Report from the 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy and 46th Infectious Diseases Society of America Joint Conference 2008

The 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC)/46th Infectious Diseases Society of America (IDSA) 2008 joint conference program included presentations on many important topics in medical mycology. This meeting report briefly highlights selected abstracts, with an emphasis on those pertaining to the diagnosis and management of invasive fungal diseases.

Diagnostics

Despite advances in antifungal drug development, crude mortality rates attributable to invasive candidiasis (IC) have been relatively stagnant for nearly two decades [1]. Improvements in primary prophylaxis have contributed to a recent slight decline in the overall incidence of invasive aspergillosis (IA) in hematopoietic stem cell transplant (HSCT) patients; however, overall mortality remains unacceptably high, with reported case fatality rates as high as 60–85% [2]. A delay in diagnosis is a critical hurdle in effective management.

Invasive candidiasis

Anna Lau (University of Sydney, Sydney, NSW, Australia) and colleagues presented an abstract on the use of multiplex-tandem polymerase chain reaction (MT-PCR) for the rapid detection of fungemia [3]. The assay was designed to identify up to 12 fungi (including seven Candida spp., Cryptococcus neoformans, Scedosporium spp., and Fusarium spp.; Aspergillus spp. were not included). The PCR assay was tested on 65 blood samples from 43 patients who also had blood cultures. The turnaround time was <2 h. When correlated with positive fungal blood cultures, 57% and 65% of PCRs on blood samples taken prior to or after fungal blood cultures, respectively, were positive [3]. Although the sensitivity of the PCR assay was low (specificity not reported), the sensitivity of positive blood cultures (current gold standard) is also low, at 50–60% [4].

In addition to rapid laboratory diagnostic tests, clinical parameters may aid in early identification of patients at the highest risk of invasive fungal disease. Preliminary results from the TREAT (Treatment with Empirical Antifungal Therapy) study were presented in a slide session [5]. This multicenter, international, retrospective study applied a clinical prediction rule (CPR) to critically ill patients at risk of IC in order to identify patients requiring empiric antifungal therapy. The CPR included four major categories: clinical criteria (fever, hypotension, leukocytosis); presence or absence of central venous catheters; use of antibacterial agents; and known major risk factors for candidiasis such as dialysis, major surgery, and use of steroids. Results from 649 cases analyzed revealed the sensitivity and specificity of the CPR alone to be 83% and 68%, respectively. However, when the CPR was combined with the Candida colonization index (CCI), the sensitivity decreased to 67% while the specificity increased to 87%; accuracy was increased from 68% to 87%, and number needed to treat decreased from 23 to 12 [5]. In another slide presentation, Playford et al. prospectively evaluated several risk prediction models for invasive candidemia in 615 intensive care unit patients, including clinical prediction rules 1 and 2 (CPR1 and CPR2, respectively), Candida score, and CCI [6]. The results indicated that combining CPR1 with the CCI resulted in an improved positive predictive value with minimal compromise in sensitivity. CPR1 with CCI improved the specificity of the rule from 79% to 90%.

Invasive aspergillosis

In a symposium on IA, Paul Verweij (University Medical Center St Radboud, Nijmegen, The Netherlands) outlined the many challenges in developing clinically viable PCR-based technologies for the diagnosis of the condition [7]. Oligonucleotide designs are strain-dependent, reagents are sometimes highly reactive resulting in oversensitive assays, nucleotide extraction is often inefficient, and the presence of inhibitors can result in poorly sensitive assays. In the same session, Marta Feldmesser (Albert Einstein College of Medicine, Bronx, NY, USA) reviewed novel approaches to the diagnosis of IA and ongoing efforts to improve galactomannan (GM) antigen detection, which is prone to variability due to differing collection and assay methods, yielding unacceptably low sensitivities [8].

Two presented studies sought to increase the yield of currently available diagnostic tools for IA [9,10]. Ray Hachem et al. (University of Texas MD Anderson Cancer Center, Houston, TX, USA) prospectively studied combination quantitative PCR (qPCR) and GM-enzyme immunoassay (GM-EIA) performance in patients with hematological malignancy diagnosed with probable or proven IA receiving anti-mould therapy versus control patients with solid organ malignancies and no evidence or risk factors for IA [9]. The results demonstrated either unacceptably low sensitivity or specificity when using PCR or GM assays alone; however, in combination, qPCR (18S) and GM-EIA sensitivity was 71% and specificity was 80%. A second study hypothesized that the yield of PCR in IA might be improved by applying it to both whole blood and serum, thereby capturing both Aspergillus DNA bound to fungal cells in tissue and host leukocytes, as well as unbound DNA in serum [10]. Whole blood and serum were collected from 16 patients with proven or probable IA. In 50% of patients, combination serum plus whole blood PCR was positive 12 days (median) before whole blood-only PCR, and 7 days before serum-only PCR. The study suggests that PCR testing on both whole blood and serum together may improve the ability to make an early diagnosis of IA.

Management of invasive fungal infections

Invasive candidiasis

A multicenter, double-blind, randomized, non-inferiority study presented at the meeting by Robert Betts (University of Rochester, Rochester, NY, USA) and colleagues, investigated standard- (70 mg initially, then 50 mg daily) versus high-dose (150 mg daily) caspofungin. The groups were similar with regard to infection site, neutropenic status, and Acute Physiology and Chronic Health Evaluation II (APACHE II) scores. There were no statistically significant differences in drug-related adverse events, successful clinical and microbiological resolution of infection, or mortality rates at 8 weeks post-treatment. The study established the safety of high-dose caspofungin but found no significant difference in efficacy between standard- and high-dose therapy [11].

Given the importance of early antifungal therapy for overall mortality rate among fungemic patients and the reduced susceptibility of Candida glabrata to fluconazole, Michael Klevay et al. (University of Iowa, Iowa City, IA, USA) compared treatment and outcomes in patients with C glabrata versus C albicans bloodstream infections (BSI) using data collected over 3 years from a prospective, multicenter, observational study of invasive fungal infections [12]. A total of 161 C glabrata patients were matched to 161 C albicans patients by age, sex, and underlying illness. The primary outcome was mortality rate; secondary outcomes included choice, dose, and timing of empiric therapy. Patients with C glabrata fungemia were less likely to receive adequate empiric fluconazole dosing and more likely to receive an echinocandin. Although the time to initiation of empiric antifungal therapy was not different between the two groups, the time to receipt of adequate treatment was longer in patients with C glabrata BSI and the risk of receiving inadequate treatment was higher in this group. This did not translate to an effect on mortality rate, which was no different between the two groups at 4 weeks [12].

Invasive aspergillosis

Voriconazole has been the first-line therapy for IA for several years. Daniel Diekema et al. (University of Iowa) undertook a surveillance study to uncover any resultant emergence of voriconazole-resistant Aspergillus spp. utilizing the SENTRY and ARTEMIS surveys. They collected 1626 clinical isolates of Aspergillus spp. over 8 years. Applying the Clinical and Laboratory Standards Institute (CLSI) breakpoint definition of ≤1 μg/mL for Candida spp. to Aspergillus spp., they found only 2.3% of isolates had a minimum inhibitory concentration (MIC) >1 μg/mL. However, overall MIC 50/90 increased from 0.25/0.5 μg/mL in 2000–2002 to 0.5/1.0 μg/mL during 2005–2007, suggesting a slow trend towards increasing MICs since the use of voriconazole in IA became widespread [13].

Given the concerns regarding voriconazole hepatotoxicity, particularly in allogeneic HSCT patients, caspofungin has been studied as first-line treatment against IA in this subpopulation. In a study by Raoul Herbrecht and colleagues (Hôpital de Hautepierre, Strasbourg, France), patients with probable and proven IA were evaluated for a complete or partial response (CR and PR, respectively), which served as primary endpoints [14]. Secondary endpoints included response to treatment at day 84, survival, and safety. Among the 24 patients who met inclusion criteria, the median duration of caspofungin treatment was 24 days. A CR or PR was seen in 42% of patients by the end of treatment. However, it is noteworthy that 50% of patients had worsening disease. At day 84, the survival rate was 48%. No significant adverse drug reactions were reported. This study suggests that caspofungin is a viable first-line therapy for Aspergillus spp. in HSCT patients in whom voriconazole may convey a higher risk of hepatic toxicity [14].

Update on cryptococcosis

Data were presented showing that cryptococcosis remains a medical mycosis of major importance. Benjamin J Park (Centers for Disease Control and Prevention [CDC], Atlanta, GA, USA) presented findings from a compelling CDC study, which estimated that the annual global burden of C neoformans meningitis among HIV patients is 957 900 cases, with a resultant number of deaths at 3 months of 624 700. The highest incidence was in Sub-Saharan Africa, followed by South/Southeast Asia (with estimated median incidences of 720 000 and 120 000 cases, respectively). Remarkably, these estimates surpass annual mortality rates from HIV-associated tuberculosis, underscoring the importance of improved diagnosis, prophylaxis/prevention, and treatment of cryptococcal disease in resource-poor settings [15].

In the same session, Brian Metzger (Albert Einstein College of Medicine) presented findings from a study linking the risk of cryptococcal disease among two separate cohorts of HIV patients with immunoglobulin M (IgM) memory B cell deficiency [16]. Remarkably, having less than 38% IgM memory B cells was significantly predictive of developing cryptococcal disease as much as 3 years prior to disease onset (odds ratio 12). This novel finding could allow identification of at-risk HIV patients who may benefit from early fungal prophylaxis determined not only by CD4⁺ T cell counts but also IgM memory B cell quantity, thereby potentially curbing the rising mortality of cryptococcal disease in HIV patients.

Although the incidence of HIV-associated cryptococcosis in the US has decreased dramatically since the introduction of highly active antiretroviral therapy (HAART), Cryptococcus-associated morbidity is a growing concern in immunocompetent hosts. Surveillance data from a Cryptococcus neoformans var. gattii outbreak in the Pacific Northwest revealed that all 11 cases studied occurred in HIV-negative patients; six of these patients had underlying comorbidities that required steroid use within 1 year of diagnosis [17], one had immunoglobulin G2 (IgG2) deficiency, and the remainder had no discernible immune compromise.

The role of fungus in chronic respiratory disease

David Goldman (Albert Einstein College of Medicine) presented a series of studies outlining serological evidence of subclinical C neoformans among children residing in an urban setting, demonstrating hyper-reactive, allergic airway inflammation caused by non-lethal pulmonary infection with Cryptococcus in a murine model, and examining the unique abilities of this fungus to induce chitinase expression, which is increasingly being recognized as an important mediator of allergic pulmonary inflammation and hyperactivity [18]. These findings provide compelling evidence for a potential link between pulmonary cryptococcal infection and urban asthma in pediatric patients. Fungal colonization has been correlated with severe, persistent asthma. David Denning (University of Manchester, Manchester, UK) presented findings from a randomized controlled trial of oral itraconazole treatment (200 mg orally twice daily) versus placebo for 32 weeks in patients with severe asthma with fungal sensitization. Patients with severe asthma who were sensitive to one or more of seven fungi by skin prick test had baseline IgE concentrations of <1000 IU/mL, and negative Aspergillus precipitins were included in the study. The results indicate statistically significant improvements in the primary outcome of Asthma Quality of Life Questionnaire (AQLQ) score as well as in secondary outcomes including rhinitis score, morning peak flow, and reduced serum IgE levels in patients treated with itraconazole versus placebo [19]. Such findings support the need for further studies delineating the role of fungi in chronic respiratory diseases.