Pre-medication practices and incidence of infusion-related reactions in patients receiving AMPHOTEC®: data from the Patient Registry of Amphotericin B Cholesteryl Sulfate Complex for Injection Clinical Tolerability (PRoACT) registry

Background Clinical studies have suggested that rates of infusion-related reactions (IRRs) may be higher with amphotericin B colloidal dispersion (ABCD) versus other forms of amphotericin B. However, these studies did not permit the use of pre-medications upfront, which are now commonly used. Objectives To describe the use of pre-medications and determine the rate of IRRs in the real-world setting. Methods PRoACT, a multicentre, worldwide observational registry, captured real-world data about pre-medication practices and IRRs in patients receiving ABCD. Eligible patients were those beginning treatment with ABCD; treatment was according to the site's standard treatment practice. Incidence of IRRs was collected during the first 10 days of ABCD therapy. Clinical response data were collected 12 weeks after treatment start. Results One hundred and seventy patients from 21 worldwide sites were included (median age 37 years; 52% male). There were a total of 1230 ABCD infusions (mean dose 2.8 mg/kg/day); 90% of the infusions (1105/1230) had pre-medication. Common pre-medications included corticosteroids, antihistamines, paracetamol (acetaminophen) and metamizole. The overall IRR rate was 12% (147/1230) and was lower in infusions with pre-medication (11%) versus no pre-medication (22%), P < 0.001. Corticosteroids were associated with a decreased incidence of IRRs (P < 0.05), while paracetamol and antihistamines were not. The most common IRRs were chills (7%), fever (7%) and rigors (5%). Clearance of the fungal infection occurred in 52% of the participants. Conclusions These data suggest a lower rate of IRRs with ABCD than previously reported. Pre-medication is associated with decreased IRR incidence. Corticosteroids in particular appear to decrease IRRs while paracetamol and antihistamines, though commonly used, do not.