Considerations, Challenges & Gaps for Future Research
In addition to the lack of robust clinical data, the major challenge associated with the pharmacoeconomics of empirical antifungal therapy is the accompanying absence of appropriate use of outcome measures. While the use of QALY is a gold standard in economic evaluation, allowing for multiple outcomes to be compiled into a single measure detailing quality and quantity of life generated by an intervention that can be compared easily across studies, QALY is not without its own methodological pitfalls. Given the relatively short duration of empiric antifungal therapy in which to calculate a QALY gain/loss, and the potential for confounding of this result by the underlying conditions of the patients being treated, QALY as an outcome measure in this setting may not be appropriate. The current use of a five-point composite outcome measure for empirical antifungal treatment has been beneficial, although the benefit of including fever resolution as a composite outcome measure remains debatable.
Inclusion of relevant clinical outcomes, whether utilizing a composite outcome or individual endpoints, is another important consideration, particularly when considering antifungal agents with different toxicity profiles. When formulating an economic evaluation, all relevant cost-driving adverse effects need to be considered and appropriate sources for these data need to be used for all the antifungal agents under evaluation.
The ECIL-1 in 2007 assessed the clinical practice of antifungal use in Europe. Ninety seven percent of clinicians at this conference considered that empiric therapy was standard practice and that the median time to commencing antifungal treatment was 5 days if it was the first febrile episode for a given patient, or 3 days if a relapse had occurred – that is, fever returned. Most suggested that there was a lack of evidence-based guidelines in this area and 84% thought there was a need for further clinical trials.
A notable gap until recently had been the limited data on evaluations comparing voriconazole and caspofungin. The most recent study by Al-Badriyeh et al. concluded that there was no significant difference in cost benefit between these two options. The data for this study, however, were obtained retrospectively. There is a need for large prospective and well-designed economic studies. Ideally, these studies should be part of any future clinical studies evaluating empirical antifungal therapy. To date, such studies are lacking. Another limitation relates to the fact that many existing economic studies are based on data from specific patient group(s) or populations. The issue at hand here is appropriate patient selection to reflect the general population encountered on a daily basis, and the appropriateness of generalizing the findings from the population used in existing studies.
Difficulty in accessing clinical data and being able to determine significant changes in each point of the composite outcome measure for empirical antifungal therapy, especially when using retrospective data, has been challenging. To determine if a breakthrough IFI has occurred, fungal cultures and serological or molecular tests are required with their associated limitations in turn-around-times; these data may not be readily accessible in a retrospective study as the tests may not have been available at the time of patient management. Furthermore, there may be a lack of thoroughly implemented clinical guidelines to standardize diagnostic testing.
Finally, there is a relative lack of pharmacoeconomic data for anidulafungin or micafungin as empiric antifungal agents. Future clinical and pharmacoeconomic studies should consider the inclusion of these agents. The use of combination antifungal therapy in highly symptomatic disease and/or high burden of disease is likely to occur. As such, both clinical and pharmacoeconomic analysis will be important in future studies of combination therapy. Published clinical and pharmacoeconomic data on combination empirical antifungal therapy are scant.