Performance of Candida albicans germ tube antibodies (CAGTA) and its association with (1 → 3)-β-D-glucan (BDG) for diagnosis daof invasive candidiasis (IC)

Abstract

Invasive candidiasis (IC) plays an important role as severe infection. Elder population, immunocompromised individuals, and intensive care unit (ICU) patients, especially when exposed to major surgery, are the most affected. IC diagnosis and treatment are difficult because of the absence of pathognomonic signs and symptoms. In addition, culture-based examination (gold standard) is known to have low sensitivity and long time to report. All these often lead to unnecessary and costly empirical antifungal therapies, burdened also by the onset of drug resistance and serious side effects for the patient. To partially overcome these problems, in recent years, novel noncultural markers have been investigated with the aim of easily and rapidly achieving an early diagnosis of IC. Such novel markers include the pan-fungal antigen (1 → 3)-β-D-glucan (BDG) and the anti–Candida albicans germ tube antibodies (CAGTAs). We retrospectively analyzed the presence of CAGTA on −80 °C stored serum samples, where the level of BDG had been previously assessed in a prospective study conducted in the Azienda Ospedaliero–Universitaria Policlinic of Modena (Pini et al. Infection 44:223–233, 2016). In particular, we selected 29 samples from proven IC episodes and 28 from non-IC cases. The 29 IC samples had been diagnosed as infections by C. albicans (n = 16), C. glabrata (n = 8), C. parapsilosis (n = 1), C. pelliculosa (n = 1), and C. tropicalis (n = 1), while 2 samples had intrasurgery biopsies positive for yeast (compatible with Candida spp.). The 28 control samples (non-IC) included 9 sera with positive blood cultures [E. faecium (n = 5), S. pneumoniae (n = 2), P. aeruginosa + A. baumannii (n = 2)] and 19 negative blood cultures. The CAGTA immunofluorescence assay was performed using 1:40, 1:80, 1:160, and 1:320 dilutions (reference dilution, as indicated by the manufacturer). According to the protocol, the samples were evaluated by the operator-dependent optical reading based on immunofluorescence positive/negative samples. In parallel, with the aim of standardizing the reading, the fluorescence images were captured, and the data were expressed as arbitrary fluorescence units (AFU). Finally, the results were interpreted as positive or negative using a cutoff provided by receiver operating characteristic (ROC) curves (Youden index). The traditional operator-dependent optical reading and the AFU measuring protocol provided comparable information with respect to the processed samples since IC and non-IC sera were correctly identified by the 2 CAGTA reading strategies in most of the cases. Interestingly, the AFU reading enabled a semiquantitative evaluation of the samples and an objective interpretation of the results. Based on the cutoff value, the AFU-based CAGTA procedure demonstrated a sensitivity of 52% and a specificity of 89%, while BDG showed a sensitivity of 90% and a specificity of 75%; the overall accuracy was 70% and 83% for CAGTA and BDG, respectively. The association of the 2 markers greatly increased both sensitivity and accuracy to 97% and 84%, respectively. As expected, when excluding non–C. albicans episodes, the sensitivity of CAGTA increased from 52% to 86%; moreover, with the exclusion of the non–deep-seated episodes, the sensitivity of CAGTA increased to 67% and reached 100% for C. albicans deep-seated candidiasis. Finally, when evaluating the influence of colonization, BDG demonstrated the most drastic decrease in specificity that dropped from 88% in noncolonized to 58% in colonized patients. With the exception of non–C. albicans episodes, CAGTA is a good marker of IC, particularly in the presence of deep-seated candidiasis. The performance of CAGTA greatly increases when used in combination with BDG.