Micafungin for Candida infections in Slovenia and Romania: A multicenter, observational, prospective study

Introduction: An echinocandin, such as micafungin, is recommended as first-line treatment for invasive Candida infections in immunocompromised patients. This multicenter, observational, prospective, noninterventional study evaluated the real-world use of micafungin in clinical practice in Slovenia and Romania, as this remains unexplored. Methodology: The primary endpoint was evaluation of micafungin use, including rationale for prescription, treatment duration, and daily dose. Secondary endpoints included recordings of patient baseline characteristics and evaluations of efficacy and safety. Across 11 centers in two countries, 118 patients (18 children [< 16 years] and 100 adults [≥ 16 years]) received micafungin for the first time according to their clinic’s standard practice. Results: Micafungin was prescribed for treatment in 57.6% of patients and for prophylaxis in 40.7% of patients. The median (range) treatment duration was 9.0 (0 – 54) days and 13.0 (2 – 6)] days, respectively. The median dose of micafungin was higher than recommended for children receiving prophylaxis or treatment for invasive candidiasis and for adults receiving prophylaxis. Fever was the most commonly observed clinical sign at baseline (16 children [88.9%] and 31 adults [31%]) and hematologic malignancy was the most frequent primary diagnosis at admission (11 children [61.1%] and 40 adults [40%]). Candida species were the most commonly identified causal agents of invasive fungal infections (2 children [11.1%] and 48 adults [48%]). Conclusions: The efficacy and safety profiles of micafungin use in Slovenia and Romania based on clinician’s own experiences in local clinical practice were consistent with those reported in other real-world studies.


Introduction
In recent years, the incidence of invasive fungal infections (IFIs) has increased worldwide. Immunocompromised individuals are at an increased risk of IFIs, including those undergoing autologous and allogeneic hematopoietic stem cell transplantation (HSCT), solid organ transplantation, chemotherapy, and major surgical procedures [1]. Candida species are the most common cause of IFIs, with Candida albicans being the most dominant strain globally (44 -70% of cases) [2,3], although regional variations exist [2][3][4][5][6][7]. Invasive candidiasis (IC) remains a leading cause of opportunistic fungal infection in hospitals and is associated with substantial morbidity and mortality [2,8], with an estimated ~700,000 cases occurring globally each year [9]. In 2016, there were an estimated 984 cases of candidemia in Romania [10]. Additionally, epidemiological data from Slovenia showed that the incidence of candidemia increased from 4.23 to 7.02/100,000 inhabitants between 2013 and 2017, with C. albicans (51.7%) and Candida glabrata (27.7%) identified as the most prevalent species among isolates [11].
Echinocandins, such as micafungin, induce fungal cell lysis by inhibiting the synthesis of 1,3-β-glucan synthase [12]. Guidelines recommend an echinocandin as first-line treatment for IC and esophageal candidiasis, and for prophylaxis in patients at high-risk for Candida infection, including those who are immunocompromised and undergoing HSCT [13][14][15][16]. Micafungin has broad-spectrum anti-fungal activity against Candida species and some Aspergillus species [17], and is approved for the treatment of adults and children (< 16 years) with IC, for the treatment of esophageal candidiasis in adults, and for prophylaxis in patients undergoing allogeneic HSCT and those expected to have neutropenia (absolute neutrophil count < 500 cells/µL) for ≥ 10 days [18] when other antifungals are not appropriate. Micafungin has demonstrated non-inferiority to caspofungin, amphotericin B, fluconazole, and itraconazole in the treatment of IFIs [19,20], and is the only echinocandin approved for antifungal prophylaxis; in clinical trials, micafungin has shown superiority to fluconazole and non-inferiority to itraconazole for fungal prophylaxis [20][21][22].
Following the launch of micafungin for clinical use in Slovenia and Romania in October 2012, this study aimed to collect and evaluate data on its use in daily hospital practice, assessed separately for prophylactic and treatment use, and focused on the rationale for prescription, and dose and treatment duration in both children and adults.

Study design
M-TREAT (Micafungin® in Routine Clinical Practice for the Treatment of IC, esophageal candidiasis or Prophylaxis of Candida Infections: A Multicenter, Observational, Prospective, Non-interventional Study; ISN/protocol number: SEE-MYC-01) was performed to evaluate the use of micafungin in routine clinical practice in Slovenia and Romania. The study included an initiation visit (Visit one), during which micafungin was prescribed, and a follow-up visit at the end of micafungin treatment or treatment withdrawal (Visit two). The therapeutic approach was not decided in advance or influenced by the protocol; each patient was treated and followed up at the investigator's discretion and according to their clinic's standard practice. Treatment decisions were made based on the patient's clinical condition and the identification of causative fungal species.

Ethics
Before the start of the study, written approval was obtained from the Independent National Ethics Committee (IEC). The study was conducted in adherence to the ethical principles of the Declaration of Helsinki, International Conference of Harmonisation Good Clinical Practice Guidelines, the European Union Clinical Trials Directive 2001/20/EC, and applicable local laws and regulations. Written informed consent was obtained from each patient or their legal representative before study enrolment. The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received.

Patients
Between December 2013 and March 2017, patients were prospectively assessed across 11 centres in Slovenia (n = 1) and Romania (n = 10). Patients prescribed micafungin for the first time in accordance with the label were included in the study. There were no exclusion criteria.

Endpoints and assessments
The primary endpoints were reasons for micafungin prescription (treatment or prophylaxis) in the overall population; duration of micafungin treatment (defined as the time from treatment initiation to discontinuation or switching to another antifungal treatment); and micafungin daily dose (measured at Visits one and two). The secondary endpoints included efficacy assessments of micafungin when used for treatment and prophylaxis (changes in clinical signs and symptoms, and microbiological and radiographic tests at Visits one and two); safety (assessed via changes in laboratory parameters at Visits one and two, and by recording adverse events [AEs] and severity [at Visit one and two]); identification of the possible causative agents of fungal infections at Visit one; prior antifungal treatments and concomitant medication use at Visit one; and treatment switching after micafungin discontinuation at Visit two. Visits one and two occurred while the patient was hospitalized. DBP: diastolic blood pressure; mmHg: millimeter of mercury; SBP: systolic blood pressure; SD: standard deviation. † Data was missing for the following patients: fever (adults, n = 1); cough (adults, n = 14); pneumonia (adults, n = 14); pleuritic pain (children, n = 1; adults, n = 17).
IFIs were classified as "proven", "probable" or "possible" based on criteria outlined in De Pauw et al. [32]. AEs were recorded from the moment the patient signed the informed consent form until the end of the study. An AE was defined as any untoward medical occurrence that did not necessarily have a causal relationship with the treatment. A treatment emergent AE (TEAE) was defined as an AE observed after starting administration of the study drug. A serious adverse event (SAE) was defined as an AE or adverse drug reaction (ADR) that resulted in death; was lifethreatening; resulted in persistent or significant disability or the ability to conduct normal life functions; resulted in birth defects; required hospitalization/led to prolongation of hospitalization; or other medically important events. ADRs were those listed in the summary of product characteristics for micafungin [18].

Diagnostic tests
Investigators recorded whether culture-based microbiological diagnostic tests, non-culture microbiological diagnostic tests (fungal biomarkers and metabolites), and radiographic diagnostic tests were performed at Visits one and two, together with their results. Microbiological culture tests were performed using samples obtained from both sterile and non-sterile sites, and included blood culture (venipuncture, central venous catheter), cerebrospinal fluid, peritoneal fluid, bronchoalveolar lavage, tracheal aspirate, and other (specified by the investigator according to practice at each participating center). Matrix assisted laser desorption ionisation time of flight (Bruker Daltonics, Coventry, UK) and antifungigrams were used to identify causative fungal agents isolated from these sites. Non-culture tests for fungal biomarkers and metabolites included galactomannan test, 1,3-β-Dglucan test, and other (specified by the investigator). Radiographic tests included chest x-ray, thoracic computed tomography (CT), cerebral CT, ultrasonography, and other (specified by the investigator). All tests were performed according to standard practice at each participating center.

Statistical analyses
Data were analyzed using SPSS version 23.0. Mean (standard deviation [SD]) is reported for continuous variables where the distribution is approximately normal. Median (range) is reported for continuous variables where the distribution is highly skewed (Shapiro-Wilk test). Frequencies and percentages were recorded for categorical variables. No imputation method was applied to account for missing data. The safety analysis set (SAF) included all patients who were prescribed micafungin and was used for assessment of all safety-related variables. The full analysis set (FAS) included all patients who received any medication during the study and was used for all other analyses. The relationship between micafungin treatment duration and change in aspartate aminotransferase (AST) and alanine aminotransferase (ALT) values from baseline to end of micafungin treatment was evaluated by Spearman correlation coefficient. Spearman correlation coefficient between treatment duration and AST/ALT change was calculated per each daily dose group (50 mg; ≥ 100 mg). Statistical significance was considered when p < 0.05.

Patient demographics and baseline characteristics
Patient characteristics at baseline are presented in  Table 2).

Reasons for micafungin prescription
Reasons for micafungin prescription were available for 116 patients ( HSCT: hematopoietic stem cell transplant; NA: not applicable. † NA: missing data, poor investigator compliance. ‡ Probable invasive fungal infection was defined according to de Pauw et al. [32]. § For two patients in Romania, no treatment indication was recorded; these patients were prophylactic cases.

Micafungin dose and treatment duration
The median micafungin dose and treatment duration are presented in Table 3. Patients received micafungin for a median (range) duration of 9 (0 -54) days for treatment and 13 (2 -69) days for prophylaxis. In Slovenia, the median (range) duration of treatment

Causative agents of fungal infections
Candida spp. were recorded in two (11.1%) children and 48 (48%) adult patients in the overall population (Table 4). Candida albicans was the most frequently isolated Candida spp.; this was the causative agent in all children who had Candida spp. recorded (two children), and in 27 adults. Other species were identified in two (11.1%) children (both Saccharomyces cerevisiae) and two (2%) adults (one case of S. cerevisiae and one case of Cryptococcus neoformans). Missing data (due to poor investigator compliance in recording data or failure to perform a diagnostic test) or no causative fungal agent (due to a negative test result) was recorded for 14 (77.8%) children and 50 (50%) adults. Ten (55.6%) children and 11 (11%) adults had possible IFI, two (11.1%) children and 22 (22%) adults had probable IFI, and no (0%) children and 29 (29%) adults had proven IFI.

Antifungal treatment switching
There were 17 switches to another antifungal agent in 13 of the 68 patients who received micafungin for treatment, and a total of 12 switches in 11 of the 48 patients who received micafungin for prophylaxis (Table 5). Reasons for treatment switching included identification of the causative fungal species, prolonged febrile neutropenia, and deteriorating condition of the patient. Patients were switched to fluconazole as de-  (38) † Some adults are included more than once as they had more than one type of infection; ‡ Other specified causative agents were Saccharomyces cerevisiae (n = 2) in children and Cryptococcus neoformans (n = 1) and Saccharomyces cerevisiae (n = 1) in adults; § Missing data (due to poor investigator compliance in recording data or no diagnostic test performed) or no causative fungal agent (due to negative test result) was recorded for 14 (77.8%) children and 50 (50%) adults; * Probable invasive fungal infection was defined according to de Pauw et al. [32].

Efficacy
Changes in diagnoses or clinical signs and symptoms of an IFI (fever, cough, pneumonia, and pleuritic pain) in patients receiving micafungin are shown in Table 6 Table 3). Ten of 38 patients had a positive culture test at Visit one, of whom five had a negative culture test at Visit two. Two patients had no culture tests recorded at Visit two. One patient with a negative culture test at Visit one had a positive culture test at Visit two.
Positive culture tests at Visit one included bronchoalveolar lavage (one positive test), tracheal aspirate (two positive tests) and other (four positive tests), including stool culture and rectal swab. At Visit two, positive culture tests were obtained from sputum culture + throat culture (one positive test) and throat swab + rectal swab (one positive test).
Among patients prescribed micafungin for treatment reasons, 58 of 68 had microbiological culture tests at Visit one, and 41 had culture tests at Visit two.

Safety
Overall, 22 TEAEs were reported in 19 (16.1%) patients. One TEAE (fever) was assessed by the investigator as possibly drug related. Three TEAEs were mild, seven moderate, and 12 severe. There were 21 serious TEAEs reported in 19 patients; 13 of these were considered unrelated to micafungin; causality was not recorded in eight cases (Table 7). Eleven patients died during the study; 10 deaths were assessed as not related to micafungin. One event was assessed as unassessable (cause of death unknown). Changes in laboratory parameters and vital signs are presented in Supplementary Table 4. There were no unexpected safety issues relating to liver or kidney function, as assessed by changes in laboratory parameters. For adult patients receiving a daily dose of 50 mg or ≥ 100 mg micafungin who had AST/ALT data available (n = 9 and 61 patients, respectively), there were no statistically significant correlations between micafungin treatment duration and change in AST (50 mg: r = 0.243, p = 0.529; ≥ 100 mg: r = 0.123, p = 0.348) or ALT levels (50 mg: r = 0.226, p = 0.559; ≥ 100 mg: r = −0.043, p = 0.740). For children, there were no statistically significant correlations between micafungin treatment duration and change in AST (r = −0.08, p = 0.845) or ALT levels (r = 0.375, p = 0.321), or between micafungin daily dose and change in AST (r = 0.311, p = 0.415) or ALT levels (r = 0.05, p = 0.897) from baseline to the end of the study (n = 9 children).

Discussion
The aim of this non-interventional, multicenter, prospective study was to collect and evaluate data on the use of micafungin in daily clinical practice in Slovenia and Romania, focusing on the reasons for prescription, dose, and treatment duration. Overall, the duration of treatment and dose of micafungin in this study deviated from the instructions supplied in the labelling from the pharmaceutical laboratory, and in some cases, the recommended treatment guidelines [13][14][15][16]18]. These findings add to the body of evidence supporting the efficacy and safety of micafungin when used in real-world settings for prophylactic and treatment purposes against Candida infections in hospitalized patients. In this study, fewer patients were enrolled in Romania than in Slovenia (39 and 79 patients, respectively) and considerably fewer patients received prophylaxis in Romania compared with Slovenia (nine and 41 patients, respectively). These differences in recruitment and treatment patterns may be reflective of differences in the real-world use of micafungin between Slovenia and Romania. The data reported here are consistent with data from previous real-world studies. For example, in the present study, micafungin was most commonly used for the treatment of Candida infections (57.6%), whilst in an observational study in France, antifungal prophylaxis with an azole, amphotericin B, or an echinocandin was recorded as the leading antifungal strategy, accounting for 76% of prescriptions in French hematological units, regardless of underlying disease [33]. However, in the observational study in France, antifungal treatment was only prescribed by hematologists, while in Slovenia, hematologists as well as other specialists, such as ICU doctors, infectologists and pediatricians, who more frequently encounter IC, were involved. Consequently, the population of patients included in M-TREAT from Slovenia and Romania was more consistent with those in other observational studies of micafungin use [17,34].
Overall, the duration of micafungin treatment was shorter than the recommended minimum duration of 14 days in both countries [18] and was shorter in Romania than in Slovenia (5 -8 vs. 11 -13 days, respectively). This may be explained by the high number of patients who received prior antifungal treatment (65 patients overall), the high number of patients who switched to another antifungal treatment after receiving micafungin (24 patients overall) and the clinical condition of the patients in addition to real-world differences in clinical practice between these two countries. Whilst guidelines recommend antifungal treatment for ≥14 days after the end of candidemia in non-neutropenic patients, this can be simplified by stepping down from an echinocandin to oral fluconazole after 5 -10 days if the patient is stable and isolated strains are susceptible to fluconazole [13][14][15][16]. It is possible that the shorter duration of treatment in both countries is representative of realworld clinical practice; however, other factors, such as the financial burden of treatment on the healthcare systems, cannot be discounted. Studies have reported variation in the duration of micafungin treatment in real-world settings; in a retrospective, observational study of Chinese patients treated for IFIs in a hospital setting, micafungin was administered for a mean treatment duration of 10.2 days [17]. In a real-world study of prophylactic micafungin use in patients with hematological malignancies, the duration of micafungin treatment was consistent with the label from the pharmaceutical laboratory [35], whilst a longer treatment duration has been observed in some studies (25 -36 days) [36,37].
In the present study, children receiving prophylaxis and treatment for IC, and adults receiving prophylaxis, were administered micafungin at doses higher than recommended by the label from the pharmaceutical laboratory [18]. More than half of patients with a recorded indication of esophageal candidiasis received a lower daily dose of micafungin than recommended in the product information and treatment guidelines [15,18]. Prophylaxis with 100 mg day −1 for some patients in Slovenia was selected based on data from a previous study in which this dose was well-tolerated [38]. In addition, guidelines from the Infectious Diseases Society of America recommend 100 mg day −1 for prophylaxis of IC in the ICU setting; however, this is a weak recommendation with low-quality evidence [15].
Another reason for the high daily doses in prophylactic patients could be that a significant proportion of patients were recorded as having immune deficiency (57 patients, including 15 children and 42 adults) and had a diagnosis of hematologic malignancy (48 patients, including 9 children and 39 adults). In case of breakthrough fever, a micafungin dose of 50 mg in severely neutropenic hematological patients may not be a consequence of non-effectiveness, but the result of micafungin underdosing. A higher dose of micafungin might lead to the resolution of fever, re-establishing the efficacy of micafungin. In previous studies, micafungin has been administered for prophylaxis at 50 mg day −1 , which is consistent with the approved use but lower than that administered in the present study [37,[39][40][41]. Higher doses of 100 mg day −1 and 150 mg day −1 have been used in other studies, including for prophylaxis in neutropenic patients and high-risk liver transplant recipients [42,43]. For treatment, doses administered to adult patients in other hospital-based studies are consistent with the approved and recommended doses and are comparable to those administered here [17,41]. The deviation in treatment duration and dose from the label and associated treatment guidelines in this study reflect the real-world use of micafungin in Slovenia and Romania and may be a consequence of socio-economic factors and clinicians' experience of the treatment of invasive fungal infections in these two countries. Furthermore, in Slovenia, different clinical wards have differing prescription policies, which may be influenced by patient age and clinical condition and could also influence treatment decisions.
In this study, C. albicans was the most frequently isolated fungal species, consistent with the findings of other observational studies (57 -60% of patients) [6,34], a study conducted using data from a North American registry (46% patients) [5] and a multicenter, retrospective study where C. albicans accounted for 37% of blood isolates [44]. However, no causative agent was identified in a considerable number of patients due to poor investigator compliance or failure to perform a diagnostic test, which reflects real-life clinical practice in the participating hospital centers.
IC is associated with high morbidity and mortality, with mortality rates of around 40% reported [2,8]. However, studies have shown that early anti-fungal treatment can improve outcomes, meaning that the prompt and accurate diagnosis of IFI is crucial [45][46][47]. Conventional diagnostic procedures such as microscopic examinations, culture and identification of micro-organisms are essential investigations; however, their performance is dependent on the possibility of obtaining samples from deep tissues, whether the samples collected are from sterile sites, the expertise of laboratory personnel and the availability of the diagnostic method in the hospital center. Furthermore, owing to the low sensitivity of blood culture, a negative result does not exclude IFIs. To overcome this, nonculture-based diagnostic tests have been developed for the diagnosis of IC that detect fungal biomarkers and metabolites [48].
The results of M-TREAT support the wellestablished safety and efficacy profile of micafungin observed in other real-world studies [17,40,41]; there was an overall reduction in the number of patients with signs and symptoms of an IFI at the end of the study and a considerable number of patients with a positive microbiological or radiographic test at visit one had a negative result by visit two. There were no grade 3/4 effects of micafungin on renal or hepatic function in this study, which is in line with previous observations [43,49,50]. The results showing a lack of correlation between micafungin daily dose, treatment duration and changes in AST/ALT levels should be interpreted with caution due to the low number of patients assessed. The label from the pharmaceutical laboratory indicates that micafungin should be used on a careful risk/benefit basis in patients with impaired liver function due to adverse hepatic events observed in rats [18]; because it is not metabolized by the liver, anidulafungin is often favoured in patients with hepatic impairment, and is the only echinocandin indicated for patients with severe liver impairment [12,51,52]. However, the results of this study suggest that micafungin is well-tolerated in patients with hepatic impairment and may be suitable as an alternative to anidulafungin. The results from the present real-life study support the favourable safety profile of micafungin observed in previous studies [17,[35][36][37][38][39][40][41][42][43].
The main strength of this observational study was that it provided the opportunity to study a wider patient group than possible in randomized-controlled trials due to a lack of strict selection criteria, i.e., all patients, with and without comorbidities, treated with micafungin for the first time were included according to local clinical practice. In addition, it allowed responses to treatment to be evaluated in a real-world setting, representative of everyday clinical practice. One limitation of this study was the risk of selection bias; no measures to reduce bias were taken as this reflected real-world situations.
The study was open-label and therefore there was no removal of bias through blinding. In addition, missing observations might have distorted study results. In this study, test samples were selected based on the availability/accessibility of patients and at the discretion of the investigator. The results were also influenced by the heterogeneity of the healthcare systems and treatment patterns in Slovenia and Romania, as well as the heterogeneity of the studied population. No standard protocol relating to treatment decisions could be implemented across the study centers. Finally, the number of patients enrolled in Romania was lower than the number enrolled in Slovenia and fewer patients received prophylaxis in Romania compared with Slovenia, providing an unbalanced dataset. Nevertheless, these data reflect the real-world use of micafungin in these countries. Additionally, as this was an observational study performed to document micafungin use in a real-world setting, there was no formal assessment of the clinical response to treatment, other than the presence of symptoms, and the results of culture and radiological tests.

Conclusions
In conclusion, this study, in which treatment decisions were based on clinicians' own experience in local clinical practice, provides information about the use of micafungin in daily clinical practice in Slovenia and Romania for the first time. The safety and tolerability of micafungin in this study were consistent with previous real-world clinical studies.  Row percentages are shown for treatment and prophylaxis separately. All negative -all possible tests were performed, and the results were all negative. All performed are negative -not all possible tests were performed, but those that were performed were negative. Microbiological culture tests recorded at start of micafungin treatment: hemoculture (venopunction), hemoculture (blood from CVC), cerebrospinal fluid, peritoneal fluid, bronchoalveolar lavage, tracheal aspirate, coproculture, rectal swab, throat swab, urocultures, CVC tip swab, wound secretion + pleural fluid culture. Microbiological culture tests recorded at the end of micafungin treatment: hemoculture (venopunction), hemoculture (blood from CVC), cerebrospinal fluid, peritoneal fluid, bronchoalveolar lavage, tracheal aspirate, coproculture, ascites culture, CVC tip swab, wound swab, pharyngeal swab, anal swab, axillary swab, sputum culture, throat swab, rectal swab. Microbiological non-culture tests recorded at the start and end of micafungin treatment: galactomannan test, 1,3-β-D-glucan test, mannan, anti-mannan antibodies. Radiographic tests recorded at the start and end of micafungin treatment: chest x-ray, thoracic CT, cerebral CT, ultrasonography (CT: computerized tomography; CVC: central venous catheter).