The following is a brief and selected overview of presentations at Focus on Fungal Infections 13 in Maui, Hawaii, March 19-21, 2003. — Stan Deresinski, MD, FACP
Training in Clinical Mycology
William Steinbach reviewed the current state of training in medical mycology—not an encouraging picture. At a time when the average age of clinical mycologists is reaching toward the stratosphere and the demand for their services is increasing, the amount of time devoted to the subject in medical school is low and decreasing. As an example, only 18% of schools offered a lecture devoted to antifungal therapy.
Clinical Mycology Laboratory
Mahmoud Ghannoum of Case Western Reserve University addressed the usefulness of a modern fungal clinical laboratory to the clinician, given the emergence of pathogens with increasing antifungal resistance and the introduction of new therapeutic agents. Because of the relatively predictable resistance of some fungal species to some antimycotic agents, species identification of many pathogens is increasingly important. It was recommended that indications for antifungal susceptibility testing include Candida species recovered from sterile sites, which should be tested against fluconazole or flucytosine. In addition, the following testing should be performed in the face of clinical failure of initial therapy:
a) Candida species and amphotericin B,
b) Cryptococcus neoformans and fluconazole, flucytosine, or amphotericin B, and
c) Histoplasma capsulatum and fluconazole.
James Ito examined the prospects for vaccine protective against invasive aspergillosis in high-risk patients.
Later in the meeting, Jan Patterson illustrated issues related to clinical decision-making in the use of susceptibility testing.
Newer Diagnostic Tests
Kieren Marr discussed the use of newer techniques for the diagnosis of invasive aspergillosis. These include the detection of circulating (1,3)-b-D-glucan by a limulus amoebocyte lysate assay, genome detection by PCR, and detection of galactomannan by ELISA. Some European studies of the BioRad galactomannan assay, currently under consideration by the US FDA, reported sensitivity and specificity each greater than 90%. Studies in the United States and at least 1 French study have, however, reported less favorable results. In the latter, the sensitivity in definite cases of invasive aspergillosis in cancer patients was 65%, the specificity was 95%, and the positive and negative predictive values were 58% and 85%, respectively (J Clin Oncol. 2002;20:1898-1906). Separately, in a poster presentation, it was reported that the sensitivity of this assay in children with aspergillosis was only 50% (7 of 14), but its specificity was 98.4%. Some of the variability in test performance is the result of the varying populations studied, as well as the cutoff used in determining ELISA positivity. Furthermore, it has been reported that the presence of circulating antibodies against Aspergillus antigen reduce the sensitivity of the antigen test.
It was reported, in a poster presentation, that corticosteroid therapy (> 2 mg/kg/d for > 7 days) was associated with an increased risk of breakthrough fungal infection during secondary antifungal prophylaxis.
The safety and tolerability of aerosolized administration of amphotericin B deoxycholate and ABLC as prophylaxis against pulmonary fungal infections were compared in a randomized trial in lung transplant recipients; ABLC was better tolerated. Recipients of conventional amphotericin were more likely than recipients of ABLC to experience an adverse effect (OR 2.16, P = .02). Discontinuations due to drug intolerance occurred in 6% of the former and 12% of the latter (P = .313). No cases of fungal pneumonia occurred in either group during the 7-week period of prophylaxis. John Perfect, in reviewing these results, as well as the overall Duke experience, concluded, "Although multicenter studies will be needed to determine efficacy results, it is our impression that aerosolized polyenes are safe and associated with a low rate of pulmonary fungal infections."
Ben de Pauw provided his perspective on early preemptive therapy for invasive candidiasis. Marcia Nucci examined the need for routine removal of venous catheters in patients with candidemia, recapitulating material from their recent publication (Clin Infect Dis. 2001;33:1959-1967). In essence, he suggested that catheters may be retained in some cases in the absence of associated tissue infection or of severe neutropenia, under circumstances in which removal is not possible and the patient is not severely septic.
In vitro studies found that, consistent with previous studies indicating the interaction between the ergosterol synthesis pathway, the calcineurin pathway, and fungal viability, both terbinafine and fenpropimorph were synergistic with either cyclosporin or FK506 against C albicans.
In an open trial, micafungin therapy was successful in the primary treatment of 60 of 68 (88%) patients with candidemia and in the salvage therapy of 39 of 51 (76%) patients.
Two patients with endovascular infection (endocarditis, septic subclavian vein thrombophlebitis) had clearance of candidemia, which had persisted during amphotericin therapy when caspofungin was added to the regimen. These observations may be related to the observation that caspofungin is active against Candida growing in biofilm.
Aspergillosis and Other Infections
Ellie Anaissie discussed risk factors for and means of prevention of invasive aspergillosis.
The incidence and outcome of subsequent Aspergillus infection in 45 patients with a history of prior invasive aspergillosis who underwent allogeneic hematopoietic stem cell transplantation, most of whom were started on antifungal therapy at least 1 month prior to transplantation, were reviewed. Post-transplant aspergillosis occurred earlier (median of day 26 vs day 54) in those with prior aspergillosis vs those without, and transplant-related death also was more frequent. Factors associated with a lowered risk of post-transplant recurrence of aspergillosis included receipt of antifungals for more than 1 month pretransplantation and resolution of chest x-ray abnormalities.
Of 16 cases of toenail onychomycosis in neutropenic patients were due to Fusarium, 4 to Aspergillus, 2 to Trichophyton, and 1 to Paecilomyces (1 was infected with both Fusarium and Aspergillus). Dissemination, particularly with Fusarium, occurred in some patients whose neutropenia was of long duration.
A fumigatus isolates from pulmonary aspergillomas were found to produce aflatoxin in vitro.
Emerging Nosocomial Mycoses
Mike Pfaller discussed recent data concerning nosocomial fungal pathogens that are emerging as we "dig deeper into the medical mycology compost heap.’" The increasing importance of Candida glabrata was discussed. Data from the Emerging Infections and Epidemiology of Iowa Organisms (EIEIO) study demonstrates that the frequency of isolation of this organism increases with increasing patient age. C glabrata is of particular concern because of its reduced susceptibility to fluconazole and amphotericin B. Furthermore, exposure to subtherapeutic concentrations of fluconazole may lead to induction of multidrug efflux pumps, leading to high-level resistance to all available azole antifungal agents. Other Candida species that exhibit antifungal resistance, albeit with much lesser frequency than C glabrata, include C krusei and C lusitaniae. The former routinely exhibits high-level resistance to fluconazole and the latter commonly develops secondary resistance to amphotericin B as a consequence of high-frequency phenotypic switching. Aspergillus terreus, the Aspergillus species second in frequency of isolation to only A fumigatus, has reduced susceptibility to amphotericin B. The most reliably active US FDA-approved antifungal agent against these species of Candida and Aspergillus is caspofungin, although the newer azoles have good activity as well. Amphotericin B remains the most reliably active against Rhodotorula spp., although the investigational agent posaconazole has significant in vitro activity against this yeast pathogen. Triazoles are most reliably active in vitro against Trichosporon spp. and are the therapeutic agents of choice. Fusarium spp. remain problematic, although voriconazole has received FDA approval for treatment of infections due to this organism, as it has for infections caused by Scedosporium apiospermum, for which it has become the treatment of choice. Posaconazole shows promise against many zygomycetes. While caspofungin MICs are high against many mould isolates, minimum effective concentrations are much lower.
In reviewing the changing epidemiology of fungal infections in transplant recipients, Barbara Alexander Lodge noted a decrease in the incidence of invasive candidiasis and an increase in filamentous mould infections, not only due to Aspergillus spp., but also to Scedosporium and Fusarium spp.
Fluconazole resistance was selected in vivo from an initially heteroresistant population of C parapsilosis during treatment of candidemia with fluconazole. The resistance was reversible during serial in vitro passage.
Crypto, Histo, Cocci, Blasto, etc
Peter Pappas and Shigeri Kohno each discussed the management of cryptococcal infection in non-AIDS patients, and Robert Bradsher reviewed the treatment of histoplasmosis, blastomycosis, coccidioidomycosis, and paracoccidioidomycosis. Rana Hajjeh presented the CDC perspective on opportunistic mycoses in AIDS, examining, in particular, cryptococcosis in Africa and penicilliosis in Southeast Asia and southern China. Particularly in Africa, the lack of diagnostic tests and affordable therapies presents a dismal circumstance.
The median time to onset of cryptococcal infection in 15 Mayo Clinic solid organ recipients was 30 months post-transplantation (range, 2-159 months), with three-fourths occurring after more than 6 months. Nine of the 15 had meningitis; 2 of the 15 patients died of their infection.
The usually recommended dose for treatment of PCP is 75-100 mg/kg/d sulfamethoxazole and 15-20 mg/kg/d trimethoprim. Sixteen AIDS patients with PCP were successfully treated with a mean of only 46 mg/kg/d of the former and 9 mg/kg/d of the latter.
Interpretation of Clinical Trial Results
John Bennett reviewed the concept of noninferiority trials and distinguished them from equivalency or superiority trials. However, he pointed out that, given appropriate statistical outcomes, superiority may, on occasion, be demonstrated in trials powered only to demonstrate noninferiority. He demonstrated these concepts by examining the results of 3 recent trials. By analysis of 2 of these, a comparison of voriconazole to amphotericin B deoxycholate in the treatment of invasive aspergillosis and a comparison of caspofungin to amphotericin in the treatment of invasive candidiasis, although both were designed as noninferiority trials, it could be concluded that the newer agent was, in fact, superior to the comparator since the 95% confidence intervals for the delta did not include zero. In contrast, a comparison of voriconazole with liposomal amphotericin B in the empiric treatment of prolonged fever in neutropenia did not demonstrate noninferiority of the former since the lower limit of the 9% confidence interval for the delta was greater than the 10% that had been the a priori agreed upon threshold.
Treatment Perspectives—Current and Future
Jack Graybill discussed his views of the appropriate use of newer antifungal agents. He indicated that the newer azoles, including voriconazole, are not the first choice in the treatment of fluconazole-resistant invasive candidiasis because of some degree of cross resistance. He indicated that voriconazole is the drug of choice in the treatment of invasive aspergillosis and that this drug also has a role in the treatment of fusariosis, pseudallescheriasis, and some other mould infections, but not for infection due to zygomycetes, against which posaconazole has potential. He suggested that the echinocandins, such as caspofungin, may have a role in the treatment of infection due to azole-resistant Candida and/or species of nonalbicans Candida likely to be azole resistant. The relative lack of toxicity and pharmacokinetic drug-drug interactions also gives an edge to echinocandins in some circumstances. He indicated that, although antagonism does not appear to be a concern, the data are insufficient to demonstrate benefit from the use of caspofungin in combination with an azole or polyene.
Tom Walsh gave his view of the future of antifungal chemotherapy. These include the development of new compounds within existing classes, as well as the development of new classes of drugs. Also important will be clinical trials enabling the rational use of antifungal agents in combination. Immunomodulatory therapies may also play a role in the future, as will new diagnostic techniques, allowing earlier institution of pathogen-specific agents. Finally, several individuals discussed the path a new drug must traverse in order to obtain the blessing of the US FDA.
Pharmacology of Antifungal Agents
Tom Walsh reviewed the pharmacokinetics and pharmacodynamics of antifungal agents, as did Paul Gubbins. Of particular note are the data suggesting that amphotericin B and caspofungin each demonstrate concentration-dependent killing against Candida spp.
Scott McConnell examined some pharmacological considerations in the choice of newer antifungal agents. Because the voriconazole AUC and Cmax is significantly reduced when the drug is taken with a high-fat meal, it is recommended that it be ingested 1 hour before or after a meal. Of note is that, in contrast to the pH effect on itraconazole absorption, no such effect is seen with voriconazole. The kinetics of this triazole are nonlinear in adults, but not in children. Of importance is a very large degree of intersubject variability in voriconazole exposure, a significant portion of which is accounted for by variations in hepatic cytochrome P450 enzymes, particularly CYP2C19. Individuals with full expression of this isoform, which is very prevalent in Asians and also present in 8% of Caucasians, have a 4-fold or greater AUC and Cmax than is seen in its absence. However, it has not, as yet, been possible to relate serum concentrations of voriconazole to efficacy. The interaction with CYP450 enzymes also accounts for a wide variety of pharmacokinetic drug-drug interactions that must be taken into account in the clinic.
A comparison of patients who received either liposomal amphotericin B or ABLC at multiple centers in Canada found no difference in associated nephrotoxicity.
The observation of modest transaminase elevations in volunteers receiving cyclosporin and caspofungin in combination in a Phase 1 study led to warnings to use caution in administering these 2 drugs together. It was, however, reported that 6 patients received cyclosporin and caspofungin simultaneously for 2-56 days without adverse effect.
Duration of Antifungal Therapy
Nina Singh examined the issue of the necessary duration of treatment of invasive infections caused by 3 different fungi. Aspergillus infections and the immune status of the affected patient are highly variable, making any firm conclusion subject to discussion. Treatment decisions must be individualized. It was, nonetheless, recommended that treatment be continued for a minimum of 10-12 weeks and for at least 4 weeks after resolution of all clinical and radiographic abnormalities. Singh recommended that cryptococcal meningitis in hematopoietic stem cell or solid organ transplant recipients be treated with amphotericin B and 5-flucytosine for at least 4 weeks after which fluconazole is administered for 8-10 weeks or until clinical and radiographic findings have resolved. Uncomplicated candidemia can be effectively managed with 2-3 weeks of antifungal therapy.
Jo-Anne van Burik discussed the potential role of granulocyte transfusions in the prevention and treatment of bacterial and fungal infections in neutropenic patients. The yield of granulocytes obtained by leukapheresis after administration of G-CSF and dexamethasone is at least 7 × 1010 cells. A meta-analysis of trials of prophylactic granulocyte infusion in leukemia and/or bone marrow transplant patients performed at a time when these pharmacologic means were not used and granulocyte yields were consequently much lower found a reduction of risk of infection and attributable mortality in recipients, at the cost of an increased risk of pulmonary infections and of CMV infections. Despite a lack of controlled therapeutic trial data, van Burik stated that the "use of granulocyte transfusions should not be limited to clinical trials because enough evidence exists that they may be beneficial when cell doses are high enough."
Immunology and Immunotherapy
Brahm Segal reviewed concepts of potential immunotherapy, particularly interferon gamma, of invasive mycoses, indicating that a randomized prospective trial will be initiated comparing voriconazole plus interferon gamma to voriconazole plus placebo in patients with invasive mould infections.
David Stevens reviewed his laboratory’s investigations in the area of host resistance to fungal infections, cytokine therapy, and the interaction between cytokines, antifungal agents, and phagocytic cells against some fungal pathogens. Among their recent findings was the demonstration that GM-CSF reversed corticosteroid induced diminution of macrophage activity against Aspergillus both in vitro and in vivo (J Infect Dis. 2003;187:705-709).
Dr. Deresinski is Clinical Professor of Medicine, Stanford; Associate Chief of Infectious Diseases, Santa Clara Valley Medical Center