Aspergillus-specific T-cells to diagnose invasive ... · 3 Leonardo Potenza MD1*, Patrizia Barozzi BSc1*, Giulio Rossi MD2, Giovanni Palazzi 4 MD 3 , Daniela Vallerini BsC 1 , Giovanni

Aspergillus-specific T-cells to diagnose invasive aspergillosis in a leukemic child 1

with liver lesions mimicking hepato-splenic candidiasis. 2

Leonardo Potenza MD1*

, Patrizia Barozzi BSc1*

, Giulio Rossi MD2, Giovanni Palazzi 3

MD3, Daniela Vallerini BsC

1, Giovanni Riva MD

1, Monica Cellini MD

3, Monica 4

Morselli MD1, Francesco Volzone MD

1, Venturelli Claudia BSc

5, Chiara Quadrelli BSc

1, 5

Luciana Di Pancrazio MD4, Maria Carmen Cano MD

3, Paolo Paolucci MD

3, Giuseppe 6

Torelli MD1, Luppi Mario MD PhD

1. 7

1Department of Oncology and Hematology, Section of Hematology,

2Department of 8

Pathology, 3Department of Pediatric Sciences, Section of Hemato-Oncology, 9

4Department of Radiology,

5Microbiology and Virology Unit, University of Modena and 10

Reggio Emilia, Azienda Ospedaliera Policlinico di Modena, Modena, Italy 11

* First two authors contributed equally to the study 12

Corresponding author: 13

Professor Mario Luppi, MD, PhD 14

Department of Oncology and Hematology 15

University of Modena and Reggio Emilia 16

Azienda Ospedaliera Policlinico 17

41100, Modena 18

Italy 19

Tel: +39 0594224641 20

Fax: + 39 0594224549 21

22

ACCEPTED

Copyright © 2008, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.Clin. Vaccine Immunol. doi:10.1128/CVI.00198-08 CVI Accepts, published online ahead of print on 30 July 2008

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

Abstract 1

A child with acute myeloid leukemia presented multiple liver lesion mimicking 2

hepato-splenic candidiasis during the neutropenic phase following the induction 3

chemotherapy. All the available diagnostic tools resulted repeatedly negative, 4

including galactomannan. An enzyme-linked-immunospot (ELISPOT) assay showed 5

a high number of Aspergillus-specific T-cells producing interleukin-10 (IL-10-TH2) 6

and a low number of Aspergillus-specific T-cells producing interferon-γγγγ (IFN-γγγγ-TH1), 7

revealing IA before the confirmatory biopsy. A progressive skewing from the 8

predominance of IL-10-TH2 to a predominance of IFN-γγγγ-TH1 was observed close to 9

the complete resolution of the infection, and foreshadowed the outcome. ELISPOT 10

holds promise for diagnosing pediatric IA. 11

Word count 95 12

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

On day 9 of induction chemotherapy (idarubicin 10 mg/mq, days 1-3; cytarabine 200 1

mg/mq, days 1-7; etoposide 100 mg/mq, days 1-5), an 11 year-old child with acute 2

myeloid leukemia (AML), presented febrile neutropenia. Complete cultural and 3

molecular examination (CCME) of blood, urine, and stool were repeatedly unrevealing. 4

The galactomannan antigenemia (GM), determined twice a week, and the complete 5

pulmonary radiological study (PRxS) resulted negative. Empirical antibiotic therapy 6

(EAT) resulted ineffective, and empirical intravenous fluconazole (12 mg/kg/day) was 7

undertaken. On day 28, the febrile patient presented pain in the right flank, with 8

abdominal tenderness. Laboratory tests showed neutrophil recovery, slight increase of 9

transaminases and marked increase of alkaline phosphatase (ALP). On day 33, an 10

abdominal ultrasound (AUS) disclosed multiple hypoechoic liver lesions, with sparse 11

hyperecoic features (Fig. 1A). PRxS persisted negative. Fluconazole was replaced by 12

Liposomal amphotericin B (L-AmB) (1,5 mg/kg e.v.), with improvement of the 13

symptoms. On day 42, the patient, showing complete remission (CR) of AML, underwent 14

a second chemotherapeutic course with the same schedule. Concomitantly, the lesions 15

persisted unmutated on AUS. On day 45, fever rose again above 38°C. CCME and GM 16

remained negative. EAT was reintroduced, and L-AmB was increased to 3 mg/kg/day. 17

On day 51, AUS revealed a slight reduction of the liver lesions. Nonetheless, fever 18

persisted until bone marrow recovery. On day 62, a third course was undertaken 19

(cytarabine 3 g/mq b.i.d., days 1-3; etoposide 125 mg/mq, days 2-5). On day 69, after a 20

further febrile episode, AUS displayed enlargement of the liver lesions (Fig. 1B). CCME 21

and GM persisted negative. We judged it appropriate to explore the patient’s Aspergillus-22

specific T cell responses by a recently described application of an enzyme linked 23

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

immunospot (ELISPOT) assay to definitely rule out an invasive aspergillosis (IA) (9). On 1

day 75 the peripheral blood sample collected by ELISPOT assay, resulted positive for a 2

high number of Aspergillus-specific T-cells producing interleukin-10 (IL-10-TH2) and a 3

low number, producing interferon-gamma (IFN-γ-TH1) (Fig. 1G). On day 80, a liver 4

biopsy disclosed hepatic IA, by A. fumigatus, on either histologic (Fig. 2A to B) or 5

cultural examination (not shown). Caspofungin (50 mg/day) was associated to L-AmB. 6

On day 84, ELISPOT confirmed a very high number of IL-10-TH2 and low number of 7

IFN-γ-TH1 (Fig. 1G). On day 95, fever resolved. On day 102, the AUS showed a slight 8

decrease of the liver lesions in size (Fig. 1C). Eight days later, ELISPOT demonstrated 9

sustained number of IL-10-TH2 and stable number of IFN-γ-TH1 (Fig. 1G). Cytarabine 3 10

g/mq b.i.d. (days 1-3) and mitoxantrone 10 mg/mq (days 3-4) were administered without 11

additional complications. On day 132, because of a smaller reduction of the liver lesions 12

on AUS (Fig. 1D), L-AmB was withdrawn. On day 166, ELISPOT persisted positive, 13

with decreased number of IL-10-TH2 and a steady number of IFN-γ-TH1 (Fig. 1G). Four 14

days later, the AUS revealed significant reduction of the liver lesions in size and number 15

(Fig. 1E). On day 190, cyclophosphamide 4 g/mq and etoposide 200 mg/mq (days 2-4) 16

were given. The neutropenia elapsed asymptomatically. On day 204, the ELISPOT 17

resulted positive only for IFN-γ-TH1 (Fig. 1G). On day 219, the AUS revealed a 18

homogenous liver without focal lesions and caspofungin was finally discontinued (Fig. 19

1F). Actually, the patient is well and in CR. 20

21

IA and invasive candidiasis (IC) represent the most common causes of infection-22

related mortality in pediatric patients with AML (10). Data from the Centre for Disease 23

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

Control and Prevention show that the mortality associated with IA has increased 357% 1

since 1980, due to the limitation of diagnostic tools in establishing a prompt and 2

undoubted diagnosis (7). Also GM, the widest used non-cultural method, is hampered by 3

a variable diagnostic accuracy, especially in children (2, 8). 4

Clinically, pulmonary infiltrates are the most common presentation of IA, with 5

liver involvement occurring in less than 15% of patients, only in the late phase of the 6

infection (3). In contrast, visceral presentation is common in IC, expecially in the chronic 7

form. The latter, also termed hepato-splenic candidiasiasis (HSC), occurs in 3% to 7% of 8

neutropenic patients, persisting for months, despite granulocyte recovery and antifungal 9

treatment. Fever, upper gastrointestinal symptoms and increased serum ALP characterises 10

HSC, presenting with four dominant ultrasonographic patterns: “wheel within a wheel”, 11

“bull’s eye”, hypoechoic lesion or echogenic foci (4). 12

Only four cases of primary liver IA have so far been reported in three adult and 13

one pediatric patients, respectively, presenting with either single or multiple lesions. In all 14

four patients, the diagnosis was biopsy proven, and preceded by a single positive GM test 15

in only one case (5, 6, 12, 13). 16

This is a further case of liver IA, mimicking HSC, on clinical and radiological 17

grounds. Of note, while all the others diagnostic tools, including GM, resulted repeatedly 18

negative, ELISPOT has been the sole suspicion of IA and shortly preceded the 19

confirmatory biopsy. We have recently reported the potential of ELISPOT assay to 20

diagnose IA in adult patients with pulmonary infiltrates (9). Also in this case of 21

extrapulmonary IA, ELISPOT has provided a kinetics of the Aspergillus-specific T cell 22

responses during the course of IA by showing: 1) a permissive response, polarized in IL-23

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

10-TH2, at the onset; 2) a stable IFN-γ-TH1 response with a progressive decrease of IL-1

10-TH2 response, which, in turn, was sufficient to countebalance the IFN-γ-TH1, and 2

avoid an excessive inflammatory response, during the regression phase; 3) a protective 3

response, polarized in IFN-γ-TH1, at the complete resolution (Fig. 2). This is reminiscent 4

of the kinetics of the Aspergillus-specific T cell responses in the mouse model of IA, and 5

consistent with the notion that the presence of Aspergillus-specific IFN-γ-TH1 responses 6

provides a better chance to survive IA in stem cell transplant patients (1, 10). 7

In conclusion, liver IA should be considered in the differential diagnosis of liver 8

lesions in neutropenic patients, also including HSC, tuberculosis and staphylococcal 9

infections. The ELISPOT assay may have a role in the diagnosis of IA also in pediatric 10

patients. The diagnostic accuracy of ELISPOT needs larger confirmatory studies. 11

12

Word Count 999 13

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

References 1

1. Hebart, H., C. Bollinger, P. Fisch, J. Sarfati, C. Meisner, M. Baur, J. 2

Loeffler, M. Monod, J. P. Latgé, and H. Einsele. 2002. Analysis of T-cell 3

responses to Aspergillus fumigatus antigens in healthy individuals and patients 4

with hematologic malignancies. Blood. 100:4521-4528. 5

2. Hope, W. W., and D. W. Denning. 2005. Laboratory diagnosis of invasive 6

aspergillosis. Lancet Infect. Dis. 5:609-622. 7

3. Hori, A., M. Kami, Y. Kishi, U. Machida, T. Matsumura, and T. Kashima. 8

2002. Clinical significance of extra-pulmonary involvement of invasive 9

aspergillosis: a retrospective autopsy-based study of 107 patients. J. Hosp. Infect. 10

50:175-182. 11

4. Kontoyiannis, D. P., M. A. Luna, B. I. Samuels, and G. P. Bodey. 2000. 12

Hepatosplenic candidiasis. Infect. Dis. Clin. North Am. 14:721-739. 13

5. Lee, T. Y., S. S. Yang, G. H. Chen, W. L. Hwang, Y. H. Lin, and J. I. Hwang. 14

2003. Hepatic abscesses caused by aspergillus fumigatus infection following 15

splenectomy and immunosuppressive therapy. J. Formos. Med. Assoc. 102:501-16

505. 17

6. Marotta, G., M. Tozzi, S. Sammassimo, M. Defina, D. Raspadori, A. Gozzetti, 18

and F. Lauria. 2005. Complete resolution of hepatic aspergillosis after non-19

myeloablative stem cell transplantation in a patient with acute myeloid leukemia. 20

Hematology. 10:383-386. 21

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

7. McNeil, M. M., S. L. Nash, R. A. Hajjeh, M. A. Phelan, L. A. Conn, B. D. 1

Plikaytis, and D. W. Warnock. 2001. Trends in mortality due to invasive 2

mycotic diseases in the United States, 1980-1997. Clin. Infect. Dis. 33:641-647. 3

8. Pfeiffer, C. D., J. P. Fine, and N. Safdar. 2006. Diagnosis of invasive 4

aspergillosis using a galactomannan assay: a meta-anaysis. Clin. Infect. Dis. 5

42:1417-27. 6

9. Potenza, L., P. Barozzi, D. Vallerini, R. Bosco, C. Quadrelli, L. Mediani, M. 7

Morselli, F. Forghieri, F. Volzone, M. Codeluppi, G. Rossi, G. Tazzioli, C. 8

Venturelli, G. Torelli, and M. Luppi. 2007. Diagnosis of invasive aspergillosis 9

by tracking Aspergillus-specific T cells in hematologic patients with pulmonary 10

infiltrates. Leukemia. 21:578-581. 11

10. Romani, L. 2004. Immunity to fungal infections. Nat. Rev. Immunol. 4: 1-23. 12

11. Sung, L., B. J. Lange, R. B. Gerbing, T. A. Alonzo, and J. Feusner. 2007. 13

Microbiologically documented infections and infection-related mortality in 14

children with acute myeloid leukaemia. Blood. 110:3532-3539. 15

12. Trachama, M., E. Roilides, N. Gompakis, K. Kanellopoulou, M. 16

Mpantouraki, and F. Kanakoudi-Tsakalidou. 2001. Case Report. Hepatic 17

abscesses due to Aspergillus Terreus in an immunodeficient child. Mycoses. 18

44:415-418. 19

13. van der Velden. W. J. F., N. M. A. Blijlevens, R. R. Klont, J. P. Donnelly, and 20

P. E. Verweij. 2006. Primary hepatic invasive aspergillosis with progression after 21

rituximab therapy for a post transplantation lymphoproliferative disorder. Ann. 22

Hematol. 85:621-623. 23

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

Figure Legends 1

Figure 1. A-G. Outcome of liver lesions and kinetics of the Aspergillus-specific T cell 2

responses in the patient. 3

A-E. Liver abdominal ultrasounds of the patient showing multiple hypoecoic nodular 4

lesions (arrows) with scanty hyperecoic features (arrow heads) at the onset of symptoms, 5

on day 33 (A); and during the antifungal treatment (B-E), on days 69, 102, 132 and 170, 6

respectively. 7

F. Abdominal ultrasound showing an homogeneous liver at the complete resolution of the 8

IA on day 219 of the treatment (F). 9

G. Kinetics of Aspergillus-specific T-cell responses, by IFN-γ and IL-10 ELISPOT assay, 10

in the patient during the entire course of liver IA. Black columns represent the number of 11

Aspergillus-specific IFN-γ-producing T-cells. Grey columns represent the number of 12

Aspergillus-specific interleukin-10 producing T-cells. Striped columns represent T-cell 13

responses in wells with PHA. Dotted columns represent T-cell responses in wells with 14

anti-CD3 human stimulating antibody (Mabtech, Nacka Strand, Sweden). Results are 15

shown as the mean of triplicate wells ± S.D. PHA = phytohemoagglutinin. SFCs = Spot 16

Forming Cells. PBMCs = peripheral blood mononuclear cells. IFN-γ= interferon gamma. 17

IL-10= interleukin 10. 18

Figure 2 A-B. Histological and immunohistochemical findings. 19

A. Hematoxylin-eosin staining (magnification, ×200) showed an extensive 20

granulomatous inflammation of the liver with dispersed birefringens filamentous 21

structures (inset at magnification, ×400). 22

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

B. Silver staining (magnification, ×200) showed septate, acutely branching, hyphae in a 1

pattern typical of aspergillosis (inset at magnification, ×400). 2

3

4

Acknowledgements 5

This study was supported by the Italian Ministry for Education, Universities and 6

Research (MIUR), Rome, Italy; the Associazione Italiana Lotta alle Leucemie, Mielomi e 7

Linfomi (AIL)-Modena-ONLUS; the Associazione Italiana per la Ricerca sul Cancro 8

(AIRC), Milan, Italy; the European Commission’s FP6 Life-Science-Health Programme 9

(INCA project; LSHC-CT-2005-018704) and the Programma di ricerca Regione-10

Università 2007-2009. We are indebted to Merck Sharp & Dohme and the staff of the 11

Medical Direction of the Azienda Ospedaliera Policlinico of Modena for the fellowship 12

support. Drs Potenza, Barozzi, Torelli and Luppi have applied for an international patent 13

regarding clinical applications of the ELISPOT assay for the diagnosis of Aspergillus 14

infection. 15

Conflict of interest or funds sourcing 16

Merck Sharp & Dohme provided funds for a fellowship support. 17

18

19

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from

ACCEPTED

on February 23, 2019 by guest

http://cvi.asm.org/

Dow

nloaded from