on a subgroup of patients (IA and controls) in whom BAL was done. IA cases were matched with two non-IA controls respecting the age, sex, and follow-up duration. For the 28 controls: 16 had acute lymphocytic leukemia (ALL), 10 had acute myelocytic leukemia (AML) and two medullar aplasia. 2.2. GM assay (GMA) levels Serum and BAL GM levels were measured using the PlateliaTM Aspergillus enzyme immunoassay (EIA) kit (BioRad, France), as described previously by Maertens et al. [11]. The assay was run twice weekly. An index cutoff value of 0.5 was chosen for positivity. The GM level at the time of diagnosis (GM0) was defined as the first GM value of more or equal to 0.5. For patients with two serum GM values, a daily GM decay value was calculated by dividing the difference between GM0 and a GM measurement obtained around 1 week after GM0 by the number of days between the tests. If a day 7 GM was unavailable, the GM value soonest after day 7 was
preferentially used for the calculation, dividing by the actual number of days between tests. Daily GM decay values were multiplied by seven to yield a 1-week GM decay value in EIA units per week. 2.3. Statistical analysis The diagnostic indices of GMA in serum and BAL were calculated. Patients with possible IA were excluded from the diagnostic indices calculation. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR) and diagnostic odds ratio (DOR) were computed with their 95% confidence interval (CI). Receiver-operator characteristics (ROC) analysis was performed for serum and BAL GM results. Cox proportional hazards modeling was used to generate adjusted HRs of time to all-cause mortality at 6 and 12 weeks. Possible predictors of mortality were included in the multivariable Cox regression model if they were associated with this outcome in the univariate analysis or previously identified in the literature as significant risk factors. We assessed the impact of the inclusion of an interaction term between GM0 and 1-week GM decay. All analyses were performed using STATA version 10 (STATA Corporation, College Station, TX).
3. Results 3.1. Patients’ characteristics Two hundred and thirty-five patients participated in this prospective study. Sixty-two patients (26.4%) developed IA; their mean age was 34.6 years (range: 1–69 years), with a 1.42 sex-ratio. Three patients were classified as proven, 55 as probable and four as possible IA. Sixteen had ALL, 42 had AML and four had medullar aplasia (Table 1). A CT scan was performed for 45 patients. The halo sign was found in 62.7% and the crescent air sign in 37.3%. Twenty-eight patients who did not develop IA (two per case) were used as controls for the evaluation of BAL-based diagnosis. The mean age of this group was 37.6 years (range: 1–84 years) with a 2.5 sex-ratio. 3.1.1. Antifungal therapy Forty-three patients (93.2%) received systemic antifungal therapy: amphotericin B (74.5%), voriconazole (6.7%) and amphotericin B followed by voriconazole (8.4%). Thirty-two patients (58.1%) started antifungal therapy during the week following GM0, nine (16.3%) on the day of GM0, six (10.9%) the day before GM0 and four (7.2%) the day following GM0. The three patients who did not receive systemic antifungal therapy died 1, 6, and 18 days after GM0. 3.1.2. GM serum parameters A total of 1198 serum samples were collected from the 235 patients included in the cohort. The ROC curve and table of statistics for cutoff GM index are shown in Fig. 1 and Table 2 respectively. The ROC analysis suggested an optimal cutoff more or equal to 0.758 which yielded 90.3% sensitivity and 91.8% specificity. Using a cutoff value of more or equal to 0.5, the sensitivity and specificity were 96.8% and 82.4% respectively. The specificity was 97.6% using a cutoff of 1.5.
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
359
Table 1 Characteristics of 58 proven/probable IA patients. Patients characteristics
IA diagnosis
Number
Sex, age (years)
Primary disease
Outcome
EORTC 2008 criteria
Direct exam
Fungal culture
GM BAL
GM Blood
CT scan
Antifungal treatment
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
M, 21 F, 47 M, 3 M, 42 M, 24 M, 21 M, 19 M, 17 M, 50 M, 18 M, 20 M, 59 M, 58 F, 25 F, 21 F, 25 F, 42 F, 42 M, 55 M, 54 F, 63 M, 50 M, 43 M, 47 M, 40 F, 56 F, 1 F, 28 F, 42 F, 38 M, 19 F, 34 M, 27 M, 12 M, 32 F, 47 M, 43 M, 10 M, 51 F, 8 F, 38 F, 38 F, 35 F, 32 M, 69 M, 17 M, 5 M, 32 M, 9 F, 36 M, 44 M, 30 F, 65 M, 25 F, 54 M, 21 M, 6 M, 36
ALL AML AML AML AML AML AML AML AML AML AML AML AML AML AML ALL AML ALL AML AML AML AML AML MA MA ALL AML AML AML AML ALL AML ALL AML AML AML AML ALL AML ALL ALL MA AML ALL AML ALL AML AML ALL AML AML AML AML AML AML ALL ALL AML
Survival Survival Survival Death Death Survival Death Death Survival Death Death Death Death Death Death Death Death Survival Death Death Death Death Death Death Survival Death Death Death Death Survival Survival Survival Survival Survival Survival Death Survival Survival Death Survival Death Survival Death Death Survival Survival Survival Survival Death Survival Survival Survival Survival Survival Survival Survival Survival Survival
Provena Proven Proven Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable Probable
Pos Pos Neg ND Neg Neg ND ND Neg Pos Neg Neg ND Pos Neg Neg ND ND ND ND Neg ND ND Neg ND ND ND ND ND ND ND Neg ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND Neg ND ND ND ND Neg ND Neg ND ND ND
A. flavus A. flavus A. flavus ND 0/1 0/1 ND ND A. flavus A. flavus 0/1 A. fumigatus ND A. flavus 0/1 0/1 ND ND ND ND 0/1 ND ND 0/1 ND ND ND ND ND ND ND 0/1 ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND 0/1 ND ND ND ND 0/1 ND 0/1 ND ND ND
1/1 ND ND ND 1/1 1/1 ND ND 1/1 1/1 1/1 1/1 ND 1/1 1/1 1/1 ND ND ND ND 1/1 ND ND 1/1 ND ND ND ND ND ND ND 0/1 ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND
4/4 11/14 2/2 7/17 6/7 2/3 4/5 4/5 7/10 9/14 3/4 4/4 2/5 10/10 4/4 8/10 2/3 5/9 4/5 2/2 1/1 2/3 4/4 11/13 6/6 16/27 2/4 5/9 2/3 5/8 4/4 3/3 6/6 7/8 4/4 2/3 3/7 3/5 2/5 2/3 3/6 3/4 4/7 2/2 2/2 4/4 2/2 3/4 1/1 5/7 1/6 3/4 1/1 3/5 4/13 3/6 1/4 4/9
Pos Pos Pos Pos Pos Pos ND Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos ND ND ND ND ND ND ND ND ND ND ND ND Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos
Ampho/Vorico Vorico Ampho/Vorico Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho No Vorico Ampho Ampho Ampho No Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Ampho Vorico No Ampho/Vorico Ampho No Ampho Ampho Ampho Ampho Ampho Ampho/Vorico Ampho Vorico Ampho Ampho Ampho Ampho Ampho Ampho Ampho/Vorico
F: female; M: male; AML: acute myelocytic leukemia; ALL: acute lymphocytic leukemia; MA: medullar aplasia; Neg: negative; ND: not done; GM: galactomannan; BAL: bronchoalveolar lavage; Direct exam Pos: presence of hyphea; CT scan Pos: positive high-resolution computerized tomography results indicate findings consistent with invasive aspergillosis. a Diagnosis of proven IA was made at biopsy; lung culture at biopsy grew A. flavus; Ampho: amphoterecin B; Vorico: voriconazole; No: not treated.
The median GM0 for the 58 proven/probable IA patients was 0.95 (range, 0.508 to 9.9). Median GM decay was 0.0001 EIA units per week (range, 1.19 to 1.07). A negative number indicates an increase in GM in the week following GM0. In patients with IA, serum-based biological diagnosis was predictive of the CT scan results in 72% of the 45 patients who
underwent a CT scan. Elevated GMA was detected on average eight days before the CT scan results. 3.1.3. GM BAL parameters GM detection was performed on 14 BAL samples from proven (1) and probable (13) IA and 28 non-IA patients considered as controls. ROC curve and table of statistics for cutoff GM index are
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
360
Fig. 1. Receiver-operator characteristic curve values for cutoff galactomannan index in blood samples for all patients. AUC: area under ROC curve; IC: confidential interval.
shown in Fig. 2 and Table 2 respectively. The diagnostic indices in BAL are listed in Table 3. Using BAL GM index more or equal to 0.5, the sensitivity and specificity of GM were 85.7% and 92.9% respectively. Microscopy and BAL culture were positive in 22.2% and 38.9% respectively. 85.7% (6/7) of positive BAL culture isolated A. flavus and 14.2% (1/7) isolated A. fumigatus.
Fig. 2. Receiver-operator characteristic curve values for cutoff galactomannan index in BAL samples for all patients. AUC: area under ROC curve; IC: confidential interval.
Table 3 Diagnostic performance in BAL samples for the diagnosis of IA in (n = 42) at-risk patients with hematological malignancies.
Sensitivity [95% CI] Specificity [95% CI] LR+ [95% CI] LR [95% CI] DOR [95% CI]
3.1.4. Correlation between BAL and serum GM Among the patients with proven/probable IA, BAL and serum GM were in agreement for 92.8% (13/14) of paired samples. BAL and serum GM indices were both positive in 12 patients, and both negative in one patient. One patient had negative BAL GM and positive serum GM. 3.1.5. Outcomes For the GM level at the time of diagnosis (GM0), the mortality rates were 51.7% and 56.87% for 6 and 12 weeks respectively. Among all proven and probable IA patients, the crude HR of GM0 for time to mortality was 1.044 (95% CI, 0.738 to 1.476) per unit increase. The HR for 1-week GM decay was 0.709 (95% CI, 0.236 to 2.130) per EIA unit decline over the week following GM0.
4. Discussion Our study analyzed the routine use of serum GM test for monitoring IA in a large population of neutropenic patients with hematological malignancies. We confirmed the high sensitivity (96.8%) and specificity (82.4%) reported in other series [11–13]. We found that GM detection in BAL increased the sensitivity of IA diagnosis from 38.9% (based on conventional mycological cultures) to 85.7%. Indeed, culture and cytology of BAL samples are known to
GMA
Direct examination
Culture
85.7 [67.4–94.3] 92.9 [83.7–97.2] 12 [4.144–33.322] 0.154 [0.058–0.389] 78.00 [10.66–571.86]
22.2 [11.8–22.2] 100 [95.5–100] +1 [2.642–Inf] 0.778 [0.778–0.923] +1 [2.862–inf]
38.9 [264–38.9] 100 [94.6–100] +1 [4.906–Inf] 0.611 [0.611–0.778] +1 [6.304–inf]
LR+: positive likelihood ratio; LR: negative likelihood ratio; DOR: diagnostic odds ratio; CI: confidence interval.
have low sensibility despite extensive fungal infection observed on pathology specimens [14–17]. Our results are in keeping with previous studies showing BAL GM sensitivity for IA diagnosis ranging from 58 to 100% and specificity from 94 to 100% [14,15,18]. We cannot conclusively define an interpretive BAL GM cutoff for diagnosis IA due to the small sample size. Our observations are in keeping with the recommendation that BAL GM more or equal to 1.0 is optimal in patients with hematological disorders [18]. Long MI et al. indicated that raising the cutoff level of positivity from 0.5 to 2 improves the specificity without decreasing sensitivity [18]. For patients with proven or probable IA, we identified one patient with negative BAL GM and positive serum GM indices. On one hand, such discordance might reflect the fact that the airway and vascular compartments express different disease stages. Using in vitro and rabbit models of IA, investigators demonstrated that
Table 2 Diagnostics performance for cutoff galactomannan index in blood and BAL samples in patients with proven and probable invasive aspergillosis as a standard. BAL
Blood GM index
Sensitivity
Specificity
PPV
NPV
Sensitivity
Specificity
LR+
LR
0.50 0.75 1 1.25 1.5
0.968 0.903 0.774 0.710 0.629
0.824 0.918 0.959 0.971 0.976
0.667 0.803 0.889 0.898 0.907
0.986 0.966 0.921 0.902 0.878
0.857 0.786 0.571 0.571 0.357
0.929 0.929 0.929 0.929 0.964
12.000 11.000 8.000 8.000 10.000
0.154 0.231 0.462 0.462 0.667
PPV: positive predictive value; NPV: negative predictive value; LR+: positive likelihood ratio; LR: negative likelihood ratio.
I. Hadrich et al. / Pathologie Biologie 60 (2012) 357–361
the appearance of GM in alveolar fluid correlates with airway cellular invasion of Aspergillus [19]. On the other hand, serum GM correlates with the subsequent penetration of hyphae through the endothelial cell layer [19,20]. We found that both GM0 and 1-week GM decay were predictive of time to all-cause mortality at 6 and 12 weeks. Each EIA unit increase in GM0 increased the hazard of time to all-cause mortality at 6 weeks. Similar results were found by Koo et al. who confirmed that the negative number indicates an increase in GM in the week following GM0. These authors also found that the increase in GM0 increases the hazard of time to all-cause mortality at 6 weeks by 25% [9]. Woods et al. used a kappa correlation coefficient test to assess the correlation between GM values and survival. They indicate that serum GM values correlate strongly with aspergillosis outcome in both neutropenic and non neutropenic adults with hematologic cancer receiving various antineoplastic therapies [8]. In conclusion, the GM assay was useful for monitoring IA in patients with hematological malignancy. BAL and serum GM testing are complementary for early diagnosis of IA. Our findings support a strong correlation between GM0 and GM decay and outcome in IA patients. Persistent GM positivity is associated with death and treatment failure, whereas a successful outcome is associated with GM normalization. More study is needed to better define the role of this biomarker. Further confirmation of this strong correlation will have significant implications for patient management and clinical trial design.
[5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
[16]
References
[17]
[1] Jantunen E, Ruutu P, Niskanen L, et al. Incidence and risk factors for invasive fungal infections in allogeneic BMT recipients. Bone Marrow Transplant 1997;19:801–8. [2] Maertens J, Verhaegen J, Lagrou K, Van Eldere J, Boogaerts M. Screening for circulating galactomannan as a noninvasive diagnostic tool for invasive aspergillosis in prolonged neutropenic patients and stem cell transplantation recipients: a prospective validation. Blood 2001;97:1604–10. [3] Denning DW, Stevens DA. Antifungal and surgical treatment of invasive aspergillosis: review of 2,121 published cases. Rev Infect Dis 1990;12:1147–201. [4] Caillot D, Couaillier JF, Bernard A, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic
[18]
[19]
[20]
361
computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19:253–9. Vandewoude KH, Blot SI, Depuydt P, et al. Clinical relevance of Aspergillus isolation from respiratory tract samples in critically ill patients. Crit Care 2006;10:R31. Yeo SF, Wong B. Current status of nonculture methods for diagnosis of invasive fungal infections. Clin Microbiol Rev 2002;15:465–84. Persat F, Ranque S, Derouin F, Michel-Nguyen A, Picot S, Sulahian A. Contribution of the (1–>3)-beta-D-glucan assay for diagnosis of invasive fungal infections. J Clin Microbiol 2008;46:1009–13. Woods G, Miceli MH, Grazziutti ML, Zhao W, Barlogie B, Anaissie E. Serum Aspergillus galactomannan antigen values strongly correlate with outcome of invasive aspergillosis: a study of 56 patients with hematologic cancer. Cancer 2007;110:830–4. Koo S, Bryar JM, Baden LR, Marty FM. Prognostic features of galactomannan antigenemia in galactomannan-positive invasive aspergillosis. J Clin Microbiol 2010;48:1255–60. De Pauw B, Walsh TJ, Donnelly JP, et al. Revised definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis 2008;46:1813–21. Maertens J, Theunissen K, Verbeken E, et al. Prospective clinical evaluation of lower cut-offs for galactomannan detection in adult neutropenic cancer patients and haematological stem cell transplant recipients. Br J Haematol 2004;126:852–60. Yao JF, Su D, Huang Y, et al. Circulating galactomannan screening for early diagnosis and treatment of invasive aspergillosis. Zhonghua Xue Ye Xue Za Zhi 2009;30:592–5. Sarrafzadeh SA, Hoseinpoor Rafati A, Ardalan M, Mansouri D, Tabarsi P, Pourpak Z. The accuracy of serum galactomannan assay in diagnosing invasive pulmonary aspergillosis. Iran J Allergy Asthma Immunol 2010;9: 149–55. Maertens J, Maertens V, Theunissen K, et al. Bronchoalveolar lavage fluid galactomannan for the diagnosis of invasive pulmonary aspergillosis in patients with hematologic diseases. Clin Infect Dis 2009;49:1688–93. Nguyen MH, Leather H, Clancy CJ, et al. Galactomannan testing in bronchoalveolar lavage fluid facilitates the diagnosis of invasive pulmonary aspergillosis in patients with hematologic malignancies and stem cell transplant recipients. Biol Blood Marrow Transplant 2011;17:1042–50. Park SY, Lee SO, Choi SH, et al. Aspergillus galactomannan antigen assay in bronchoalveolar lavage fluid for diagnosis of invasive pulmonary aspergillosis. J Infect 2010;61:492–8. Reichenberger F, Habicht J, Matt P, et al. Diagnostic yield of bronchoscopy in histologically proven invasive pulmonary aspergillosis. Bone Marrow Transplant 1999;24:1195–9. Luong ML, Filion C, Labbe AC, et al. Clinical utility and prognostic value of bronchoalveolar lavage galactomannan in patients with hematologic malignancies. Diagn Microbiol Infect Dis 2010;68:132–9. Hope WW, Kruhlak MJ, Lyman CA, et al. Pathogenesis of Aspergillus fumigatus and the kinetics of galactomannan in an in vitro model of early invasive pulmonary aspergillosis: implications for antifungal therapy. J Infect Dis 2007;195:455–66. Mennink-Kersten MA, Donnelly JP, Verweij PE. Detection of circulating galactomannan for the diagnosis and management of invasive aspergillosis. Lancet Infect Dis 2004;4:349–57.