T2Candida MR as a predictor of outcome in patients with suspected invasive candidiasis starting empirical antifungal treatment: a prospective pilot study

Abstract

We assessed the potential role of T2Candida MR (T2MR) and serological biomarkers [β-d-glucan (BDG) or Candida albicans germ tube antibodies (CAGTA)], alone or in combination with standard cultures, for identifying patients with suspected invasive candidiasis (IC), who may benefit from maintaining antifungal therapy. Prospective observational multicentre study including all adult patients receiving empirical antifungal therapy for suspected IC, from January to June 2017. CAGTA, BDG and T2MR were determined at baseline and at +2 and +4 days after enrolment. Primary endpoint was the diagnostic value of CAGTA, BDG and T2MR, alone or in combination with standard culture, to predict diagnosis of IC and/or mortality in the first 7 days after starting antifungal therapy (poor outcome). Overall, 14/49 patients (28.6%) had a poor outcome (7 died within the first 7 days of antifungal therapy, whereas 7 ended with a diagnosis of IC). CAGTA [3/14 (21.4%) versus 8/35 (22.9%), P = 1] and BDG [8/14 (57.1%) versus 17/35 (48.6%), P = 0.75] results were similar in poor- and good-outcome patients. Conversely, a positive T2MR was associated with a higher risk of poor outcome [5/14 (35.7%) versus 0/35 (0.0%) P = 0.0001]. Specificity and positive predictive value of a positive T2MR for predicting poor outcome were both 100%, with a negative predictive value of 79.6%. After testing the combinations of biomarkers/standard cultures and T2MR/standard cultures, the combination of T2MR/standard cultures showed a high capacity to discriminate patients with poor outcome from those with good clinical evolution. T2MR may be of significant utility to identify patients who may benefit from maintaining antifungal therapy.