Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

PRIORITY PRESS - 52nd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC)

San Francisco, California / September 9-12, 2012

San Francisco - The opportunity to improve outcomes was outlined for 3 life-threatening fungal infections at the 2012 ICAAC. One was a re-exploration of optimal therapy for invasive aspergillosis; another was an evaluation of improving survival rates over the last 10 years in invasive fusariosis; and the third, based on the experience in British Columbia, was a retrospective analysis of the characteristics and outcome predictors in Cryptococcus gattii. In all 3 cases, optimal strategies for early detection and antifungal therapy selection were updated with the goal of improving outcomes, including mortality reductions. In these and other fungal infections, there are no consistently dependable markers to confirm the pathogen in advance of mycologic cultures, but strategies for identifying probable infections to direct presumptive therapy were suggested by sets of characteristics that appear to correspond with specific infection types.

Chief Medical Editor: Dr. Léna Coïc, Montréal, Quebec

In a series of reports discussed here at the 2012 ICAAC, incremental progress in the control of fungal diseases in hospitalized patients was described in 3 types of infections which impose a substantial risk of death. In invasive aspergillosis (IA), the impact of more recently adopted diagnostic criteria was applied to a landmark study conducted before the criteria existed to determine if the conclusions were valid. In invasive fusariosis (IF), the steep improvement in survival rates across centres was dissected. In Cryptococcus gattii, several predictors of disease were identified to accelerate diagnosis and implementation of therapy.

Revisiting Invasive Aspergillosis Treatment Results

The IA study was based on a reanalysis of a landmark study of voriconazole and amphotericin B. Due to the fact that the initial study was conducted using older criteria for classifying these infections, the data were recalculated using current definitions for definite, probable and possible infections. According to Dr. Raoul Herbrecht, University of Strasbourg, France, senior author of the original study (Herbrecht et al. N Engl J Med 2002;347:408-15), the changes in the definitions created a potential for altering the relative strengths of these agents against different levels of diagnostic certainty.

In the original study, both the success rate (52.8% vs. 31.6%; P<0.05) and the survival rate (70.8% vs. 57.9%; P=0.02) at 12 weeks favoured voriconazole over amphotericin B in a modified intention-to-treat analysis, but criteria for classifying IA patients have since changed. Of these, the 4 most important changes were that halo and crescent signs, thought to signal a probable infection at the time of the initial study, are only considered possible signs of infection now; a positive bronchoalveolar lavage (BAL) once considered a sign of definite infection now only identifies probable infections; a positive serum galactomannan (GM), now a sign of probable infection, was not considered at all; and a nodule without halo or mycology, now a possible IA, was an exclusion criterion.

The goal of this reanalysis was to evaluate whether results were altered using modern criteria for patient selection and classification. The secondary objectives included comparisons of the relative efficacy of voriconazole and amphotericin B in different patient subgroups, such as those with definite IA vs. those with possible IA. All the data were reconsidered in the context of underlying diagnosis, mycological data including GM results and radiological findings. Cases were individually recategorized by 5 investigators blinded to the treatment before seeking consensus from these independent assessments.

In the mycologically documented cohort of 123 patients, favourable responses were achieved in 49.6% of those randomized to voriconazole and 25.5% of those randomized to amphotericin B (P=0.0002). In the possible cohort of 54 patients, the relative advantage was also large and statistically significant (64.8% vs. 39.2%; P=0.011), producing an overall favourable outcome in 54.2% and 29.8%, respectively (P<0.0001). It is notable that there was full consensus among all 5 investigators in 83% of the cases, disagreement by a single member in 15% and disagreement by 2 members in 3%.

The differences in survival at 12 weeks also remained significant under the revised classification system. Overall, the figures were 72.9% and 59.6% for voriconazole and amphotericin B (P=0.006). Graphically, the voriconazole advantage was similar in those with probable or definite IA and those with possible IA, but the difference in those with possible IA did not reach statistical significance.

The results provide a reconfirmation that “voriconazole is the most active agent in IA,” according to Dr. Herbrecht. Although relative tolerability was not discussed in this analysis, one member of the audience pointed out in a discussion that initial results associated voriconazole with substantially better tolerability. He asked Dr. Herbrecht whether voriconazole was more potent or simply better tolerated, allowing more patients to complete a full dose regimen. Dr. Herbrecht replied, “I think both are true.”

Improved Outcomes in Invasive Fusariosis

Two databases combined to explore outcomes in 158 patients with IF documented a large improvement in survival when the years 1985 to 2000 were compared to 2001 to 2011. The patient data were drawn from 21 centres in Europe, Brazil, the US and Canada. The primary outcome was survival at 90 days, which climbed from 16% in the first era to 49% in the second era.

“We have been impressed with recent case reports that suggested that outcomes have been improving,” Dr. Mario Nucci, Federal University, Rio de Janeiro, Brazil, told delegates. “We conducted this study to seek some objective evidence of better survival and to understand the reasons.”

Data on patient characteristics, underlying conditions and treatments were collected and compared for the 2 eras in univariate and multivariate analyses. While such factors as use of corticosteroids and persistent neutropenia were poor prognostic factors in both eras, the major change was the switch in antifungals. While amphotericin B was used in 81% of those treated in the first era, it was used in only 23% of patients in the second era. Conversely, voriconazole was not used in the first era but was employed in 42% of patients in the second. The importance of this difference emerged in the multivariate analysis that indicated voriconazole was associated with a better outcome regardless of the minimum inhibitory concentration.

“In a retrospective analysis like this, we cannot control for a variety of improvements in patient care, but there was a substantial improvement in outcome with IF, and the change in primary therapy appears to have been a factor,” Dr. Nucci reported.

C. gattii: The Canadian Experience

The study of C. gattii was also a retrospective chart review. Once considered a tropical pathogen, this infection was first diagnosed in British Columbia in 1999. A report from Dr. Peter Philips, Head, Division of Infectious Diseases, St. Paul’s Hospital, Vancouver, provided data on 171 cases collected since that time of which 152 occurred in patients who were immunocompromised. Of these, 111 were characterized as confirmed and 41 were considered probable. The majority of those immunocompromized had lung infections alone (74%) while another 15% had lung and central nervous system (CNS) infections and 11% had CNS infections alone.

“Mortality was associated with multiple chest cryptococcomas (HR 3.32; P=0.0236), CNS involvement (HR 2.91, P=0.0146) and advancing patient age (HR/10 years 2.06, P<0.0001),” Dr. Phillips reported. The all-cause mortality in the cohort of patients was 19% at one year, but Dr. Phillips reported that 96% of these deaths occurred within 6 months. Most importantly, he suggested that in areas where C. gattii is active, outcome may depend on rapid recognition.

“Although we did not find that comorbidities reduced the risk of survival, our data suggests that in endemic areas, a provisional diagnosis of a lung or brain malignancy should prompt cultures for C. gattii,” Dr. Phillips suggested.

Summary

Invasive fungal infections remain one of the most feared complications in immunocompromised hospitalized patients. Strategies for rapidly recognizing these infections and initiating an appropriate therapy continue to improve, but mortality rates remain substantial because definitive signs of early disease remain elusive. Efforts to refine signs and symptoms of fungal infections remain the most promising approach to improving outcomes. The data presented at ICAAC on IA, IF and C. gattii are representative of the incremental advances based on trials and retrospective analyses while the effort to develop definitive signs of disease in advance of culture results continues. Many fungal infections are readily controlled with early initiation of an appropriate antifungal agent, emphasizing the need to apply multiple strategies for aggressively screening suspected cases.