Lack of Cross-Hepatotoxicity between Voriconazole and Posaconazole

Tothe Editor—A patient with intracerebral aspergillosis developed progressive cholestatic hepatitis while receiving voriconazole therapy, which resolved after posaconazole was substituted for voriconazole. Spellberg et al. [1] previously reported a lack of cross-hepatotoxicity between fluconazole and voriconazole when treating a patient with coccidioidal meningitis. Our case report further demonstrates the possibility of the cautious use of different azoles in the presence of significant hepatotoxicity. A 70-year-old male patient with chronic lymphocytic leukemia, hypogammaglobulinemia, and steroid-induced type II diabetes developed multiple intracerebral abscesses. He declined a diagnostic brain biopsy and initiated empirical broad-spectrum antibiotic therapy with meropenem and vancomycin. Over the next 2 weeks, the patient developed bilateral endophthalmitis and subsequently underwent a vitrectomy of his left eye. A septate fungus was isolated from a vitreous fluid specimen, and combination antifungal therapy with liposomal amphotericin B and voriconazole was commenced. At this time, the patient's alkaline phosphatase (ALP) level was 126 U/L, and his γ-glutamyl transpeptidase (GGT) level was 76 U/L (figure 1). After 3 weeks of treatment, the patient's voriconazole concentration was 2.0 mg/L, which is within the reported therapeutic range [2]. Because of slow sporulation of the fungus, however, there was a delay in fungal identification. Four weeks after the vitrectomy, Aspergillus fumigatus was identified using a panfungal PCR assay targeting the internal transcribed spacer 1 region of the ribosomal DNA gene cluster and product sequencing [3]. After identification of the fungus, liposomal amphotericin B therapy was ceased, and the patient was discharged from the hospital, receiving oral voriconazole (200 mg twice daily). At hospital discharge, the patient's ALP level was 264 U/L, and his GGT level was 317 U/L. Changes in liver function test values after initiation of antifungal therapy. ALT, alanine transaminase; ALP, alkaline phosphatase; GGT, γ-glutamyl transpeptidase. Over the next 4 weeks, the patient developed an asymptomatic increase in his cholestatic enzyme levels. Both his ALP and GGT levels became markedly elevated, reaching peaks of 1201 U/L and 1657 U/L, respectively (figure 1). The patient's alanine transaminase level was mildly elevated (104 U/L), and his bilirubin level remained normal. There was no evidence of biliary duct obstruction on ultrasound. The only recently introduced medication still being received was voriconazole, and at this time, the patient's serum voriconazole concentration was 0.6 mg/L. Because of the persistent increase in liver function test values, voriconazole therapy was ceased 8 weeks after the commencement of antifungal therapy (including 4 weeks of voriconazole as the only antifungal therapy). The patient refused recommencement of parenteral antifungal therapy, and therefore, posaconazole therapy (400 mg twice daily) was initiated immediately in substitution of voriconazole therapy. After the change in therapy, the patient's ALP and GGT levels decreased steadily toward the baseline levels. After 4 weeks of posaconazole therapy, the patient's ALP level was 226 U/L, and his GGT level was 190 U/L. The patient's intracerebral aspergillosis has continued to improve clinically. Hepatotoxicity related to azoles can range from minor liver function abnormalities (reported, to some extent, in relation to all azoles) to life-threatening hepatitis. Voriconazole-related hepatotoxicity has more commonly been reported as transaminitis than as cholestasis and has occurred with a frequency of up to 12.4% in clinical [4] and randomized, controlled trials [5]. Liver function abnormalities are usually reversible after dosage reduction or drug cessation [4, 5]. Severe life-threatening hepatitis associated with voriconazole use is much less common and has been reported with a frequency of <1% in clinical trials [6]. There has been some evidence to support a weak correlation between hepatotoxicity and serum voriconazole levels, although the threshold plasma concentration of voriconazole has not been established [7]. The hepatotoxicity experienced by our patient did not correlate with his voriconazole plasma level. There is concern about a class hepatotoxicity effect when switching between azoles. Despite a cholestatic reaction related to voriconazole use in our patient, another azole, posaconazole, was successfully substituted without duplication of the same reaction. Potential conflicts of interest. H.F. and T.G.: no conflicts.