Hormographiella aspergillata: an emerging mould in acute leukaemia patients?

Transcription

1 ORIGINAL ARTICLE MYCOLOGY Hormographiella aspergillata: an emerging mould in acute leukaemia patients? A. Conen 1, M. Weisser 2, D. Hohler 3, R. Frei 3 and M. Stern 4 1) Division of Infectious Diseases and Hospital Epidemiology, Cantonal Hospital, Aarau, 2) Division of Infectious Diseases and Hospital Epidemiology, 3) Clinical Microbiology and 4) Division of Haematology, University Hospital, Basel, Switzerland Abstract We describe three invasive mould infections due to Hormographiella aspergillata occurring within 1 year in patients undergoing treatment for acute leukaemia. All patients presented with pulmonary infiltrates; one patient additionally had cerebral and ocular involvement. Diagnostic procedures included bronchoalveolar lavage in all, and video-assisted thoracoscopic surgery in two patients. Susceptibility testing was performed by E-test and detected low minimal inhibitory concentrations for voriconazole and amphotericin B. All patients received systemic antifungal therapy; however, all of them died. Despite this cluster of three cases of an unusual mould infection, no hospital source was detected. Keywords: Chemotherapy, leukaemia, mould infection, non-aspergillus mould, stem cell transplantation Original Submission: 15 February 2010; Revised Submission: 17 April 2010; Accepted: 29 April 2010 Editor: E. Roilides Article published online: 18 May 2010 Clin Microbiol Infect 2011; 17: /j x Corresponding author: A. Conen, MD, Cantonal Hospital Aarau, Tellstrasse, CH-5001 Aarau, Switzerland anna.conen@ksa.ch Introduction Invasive fungal infections are a major cause of morbidity and mortality in haematological patients undergoing chemotherapy or hematopoietic stem cell transplantation (HSCT) for acute leukaemia [1]. In recent years, emergence of non- Aspergillus moulds such as Zygomycetes, Fusarium, Scedosporium, and Basidiomycetes has been observed. Whether the increasing use of broad-spectrum antifungal prophylaxis with newer azoles plays a role in this evolution remains unclear. Other possible explanations are the introduction of molecular biology techniques allowing the exact identification of moulds and reporting bias. Hormographiella aspergillata is the asexual form of Coprinus cinereus (recently reclassified as Coprinopsis cinerea), a non-aspergillus mould belonging to the class of Basidiomycetes that occurs in compost and sewage and has previously been implicated in human infections on only a few occasions. Five individual cases in patients after chemotherapy for a haematological malignancy have been reported over the course of 10 years [2 6]. In the present study, we describe three cases of invasive H. aspergillata infection occurring within 1 year at our institution. Hospital Setting The isolation ward at the Division of Haematology at Basel University Hospital contains twelve single-bed rooms equipped with special high-efficiency particle air filters. Each year, approximately 50 allogeneic transplants, 40 autologous transplants and 50 cycles of chemotherapy for acute leukaemia are performed. According to infection surveillance data, 34 probable/proven invasive mould infections were diagnosed between 2003 and 2008, giving an annual rate of approximately five or six cases (Weisser M, Blattler L, Elzi L, et al, unpublished data). Antimicrobial prophylaxis consisted of cotrimoxazole, fluconazole and, in the case of positive herpes simplex virus serology, valacyclovir. No primary prophylaxis against moulds was administered. Screening for mould infections was performed by twice-weekly measurement of serum galactomannan (Platelia Aspergillus; Bio-Rad Laboratories, Hercules, CA, USA; using the standard cut-off value of 0.5 optical densities) and weekly pulmonary computed tomography (CT) scanning. Surveillance fungal cultures of air filters and floor specimens were carried out monthly. Journal Compilation ª2010 European Society of Clinical Microbiology and Infectious Diseases

2 274 Clinical Microbiology and Infection, Volume 17 Number 2, February 2011 CMI Microbiological Investigations Blood cultures were performed by use of the BacT/ALERT system (biomérieux, Hazelwood, MO, USA) and a pair of aerobic/anaerobic bottles (FA/FN). For fungal cultures from other clinical specimens, Sabouraud agar containing chloramphenicol and gentamicin was inoculated and incubated for 3 weeks. Within 3 5 days of incubation, white to cream-coloured colonies were observed. Growth was faster at 37 C than at 28 C. Microscopy of the aerial mycelium showed septate conidiophores from which clusters of smooth-walled, hyaline and cylindrical arthroconidia were produced (Fig. 1a). In addition, clamps and, after 6 weeks of incubation, sclerotial bodies were observed. The identification of H. aspergillata was molecularly confirmed in each patient by amplification and sequencing of the D2 region of large-subunit ribosomal RNA gene using an ABI Prism 3130 sequencer and MicroSEQÒ D2 LSU rdna Fungal Identification kit (Applied Biosystems, Foster City, CA, USA) [7]. Susceptibility testing (a) (b) was performed by E-test (AB Biodisk, Solna, Sweden) in accordance with manufacturer s instructions and the Clinical and Laboratory Standards Institute (CLSI) [8]. Case 1 A 41-year-old female patient with acute myeloid leukaemia (AML) was scheduled for double allogeneic peripheral blood HSCT in October After conditioning with melphalan, a first T-cell depleted transplant from a human leukocyte antigen-identical sibling donor was administered. Engraftment was promptly achieved on day 9; however, the patient rejected the first transplant on day 15 and went on to a second T-cell replete transplant after myeloablative conditioning. On the day of the second transplant, the patient developed neutropenic fever. A consolidation with halo sign suspicious of invasive fungal infection was detected on pulmonary CT scanning. Bronchoalveolar lavage (BAL) was carried out but failed to reveal a causative microorganism. Serum galactomannan levels were within the reference range. According to local and national guidelines [9,10], an empirical treatment for possible pulmonary mould infection with intravenous voriconazole (4 mg/kg body weight twice-daily after loading dose) was initiated. No serum levels of voriconazole were performed to detect possible subtherapeutic serum levels. Repeat CT on day 18 after the second transplant and 3 days after engraftment showed progression of pulmonary infiltrates. Antifungal therapy was empirically changed to oral posaconazole (400 mg twice-daily). On day 24, acute graftversus-host disease (GvHD) grade III was diagnosed, and methylprednisolone was started. Because of increasing liver function tests under posaconazole therapy, antifungal treatment was switched to caspofungin on day 30. Weekly CT scans of the lungs showed lesions of constant size. On day 37, the patient developed amaurosis, and signs of intracranial hypertension. Cranial CT scanning displayed extensive cerebellar haemorrhage, and the patient died on day 41. Autopsy findings included extensive invasive mould infection of the lungs, left eye and both cerebral hemispheres. A blood culture taken during autopsy grew H. aspergillata. Case 2 FIG. 1. (a) Lactophenol cotton blue tape-mount of Hormographiella aspergillata culture displaying conidiophores and arthroconidia. (b) Pulmonary computed tomography scan of patient two showing bilateral ground glass infiltration and a pulmonary nodule compatible with invasive mould infection. A 63-year old female patient with secondary AML received induction chemotherapy in August On day 23, the patient developed neutropenic fever and a nonproductive cough. Pulmonary CT showed nodular infiltrates. Treatment with intravenous voriconazole for possible pulmonary mould

3 CMI Conen et al. H. aspergillata in acute leukaemia 275 infection (4 mg/kg body weight twice-daily after loading dose) was initiated. BAL and serum galactomannan testing were negative. Nine days later, the second course of chemotherapy was started. Follow-up CT showed a progression of the pulmonary infiltrates. BAL was repeated and remained negative, as did galactomannan. Antifungal therapy was empirically switched to oral posaconazole (400 mg twice-daily) in the still neutropenic patient. Prior, voriconazole serum levels were performed and demonstrated therapeutical serum levels (1 5 mg/l). Pulmonary CT 2 weeks later showed substantial progression of infiltrates spreading to the contralateral lung (Fig. 1b). Antifungal therapy was switched to high-dose intravenous liposomal amphotericin B (5 mg/kg body weight daily) and two of the largest pulmonary nodules were operatively removed by video-assisted thoracoscopic surgery (VATS). Histological evaluation of the pulmonary tissue revealed angioinvasive hyphae. Microbiological culture grew H. aspergillata (Fig. 1a). Based on in vitro susceptibility testing and concomitant renal failure, antifungal treatment was switched back to intravenous voriconazole (4 mg/kg body weight twice-daily after loading dose), with therapeutic serum levels thereafter. Under this treatment regimen, follow-up pulmonary CT showed marked improvement after recovery of haematological parameters, 2 months later. Because the patient did not enter remission from her AML, no further chemotherapy was administered and the patient died from disease progression. Case 3 A 55-year old male patient with relapsed AML was hospitalized for cord-blood transplantation in November During re-induction chemotherapy prior to cord-blood HSCT, the patient had developed a proven pulmonary infection with Aspergillus fumigatus, which was treated with intravenous voriconazole (4 mg/kg body weight twice-daily after loading dose) after operative resection. Pulmonary CT at admission showed no signs of persistent fungal infection. After myeloablative conditioning, double cord-blood transplantation was performed. The patient failed to engraft and therefore a second cord-blood transplant was scheduled after reduced intensity conditioning. One day before the second transplant, when the patient was still under treatment with intravenous voriconazole, a routine CT scan showed a pulmonary nodule, suspicious of invasive fungal infection. Antifungal treatment was switched to high-dose intravenous liposomal amphotericin B (5 mg/kg body weight) and pulmonary wedge resection was performed. Prior, voriconazole serum levels were performed and were within the therapeutic range (1 5 mg/l). Histological evaluation of the pulmonary tissue showed angioinvasive hyphae and microbiological cultures grew H. aspergillata. Antifungal treatment was switched back to intravenous voriconazole (4 mg/kg body weight twice-daily after loading dose) based on in vitro susceptibility testing results. Neutrophil engraftment occurred on day 32 after the second transplant. Follow-up CT at this time point showed persistent nodules in both lungs. Soon thereafter, the patient developed multi-organ failure in the context of acute GvHD to which he succumbed. Discussion We report three cases of pulmonary invasive mould infection with H. aspergillata in immunocompromised patients treated for AML. All patients had undergone prolonged periods of severe neutropenia; in one patient cerebral and ocular involvement was additionally found on autopsy. This patient had positive blood cultures with H. aspergillata, an unusual finding in invasive mould infections, which has especially been documented for invasive infections with Fusarium spp., where 40 50% of blood cultures in disseminated infection are described to be positive [11]. To our knowledge, this is the first published case of fungaemia with H. aspergillata. Another remarkable finding is that one the three patients in this series was co-infected with A. fumigatus (case 3). Of the five patients previously described in the literature, coinfection with another mould was documented in three [2,5,6]. The reason for this is unclear; however, it is possible that H. aspergillata infection occurs predominantly in patients with very severe immunosuppression, comprising a population at high risk of carrying concomitant infection with other moulds. BAL was carried out in all three patients after CT scanning had shown pulmonary infiltrates, but failed to identify a causative microorganism, confirming its low sensitivity for the diagnosis of invasive mould infection in immunocompromised patients [12]. By contrast, specimens collected by VATS grew H. aspergillata in both cases where it was performed, corroborating the high diagnostic yield of this procedure [13]. Serial serum galactomannan measurements remained negative in all three patients. Galactomannan, a component of the cell wall of Aspergillus spp., the most frequent mould isolated in haematological patients, is not produced by non-aspergillus moulds, and should not be used as an isolated screening strategy for mould infection. Because of the rarity of cases reported [2 6], there is no established treatment for H. aspergillata. Standardized antifungal susceptibility testing of filamentous fungi has been

4 276 Clinical Microbiology and Infection, Volume 17 Number 2, February 2011 CMI established only recently [8] and breakpoints with proven clinical relevance have yet to be defined. To date, in vitro susceptibility data in filamentous fungi cannot be translated into clinical practice without reservation. In all patients in this series, in vitro antifungal susceptibility testing was performed and showed low Minimal Inhibitory Concentrations (MIC) for voriconazole and amphotericin B, whereas caspofungin and fluconazole revealed limited and no activity; the MIC for posaconazole could not be determined in two patients as a result of distinct trailing growth within the inhibition zone, and demonstrated an elevated MIC in one patient with 2 mg/l (Table 1). These results are in line with three previously described case reports [2,3,6] and with data generated from isolates stemming from different natural substrates [14,15]. An evaluation of the in vivo efficacy of the antifungal treatment administered in this series is difficult as a result of not only the lack of autopsy in two patients, but also the progression of the underlying haematological disease with consecutive death: The first patient died as a direct consequence of fungal infection and was under treatment with caspofungin at that time, after severe hepatotoxicity due to azole treatment. Echinocandins are ineffective against many non-aspergillus moulds, including H. aspergillata [16]; however, in this case, the causative microorganism was only identified and tested for susceptibility to antifungal agents after the patient had died. Patients two and three showed radiological improvement after treatment with second generation azoles but died from progression of the underlying haematological disease and allogeneic transplant-related complications, respectively, precluding a definitive appraisal of the efficacy of antifungal treatment administered. The unusual cluster of three H. aspergillata cases at our division prompted an epidemiological investigation, including fungal cultures of environmental specimens and air particle filters, although this did not detect a common source of infection. All patients had been treated in different rooms and stemmed from different regions within Switzerland. No clear seasonality could be determined because of the small case number in the series. In the 12 months following the third case, no further infection with H. aspergillata was found; therefore, a hospital source appears unlikely. TABLE 1. Minimal inhibitory concentrations of antifungals against Hormographiella aspergillata patients isolates Compound Case 1 Case 2 Case 3 Amphotericin B 0.5 mg/l 0.5 mg/l 0.5 mg/l Caspofungin 32 mg/l not done >32 mg/l Voriconazole mg/l mg/l 0.25 mg/l Fluconazole 256 mg/l Not done >256 mg/l Posaconazole Not interpretable 2 mg/l Not interpretable Acknowledgements The authors would like to thank Andrej Trampuz for helpful discussions. The data described in this manuscript were presented at the ECCMID meeting held in Vienna in April Transparency Declaration All authors declare that there are no conflicts of interest. References 1. Marr KA, Carter RA, Crippa F, Wald A, Corey L. Epidemiology and outcome of mould infections in hematopoietic stem cell transplant recipients. Clin Infect Dis 2002; 34: Lagrou K, Massonet C, Theunissen K et al. Fatal pulmonary infection in a leukaemic patient caused by Hormographiella aspergillata. J Med Microbiol 2005; 54: Verweij PE, van Kasteren M, van de Nes J, de Hoog GS, de Pauw BE, Meis JF. Fatal pulmonary infection caused by the basidiomycete Hormographiella aspergillata. J Clin Microbiol 1997; 35: Surmont I, Van Aelst F, Verbanck J, De Hoog GS. A pulmonary infection caused by Coprinus cinereus (Hormographiella aspergillata) diagnosed after a neutropenic episode. Med Mycol 2002; 40: Nenoff P, Friedrich T, Schwenke H, Mierzwa M, Horn LC, Haustein UF. Rare fatal simultaneous mould infection of the lung caused by Aspergillus flavus and the Basidiomycete coprinus sp. in a leukemic patient. J Med Vet Mycol 1997; 35: Abuali MM, Posada R, Del Toro G et al. Rhizomucor variabilis var. regularior and Hormographiella aspergillata infections in a leukemic bone marrow transplant recipient with refractory neutropenia. J Clin Microbiol 2009; 47: Hall L, Wohlfiel S, Roberts GD. Experience with the microseq d2 large-subunit ribosomal DNA sequencing kit for identification of filamentous fungi encountered in the clinical laboratory. J Clin Microbiol 2004; 42: CLSI. Reference method for broth dilution antifungal susceptibility testing of filamentous fungi: approved standard., 2nd edn. Wayne, PA: Institute CaLS, 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: Fluckiger U, Marchetti O, Bille J et al. Treatment options of invasive fungal infections in adults. Swiss Med Wkly 2006; 136: Dignani MC, Anaissie E. Human fusariosis. Clin Microbiol Infect 2004; 10 (suppl 1): Reichenberger F, Habicht J, Matt P et al. Diagnostic yield of bronchoscopy in histologically proven invasive pulmonary aspergillosis. Bone Marrow Transplant 1999; 24: Reichenberger F, Habicht J, Kaim A et al. Lung resection for invasive pulmonary aspergillosis in neutropenic patients with hematologic diseases. Am J Respir Crit Care Med 1998; 158:

5 CMI Conen et al. H. aspergillata in acute leukaemia Gene J, Guillamon JM, Guarro J, Pujol I, Ulfig K. Molecular characterization, relatedness and antifungal susceptibility of the Basidiomycetous hormographiella species and coprinus cinereus from clinical and environmental sources. Antonie Van Leeuwenhoek 1996; 70: Gonzalez GM, Sutton DA, Thompson E, Tijerina R, Rinaldi MG. In vitro activities of approved and investigational antifungal agents against 44 clinical isolates of Basidiomycetous fungi. Antimicrob Agents Chemother 2001; 45: Diekema DJ, Messer SA, Hollis RJ, Jones RN, Pfaller MA. Activities of caspofungin, itraconazole, posaconazole, ravuconazole, voriconazole, and amphotericin B against 448 recent clinical isolates of filamentous fungi. J Clin Microbiol 2003; 41: