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 - 37th Annual Meeting of the European Group for Blood and Marrow Transplantation

Paris, France / April 3-6, 2011

In the early 1990s, allogeneic stem cell transplantation (SCT) recipients experienced improved survival with anti-Candida prophylaxis. However, the medical community observed a subsequent increase in mould infections, most notably invasive aspergillosis (IA). “This was largely related to disease that occurred late after transplantation and during graft-versus-host-disease (GVHD),” Dr. Kieren Marr, Professor of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, told EBMT delegates.

IA is still associated with high mortality rates. As recently as 2007, mortality 3 months after IA diagnosis was 30%, “so the infection is still associated with poor outcomes,” she observed.

High Risk for IA

Patients undergoing allogeneic SCT are at high risk for IA among other invasive fungal infections (IFIs). Dr. Marr cited 2 studies evaluating antifungal prophylaxis in allogeneic SCT populations with fluconazole and itraconazole. While itraconazole was associated with significantly fewer IA infections compared with fluconazole, she reported that the former was not associated with improved survival odds (Blood 2004;103(4):1527-33, Ann Intern Med 2003;138(9):705-13).

“This is very concerning,” Dr. Marr remarked, noting that the lack of a significant difference may be due to “informative censoring”. She explained that if a toxic drug is given long enough, higher-risk patients may be lost to follow-up and eliminated from the statistical analysis due to drug-related toxicities or death. “And this is not what we are trying to do when we give a drug prophylactically,” Dr. Marr emphasized.

Attempting to eliminate the potentially misleading effects of informative censoring, Dr. Marr and colleagues subsequently designed a double-blind study comparing the antifungal voriconazole to fluconazole using fungal-free survival (FFS) as their primary end point measured at 6 months post-transplantation. Study drug was given for 100 days (180 days under specific circumstances), voriconazole at a relatively low dose of 200 mg b.i.d. and fluconazole at 400 mg/day. Empirical antifungal therapy was permitted for suspected IFI for a period of up to 2 weeks. Patients were largely standard-risk status, most had received match-related allogeneic transplants and the most common underlying disease was acute myeloid leukemia (AML).

Between days 0 and 180 post-transplantation, the number of IFIs was numerically lower in the voriconazole arm (22) than in the fluconazole arm (33) but the difference between them was not statistically significant. “Numerically, this difference was driven by fewer Aspergillus infections in the voriconazole arm vs. the fluconazole arm,” Dr. Marr told delegates.

At 6 months, there was no difference in FFS between treatments, although a post-hoc analysis of patients with AML found that voriconazole was associated with a better FFS and fewer IFIs than fluconazole, she added. As she observed, choosing FFS at 6 months post-transplant may have been an “unfortunate end point,” as other events unrelated to IFIs may have led to death. She advised delegates, “It is important to keep this in mind as it may have affected our conclusions.”

A differently designed trial comparing posaconazole with fluconazole given upon diagnosis of symptomatic GVHD in patients with mostly chronic myeloid leukemia showed that posaconazole was superior to fluconazole in preventing the probability of IFIs, although there was no difference in survival between the 2 groups.

“When you measure FFS, there are a number of events that can happen that dilute our ability to see a difference in the voriconazole arm, especially when measuring the end point very late after transplantation,” Dr. Marr stated. “But I personally believe that studies are suggesting that we can prevent [fungal] infections by administering drugs prophylactically.”

Dr. Tony Pagliuca, Transplant Director, King’s College Hospital, London, UK, reviewed 2 studies here at the EBMT: the still unpublished IMPROVIT study of primary prophylaxis and the VOSIFI study of targeted secondary prophylaxis in allogeneic SCT recipients with a proven or probable previous fungal infection. “IMPROVIT was the first international study to compare 2 active [antifungal] agents,” Dr. Pagliuca observed.

Primary Prophylaxis Study Findings

IMPROVIT randomized a total of 489 allogeneic SCT patients to primary prophylaxis with either voriconazole or itraconazole for at least 100 days (and 180 days in patients with active GVHD). The primary composite end point was survival at 180 days and no proven or probable breakthrough IFI. The primary analysis was intended to demonstrate non-inferiority of voriconazole and if non-inferiority was shown, then superiority was tested.

At 180 days, 49% of patients in the voriconazole arm had achieved the primary end point compared to 33% of patients in the itraconazole arm, a statistically significant difference (P<0.001). Voriconazole also met study criteria for both non-inferiority and superiority to itraconazole, Dr. Pagliuca added. As in other studies, survival at 180 days was similar in both arms at 84% and 85% for voriconazole and itraconazole, respectively.

Yet there was a “clear difference” in duration of prophylaxis, the median duration in the voriconazole arm being 96 days vs. a significantly shorter 68 days in the itraconazole arm. Patients in the voriconazole arm were also significantly less likely to receive other antifungals than those in the itraconazole arm, a significant proportion of whom crossed over to voriconazole or other antifungals. Time to emergent IFIs was also about 30 to 40 days earlier for patients on itraconazole, although few patients in either arm developed an IFI at approximately 2% each.

“The superiority of voriconazole in this study was driven by better long-term toxicity,” Dr. Pagliuca remarked. And “Long-term primary prophylaxis with voriconazole is a safe and effective option for the prevention of IFI in the allogeneic transplant population.”

Secondary Prophylaxis Study Findings

Regarding higher-risk SCT recipients with a previous fungal infection, Dr. Pagliuca also reviewed the recently published VOSIFI data (Haematologica 2010;95(10):1762-8), in which 45 patients with proven or probable IFI in the previous 12 months received prophylactic voriconazole within 48 hours post-conditioning chemotherapy for 100 days. Of the cohort, aspergillosis was probable in 58% and proven in 13% of patients.

At 12 months, the cumulative incidence rate of IFIs was 6.7% compared to an expected recurrence rate approaching 30%. Only 1 out of 45 patients died during follow-up and no cases of aspergillosis were observed following transplantation.

“VOSIFI was the first prospective trial of secondary prophylaxis in this high-risk population and results show that voriconazole appears to be safe and effective in protecting patients from recurrence of IFI following allogeneic SCT,” Dr. Pagliuca concluded.

Pros and Cons of Primary Prophylaxis

Although antifungals have been shown to protect against IFIs, initiation of primary prophylaxis needs to be clarified in the context of IFI risk and capacity of rapid diagnosis in a transplant centre. According to Dr. Marr, not all at-risk patients should receive primary prophylaxis. “There is a relative risk in using these drugs and we have to appreciate that that risk may exceed the risk for an IFI,” she stated.

Assessment of individual susceptibility to the toxicities of antifungals is key here, Dr. Marr stressed. If a patient already has hepatic abnormalities, then administering a drug that could potentially increase hepatotoxicity is an important variable.

Dr. Pagliuca suggested that all at-risk patients should receive primary prophylaxis but emphasized both the risk of exposing patients to mould infections and diagnostic capabilities. “It really is a philosophical argument, whether you are a ‘primary prophylactor’ or you are a ‘diagnostically-driven early treater,’” he said. If a centre has all the necessary technology and expertise to diagnose IFI early, “you might go down the latter route and be a diagnostically-driven early treater,” he noted. On the other hand, if a centre has neither the technology nor the expertise to intervene quickly when an IFI is suspected, “you may choose primary [prophylaxis],” he added.

An analysis of the most comprehensive fungal infection database in Europe confirms that galactomannan ELISA using either serum or bronchoalveolar lavage fluid is the most sensitive test available for the diagnosis of IA, the latter of which has a sensitivity in excess of 80% when using a cutoff of 0.5.