Bram Lestrade

Our studies have provided new insights into the epidemiology and clinical implications of azole resistance in invasive aspergillosis. The main findings were:

1. Triazole resistance has a major impact on the clinical management of patients with aspergillus disease; it is associated with excess mortality compared with azole-susceptible invasive aspergillosis.

2. Surveillance studies on resistance in aspergillosis vary greatly in their research procedures and reporting methods, thus hampering unification of data.

3. In the Netherlands a paradoxical trend towards lower voriconazole MICs has been found among azole resistant A. fumigatus strains.

A fumigatus is an omnipresent mould. Its spores are released in the air and inhaled. In healthy people, defensive mechanisms prevent the spores to grow out into fungal threads. However in the immune compromised patient they can invade the tissues of the lower respiratory tract. A fumigatus can mimic cancer; it is able to spread metastatically to other organs. It has a fatality rate of twenty to thirty percent. The preferred therapy for treatment is voriconazole. Resistance to this agent is increasingly found. Case reports point to a mortality of more than fifty percent, but direct control groups were not included in those studies. In this thesis, it was found that, compared to a carefully selected control group of patients with invasive aspergillosis with susceptible strains, the mortality in the group with voriconazole resistant strains was more than twenty percent higher.

The Netherlands has a relatively high resistance percentage. Since 2016 it is well above ten percent, showing peaks in 2017 and 2018 of almost fifteen percent. In 2017 our national treatment guidelines changed: monotherapy of voriconazole was no longer advised for invasive aspergillosis. A second agent was to be added to cover potentially resistant strains. At the time of initial diagnosis, it is usually not yet known what kind of strains are involved. Susceptibility testing takes several days and often we will not become informed at all. At first glance double therapy may seem logical, however the second agent can lead to unwanted side effects and it is less powerful in comparison with voriconazole. Resistance is not evenly distributed across hospitals and differs between departments as well. In this thesis it was confirmed again that surveillance of triazoles resistance has not yet become common practice. Not every hospital is engaged in a surveillance program. The depth of surveillance studies varies as well. Only a few hospitals are able to present resistance percentages at a detailed level: in which patient groups does it occur and to what extent? Patients in some regions may be unjustifiably receiving toxic double therapy. Conversely, if the resistance percentage of voriconazole is locally higher than expected, the value of voriconazole would become questionable. The choice of the second agent might also be dependent on the estimate of the percentage of intermediately sensitive strains: echinocandins for example are known to provide synergy in these conditions (based on animal experiments). In short, the choice of the best regimen depends on detailed local epidemiological knowledge. Therefore hospitals should invest time and energy in resistance surveillance.

Invasive aspergillosis is relatively rare. In academic centers it is found ten to thirty times a year, in other hospitals the frequency is usually much lower. In the Netherlands the general resistance percentage of triazoles fluctuates between ten and twenty percent. In other countries reported figures are usually well below ten percent. Due to its rarity, research on management strategies in invasive aspergillosis with resistant strains is a laborious process. Forces must be joined. As yet there is little unity in research approaches. Pathways already start to diverge with the definition of resistance. There are several kinds of triazoles and they do not always show cross-resistance. Resistance to a specific agent is not a black and white phenomenon, it comes in grades. A strain can show “intermediate” susceptibility to voriconazole while having a high degree of resistance to other triazoles. If such a patient is treated with voriconazole, should we consider him or her as a “victim” of resistance? The answer is unsure. In this thesis it has been shown that there is an increase of relatively voriconazole susceptible strains in the most common resistance genotype. Unfortunately the clinical significance of this phenomenon is not yet known. A pathogen may appear to be susceptible in laboratory conditions and nevertheless in clinical practice it may be unresponsive to therapy. Under the pressure of antifungals resistance might perhaps develop in strains that are already affected at a genetic level. Experts have proposed a resistance threshold of ten percent above which one should no longer treat with triazole monotherapy. However, the experts have not specified in which groups resistance should be measured precisely, nor have they specified criteria for resistance. In the Netherlands overall triazole resistance in 2018 was above ten percent in individual patients regardless of the underlying disease. However, “highlevel” resistance to voriconazole was still below that limit. New research is needed to show us if and under what conditions the use of voriconazole is safe when strains, despite altered genotypes, have relatively low resistance levels for voriconazole. To speed up discoveries (to gain sufficient power) it would be desirable if researchers tried to collect their data in a uniform manner and share them in their original form (without editing) so that we can build a coherent powerful data set, ultimately leading to a more refined resistance threshold.