Optimising the use of high-cost antifungal agents in clinical practice

Abstract

Fungal infections are often associated with prolonged hospital stay and high mortality, and are costly to the healthcare system. The incidence of fungal infections has increased in tandem with the growing number of immunocompromised patients in recent years, underlining the need for novel antifungal agents. The 2000s were the most prolific decade for the development of antifungal drugs, with the introduction of newer triazoles (voriconazole, posaconazole) and echinocandins (caspofungin, micafungin, anidulafungin), providing a wider choice of treatment options. These newer antifungal agents, however, are much more expensive, and thus, significantly increase hospital drug expenditure. Relative health returns of these high-cost alternatives need to be determined given that healthcare services have limited resources. Accordingly, this thesis was geared towards deriving critical information to facilitate decision making by the healthcare providers in optimising the use of these high-cost antifungal agents for the prophylaxis and treatment of fungal infections. Three main areas were explored: voriconazole prophylaxis in lung transplant (LTx) recipients, cost-effective use of echinocandins in treating candidaemia and invasive candidiasis (IC) and the potential use of caspofungin eye drops for the treatment of fungal keratitis. Within the context of antifungal prophylaxis in lung transplantation, there is no consensus with respect to the choice of agent and strategy due to the paucity of data. An anonymous web-based survey was conducted to determine antifungal prophylactic practice in LTx centres world-wide. Findings from the survey indicated that voriconazole was the most commonly used agent for both universal and pre-emptive prophylaxis within the first six months post-LTx. A retrospective, single-centre, observational study was then undertaken to investigate the efficacy and safety of voriconazole pre-emptive prophylaxis in 62 LTx recipients. Six months after initiation of voriconazole prophylaxis, the majority (75.8%) had successful eradication of fungal colonisation; one patient developed proven fungal infection. Sixteen (25.8%) had died by 12 months after commencing prophylaxis, half due to Bronchiolitis Obliterans Syndrome. Chronic rejection within the 30 days prior to voriconazole prophylaxis was significantly associated with mortality at the 6-month and 12-month end-points. Ten patients (16.1%) had drug-related hepatotoxicity. The second focus was to determine the cost-effectiveness of echinocandins compared to other antifungal agents in treating candidaemia and IC. Accordingly, pharmacoeconomic comparisons between echinocandins (caspofungin, micafungin, anidulafungin), liposomal amphotericin B (LAmB) and fluconazole in the Australian setting were conducted. The comparisons employed decision analytical models based on treatment pathways derived from published randomised clinical trials, and utilised expert panels from Australia. Analyses were from a hospital perspective and cost inputs were obtained from the latest Australian resources. The robustness of the economic models was evaluated by sensitivity analyses. Hospitalisation was revealed as the major cost driver. Anidulafungin was a cost-effective option compared to fluconazole, with an incremental cost-effectiveness ratio of AU$22,003 per life-year saved. Micafungin was expected to have a total net cost-saving of AU$10,957 per patient over LAmB, whereas it was cost-equivalent to caspofungin, with variation in drug acquisition cost likely to be a critical factor. The final aim was to investigate the potential use of caspofungin eye drops as a novel strategy to treat fungal keratitis, with the intention of improving clinical outcomes and saving costs. The successful use of intrastromal and topical caspofungin 0.5% in combination with voriconazole in a case of refractory atypical Alternaria keratitis in a local institution was reported. The utility of caspofungin eye drops in fungal keratitis, however, remains limited, partly due to the lack of data on penetration into the human eye. A liquid chromatography/mass spectrometry assay was then developed and validated to determine the concentrations of caspofungin in human aqueous humour. An open-label study demonstrated that caspofungin 0.5% eye drops resulted in low penetration into the human eye in the absence of inflammation or corneal abrasion, but were generally well tolerated. The eye drops were shown to be stable for at least 28 days at 4 ± 1oC, affording economical use in clinical practice. This thesis has explored a number of important aspects related to the optimal and cost-efficient use of high-cost antifungal agents. Significantly, the data will facilitate informed decision making by clinical practitioners and healthcare policy makers about the use of these high-cost antifungal agents.