- Frederic Tissot1⇑,
- Samir Agrawal2,
- Livio Pagano3,
- Georgios Petrikkos4,
- Andreas H. Groll5,
- Anna Skiada6,
- Cornelia Lass-Flörl7,
- Thierry Calandra1,
- Claudio Viscoli8 and
- Raoul Herbrecht9
- 1Infectious Diseases Service, Centre Hospitalier Universitaire Vaudois and University of Lausanne, Switzerland
- 2Division of Haemato-Oncology, St Bartholomew’s Hospital and Blizard Institute, Queen Mary University, London, UK
- 3Hematology, Catholic University of Sacred Heart, Roma, Italy
- 4School of Medicine, European University Cyprus, Engomi, Cyprus
- 5Infectious Disease Research Program, Center for Bone Marrow Transplantation and Department of Pediatric Hematology/Oncology, University Children’s Hospital, Münster, Germany
- 61st Department of Medicine, University of Athens, Greece
- 7Division of Hygiene and Medical Microbiology, Medical University of Innsbruck, Austria
- 8University of Genova (DISSAL), Infectious Disease Division, IRCCS San Martino-IST, Genova, Italy
- 9Oncology and Hematology, Hôpitaux Universitaires de Strasbourg and Université de Strasbourg, France
The European Conference on Infections in Leukemia (ECIL) provides recommendations for diagnostic strategies and prophylactic, pre-emptive or targeted therapy strategies for various types of infection in patients with hematologic malignancies or hematopoietic stem cell transplantation recipients. Meetings are held every two years since 2005 and evidence-based recommendations are elaborated after evaluation of the literature and discussion among specialists of nearly all European countries. In this manuscript, the ECIL group presents the 2015-update of the recommendations for the targeted treatment of invasive candidiasis, aspergillosis and mucormycosis. Current data now allow a very strong recommendation in favor of echinocandins for first-line therapy of candidemia irrespective of the underlying predisposing factors. Anidulafungin has been given the same grading as the other echinocandins for hemato-oncological patients. The beneficial role of catheter removal in candidemia is strengthened. Aspergillus guidelines now recommend the use of either voriconazole or isavuconazole for first-line treatment of invasive aspergillosis, while first-line combination antifungal therapy is not routinely recommended. As only few new data were published since the last ECIL guidelines, no major changes were made to mucormycosis recommendations.
The European Conference on Infections in Leukemia (ECIL) is the result of a collaboration between the European Organization for Research and Treatment of Cancer (EORTC), the European Society for Blood and Marrow Transplantation (EBMT), the European Leukemia Net (ELN), and the International Immunocompromised Host Society (ICSH). First recommendations for the treatment of Candida and Aspergillus infections in hematologic patients were published in 2007 after the first conference (ECIL-1) and have then been updated at ECIL-2 and ECIL-3.1,2 First recommendations for the diagnosis and treatment of mucormycosis have been published after ECIL-3.3 ECIL-4 updates for antifungal therapy were only available as slides on the websites of these participating societies without publication of a manuscript in consideration of the lack of substantial new data and the limited modifications compared to the latest publication.
With respect to the targeted treatment of fungal infections, the goals for ECIL-5 were to update the recommendations with analysis of the new data for invasive candidiasis, aspergillosis and mucormycosis in hematologic patients. The update was also necessary to change the prior 5-level grading (A to E) used during the ECILs 1 to 4 for the strength of recommendations for Candida and Aspergillus infections into the 3-level grading (A to C) already used during ECIL-3 for the first recommendation for mucormycosis (Table 1).1–3 The grading for quality of evidence has not been modified.
The ECIL-5 meeting was held in September 2013 and involved 57 experts from 21 countries, including 3 non-European countries. Slides of the conclusions of the ECIL-5 were made available on the websites of the EORTC, EBMT, ELN, and ICHS. The ECIL-6 meeting was held in September 2015 with the presence of 55 experts from 24 countries, including 4 non-European countries (see list of collaborators at the end of this Review).
At both the ECIL-5 and the ECIL-6 meetings, the antifungal therapy working group made a search for new publications regarding treatment of invasive candidiasis, aspergillosis and mucormycosis. The group was divided into three subgroups, each being responsible for one of each fungal infection type. The literature search was performed in Pubmed and Cochrane databases. Abstracts presented at major congresses during the previous two years were also retrieved and integrated into the ECIL recommendation. All recommendations referring to an abstract, however, were classified as provisional until the publication of the final manuscript.
The working group presented its recommendations during the plenary session at the ECIL-5 meeting and then incorporated the suggestions coming from the assembly. In cases in which full consensus was not obtained, the decision was put to the vote, and the final decision was based on a majority of votes from the full ECIL-5 assembly. The updated recommendations were presented on the next day during a second plenary session for final approval. Recommendations were graded on the basis of the strength of recommendations (3-level scale: A, B, or C) and quality of evidence (3-level scale: I, II, or III), as detailed in Table 1.
The manuscript of the ECIL-5 was put on hold after a debate arose on differences between ECIL and European Society for Clinical Microbiology and Infectious Diseases (ESCMID) / European Confederation of Medical Mycology (ECMM) recommendations on guidelines for prophylaxis and treatment of invasive aspergillosis (draft presented at the ECCMID 2014).4 Two joint meetings were subsequently held (December 2014 and April 2015) to identify the differences and the exact reasons for these differences. The aim was not to modify the recommendations made by each of the two groups but rather to add explanations on the differences in the manuscript. For further clarification, a joint presentation was given at the ECIL-6 by members of the ECIL group and of the ESCMID/ECMM group. This resulted in a delay in publication of the ECIL-5 recommendations and during the ECIL-6 plenary session, the ECIL assembly approved a new search for publications or abstracts until September 2015 with inclusion of all relevant data on aspergillosis, candidiasis and mucormycosis for a full update of the guidelines. Final approval by the majority of the members of the group was obtained in Autumn 2015. The current manuscript includes updates from both the ECIL-5 and the ECIL-6 and is called “ECIL-6 guidelines for the treatment of invasive candidiasis, aspergillosis and mucormycosis in leukemia and hematopoietic stem cell transplant patients”.
Like previous ECIL recommendations, the current guidelines for invasive candidiasis cover the hematologic population as well as the general population of patients. Although hematologic patients are the main focus of the recommendation, this distinction is maintained because available data from the original randomized controlled trials mainly include non-neutropenic patients. Chronic infections are not considered. Twenty-two major publications were identified (Tables 2 and 3).5–26 Fifteen reported primary results from clinical trials.5–11,13–17,19,20 One publication analyzed results of a subgroup of cancer patients from a previously published trial.12 One publication reported the analysis of pooled data from 2 trials previously published with a focus on patients with an underlying malignancy.21 All these studies were published before the ECIL-4. Since then, 5 studies have been identified, including one patient-level quantitative review of 7 published trials on invasive candidiasis, one pooled patient-level data analysis from 5 prospective trials on anidulafungin, one systematic review of 17 randomized clinical trials focusing on invasive candidiasis in neutropenic patients, one prospective non-comparative trial evaluating a strategy of early oral switch from anidulafungin for invasive candidiasis, and one observational study comparing the initial use of echinocandin-based versus azole-based regimen for C. parapsilosis candidemia.22–26 These publications were the reasons for the change in guidelines. Characteristics of these studies and main results are shown in Tables 2 and 3.
The number of neutropenic patients included in each of these studies was low and limited the level of evidence of the recommendation for this group of patients. The review published by Andes et al. showed that, in the univariate analysis, neutropenia was one of the factors significantly and negatively associated both with clinical outcome and with survival.22 In the multivariate analysis, however, the effect of neutropenia disappeared, but there was a significant association of immunosuppressive therapy (including steroids) with lower survival. Other factors significantly associated with lower survival were the APACHE score, infection by C. tropicalis and age, while treatment with an echinocandin [Odds Ratio (OR) 0.65, 95%CI: 0.45–0.94; P=0.02] and catheter removal were both significantly associated with better survival (OR 0.50, 95%CI: 0.35–0.72; P=0.0001).
Based on the patient-level quantitative analysis by Andes et al., echinocandins must be considered as first-line choice for invasive Candida infections before species identification (Table 4).22 The strength of recommendation is the same (A) for anidulafungin, caspofungin and micafungin and is also the same for the overall and the hematologic population. However, the quality of evidence is lower for hematologic patients (II) compared to the overall population, as the number of neutropenic patients recruited in the clinical trials was low. A recent communication on a patient-level pooled analysis of one randomized clinical trial and 4 open label studies focusing on anidulafungin in 46 neutropenic patients with candidemia showed comparable response and survival rates to those observed with caspofungin and micafungin in other studies.25 Therefore, the grading is now similar (A II) for all three echinocandins for the treatment of invasive candidiasis in hematologic patients.
Liposomal amphotericin B has also been graded A I for the overall population and A II for hematologic patients due to similar efficacy in comparison to micafungin.15,21 However, its safety profile is less favorable and therefore liposomal amphotericin B should be considered as an alternative in case of contraindication to echinocandins. Fluconazole and voriconazole are potential alternatives for first-line treatment in the overall population provided there is no previous exposure to azoles and the infection is not severe (fluconazole).
After species identification, susceptibility testing should guide the treatment. In general, echinocandins remain the drug of choice, except for C. parapsilosis where fluconazole is more appropriate (Table 5). However, a recent observational study reported no difference in 30-day mortality and persistent candidemia at 72 hours of an echinocandin-based regimen compared to an azole-based therapy for patients with C. parapsilosis candidemia.26 Therefore, the continuing use of echinocandins might be considered in patients with a clinical and microbiological response. When Candida species is azole-susceptible, step-down to fluconazole can be considered in stable patients after five days of intravenous (iv) therapy.24 In patients with Candida krusei infection, switch to oral voriconazole is an option.
Although the role of catheter removal in the management of candidemia has long been controversial, most recent studies suggest a beneficial effect on outcome. 6–8,10,11,15,16,20,26–33 Garnacho-Montero et al. showed in a large number of candidemia that early adequate therapy and removal of central venous line were independently associated with lower mortality.34 The patient-level quantitative analysis by Andes et al. also demonstrated in a multivariate analysis that removal of catheter was associated with a decreased mortality (OR 0.50; 95%CI: 0.35–0.72; P=0.0001).22 The recommendation is, therefore, to rapidly remove the catheter in the overall population (grade A II) as well as in hematologic patients (grade B II) irrespective of the Candida species. If central venous catheter cannot be removed, treatment should include an echinocandin or a lipid formulation of amphotericin B due to their better activity on Candida biofilms.35–37
Invasive Aspergillus infections
Nine prospective trials (only 4 being randomized comparative trials) had been published before the ECIL-4 and provided the basis of the previous guidelines for first-line therapy in invasive aspergillosis (Table 6).38–46 An additional paper reported a post-hoc analysis of the trial comparing standard dose of liposomal amphotericin B to high-dose liposomal amphotericin B.47 This post-hoc analysis comparing outcome in possible versus mycologically documented aspergillosis underscored the limited number of mycologically documented infections but did not lead to any change in the grading for liposomal amphotericin B. A second post-hoc analysis was performed on the voriconazole versus amphotericin B deoxycholate trial.48 Integration of the results of baseline galactomannan detection tests performed after primary analysis and re-categorization according to the 2008 EORTC/MSG definition criteria allowed more mycologically documented cases of invasive aspergillosis to be identified.49 Conclusions of this post-hoc analysis were similar to those of the primary analysis and therefore its results did not affect the grading for voriconazole and for amphotericin B deoxycholate.
At the time of the ECIL-5, results from the comparative study of voriconazole plus anidulafungin versus voriconazole plus placebo were only available in abstract form. The results have been discussed with a provisional grading that could be transformed in a definite grading, as no additional data available in the full paper suggested a need for change in provisional recommendations.50 This study failed to reach the primary endpoint of decreased all-cause mortality at week 6 (difference of −8.2% in favor of combination; P=0.087). However, in a subgroup of patients with an invasive aspergillosis documented by positive galactomannan in either serum or bronchoalveolar lavage, 6-week all-cause mortality was lower in patients receiving combination therapy (difference of −11.6% in favor of combination; P=0.037). A large majority of the ECIL members felt that this subgroup analysis, that had not been originally planned, was not sufficient to give a stronger recommendation although this subgroup included 80% of the modified intent-to-treat population. Therefore, the combination of voriconazole plus anidulafungin was graded C I for primary therapy of invasive aspergillosis while all other combinations were graded C III in the absence of well-designed studies for first-line therapy.
Table 6 summarizes the main characteristics and results of the various studies. Importantly, very few studies had a large number of patients with a mycological documentation.40,41,50 As shown by the 2 post-hoc analyses, survival was substantially lower in mycologically-documented infections compared to possible cases.47,48 Therefore, studies with a limited number of documented cases cannot lead to the strongest recommendations. As no study specifically addressed management of breakthrough aspergillosis after failure of posaconazole or voriconazole prophylaxis, no recommendation could be made on this issue.
The clinical trial comparing the new triazole isavuconazole versus voriconazole for primary therapy of invasive aspergillosis could not be discussed during the ECIL-5 as results were only presented as an abstract in 2014. However, the group could review the data from these abstracts during the ECIL-6 meeting. Isavuconazole appears to be as effective as voriconazole for the treatment of invasive aspergillosis and has a better safety profile. Therefore, a grade A I similar to the grading for voriconazole has been given to isavuconazole (Table 7). As the full paper was published shortly after the meeting, and confirms the results, the provisional grading attributed during the meeting has been transformed into a definite grading in this manuscript.51
Currently, amphotericin B deoxycholate is considered to have no role in the treatment of invasive aspergillosis when more effective and less toxic agents are available. Its limited efficacy and its poor safety profile led to a recommendation against its use. No substantial change has been made for second-line therapy in the absence of new data (Table 8).
Diagnostic and therapeutic strategies were discussed during the ECIL-5 and the ECIL-6. Rhizopus, Mucor, Lichtheimia (previously classified as Absidia), Cunninghamella, Rhizomucor, Apophysomyces, and Saksenaea are the genera most frequently involved in human disease.52 Cunninghamella species is more virulent in experimental models and may be associated with a higher mortality rate in patients.53 So far, there has not been enough evidence that identification of mucormycosis to the genus and/or species level helps guide antifungal treatment.54,55 Species identification remains, nevertheless, important for outbreak investigations.56 However, the differentiation between mucormycosis and other invasive mold infection is of critical importance as it has major therapeutic implications.
While epidemiological aspects and some clinical (sinus disease, concomitant diabetes, occurrence under voriconazole therapy) and radiological (reverse halo sign on chest CT-scan) factors may help to suspect mucormycosis, the diagnosis remains difficult and biopsy of the lesion is often required. Identification of the pathogen most often comes from microscopic, culture and/or histopathological examination of relevant samples. New diagnostic approaches include molecular testing on serum and various other clinical samples including formalin-fixed tissues, MALDI-TOF and Mucorales-specific T-cell detection.57–64 Although these new approaches are very promising for an earlier diagnosis, no grading for their use can be given yet due to the lack of data.
Amphotericin B, posaconazole and isavuconazole are the most potent agents in vitro.65–67 Currently, no validated minimum inhibitory concentration breakpoints for any of the drugs are available and thus determination of susceptibility categories is not possible for the agents of mucormycosis. The ECIL-3 recommendations for the treatment of mucormycosis were mostly based on retrospective studies, registry data and small prospective non-controlled studies.3,68–77 Few new data are available for the treatment of mucormycosis since the ECIL 4 and, therefore, the current recommendations are very similar (Table 9).
A prospective non-comparative trial assessed the efficacy and safety of first-line therapy with high-dose liposomal amphotericin B given at 10 mg/kg/day combined with surgery when appropriate.78 This trial demonstrated efficacy of high-dose liposomal amphotericin B plus surgery in mucormycosis with a survival rate of 62% at week 12. The only factor associated with mortality was the presence of hematologic malignancy or cancer (HR: 3.15; 95%CI: 1.12–8.91; P=0.02). Renal impairment of any degree was observed in 40% of the patients but was transient in most of them. These results confirm the beneficial role of liposomal amphotericin B but do not yet allow any recommendation for the administration of such a high dose of 10 mg/kg/day.
A short paper presented data from a retrospective analysis of a combination of posaconazole and a lipid formulation of amphotericin B.79 Thirty-two patients received this combination of posaconazole with liposomal amphotericin B (n=27) or amphotericin B lipid complex (n=5). Only 3 of them were treated with this combination in first line. Overall response rate was 56% but a large proportion of patients (59%) died before day 90. The low number of patients, the retrospective nature of the study, and the high mortality rate at day 90 only allowed for a B III recommendation for this combination for salvage therapy of mucormycosis (Table 10).
Discussion and conclusions
An update of the ECIL antifungal treatment recommendations was needed as there were important new data, and also because of necessary changes in the ECIL grading system so as to be in harmony with other ECIL recommendations. The most important data for invasive candidiasis came from a large review of patients included in 7 major trials.22 The multivariate analysis now allows a very strong recommendation in favor of an echinocandin for the first-line therapy of candidemia irrespective of the underlying predisposing factors. The controversy on the beneficial role of catheter removal can now be considered to be resolved. The most interesting new data were the publication of a first-line combination study in invasive aspergillosis and the results of a randomized comparative trial comparing isavuconazole to voriconazole. Aspergillus guidelines now include the results of these 2 clinical trials and should help clinicians in their treatment decision making. Since few new data have been published since the last ECIL guidelines, no major changes were made to mucormycosis management. Importantly, the posology and indication of antifungal agents reported in the current guidelines do not necessarily reflect those licensed by the European Medicines Agency (EMA), but are the result of a consensus-based analysis of available literature within the ECIL group.
There has been controversy about some discrepancies between the ECIL-5 and the ESCMID recommendations for invasive aspergillosis in hematologic patients. These differences were identified during a joint meeting and an ESCMID representative was invited to discuss them at the ECIL-6 meeting. Most differences were minor and mostly reflected a difference in grading system. The ECIL Aspergillus recommendations are restricted to hematologic patients who represent more than 90% of the patients included in the major clinical trials.41,43,50,51 No subgroup of hematologic patients deserving specific recommendation for Aspergillus infection treatment has been identified by the ECIL group. In contrast, the ESCMID group had a broader approach considering all other conditions predisposing to invasive aspergillosis, grading the diagnostic procedures, and including environmental measures in the prevention, also providing a grade for specific infection sites. In addition, the ESCMID group also segregated the hematologic patients into subgroups and provided specific grading for each of them, with usually weaker recommendations when there was not a sufficient number of patients with these specific underlying conditions included in the clinical studies. Finally, and importantly, some data were not available at the time of the ECIL-5 meeting but were in the public domain when the ESCMID group met. In September 2015, the ECIL-6 group was able to incorporate the new data, and this has helped to reduce the apparent differences with the ESCMID guidelines. Therefore, neither the ECIL group nor the ESCMID group felt any change other than this update was required.
The authors and contributors thank the group GL-Events, Lyon, France, for the organization of the meeting. They also thank Valérie Rizzi-Puechal (Pfizer), France; Markus Rupp (MSD), Germany; Sonia Sanchez (Gilead Sciences), UK; Anne-Therese Witschi (Basilea), Switzerland; Lorraine Tweddle (Astellas), UK.
Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/102/3/433
Collaborators: Agrawal Samir, United Kingdom; Akova Murat, Turkey; Alanio Alexandre, France; Aljurf Mahmoud, Saudi Arabia; Averbukh Dina, Israel; Barnes Rosemary, United Kingdom; Blennow Ola, Sweden; Bochud Pierre Yves, Switzerland; Bouza Emilio, Spain; Bretagne Stephane, France; Bruggemann Roger, Netherlands; Calandra Thierry, Switzerland; Carratalà Jordi, Spain; Castagnola Elio, Italy; Cesaro Simone, Italy; Cordonnier Catherine, France; Cornely Oliver, Germany; Dalianis Tina, Sweden; De La Camara Rafael, Spain; Donnelly Peter Joseph, Netherlands; Drgona Lubos, Slovakia; Duarte Rafael, Spain; Einsele Hermann, Germany; Engelhard Dan, Israel; Ferreira Isabelina, Portugal; Fox Christopher, United Kingdom; Girmenia Corrado, Italy; Groll Andreas, Germany; Hauser Philippe, Switzerland; Heldal Dag, Norway; Helweg-Larsen Jannik, Denmark; Herbrecht Raoul, France; Hirsch Hans, Switzerland; Hubacek Petr, Czech Republic; Johnson Elizabeth M, United Kingdom; Kibbler Christopher, United Kingdom; Klyasova Galina, Russian Federation; Koskenvuo Minna, Finland; Kouba Michal, Czech Republic; Kullberg Bart-Jan, Netherlands; Lagrou Katrien, Belgium; Lewis Russel, Italy; Ljungman Per, Sweden; Maertens Johan, Belgium; Mallet Vincent, France; Marchetti Oscar, Switzerland; Martino Rodrigo, Spain; Maschmeyer Georg, Germany; Matos Olga, Portugal; Melchers Willem, Netherlands; Mikulska Malgorzata, Italy; Munoz Patricia, Spain; Nucci Marcio, Brazil; Orasch Christina, Switzerland; Padoin Christophe, France; Pagano Livio, Italy; Pagliuca Antonio, United Kingdom; Penack Olaf, Germany; Petrikkos Georgios, Greece; Racil Zdenek, Czech Republic; Ribaud Patricia, France; Rinaldo Christine H, Norway; Robin Christine, France; Roilides Emmanuel, Greece; Rovira Montserrat, Spain; Schellongowski Peter, Austria; Sedlacek Petr, Czech Republic; Sinko Janos, Hungary; Skiada Anna, Greece; Slavin Monica, Australia; Styczynski Jan, Poland; Tissot Frédéric, Switzerland; Van Boemmel Florian, Germany; Viscoli Claudio, Italy; Von Lilienfeld-Toal Marie, Germany; Ward Katherine, United Kingdom.
- Received August 5, 2016.
- Accepted December 20, 2016.
- Copyright© Ferrata Storti Foundation
Material published in Haematologica is covered by copyright. All rights are reserved to the Ferrata Storti Foundation. Use of published material is allowed under the following terms and conditions:
https://creativecommons.org/licenses/by-nc/4.0/legalcode. Copies of published material are allowed for personal or internal use. Sharing published material for non-commercial purposes is subject to the following conditions:
https://creativecommons.org/licenses/by-nc/4.0/legalcode, sect. 3. Reproducing and sharing published material for commercial purposes is not allowed without permission in writing from the publisher.