Serum galactomannan (GM) assay for invasive aspergillosis (IA) in acute leukemia (AL) and hematopoietic stem cell transplantation (HSCT).

Abstract

6538 Background: IA is a major cause of mortality in AL and HSCT. The objectives of this study were to determine the yield of GM assay for the diagnosis of probable IA, its temporal relationship with high-resolution computed tomographic scan (HRCT) scans and correlation with mortality in patients of AL and recipients of HSCT. Methods: Consecutive neutropenic episodes among inpatients aged ≥ 15 years with a diagnosis of AL and recipients of HSCT were prospectively evaluated over 1½ years. All patients received oral itraconazole or thrice weekly IV amphotericin B (AmB) as prophylaxis. Weekly serum GM assay was performed for all subjects. Optical density index (ODI) > 0.5 for two or more samples was defined as positive. Patients with persistent fever for ≥ 96 hours on intravenous antibiotics were evaluated with HRCT chest and received empiric AmB. IA was diagnosed according to EORTC 2008 guidelines and treated with therapeutic doses of voriconazole or AmB. Results: Of the 150 episodes enrolled, AL induction constituted 51.3% (AML 35.3%, ALL 16%), high-dose cytarabine 20%, autologous HSCT 24.7%, allogeneic HSCT 2.7% and cladribine for hairy cell leukemia 1.3% episodes. Forty-three episodes (28.7%) were diagnosed with IA: possible 25 (16.7%), probable 17 (11.3%) and proven 1 (0.7%). The yield of GM assay in diagnosing probable IA was 17/42 (39.5%). False-positivity of a single test was 54.2% but decreased to 11.2% with two results. Assessment of sensitivity was impaired due to the lack of proven cases. The lone proven case had 1/1 sample positive (ODI 2.17). In 88.2% of probable IA episodes GM was positive before HRCT at a median of 10 days (range, 1-16) earlier, while HRCT was positive before GM in the remaining 11.8% episodes at median of 3.5 days (range, 1-7) earlier. Of the 23 fatal episodes (15.3%), 17 had a diagnosis of IA. Considering only those episodes with ≥2 samples tested, fatality was 31.0% in those with at least 2 values positive for GM, compared to 13.2% in the episodes with < 2 GM samples positive (OR 2.96, 95% CI 1.09 – 8.00; p=0.04). Conclusions: In a population of high-risk patients, GM assay could identify patients with probable IA earlier than HRCT chest and also predicted a higher risk of death.