Abstract Body

Talaromycosis (formerly penicilliosis) is an invasive fungal infection endemic in Southeast Asia, and is a leading cause of death in patients with advanced HIV disease. We have demonstrated in the recent IVAP trial the superiority of Amphotericin B over Itraconazole with regard to mortality, clinical response, and blood fungal clearance. Here, we investigated if early fungicidal activity in blood is a useful surrogate endpoint for all-cause mortality for HIV-associated talaromycosis.

IVAP trial enrolled 440 patients across 5 hospitals in Vietnam from October 2012 to December 2016. 391 (87%) patients who had at least 3 measurements of blood fungal load (in colony forming units [CFUs] per ml) during the first 2 weeks were included in the analysis. We used nonlinear regression to characterize individual fungal clearance dynamics with single and double exponential curves that were best fitted to the data. Individual parameters estimated from the model fit include: peak value of log fungal count (a), time to reach peak fungal load (c, days), and rate of fungal decline (br, log fungal count/ml/day). The parameters were used in a multivariate survival analysis to predict the risk of death over 6 months using Cox proportional hazard model, adjusting for age, antifungal treatment, antiretroviral status, intravenous drug use, baseline CD4 count, log fungal count at enrollment, and dyspnea requiring oxygen.

The median adjusted R² of the fitted curves is 0.94 (IQR: 0.79-1.00); 74.4% of fitted curves had an adjusted R² > 0.8. In the multivariate Cox regression model, older age (+5 years) and time to reach peak blood fungal load were independent predictors of death, HR=1.25, 95% CI: 1.07-1.47, p=0.005 and HR=1.23, 95% CI: 1.07-1.41, p = 0.002, respectively. The risk of death due to a delay in fungal clearance (of > 0 day)was statistically significant in patients who received Itraconazole, (HR=1.31, 95% CI: 1.12-1.53, p<0.001) but not in patients receiving Amphotericin B (HR=1.17, 95% CI: 0.86-1.58, p=0.313) (Figure).

Time to initiation of blood fungal clearance is an independent predictor of 6-month mortality and can potentially be used as an outcome measure for early phase clinical trials to efficiently assess novel therapeutic strategies for HIV-associated talaromycosis.