High incidence of false-positive Aspergillus galactomannan
test in multiple myeloma
Yasuo Mori,1,2Yoji Nagasaki,1,2Kenjiro Kamezaki,1Katsuto Takenaka,1Hiromi Iwasaki,2Naoki Harada,1
Toshihiro Miyamoto,1Yasunobu Abe,3Nobuyuki Shimono,1Koichi Akashi,1,2and Takanori Teshima2
Invasive aspergillosis (IA) remains one of the most significant causes
of morbidity and mortality in patients with hematological malignancies
undergoing chemotherapy and hematopoietic stem cell transplantation
(HSCT), mainly due to the difficulty in its early diagnosis. Monitoring
of galactomannan (GM) antigen, an exoantigen of Aspergillus, in the
blood by sandwich ELISA is a useful and noninvasive method for early
diagnosis of IA. The GM test has a sensitivity of 67–100% with a specif-
icity of 81–99% in neutropenic patients and allogeneic transplant recipi-
ents [1–3]. Although it has been widely used as a diagnostic criterion
for IA [4,5], one of the major limitations of this assay is false-positivity,
particularly in pediatric patients , patients with graft-versus-host dis-
ease (GVHD) [6,7], and those taking dietary GM [8,9] or fungus-derived
antibiotics, such as piperacillin-tazobactam (PIPC/TAZ) [10–12].
Multiple myeloma results from malignant proliferation of a single clone of
plasma cells, which produces a monoclonal immunoglobulin. Opportunistic
infection is a major cause of death in patients with myeloma [13,14]. The
risk for infection primarily resides during periods of chemotherapy-induced
neutropenia or in the terminal stages of the disease. Therefore, monitoring
of Aspergillus is recommended during chemotherapy-induced neutropenia.
One hundred twenty-four patients with hematological disorders hospitalized
in our institution from April 2007 to September 2009 were analyzed retrospec-
tively. The clinical characteristics of these patients are summarized in Sup-
porting Information Table 1. Twenty-seven patients had plasma cell associated
disorders (IgG type myeloma: 14, IgA type myeloma: 1, IgD type myeloma: 1,
Bence-Jones type myeloma: 3, lymphoplasmacytic lymphoma/Waldenstrom
macroglobulinemia: 3, plasma cell leukemia: 1, primary AL amyloidosis: 1,
and POEMS syndrome: 2). The remaining 97 patients were diagnosed as
acute leukemia (MDS/AML: 32, ALL: 15) or malignant lymphoma (n 5 50).
Out of the 124 patients, those receiving cytotoxic chemotherapy, autologous
peripheral blood stem cell transplantation (PBSCT), and allogeneic HSCT
were 81, 10, and 28, respectively. One hundred eleven patients received anti-
fungal prophylaxis, mostly with FLCZ or ITCZ. Eight patients were adminis-
tered PIPC/TAZ at sampling time points. Seventy patients were low risk for
the development of IA, whereas the remaining 54 patients were high risk.
In 21 of the 124 (16.9%) patients, GM antigenemia was positive at least
two consecutive times and their characteristics are shown in Table I. How-
ever, only seven of the 21 patients showed clinical features of IA and were
diagnosed with probable IA in the lung (Cases 1–7). Clinical features were
relieved with treatment with VRCZ in these patients, confirming a diagnosis
of IA. All of the seven patients had received antifungal prophylaxis, but not
antibiotics known to cause false-positive results, such as PIPC/TAZ [10–12],
amoxicillin/clavulanic acid, or amoxicillin . Four of the 7 patients were
high risk for the developing IA and three were low risk. Possible IA was diag-
nosed in four patients in the absence of positive GM results. On the other
hand, no proven or probable IA was detected in 103 patients with negative
Fourteen of the 21 (66.7%) patients with positive GM antigenemia did not
satisfy the diagnostic criteria of proven or probable IA (Cases 8–21 in Table
I); thus, their episodes were considered to be false-positive. None of the 14
patients were treated with antibiotics potentially causing false-positivity of
GM test. Antifungal prophylaxis had been given in all 14 episodes (FLCZ in
5, ITCZ in 5, and MCFG in 4). These patients did not show any clinical fea-
tures suggestive of IA. Chest CT scans did not show any abnormal findings
in 10 patients, whereas four patients showed abnormalities in the lung which
were not suspicious of IA. Diagnosis of these lung lesions were history of
pneumoconiosis in one patient, proven bacterial pneumonia in one, and idio-
pathic interstitial pneumonia in the remaining two patients. These lesions
were not deteriorated without antifungal treatment. (1?3)-b-D-glucan was
negative in all patients showing false-positive GM test. With a median follow-
up of 10 months (range: 1–19 months), these patients did not develop fungal
infection without treatment. On the other hand, false-negative GM results
were obtained in four patients (3.2 %). In this study, the sensitivity, specific-
ity, PPV, and NPV of the GM ELISA test were 63.6% (7/11), 87.6%
(99/113), 33.3% (7/21), and 96.1% (99/103), respectively.
Surprisingly, 11 out of the 14 patients showing the false-positive results
had diagnosis of multiple myeloma. The false-positivity of GM antigenemia
was significantly higher in myeloma patients (11/22, 50%) than those with
other hematological malignancies (3/102, 2.9%) (P < 0.001). Moreover, in
myeloma patients false-positive results were exclusively detected in those
with IgG subtype. Thus, rate of false-positivity was extremely high in patients
with IgG myeloma (11/14, 78.6%). We could not find any difference in the
characteristics of IgG myeloma patients with or without GM false-positivity,
including serum levels of IgG (5083 ± 2077 mg/dl versus 4713 ± 3729 mg/
dl). In addition, the GM test remained to be false-positive even after normal-
ization of IgG levels by chemotherapy in eight of the 11 patients. We also
evaluated GM antigenemia in frozen serum samples collected before
chemotherapy in three myeloma patients who showed false-positivity after
chemotherapy to rule out the possibility that administration of myeloma-
specific chemotherapy is associated with the false-positivity, and confirmed
GM positivity before chemotherapy in these samples.
In a univariate logistic regression analysis, IgG myeloma and low-risk cate-
gory were strongly associated with false-positive GM antigenemia. Sex, type
of treatment, antibiotics, corticosteroid usage, and serum levels of immunoglo-
bulins were also significant or marginally associated with false-positivity. Multi-
variate analysis confirmed diagnosis of IgG myeloma as the only independent
risk factor for false-positivity (odds ratio, 59.41; 95% confidence interval,
8.19–431.0; P < 0.001) (Supporting Information, Table 2). In patients with
other diseases, the GM assay showed a high sensitivity (7/11, 63.6%), specif-
icity (96/99, 97.0%), PPV (7/10, 70%), and NPV (96/100, 96%). In contrast,
for patients with IgG myeloma, specificity and PPV of the assay were very low
(3/14, 21.4%, and 0/11, 0%), whereas NPV was 100% even in this cohort.
A recent meta-analysis addressing the accuracy of a GM assay for diag-
nosing IA confirmed the clinical usefulness of this test with a sensitivity of
71% and a specificity of 89% . Although our study demonstrated similar
sensitivity (7/11:64%) and specificity (103/117:88%) of the GM test, PPV
(33%) was lower, compared with previous studies that demonstrate 40–60%
PPV [7,17–19]. This difference is due to an unexpectedly high incidence of
GM antigen false-positivity (11.3%) in our study. It should be noted, how-
ever, that screening of GM antigen was performed less frequently in this
study compared with previous studies, where GM antigenemia was eval-
uated two to three times per week [1,2,20], and such a frequent monitoring
is ideal to assure the optimal PPV and NPV.
Diagnosis of multiple myeloma is a major risk factor for GM false-positivity.
In particular, the false-positivity was exclusively observed in patients with IgG
myeloma and was not observed in patients with other types of plasma cell
disorders. These observations should be confirmed in a larger study because
some studies previously reported the usefulness of GM antigen assay as a
diagnostic tool for IA among patients with hematological malignancies includ-
ing IgG myeloma [21,22], and only small numbers of patients with plasma
cell disorders other than IgG myeloma were included in this study. Low-risk
category of developing IA was a risk for false-positive results in a univariate,
but not multivariate analysis. IgG myeloma remained a strong risk for false-
positivity even after the compensation by the risk categorization.
Mechanisms of high frequency of GM false-positivity in myeloma patients
remain to be investigated. (1?3)-b-D-glucan, which is released from the fun-
gal cell wall, is also widely used to support diagnosis of fungal infections
C 2010 Wiley-Liss, Inc.
American Journal of Hematology
and adopted as one of the microbiological criterion for probable IA in the
revised EORTC/MSG definition . A previous study reported that high lev-
els of immunoglobulins interfere with the measurement of (1?3)-b-D-glucan
by causing precipitation of insolubilized proteins and increase the nonspe-
cific optical density levels of reaction fluid , although (1?3)-b-D-glucan
was negative in patients showing false-positive GM test in this study. This
phenomenon has not been reported in the GM assay. However, serum levels
of IgG were not directly associated with the false-positivity; IgG levels did
not differ between IgG myeloma patients with and without false-positivity. In
addition, the GM test remained to be false-positive even after normalization
of IgG levels by chemotherapy in eight of the 11 patients.
Causative role of PIPC/TAZ, amoxicillin/clavulanic acid, and amoxicillin in
GM false-positivity has been well documented [10–12,15]; therefore, collec-
tion of samples before infusion of these antibiotics and the use of a relatively
higher cut-off level (>0.7) are recommended in patients receiving these
agents . In this study, no patients with GM false-positivity received these
antibiotics at the time of sampling. It has been hypothesized that dietary
contamination by GM causes GM false-positivity by the translocation of diet-
ary GM into the systemic circulation through the disrupted intestinal mucosal
barrier, especially in patients with gastrointestinal GVHD after allogeneic
HSCT [6,7,9]. In our cohort, 1(3.6%) of the 28 patients who underwent allo-
geneic HSCT showed false-positivity. This patient with lymphoma had acute
GVHD involving in the skin and intestine. A previous study demonstrated
that false-positive results were preferentially observed in patients with febrile
neutropenic sepsis , although subsequent study was unable to replicate
this result . A recent study revealed that serum GM antigen levels was
significantly higher in severely neutropenic patients (< 0.1 3 109/L) than in the
other patients . However, we did not find such an association in this study.
In conclusion, the incidence of false-positive GM antigenemia was high in
patients with IgG myeloma. Although the results should be confirmed in a
prospective study including larger numbers of patients, positive results of
GM antigenemia may be interpreted with caution, and intimate survey
including CT scan or other microbiological markers will be recommended in
1Medicine and Biosystemic Science, Kyushu University Graduate School of
Medical Sciences, Fukuoka Japan;2Center for Cellular and Molecular Medicine,
Kyushu University Graduate School of Medical Sciences, Fukuoka Japan;
3Medicine and Bioregulatory Science, Kyushu University Graduate School of
Medical Sciences, Fukuoka Japan
Additional Supporting Information may be found in the online
version of this article.
*Correspondence to: Takanori Teshima, Center for Cellular and
Molecular Medicine, Kyushu University Hospital, 3-1-1 Maidashi,
Higashi-ku, Fukuoka 812-8582, Japan
Contract grant sponsor: Health and Labor Science Research Grants;
Japan Society for the Promotion of Science
Conflict of interest: Nothing to report.
Published online 2 March 2010 in Wiley InterScience
1. Herbrecht R, Letscher-Bru V, Oprea C, et al. Aspergillus galactomannan
detection in the diagnosis of invasive aspergillosis in cancer patients. J Clin
2. Maertens J, Verhaegen J, Lagrou K, et al. Screening for circulating galacto-
mannan as a noninvasive diagnostic tool for invasive aspergillosis in pro-
longed neutropenic patients and stem cell transplantation recipients: A pro-
spective validation. Blood 2001;97:1604–1610.
3. Wheat LJ. Rapid diagnosis of invasive aspergillosis by antigen detection.
Transpl Infect Dis 2003;5:158–166.
4. Ascioglu S, Rex JH, De Pauw B, et al. Defining opportunistic invasive fungal
infections in immunocompromised patients with cancer and hematopoietic stem
cell transplants: An international consensus. Clin Infect Dis 2002;34:7–14.
5. De Pauw B, Walsh TJ, Donnelly JP, et al. Revised definitions of invasive fun-
gal disease from the European Organization for Research and Treatment of
Cancer/Invasive Fungal Infections Cooperative Group and the National Insti-
tute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG)
Consensus Group. Clin Infect Dis 2008;46:1813–1821.
6. Hamaki T, Kami M, Kanda Y, et al. False-positive results of Aspergillus
enzyme-linked immunosorbent assay in a patient with chronic graft-versus-
host disease after allogeneic bone marrow transplantation. Bone Marrow
TABLE I. Characteristics of 21 Patients with Positive GM Test
Risk of IA
nodules with halo
consolidation with pleural pain
consolidation with pleural pain
old inflammatory change
MDS, myelodysplastic syndrome; AML, acute myelogenous leukemia; ML, malignant lymphoma; MM, multiple myeloma; CTx, chemotherapy; allo-SCT, allogeneic stem cell transplantation; C.O.I, cut-off index; IP, interstitial pneumonia; F-P,
false-positive; CFPM, cefepime; MEPM, meropenem; AMK, amikacin; CAZ, ceftazidime; VCM, vancomycin; CZOP, cefozopran; PIPC/TAZ, piperacillin/tazobactam; CPFX, ciprofloxacin; PSL, prednisolone; mPSL, methylprednisolone; DEXA,
dexamethasone.bAdministrated at initial positive sampling.
450American Journal of Hematology
7. Asano-Mori Y, Kanda Y, Oshima K, et al. False-positive Aspergillus galacto- Download full-text
mannan antigenaemia after haematopoietic stem cell transplantation. J Anti-
microb Chemother 2008;61:411–416.
8. Ansorg R, Van Den Boom R, Rath PM. Detection of Aspergillus galacto-
mannan antigen in foods and antibiotics. Mycoses 1997;40:353–357.
9. Murashige N, Kami M, Kishi Y, et al. False-positive results of Aspergillus
enzyme-linked immunosorbent assays for a patient with gastrointestinal graft-
versus-host disease taking a nutrient containing soybean protein. Clin Infect
10. Sulahian A, Touratier S, Ribaud P. False positive test for Aspergillus antigene-
mia related to concomitant administration of piperacillin and tazobactam.
N Engl J Med 2003;349:2366–2367.
11. Viscoli C, Machetti M, Cappellano P, et al. False-positive galactomannan pla-
telia Aspergillus test results for patients receiving piperacillin-tazobactam. Clin
Infect Dis 2004;38:913–916.
12. Orlopp K, Von Lilienfeld-Toal M, Marklein G, et al. False positivity of the
Aspergillus galactomannan Platelia ELISA because of piperacillin/tazobactam
treatment: Does it represent a clinical problem? J Antimicrob Chemother
13. Blade J, San Miguel JF, Fontanillas M, et al. Increased conventional che-
motherapy does not improve survival in multiple myeloma: Long-term
results of two PETHEMA trials including 914 patients. Hematol J 2001;2:
14. Augustson BM, Begum G, Dunn JA, et al. Early mortality after diagnosis
of multiple myeloma: Analysis of patients entered onto the United king-
dom Medical Research Council trials between 1980 and 2002--Medical
Research Council Adult Leukaemia Working Party. J Clin Oncol 2005;23:
15. Mattei D, Rapezzi D, Mordini N, et al. False-positive Aspergillus galacto-
mannan enzyme-linked immunosorbent assay results in vivo during amoxicil-
lin-clavulanic acid treatment. J Clin Microbiol 2004;42:5362–5363.
16. Pfeiffer CD, Fine JP, Safdar N. Diagnosis of invasive aspergillosis using a gal-
actomannan assay: A meta-analysis. Clin Infect Dis 2006;42:1417–1427.
17. Kawazu M, Kanda Y, Nannya Y, et al. Prospective comparison of the diagnos-
tic potential of real-time PCR, double-sandwich enzyme-linked immunosorb-
ent assay for galactomannan, and a (1-->3)-beta-D-glucan test in weekly
screening for invasive aspergillosis in patients with hematological disorders.
J Clin Microbiol 2004;42:2733–2741.
18. Alhambra A, Cuetara MS, Ortiz MC, et al. False positive galactomannan
results in adult hematological patients treated with piperacillin-tazobactam.
Rev Iberoam Micol 2007;24:106–112.
19. Lai CC, Hsu HL, Lee LN, Hsueh PR. Assessment of Platelia Aspergillus
enzyme immunoassay for the diagnosis of invasive aspergillosis. J Microbiol
Immunol Infect 2007;40:148–153.
20. Maertens J, Verhaegen J, Demuynck H, et al. Autopsy-controlled prospective
evaluation of serial screening for circulating galactomannan by a sandwich
enzyme-linked immunosorbent assay for hematological patients at risk for
invasive Aspergillosis. J Clin Microbiol 1999;37:3223–3228.
21. Miceli MH, Grazziutti ML, Woods G, et al. Strong correlation between serum asper-
22. Woods G, Miceli MH, Grazziutti ML, et al. Serum Aspergillus galactomannan
antigen values strongly correlate with outcome of invasive aspergillosis: A
study of 56 patients with hematologic cancer. Cancer 2007;110:830–834.
23. Yoshida K. [Recent advances of serodiagnosis for systemic fungal infections].
Nippon Ishinkin Gakkai Zasshi 2006;47:135–142.
24. Swanink CM, Meis JF, Rijs AJ, et al. Specificity of a sandwich enzyme-linked
immunosorbent assay for detecting Aspergillus galactomannan. J Clin Micro-
25. Cordonnier C, Botterel F, Ben Amor R, et al. Correlation between galacto-
mannan antigen levels in serum and neutrophil counts in haematological
patients with invasive aspergillosis. Clin Microbiol Infect 2009;15:81–86.
Mixed phenotype acute leukemia with t(11;19)(q23;p13.3)/
MLL-MLLT1(ENL), B/T-lymphoid type: A first case report
Mojdeh Naghashpour,1* Jeffrey Lancet,2Lynn Moscinski,1and Ling Zhang1
The majority of cases of acute leukemia belong to a specific lineage ori-
gin, either lymphoid or myeloid, and therefore are classified as acute lym-
phoblastic leukemia (ALL) or acute myelogenous leukemia (AML), based
on morphologic features and cytochemical and immunophenotypic pro-
file of the blast cells. A minority of acute leukemias however, show no
clear evidence of differentiation along a single lineage. These are now
classified under acute leukemias of ambiguous lineage by the most
recent WHO classification and account for <4% of all cases of acute leu-
kemia . They include leukemias with no lineage specific antigens
(acute undifferentiated leukemias) and those with blasts that express anti-
gens of more than one lineage to such degree that it is not possible to
assign the leukemia to any one particular lineage with certainty (mixed
phenotype acute leukemias). The latter can either be leukemias with two
distinct populations of blasts, each expressing antigens of a different lin-
eage (historically referred to as ‘‘bilineal’’ leukemias) or a single blast
population expressing antigens of multiple lineages (historically referred
to as ‘‘biphenotypic’’ acute leukemias) . Acute leukemias of ambiguous
lineage mayharbora variety
t(9;22)(q34;q11) or translocations associated with mixed lineage leuke-
mias (MLL) gene, i.e., t(11;V)(q23;V), occur frequently enough and are
associated with distinct features, that are considered as separate entities
according to the recent WHO classification. Co-expression of myeloid
and B-lymphoid antigens is most common in mixed phenotype acute leu-
kemia (MPAL), followed by co-expression of myeloid and T-lymphoid anti-
gens, accounting for 66–70% and 23–24% of MLLs, respectively. Co-
expression of B- and T-lineage associated antigens or antigens of all
three lineages is exceedingly rare, accounting for <5% of MLLs [3,4]. The
requirements for assigning more than one lineage to a single blast popu-
lation has been established by current WHO classification .
Chromosomal rearrangements of mixed lineage leukemias (MLL) gene on
chromosomal segment 11q23 occur in a subset of leukemias with poor prog-
nosis and are seen in pediatric, adult and therapy-related acute leukemias .
of geneticlesions. Those with
MLL translocations can be found in de novo AML and lymphoid lineages
(ALL), secondary myelodysplastic syndrome, as well as mixed phenotype
acute leukemias [6–8]. Among them, mixed phenotypic acute leukemia has a
high incidence of chromosomal abnormalities involving MLL gene [2,3]. Wild-
type MLL is a transcriptional regulatory factor involved in the maintenance of
clustered homeobox (Hox) gene expression, particularly during hematopoietic
development . MLL translocations generate an in-frame fusion protein, in
which the N-terminal portion of the MLL protein is fused to the C-terminal
region of a partner protein, resulting in a chimeric protein with oncogenic prop-
erties . Leukemias that harbor MLL translocations are a unique biologic
subgroup that co-express myeloid and lymphoid associated genes and an
overall gene expression profile that characterize those of the precursor cells
[11,12]. To date, more than 51 partner genes have been characterized at the
molecular level and an additional 35 genetic loci have been identified by cyto-
genetic analysis . The most frequent translocations, t(4;11)(q21;23) involv-
t(11;19)(q23;p13.3) involving MLLT1 (ENL) account for 80% of investigated
leukemia samples . The characteristic of the fusion partner seems to
determine a bias for generation of myeloid versus lymphoid or mixed lineage
type leukemia. While MLL-MLLT3 and MLL-MLLT2 are predominantly associ-
ated with a myeloid or a lymphoid/mixed lineage phenotype, MLL-MLLT1
occurs equally likely in both leukemia subclasses . To our knowledge how-
ever, B/T-lymphoid MLL harboring t(11;19)(q23;p13.3) has not been reported.
A previously healthy 37-year-old male, presented to a local primary care
physician with flu-like symptoms. A complete blood count (CBC) was within
normal limits, with a normal differential count, at initial visit. He was
treated conservatively without relief of his symptoms. A CBC, repeated 1
month later revealed a white blood cell count 10.6 3 109/L, including 3%
circulating blasts, hemoglobin 8.6 g/dL, and a platelet count of 179 3
109/L. His absolute neutrophil count was 0.4 3 106/L. With an uncertain
diagnosis, the patient was referred to our institution for further consulta-
American Journal of Hematology 451