← Back to July 2013

Universal Prophylaxis or Bio-marker Guided Antifungal Therapy: Which Road to Take?

links image
Orla Morrissey, MD
Alfred Health
Melbourne, AUSTRALIA

Invasive aspergillosis (IA) is a major cause of mortality in patients undergoing allogeneic stem cell transplantation (SCT) or chemotherapy for acute leukaemia, due mainly to the limited ability of culture and histology to make an accurate or early diagnosis. Consequently, clinicians rely on empiric antifungal therapy (EAFT) whenever IA is suspected or prevention through antifungal prophylaxis. Whilst these strategies have been successful in reducing IA-related mortality they have a number of limitations including drug interactions, emerging antifungal resistance, breakthrough invasive fungal infections (IFI) and overtreatment with expensive and toxic antifungal drugs.

New diagnostic tests including Aspergillus galactomannan (GM) enzyme-linked immunosorbent assay (ELISA) and PCR have been developed and have been variably incorporated into pre-emptive or biomarker-based strategies and compared to the strategy of administering antifungal therapy empirically in the setting of persistent fevers despite broad spectrum antibiotics.

Instead, we compared a bio-marker strategy (that used both Aspergillus GM-ELISA and PCR independently to direct antifungal therapy) to the traditional culture- and histology-based strategy to determine the impact on EAFT use, early diagnosis of IA and survival (1).

Adult patients undergoing allogeneic SCT or chemotherapy for acute leukaemia with no history of IFI were randomised 1:1 to the bio-marker strategy or culture- and histology-based strategy. Patients were followed for 26 weeks or until death, if earlier. Patients assigned to the bio-marker strategy had twice weekly Aspergillus GM-ELISA and PCR as in-patients and weekly as out-patients. The results of the assays determined the time of high-resolution computed-tomography (HRCT) scanning and whether antifungal therapy was given for probable or possible IA. For those assigned to the culture- and histology-based strategy cultures of blood, urine and sputum (if available) and faeces (if clinically indicated) in addition to, HRCT scan of chest were performed in those who had persistent fevers for 72-hours and antifungal therapy could be instituted empirically. Bronchoscopy and CT-guided or open lung biopsies were performed as per institutional protocols. The results of these tests determined the type of antifungal therapy that was continued. The primary end-point was the proportion of patients in each arm who had at least one course of EAFT during the 26 weeks of follow-up. As the trial was open-label the primary and mortality endpoints were adjudicated by an independent data review committee. Analysis was intention-to-treat and included all enrolled patients.

We found that our bio-marker strategy significantly reduced the amount of EAFT administered. The bio-marker strategy was capable of differentiating between those with persistent fevers who had IA and those with persistent fevers who didn't have IA, with certainty. The bio-marker strategy made significantly more diagnoses of IA. Patients in the culture- and histology-based strategy also had Aspergillus GM-ELISA and PCR testing performed but the results were withheld. Post-hoc analysis of the test results indicated that the incidence of IA was the same when Aspergillus GM-ELISA and PCR were used to diagnose IA in the culture- and histology-based strategy arm; indicating that these assay are more sensitive than culture and histology. Whilst the study was not powered to detect a significant mortality difference between the two arms, mortality rates were 31% lower in the bio-marker strategy arm. This was attributed to the ability of Aspergillus GM-ELISA and PCR to make an earlier diagnosis than culture and histology. Sub-group analysis was performed according to type of antifungal prophylaxis used. With voriconazole and posaconazole prophylaxis the significant reduction in EAFT use was no longer evident in the bio-marker strategy arm. Given that we found only one case of IA in those on voriconazole or posaconazole prophylaxis and that no cases of IA were diagnosed by radiological means only in the bio-marker strategy arm indicated that voriconazole and posaconazole are highly effective as prophylaxis. As a result we concluded that screening once to twice weekly with Aspergillus GM-ELISA and PCR is not needed in those on voriconazole or posaconazole prophylaxis.

So how do the findings of this study apply to the lung transplant population? Invasive aspergillosis is the most common fungal infection in this population and is also associated with high mortality rates. Two types of prophylaxis are used in lung transplant centres world-wide; namely universal and pre-emptive prophylaxis. Universal prophylaxis is administered to all lung transplant recipients for 3-6 months post transplantation. The regimen most commonly contains voriconazole targeting Aspergillus (2). This strategy is associated with unnecessary costs and drug-related toxicity since not all lung transplant recipients are at-risk or have the same risk of IA. This issue is all the more critical given the recently demonstrated association between voriconazole and squamous cell carcinoma in this population (3).

With the pre-emptive prophylactic approach, antifungal therapy is instituted based on the detection of a mould isolate (most commonly Aspergillus) in a surveillance bronchoscopy lavage (BAL) specimen (4). However, it has been reported that this strategy is associated with high rates of breakthrough IFI which may be due to the poor sensitivity of culture (5). Whilst it has been shown that GM-ELISA has poor sensitivity in serum in lung transplant recipients it has better sensitivity in BAL fluid (5). Furthermore, a positive BAL GM result is likely consistent with active infection given that GM is only released from growing hyphae and not dormant spores. A positive BAL Aspergillus PCR alone is most likely just colonisation with dormant spores and not active infection; thus, it may not require treatment. These assays incorporated into a pre-emptive strategy may allow us to refine and improve our ability to determine who should and more importantly, who shouldn't get antifungal prophylaxis.

Husain et al, reported at the most recent ISHLT meeting in Montreal (April 24-27 2013) that a pre-emptive strategy incorporating BAL GM-ELISA is safe and effective in targeting antifungal therapy and has utility in detecting true Aspergillus-related active infection. No adverse effect on survival was seen. Whilst this work represents a very important step in sorting out the precise role of these assays in the lung transplant setting a randomised controlled of a pre-emptive versus universal prophylaxis is urgently needed.

Disclosure Statement: Orla Morrissey has been a member of advisory boards for, received investigator-initiated grants from and given lectures for Gilead Sciences, Pfizer, Merck, Sharp and Dohme and Orphan Australia.


  1. Morrissey CO, Chen SC-A, Sorrell TC, et al. Galactomannan and PCR versus culture and histology for directing use of antifungal therapy for invasive aspergillosis in high-risk haematology patients: a randomised controlled trial. Lancet Infect Dis 2013; 13: 519-528.
  2. Neoh CF, Snell GI, Kotsimbos T, et al. Antifungal prophylaxis in lung transplantation-a worldwide survey. Am J Transplant 2011; 11: 3621366.
  3. Feist A, Lee R, Osborne S, et al. Increased incidence of cutaneous squamous cell carcinoma in lung transplant recipients taking long-term voriconazole. J Heart Lung Transplant 2012; 31: 1177-1181.
  4. Silveira FP, Husain S. Fungal infections in lung transplant recipients. Curr Opin Pulm Med 2008; 14: 211-218.
  5. Luong ML, Clancy CJ, Vadnerkar A, et al. Comparison of an Aspergillus real-time polymerase chain reaction assay with galactomannan testing of bronchoalveolar lavage fluid for the diagnosis of invasive pulmonary aspergillosis in lung transplant recipients. Clin Infect Dis 2011; 52: 1218-1226.

Share via:

links image    links image    links image