Voriconazole in the Treatment of Invasive Mold ... - Springer Link

1 downloads 0 Views 136KB Size Report
Voriconazole in the Treatment of Invasive Mold Infections in Transplant Recipients. Published online: 24 June 2003. Springer-Verlag 2003. Abstract Mortality ...
Eur J Clin Microbiol Infect Dis (2003) 22:408–413 DOI 10.1007/s10096-003-0960-0

ARTICLE

J. Fortffln · P. Martn-Dvila · M. A. Snchez · V. Pintado · M. E. Alvarez · A. Snchez-Sousa · S. Moreno

Voriconazole in the Treatment of Invasive Mold Infections in Transplant Recipients Published online: 24 June 2003  Springer-Verlag 2003

Abstract Mortality due to invasive mold infections in solid organ transplant recipients is very high despite therapy with amphotericin B, including lipid formulations. Voriconazole is a triazole with a good activity against molds, including Aspergillus spp. and Scedosporium spp. Experience with voriconazole is limited, but preliminary results in patients with these infections are promising. Reported here is the experience with voriconazole administered on a compassionate-use basis to five patients with invasive mold infections: four solid organ recipients and one patient with an autoimmune disorder. Four patients had invasive Aspergillus fumigatus infection (3 lung infections, 1 abdominal infection) and one had invasive ocular Scedosporium apiospermum infection. The MIC of voriconazole was 1 g/ml for all isolates (NCCLS performance standards for microdilution assay, proposed standard M38-P). Voriconazole was administered as primary therapy in a patient with Scedosporium infection and, in patients with Aspergillus infections, after persistence of positive culture despite a cumulative dose of 3 g of a lipid formulation of amphotericin B. Voriconazole was administered for a median time of 80 days (range, 60–90 days). No visual disturbances were observed, but one patient presented a moderate increase in liver enzymes. An increase in the levels of immunosuppressive drugs (tacrolimus or cyclosporine) was detected in all patients during coadministration with voriconazole. A clinical response was observed in all patients (complete response, n=3; partial response, n=2), and a microbiological response was observed in all but one patient. Furthermore, a good relationship between the MIC of voriconazole and outcome was observed. Voriconazole is an effective and safe therapy for treatment of invasive J. Fortffln ()) · P. Martn-Dvila · M. A. Snchez · V. Pintado · M. E. Alvarez · A. Snchez-Sousa · S. Moreno Servicio de Enfermedades Infecciosas, Hospital Ramn y Cajal, Crtra Colmenar Km 9.1, 28034 Madrid, Spain e-mail: [email protected] Tel.: +34-913368709 Fax: +34-913368792

mold infections in solid organ recipients. To avoid toxicity with this drug, however, the dosing of immunosuppressive drugs must be reduced.

Introduction Aspergillosis and other invasive mold infections are severe complications in immunosuppressed patients. Although the incidence of these infections is low, the resulting mortality in immunosuppressed patients exceeds 80%, despite therapy with amphotericin B (AmB), including lipids formulations [1, 2, 3]. Treatment with amphotericin B desoxycholate is limited by poor toleration and nephrotoxicity [4, 5]. Lipid formulations of amphotericin B are associated with less nephrotoxicity, but none has been proven more efficacious than amphotericin B desoxycholate [6], despite some encouraging reports [7, 8, 9]. Moreover, the unpredictable bioavailability of itraconazole following oral administration made this route an unsuitable alternative. Voriconazole (Pfizer, USA) is a novel wide-spectrum triazole antifungal agent that is active against various yeast and molds, including Aspergillus spp. (geometric mean MIC, 0.4 mg/dl), and compares favorably with amphotericin B [10, 11, 12, 13]. It is fungicidal in vitro for a majority of isolates and efficacious in animal models, regardless of immune status, usually sterilizing tissues in experimental systemic and pulmonary aspergillosis [13, 14, 15, 16]. It is fungicidal for other pathogens, including Scedosporium spp. and Fusarium spp., and has shown potent in vitro activity against Candida spp., including Candida krusei and less susceptible isolates. Voriconazole has poor activity against Zygomycetes. It can also be given orally and intravenously, making therapy switching easier. We present here our experience with voriconazole administered on a compassionate-use basis to five patients with invasive mold infections in our institution.

409

Materials and Methods The medical charts of each of the five patients were reviewed, and clinical, haematological, chemistry, and microbiological data were analysed. Standard criteria for diagnosis of invasive mold infections and definitions of cure were applied [17]. Antifungal Susceptibility Testing Antifungal susceptibility testing was performed as outlined in the National Committee for Clinical Laboratory Standards (NCCLS) performance standards for microdilution assay [18]. Three antifungal agents were tested against each isolate: voriconazole, itraconazole, and amphotericin B. The concentration of drugs tested ranged from 0.03 to 16 g/ml. Fungi were grown on potato dextrose agar to induce conidia and sporangiospore formation and incubated at 35C for 7 days. The inoculum suspensions were prepared in accordance with the NCCLS guidelines. These suspensions were diluted 1:50 in standard medium (RPMI) to yield the working inoculum, approximately 0.4104 to 5104 cfu/ml. A volume of 0.1 ml of these suspensions was inoculated into each microdilution tray. Microdilution trays were read visually after 48 h of incubation at 35C for Aspergillus spp. and after 48 and 72 h for Scedosporium spp. The MIC of amphotericin B was defined as the lowest drug concentration that prevented any discernible growth. The MICs of the azoles corresponded to either prominent inhibition (50% reduction of growth compared to the total growth in the control well [MIC-2]) or complete inhibition (100% inhibition of growth [MIC-0]) [19, 20]. Case Reports Two liver transplant recipients (patients 1 and 2) required retransplantation after failure of a previous graft. Patient 1 was a male liver transplant recipient who had undergone transplantation due to alcoholic cirrhosis. In the postoperative period, he presented hemodynamic instability and renal failure that required dialysis. A cytomegalovirus infection was documented. The patient developed a pancreatitis and secondary intra-abdominal abscesses that required drainage and wide-spectrum antibiotics. Culture of three consecutive samples obtained by percutaneous drainage showed Aspergillus fumigatus in all three. Surgical revision was performed, and lipid-complex AmB was started. After a cumulative dose of 2.8 g had been administered, Aspergillus fumigatus persisted in new intra-abdominal samples. Patient 2 was a cirrhotic woman positive for hepatitis C virus (HCV) who, 1 month after liver transplantation, developed a thrombosis of the hepatic artery. A retransplantation was performed. The new graft had to be retransplanted the next day because a tumour was documented in the donor. The recipient received four liver grafts in 2 months. She developed the following complications: bacterial pneumonia, biliary tract infection, and cytomegalovirus infection. She also presented hemodynamic instability and acute renal failure requiring dialysis. Several lung infiltrates developed, and Aspergillus was cultured from bronchoalveolar lavage specimens. Liposomal AmB was started. After the patient had received a cumulative dose of 2 g, a new bronchoalveolar lavage culture yielded Aspergillus fumigatus, and pulmonary lesions remained unchanged. In both of these cases, voriconazole was started at a dose of 200 mg b.i.d. and was administered for a period of 60 days. Clinical response was observed in both patients: in patient 1, the size of the intra-abdominal collection decreased progressively, and in patient 2, radiological resolution of the lesions was observed. Cultures obtained 2 weeks after starting voriconazole showed no evidence of Aspergillus. Neither visual disturbances nor increases in liver enzymes were observed during voriconazole therapy. Concomitant administration of voriconazole increased the levels of both cyclosporine and tacrolimus. The daily dose of the immunosup-

pressive agents needed to be reduced in order to maintain levels in the therapeutic ranges. Patient 3 was a 24-year-old male with cystic fibrosis who underwent bipulmonary transplantation. Before transplantation, he had received itraconazole for 6 months due to bronchial colonization with Aspergillus fumigatus. A week after transplantation he developed a right-lung infiltrate and pleural effusion that did not respond to antirejection therapy and antibiotics. Pleural fluid cultures grew Aspergillus fumigatus, and liposomal AmB therapy was started. After a cumulative dose of 3 g of AmB, pulmonary infiltrates persisted and Aspergillus fumigatus was isolated again from pleural effusion cultures. In addition, a single colony of Scedosporium apiospermum was observed in a culture of a bronchoaspirated sample. Voriconazole (200 mg, b.i.d.) was started and maintained for a period of 90 days. The pleural effusion resolved, and respiratory function improved. A progressive resolution of lung infiltrates was observed, and fibrobronchoscopic studies did not reveal macroscopic findings suggestive of aspergillosis; however, cultures of bronchoaspirated samples obtained 2 months after the start of voriconazole therapy continued to yield Aspergillus fumigatus. Tracheo-bronchial colonization with Aspergillus was suspected, and aerosolised amphotericin B was added to treatment. On day 90 of voriconazole therapy, fungal cultures from bronchoalveolar lavage samples were sterile. No adverse event except increasing plasma levels of tacrolimus were observed during voriconazole therapy. Patient 4 was a 59-year-old man who had undergone kidney transplantation due to chronic glomerulonephritis. His early posttransplant outcome was favourable, and he was discharged 20 days after the transplant procedure. On day 30 post transplantation, he developed a sudden ophthalmic loss with total blindness in his left eye. Ophthalmic examination of this eye showed a best-corrected visual acuity of counting fingers and a severe anterior chamber inflammatory reaction (4+ flare and cells). Fundus examination showed a whitish cotton-like preretinal exudation over the posterior pole surrounded by small necrotizing retinitis foci and intra-retinal haemorrhages partially obscured by severe vitritis. Ophthalmic examination of the right eye was completely normal. Culture of the aqueous humour culture was negative, and empirical treatment with liposomal AmB was commenced. However, 3 days later, a progressive worsening of the ophthalmic condition was observed, and a pars plana vitrectomy was performed. A nondiluted sample of vitreous fluid and the fluid collected during vitrectomy were sent for culture, and both grew Scedosporium apiospermum. No history of ocular trauma was present. The exploration was normal, and no source of fungal infection was documented. A transthoracic echocardiogram was normal, and no abnormal features were observed in thoracic-abdominal computed tomography. No symptoms of peritonitis were present, but the peritoneal dialysis catheter was removed and cultured. Cultures of both the catheter and the ascitic fluid were negative. No other symptoms were observed, and Scedosporium was not isolated from any other site. Voriconazole (200 mg, b.i.d.) was administered for a period of 90 days. Funduscopic findings after therapy demonstrated an absence of active inflammation, although only a 30% improvement in visual acuity was observed. Tolerance of voriconazole was excellent, with no visual disturbances or liver function abnormalities observed. Patient 5 was a 45-year-old woman with diabetes mellitus, primary biliary cirrhosis, and autoimmune anaemia who was receiving therapy with corticosteroids and cyclosporine. She had been diagnosed with pulmonary aspergillosis and had multiple cavitations, and culture of bronchoalveolar lavage and transbronchial biopsy samples yielded Aspergillus fumigatus. The patient had received conventional AmB and a cumulative dose of 2 g of liposomal AmB. She relapsed 6 months later and received another course of liposomal AmB (cumulative dose, 3 g), followed by itraconazole (400 mg, once daily). Itraconazole was stopped 2 months later because of severe anaemia, and liposomal AmB (50 mg, once weekly) was reintroduced. The patient was readmitted 2 months later because of fever and new radiological pulmonary infiltrates. Bronchoalveolar cultures again yielded Aspergillus fumigatus. Voriconazole (200 mg, b.i.d.) was administered. Initial

liver transplant, alcoholic cirrhosis

liver transplant, HCV cirrhosis

Patient 2

CMV infection, pancreatitis, bacterial pneumonia, intra-abdominal abscesses biliary tract infection, CMV infection Retransplantation yes yes (3 grafts) Renal failure (Cr >2.5 mg/dl) yes (dialysis required) yes (dialysis required) Mold infection A. fumigatus A. fumigatus Clinical manifestation abdominal abscesses lung infiltrates Days after transplant (from 30 24 last transplant) Therapy before VCZ Ambisome Ambisome Cumulative dose of AmB 2.8 g (3 mg/kg/day) 3 g (3 mg/kg/day) Indication for VCZ positive cultures in control similar infiltrates and samples positive cultures Duration of VCZ therapy 60 days (14 i.v.) 60 days (40 i.v.) VCZ toxicity observed Visual no no Hepatotoxicity no no Other "FK levels "FK levels Response Clinical cure yes yes Microbiological cure yes yes Outcome (at 1 year) alive alive; recurrence of HCV

Predisposing factors

Basal condition

Patient 1

none

no transplant related S. apiospermum ocular infection 30 Ambisome (2 days) 0.2 g primary antifungal therapy 90 days (0 i.v.) no no "FK levels improved probable alive; loss of 70% of visual acuity

no no A. fumigatus lung infiltrates, empyema 7 ITZ, Ambisome 3 g (3 mg/kg/day) similar infiltrates and positive cultures 90 days (30 i.v.) no no "FK levels yes yes alive; ischemic cortical blindness (related to surgery)

kidney transplant, glomerulonephritis

Patient 4

Aspergillusbronchial colonization, lung infiltrates

lung transplant, cystic fibrosis

Patient 3

Table 1 Characteristics of five patients with invasive mold infections treated with voriconazole

improve no deceased; ketoacidotic coma

no yes renal failure

ITZ, Ambisome 5 g (3 mg/kg/day) similar infiltrates and positive cultures 90 days (0 i.v.)

no; transplant patient no A. fumigatus lung infiltrates –(transplant patient)

autoimmune anemia, corticosteroids and cyclosporine therapy, diabetes mellitus, primary biliary cirrhosis corticosteroids and cyclosporine therapy

Patient 5

410

411 Table 2 Minimum inhibitory concentrations (MICs in g/ml) of amphotericin B, itraconazole and voriconazole forAspergillus spp. and Scedosporium spp

Isolate no.

Organism (patient no.)

1 2 3 4 5 6

A. fumigatus(no. 1) A. fumigatus( no. 2) A. fumigatus(no. 3) S. apiospermun (no. 3) S. apiospermun (no. 4) A. fumigatus(no. 5)

improvement during the first 3 months of therapy was followed by a new microbiological and clinical relapse of aspergillosis. Liver enzymes increased slightly during voriconazole treatment, and the patient developed renal failure due to high serum levels of cyclosporine. Cyclosporine was discontinued, and the patient was discharged with normal renal function. She was readmitted 2 weeks later in a ketoacidotic coma after voluntary withdrawal of insulin therapy. She died 2 days later.

Results Table 1 describes the five cases of invasive mold infections (Aspergillus fumigatus, n=4; Scedosporium apiospermum, n=1) treated with voriconazole on a compassionate-use basis during the last 5 years in our institution. The MIC of voriconazole was 1 g/ml for all Aspergillus fumigatus strains (Table 2). Scedosporium apiospermum was recovered from two patients (patients 3 and 4). In patient 3, a colony of Scedosporium apiospermum (MIC of voriconazole, >16 g/ml) was isolated in mixed culture with Aspergillus fumigatus. The former mold was interpreted as colonization. Clinical response was observed in all cases (complete response in patients 1, 2, and 3 and partial response in patients 4 and 5). Microbiological response was observed in all but one patient (patient 5), in whom Aspergillus persisted in cultures of respiratory samples 3 months after the start of voriconazole therapy. Voriconazole was well tolerated. A mild increase in liver enzyme values was noted in one patient with cirrhosis (patient 5). Interference with cyclosporine or tacrolimus was observed in all patients. The increase in levels of immunosuppressive drugs caused acute renal failure in patient 5. Voriconazole was administered mainly by the oral route and allowed patients to finish treatment as outpatients.

Discussion We report here our experience with voriconazole in the treatment of severe mold infections in immunosuppressed patients in our institution. This was not a comparative study, and the contribution of AmB, which was administered previously to most patients, to outcome is unknown. Although the rate of response to treatment with AmB in invasive aspergillosis is ffi35% [1, 6, 21], mortality associated with AmB therapy (including lipid

Amphotericin B

1 0.5 0.25 >16 0.5 0.5

Itraconazole

Voriconazole

MIC-2

MIC-0

MIC-2

MIC-0

0.03 0.03 0.03 >16 0.06 0.03

0.06 0.06 0.06 >16 1 0.12

0.125 0.125 0.125 >16 0.06 0.5

0.125 0.125 0.25 >16 0.125 1

formulations) in 12 liver transplant patients with invasive aspergillosis treated previously in our institution was 100%. Two recent studies have confirmed the role of voriconazole in invasive aspergillosis [22, 23]. A noncomparative, open, and multicentre study performed in 116 immunosuppressed patients, most with haematological disorders that included haematopoietic stem cell transplants, with proven or probable invasive aspergillosis showed a good response rate of 48% (14% complete response and 34% partial response) and a rate of stable response of 21% [22]. Another study has confirmed that voriconazole is more effective for therapy of invasive aspergillosis than AmB followed by other licensed antifungal agents [23]. In that randomised, open, and international study, the 391 patients enrolled received voriconazole or AmB. In an analysis performed at week 12, complete and/or partial response was observed in 52.8% of patients receiving voriconazole and in 31.6% of patients receiving AmB, with a higher survival rate found for patients who received voriconazole (70.8% vs. 57.9%, hazard ratio 0.59; 95%CI, 0.40–0.88). Finally, Walsh et al. [24] have compared voriconazole and liposomal AmB in neutropenic patients with fever, reporting similar results except for less breakthrough fungaemia in patients receiving voriconazole. One of the patients receiving voriconazole had an invasive ocular infection produced by Scedosporium apiospermum. A vitrectomy was required in addition to voriconazole therapy. The experience with voriconazole in Scedosporium infections is very limited. However, successful therapy of infections due to Scedosporium apiospermum has been reported [25, 26, 27], even in cases with central nervous extension [28]. Microbiologically, voriconazole is more active than AmB, itraconazole, and ketoconazole against Scedosporium apiospermum and is as active as miconazole, but it shows lower activity against Scedosporium prolificans [12]. Antifungal susceptibility tests were performed on the isolates obtained from our patients. Recently, the NCCLS proposed standard parameters for testing the fungistatic antifungal activity (MICs) of established agents against filamentous fungi [18]. However, the correlation between MIC results and treatment outcome is not well defined. The MICs of voriconazole for the isolates from our patients were low, and clinical response was observed. Some authors have suggested the use of other in vitro predictors of fungicidal activities, like the minimum

412

fungicidal concentration (MFC) or the lethal concentration [19]. Considering MFC as the lowest drug dilution that results in fewer than three colonies (99.9% killing), Espinel-Ingroff et al. [19] have shown that voriconazole had similar or better fungicidal activity (MFC90 1–2 g/ ml) than itraconazole or AmB against Aspergillus spp., with the exception of Aspergillus terreus. Similarly, in that study, the mean MFC of voriconazole for Scedosporium spp. (2.52 g/ml) was lower than the MFC of AmB (5.75 g/ml) or itraconazole (7.5 g/ml). In our patients, the MFC is presented as the MIC-0, and no differences with respect to the MIC (MIC-2 in our patients) of voriconazole were observed, although slight differences were observed for itraconazole (2 dilutions in 3 strains). Voriconazole is rapidly absorbed after oral administration and exhibits nonlinear kinetics with disproportionate rises in plasma concentrations with increasing doses that permit rapid switching of therapy in patients, allowing treatment to be completed by the oral route. Voriconazole is extensively metabolised by the liver, with