Clinical and technical evidence

A literature search was carried out for this briefing in accordance with the interim process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting

Published evidence

Four studies were identified that reported the use of Fungitell for antifungal treatment stratification. This included 1 prospective cohort study (Nucci et al. 2016), 2 retrospective cohort studies (Posteraro et al. 2016, Prattes et al. 2014) and a randomised parallel pilot study (Hanson et al. 2012) in a total of 4,708 patients. One systematic review and meta-analysis was selected for inclusion on the basis of providing pooled diagnostic accuracy outcomes solely on the Fungitell test from the most recent and largest number of studies (He et al. 2015). None of the studies in the meta-analysis overlaps with the 4 primary studies selected for review in this briefing.

Overall assessment of the evidence

Overall, the evidence on utility and resource outcomes from the use of the Fungitell test for antifungal treatment stratification is of limited quality. Only 1 study (Hanson et al. 2012) included a comparator group, and no outcomes were compared between the antifungal strategies. Two studies (Posteraro et al. 2016, Prattes et al. 2014) provided data from a real-world setting, but these were limited by their retrospective study design.

The primary studies applied strict inclusion criteria and had fairly small sample sizes. The studies were non-UK-based, primarily limited to adults at risk of candidaemia and single-centre studies. The study populations therefore may not be representative of the eligible population and results may not be generalisable to current NHS practice or other patient populations.

The systematic review and meta-analysis by He et al. (2015) reported the pooled diagnostic specificity (0.76; 95% confidence interval 0.74 to 0.78), which is a limitation of the Fungitell test. False-positive results could contribute to the unnecessary prescription of antifungal treatment, or its unnecessary continuation.

The current evidence base suggests that Fungitell has the potential to improve current care pathways for patients with invasive fungal diseases. However, there is a need for randomised controlled trials to answer the question of whether BDG-based antifungal strategies could benefit patients in terms of efficacy, exposure to antifungal therapy and costs compared with other untargeted treatments.

Table 2 summarises the clinical evidence as well as its strengths and limitations

Table 2 Summary of the selected studies

Nucci et al. (2016)

Study size, design and location

n=85 ICU patients at risk of developing candidaemia (57 BDG positive, 7 candidaemia, 21 negative). All patients were tested using blood cultures and BDG tests.

Multicentre prospective cohort study in 4 tertiary care hospitals, Brazil.

Intervention and comparator(s)

BDG (Fungitell)-based stopping of empirical antifungal treatment.

Three consecutive negative BDG results were considered negative for candidaemia, and antifungal therapy was stopped. BDG positive and candidaemia patients (confirmed by at least 1 positive blood culture) continued antifungal treatment.

Single arm, no comparator.

Key outcomes

All 21 patients with baseline negative BDG stopped antifungal therapy on day 4, none of whom developed recurrent candidaemia during the follow up period of 30 days. However, 3 patients received another antifungal treatment after day 4 of the study.

No patients developed recurrent candidaemia.

The median durations of antifungal therapy for the BDG-negative group, BDG positive group and candidaemia group were 3, 10 and 14 days respectively (p<0.001).

Strengths and limitations

The multicentre methodology provides a more generalisable population. Consecutive enrolment reduces the potential for selection bias. The outcomes were also prospectively defined.

The modified prediction rule resulted in more selective criteria and only 4% of patients in the ICUs were eligible.

Posteraro et al. (2016)

Study size, design and location

n=279 ICU patients at high risk of invasive candidiasis.

198 met the eligibility criteria and were included in study – 63 were BDG positive (47 candidaemia, 16 probable candidaemia) and 135 were BDG-negative.

Retrospective observational study in a tertiary care centre, Italy.

Intervention and comparator(s)

BDG (Fungitell) based diagnostic and therapeutic algorithm for antifungal treatment of invasive candida infection.

Candidaemia was confirmed by blood cultures (proven candidaemia) or positive biomarkers such as BDG or mannan/antimannan plus high-risk factors (probable candidaemia).

Single arm, no comparator.

Key outcomes

The median days of antifungal therapy in BDG positive patients was 10 compared to 5 in BDG-negative patients (p=0.04).

In the non-candidaemia group, 135/151 patients were BDG-negative, however qualified for empiric antifungal treatment. Of these, only 25 received antifungal therapy, which would have reduced antifungal use from an estimated 89.4% to 16.5% (difference = 72.9%). Of these 25 patients, 14 patients had delayed BDG results and received antifungal therapy until negative BDG results became available.

The therapeutic approach had little impact in the ICU mortality (112 out of 198 patients died with 21 deaths attributable to candida septicaemia).

Strengths and limitations

Large study population and retrospective analysis of prospectively collected data. This provides resource outcomes in a real-world setting for using Fungitell results to rule out invasive fungal disease.

Definitions for patients requiring or not requiring antifungal therapy were unclear. It is unclear whether study outcomes were defined prospectively or retrospectively.

Prattes et al. (2014)

Study size, design and location

n=66 adult patients with suspected invasive fungal infections.

Retrospective cohort study in a Medical University Hospital, Austria.

Intervention and comparator(s)

Measurement of BDG (Fungitell) in addition to routine diagnostic measures.

Single-arm study, no comparator.

Key outcomes

Antifungal therapy was started in 40 patients. BDG results led to stopping systemic antifungal therapy in 13 patients, none of whom developed IFIs.

In 26 patients, no antifungal therapy was started. BDG results led to starting antifungal therapy in 7 of these patients.

Overall, BDG results confirmed the initial clinical decision in 46 patients (27 receiving antifungal therapy, 19 without).

The test predicted 77% (10/13) of suspected, probable and proven IFI cases.

Strengths and limitations

The study provides some, although limited, clinical outcomes in a real-world setting for early stopping of antifungal therapy according to the BDG (Fungitell) results.

A higher cut-off was used to define a positive BDG result in comparison to the manufacturer's protocol (120 pg/ml vs 80 pg/ml). The study is limited in terms of the retrospective nature and low sample size, particularly in candidaemia cases. No repeat testing was performed, except on indeterminate results.

Hanson et al. (2012)

Study size, design and location

n=64 ICU patients at risk of invasive candidiasis (1 proven and 5 probable cases of invasive candidiasis).

Single centre, randomised, non-blinded parallel group pilot study in a medical ICU, US.

Intervention and comparator(s)

Pre-emptive antifungal therapy based on BDG (Fungitell) results (active surveillance group, n=47), compared with a standard care, empiric antifungal therapy group with physicians blinded to the BDG results (standard-care group, n=17).

Two subjects in the active surveillance group were excluded because of icteric sera.

Key outcomes

Fungitell performed best when 2 sequential specimens used the ≥80 pg/ml cut-off value for a positive result. It had overall sensitivity, specificity, positive and negative predictive values of 100%, 75%, 30% and 100% respectively.

Treatment with pre-emptive antifungal therapy had a significant effect on median glucan concentrations (p<0.001) with a quicker decline compared to no antifungal therapy (slope: 2.7 vs −0.2, p=0.06)

Twenty-one subjects received pre-emptive therapy for a median duration of 13 days. In all, 10 (48%) subjects experienced 15 adverse events that were possibly related to the drug. No serious drug-related adverse events were observed.

Strengths and limitations

Effort was made to randomise patients in a 3:1 ratio to the active surveillance and standard-care groups.

The sample size was not statistically based as no power calculations were performed. There were a small number of subjects with proven/probably invasive candidiasis primarily because of poor accrual. No comparisons were made between the active surveillance and standard-care groups. It was also evident that the pre-emptive protocol was not fully adhered to (86.7% of cases).

He et al. (2015)

Study size, design and location

n=4,214 subjects from 28 studies (n=2,821 from 18 studies were specific to Fungitell in 7 case-control and 11 cohort studies).

Systematic review and meta-analysis.

Intervention and comparator(s)

BDG tests, including Fungitell, Fungitec G-assay and Wako assay.

Reference standards included EORTC/MSG criteria, histopathologic examination and microbiological culture.

Key outcomes

The pooled sensitivity, specificity and diagnostic odds ratio for the Fungitell test were 0.75 (95% CI 0.71 to 0.79), 0.76 (95% CI 0.74 to 0.78) and 11.50 (95% CI 6.56 to 20.15) respectively. The pooled AUC-SROC was 0.8855. When stratified analysis was done based on assay type, the AUC-SROC was 0.8514.

The x2 and I2 tests for heterogeneity of the 18 studies were 67 (p<0.00001) and 75% respectively.

They reported that the cut-off value of 60 pg/ml had better diagnostic accuracy than the 80 pg/ml cut-off value (AUC-SROC 0.8973 vs 0.8726) for the Fungitell test.

Strengths and limitations

Methods were clearly described, reproducible and appropriate to the clinical question.

Studies included in the meta-analysis covered a range of different IFD diagnoses; therefore diagnostic accuracy results may not be generalisable to different types of infections. Pooled negative and positive predictive value results were not included. Only full-text and English-language studies were included, which may have led to an omission of relevant studies.

Abbreviations: AUC-SROC, area under the summary receiver operating curve; BDG, beta-D-glucan; CI, confidence interval; EORTC/MSG, European Organisation for Research and Treatment of Cancer/Mycoses Study Group, ICU, intensive care unit; IFD, invasive fungal disease; IFI, invasive fungal infection.

Recent and ongoing studies