An inhibition enzyme-linked immunosorbent assay (ELISA) capable of detecting 10 ng of aspergillus carbohydrate antigen/ml of serum was developed. When retrospectively applied to the sera of 19 patients with invasive aspergillosis, the ELISA detected antigen in 11 patients. None of 14 healthy controls or 28 patients with a variety of other infections were positive for circulating antigen. A rabbit model of invasive aspergillosis was also developed. Daily fungal cultures of blood were negative in the rabbits, as in human disease, but antigen was detected in increasing amounts up to the time of death. This ELISA is a sensitive, specific, and easily performed assay for circulating aspergillus antigen that should facilitate early diagnosis of invasive aspergillosis, heretofore, seldom made without invasive tests.

A modified enzyme-linked immunosorbent assay (ELISA) for crude carbohydrate antigen was used to evaluate the kinetics of aspergillus antigenemia and to determine the effect of therapy on circulating antigen levels in an experimental model. The ELISA was rapid, simple to perform, and able to detect less than 10 ng of antigen/mL of serum. The model was also used to evaluate the effect of temporary and persistent immunosuppression on experimental disease. Antigen levels rose progressively in untreated control rabbits; all 15 animals had significant antigenemia. Treated animals had markedly reduced antigen levels, but nine of 13 rabbits had detectable antigen after 72 h of therapy, a result that correlated with persistent disease. Therapy begun 24 h after challenge in temporarily immunosuppressed animals was more likely to sterilize tissues than was therapy begun 48 h after challenge. Therapy in persistently immunosuppressed rabbits was less effective and may require improved antifungal regimens to be successful.

An immunosuppressed rabbit model of invasive aspergillosis was used to evaluate a novel micellar preparation of cholesterol sulfate complexed to amphotericin B. The acute LD50 of amphotericin B-deoxycholate was 5.1 mg/kg versus 20 mg/kg for the amphotericin/cholesterol-sulfate complexes. Amphotericin B-deoxycholate given iv at a dose of 1.5 mg/kg was more effective in sterilizing liver and kidney than the amphotericin/cholesterol-sulfate complexes given iv at 1.5-4.5 mg/kg, but infection persisted in the lungs of all rabbits treated with those doses. Infection persisted even when the rabbits were given a lethal dose of amphotericin B-deoxycholate (4.5 mg/kg), but a dose of 15 mg/kg of the amphotericin/cholesterol-sulfate complexes sterilized tissues and was associated with no acute lethality. Equivalent doses of the amphotericin/cholesterol-sulfate complexes were less effective than amphotericin B-deoxycholate, but a fourfold decrease in acute lethality improved the therapeutic index of amphotericin B. The amphotericin/cholesterol-sulfate complexes appear to be an improved means of amphotericin B delivery and may improve therapy for invasive aspergillosis.

The utility of serum Aspergillus antigen in invasive aspergillosis was determined by identifying patients with >50 ng/mL Aspergillus carbohydrate antigen by ELISA. Patients were identified from a university hospital over a 65-month period. Nineteen patients with antigenemia had proven invasive aspergillosis, 16 had probable invasive aspergillosis, and 14 had an indeterminate diagnosis. There were 5 patients with false-positive results. Antigen levels were higher in disseminated infection than in invasive pulmonary aspergillosis (median levels, 500 and 121 ng/mL, respectively). Survival also correlated with antigenemia. Serial samples were obtained from 31 of 35 patients with proven or probable invasive aspergillosis. Fifteen of 19 patients with rising or persistent antigenemia died, whereas only 1 of 12 patients who cleared antigenemia died. Higher antigen levels were useful in predicting disseminated disease, and the course of antigenemia correlated with clinical outcome. Antigen detection may be a useful addition in the management of invasive aspergillosis.

Combination antifungal therapy was assessed in an immunosuppressed rabbit model of invasive aspergillosis. Treatment with fluconazole, amphotericin B, or a combination of both significantly prolonged survival of animals lethally challenged with Aspergillus fumigatus. High-dose amphotericin B was the most effective therapy for invasive aspergillosis. Although no antagonism was seen when fluconazole was given prophylactically or therapeutically in combination with amphotericin B, combination therapy did not augment the antifungal activity of amphotericin B. Animals given a sublethal challenge of A. fumigatus had lower mortality rates when given amphotericin B, fluconazole as treatment or prophylaxis, or various combination therapies. Only animals treated with flucytosine had mortality rates comparable to those of controls. No antagonism was observed with combinations of fluconazole and amphotericin B, flucytosine and amphotericin B, or fluconazole and flucytosine. These observations provide evidence that fluconazole, flucytosine, and amphotericin B used in various combinations are not antagonistic and may provide some insight into the treatment of invasive aspergillosis in humans.