Abstract
The commercially developed PathoNostics AsperGenius species assay is a multiplex real-time PCR capable of detecting aspergillosis and genetic markers associated with azole resistance. The assay is validated for testing bronchoalveolar lavage fluids, replacing the requirement for culture and benefiting patient management. Application of this assay to less invasive, easily obtainable samples (e.g., serum) might be advantageous. The aim of this study was to determine the analytical and clinical performance of the AsperGenius species and resistance assays for testing serum samples. For the analytical evaluations, serum samples were spiked with various concentrations of Aspergillus genomic DNA for extraction, following international recommendations. For the clinical study, 124 DNA extracts from 14 proven/probable invasive aspergillosis (IA) cases, 2 possible IA cases, and 33 controls were tested. The resistance assay was performed on Aspergillus fumigatus PCR-positive samples when a sufficient fungal burden was evident. The limits of detection of the species and resistance assays for A. fumigatus DNA were 10 and ≥75 genomes/sample, respectively. Nonreproducible detection at lower burdens was achievable for all markers. With a positivity threshold of 39 cycles, the sensitivity and specificity of the species assay were 78.6% and 100%, respectively. For 7 IA cases, at least one genetic region potentially associated with azole resistance was successfully amplified, although no resistance markers were detected in this small cohort. The AsperGenius assay provides good clinical performance with the added ability to detect azole resistance directly from noninvasive samples. While the available burden will limit application, it remains a significant advancement in the diagnosis and management of aspergillosis.
INTRODUCTION
Several recent clinical studies have shown that PCR testing can aid in the diagnosis of invasive aspergillosis (IA), particularly when combined with antigen testing (1–3). Reductions in empirical therapy and an earlier diagnosis, leading to improved patient outcomes, have been shown (1–3). PCR testing has been hampered by the lack of commercially manufactured assays, resulting in limited acceptance and exclusion by disease-defining criteria (4). The development of commercial Aspergillus PCR systems will help standardize methodology, increase accessibility, and provide quality control assurance through commercial manufacture.
Several commercial PCR assays with the capacity to detect Aspergillus have been developed (MycAssay Aspergillus, Renishaw Fungiplex, Roche SeptiFast, AdemTech MycoGenie, and Abbott Plex-ID) (5–9). Many assays focus on detection in respiratory specimens, and while clinical performance looks favorable, data are compromised by limited numbers and sample types (7–9).
While potentially fatal, IA is a low-incidence disease, and different testing strategies have been applied. High-risk patients can be screened using sensitive biomarker assays, performed on easy-to-obtain specimens such as serum or plasma to exclude IA, preventing unnecessary empirical therapy. Alternatively, in symptomatic patients, a diagnostic route targeting respiratory (bronchoalveolar lavage [BAL] fluid) or biopsy samples can utilize the high specificity of the assays to confirm IA. Commercial assays require validation in more than one sample type to demonstrate robustness and ensure satisfactory performance in a range of testing strategies.
The recently developed PathoNostics AsperGenius species assay (PathoNostics, Maastricht, the Netherlands) is a multiplex real-time PCR assay that can detect Aspergillus fumigatus, A. terreus, and Aspergillus species, together with an internal control to monitor for sample inhibition, in a single reaction. It was first evaluated using BAL fluid samples for which it generated a sensitivity and specificity of 88.9% and 89.3%, respectively, in patients with hematological disorders (8). The assay contains a second multiplex reaction, the AsperGenius resistance assay, with the ability to detect the four mutations (TR34, L98H, Y121F, and T289A), representing the prevalent types L98H/TR34 and TR46/Y121F/T289A that are associated with resistance to azole antifungal drugs in the CYP51A gene of A. fumigatus. On testing of BAL fluid samples, the assay was able to determine mutation profiles for 8/11 proven/probable cases of IA directly from the clinical sample (8). Since cultures may be positive in <20% of cases of proven/probable IA and other assays do not have the capacity to determine resistance, this PCR assay has potential benefits for both the diagnosis and the management of patients with IA (3).
This current study was designed to establish the analytical and clinical performance of the AsperGenius species assay and to determine whether the circulatory burden was sufficient to detect markers of azole resistance in serum samples using standardized methods in line with international recommendations (10).
MATERIALS AND METHODS
Study design.
The study was designed to evaluate the performance of the AsperGenius assays in testing of serum samples and was performed in two parts: an analytical study to determine the limit of detection (LOD), linear range, and efficiency of amplification and a clinical study to determine performance (sensitivity/specificity, etc.) in testing of serum samples from a population with hematological disorders at high risk of IA.
Analytical study.
The analytical evaluation focused on A. fumigatus as the primary pathogen associated with IA. However, as A. terreus is identified by the assay, because of its intrinsic resistance to amphotericin B, a limited performance assessment was done. The analytical specificity (detection range/cross-reactivity) has been previously reported and was not investigated further (8). The A. fumigatus probe detected A. fumigatus and other members of the section Fumigati (e.g., A. lentulus), the A. terreus probe was specific for this species, and the generic Aspergillus probe detected the section Fumigati, A. terreus, A. niger, A. flavus, A. versicolor, and A. felis, with in silico sequence analysis predicting the detection of A. nidulans (8).
Two automated nucleic acid extraction systems were evaluated following the manufacturer's instructions. The Qiagen EZ1 Advance XL DSP virus kit was the primary platform for evaluating performance for the detection of both A. fumigatus and A. terreus. Performance for the detection of A. fumigatus was also determined using the bioMérieux easyMAG Generic 2.01 protocol. All nucleic acid extractions were eluted in 60 μl.
Simulated serum samples were prepared using sterile filtered defibrinated horse serum that was divided into 0.5-ml aliquots and spiked with various concentrations of A. fumigatus DNA (local clinical strain) to achieve final serum burdens of 10,000, 1,000, 500, 100, 75, 50, 25, 10, 5, and 1 genomes/0.5-ml sample. The number of replicates tested using the EZ1 platform is shown in Table 1. The EZ1 extractions were performed over four experiments; each time one serum aliquot was retained to provide a negative control. For the detection of A. fumigatus with the bioMérieux easyMAG protocol, all concentrations, including negative controls, were tested in triplicate.
TABLE 1.
Performance of the Pathonostics AsperGenius species assay in testing of serum samples (0.5 ml) spiked with various quantities of A. fumigatus genomic DNA and extraction using the Qiagen EZ1 DSP virus kit
| Fungal load (gea/sample) | Pathonostics AsperGenius target |
|||||||
|---|---|---|---|---|---|---|---|---|
|
A. fumigatus |
Aspergillus spp. |
A. terreus |
Internal control |
|||||
| No. of positives/total | Mean (SD) Cq | No. of positives/total | Mean (SD) Cq | No. of positives/total | Mean (SD) Cq | No. of positives/total | Mean (SD) Cq | |
| 10,000 | 3/3 | 25.17 (0.17) | 3/3 | 24.63 (0.05) | 0/3 | 3/3 | 28.6 (0.13) | |
| 1,000 | 3/3 | 28.16 (0.22) | 3/3 | 27.55 (0.12) | 0/3 | 3/3 | 30.1 (0.23) | |
| 500 | 3/3 | 29.27 (0.22) | 3/3 | 28.50 (0.19) | 0/3 | 3/3 | 29.3 (1.08) | |
| 100 | 5/5 | 32.32 (0.48) | 5/5 | 31.67 (1.64) | 0/5 | 5/5 | 31.0 (0.64) | |
| 75 | 5/5 | 33.32 (0.75) | 5/5 | 31.70 (0.62) | 0/5 | 5/5 | 33.4 (0.30) | |
| 50 | 5/5 | 33.15 (0.42) | 5/5 | 31.58 (0.41) | 0/5 | 5/5 | 31.7 (0.29) | |
| 25 | 5/5 | 35.10 (0.21) | 5/5 | 32.76 (1.86) | 0/5 | 5/5 | 31.4 (0.50) | |
| 10 | 7/8 | 37.39 (3.17) | 7/8 | 34.69 (1.42) | 0/8 | 8/8 | 31.0 (1.30) | |
| 5 | 7/8 | 38.62 (2.69) | 7/8 | 35.89 (1.58) | 0/8 | 8/8 | 32.1 (0.51) | |
| 1 | 3/8 | 41.37 (1.78) | 3/8 | 36.50 (1.28) | 0/8 | 8/8 | 30.8 (0.57) | |
| 0 | 0/4 | 0/4 | 0/4 | 4/4 | 30.5 (2.29) | |||
ge, genome equivalents.
Performance of A. terreus was determined using a limited range of fungal burdens: 50 (n = 1), 10 (n = 3), and 5 genome equivalents (ge)/0.5 ml (n = 3) plus a negative-control serum sample using the EZ1 platform.
To avoid airborne contamination, all manual processes required took place in a class II laminar flow cabinet.
Clinical study and patient population.
Serial DNA extracts from clinical serum samples from patients in whom fungal disease status had been previously defined using the revised European Organization for Research and Treatment of Cancer (EORTC) criteria were selected (4). All samples had been sent for PCR and antigen screening as part of the routine neutropenic care pathway after which DNA was stored at −80°C for quality control and performance assessment purposes (3). Once defrosted, DNA extracts were stored at 2 to 8°C until testing. The study was a performance assessment of the AsperGenius assay with an anonymous, retrospective case/control design and no impact on patient management and thus did not require ethical approval.
There were 16 cases of proven/probable/possible IA tested, including 14 cases (54 extracts) of proven/probable IA and 2 cases of possible IA (18 extracts); the median number of samples tested per case was 3 (range, 2 to 12). For cases of IA, attempts were made to select extracts obtained within ±2 weeks of radiological evidence supporting the diagnosis, although 33% of extracts were outside this time frame. Seventy extracts from 33 patients without fungal disease were included as controls; the median number of extracts tested per control patient was 2 (range, 2 to 4). Patient demographics are shown in Table 2.
TABLE 2.
Patient demographics and associated EORTC-MSG diagnosis of IAa
| Demographic | Proven/probable IA (n = 14) | Possible IA (= 2) | NEF (n = 33) |
|---|---|---|---|
| Male/female | 9/5 | 1/1 | 17/16 |
| Age (median [range]) (yr) | 60.5 (18–87) | 54 (38–70) | 55 (18–78) |
| Underlying condition (No.) | AML/MDS (9) | AML (2) | AML/MDS (16) |
| Lymphoma (1) | Lymphoma (6) | ||
| CML (1) | CML/CLL (4) | ||
| Myeloma (1) | Myeloma (3) | ||
| ALL (1) | ALL (1) | ||
| AA (1) | AA (1) | ||
| Other (2) | |||
| Allogeneic stem cell transplantation (No.) | Yes (4) | No (2) | Yes (21) |
| No (10) | No (12) | ||
| Fungal prophylaxis (No.) | Fluconazole (4) | Fluconazole (1) | Fluconazole (24) |
| Itraconazole (6) | None (1) | Itraconazole (4) | |
| AmBisome (1) | None (5) | ||
| None (3) | |||
| Fungal disease manifestation (No.) | Proven disseminated IA (1) | Possible IPA (2) | None |
| Probable disseminated IA (1) | |||
| Probable IPA (9) | |||
| Probable IPA/sinusitis (3) |
AA, aplastic anemia; AML, acute myeloid leukemia; ALL, acute lymphoblastic leukemia; CML, chronic myeloid leukemia; CLL, chronic lymphoblastic leukemia; MDS, myelodysplastic syndrome; Lymphoma, Hodgkin's, non-Hodgkin's lymphoma, and diffuse large B cell lymphoma; MSG, Mycoses Study Group; Other, other nonspecified hematological malignancy including myelofibrosis and a nonspecified hematological malignancy; IA, invasive aspergillosis; IPA, invasive pulmonary aspergillosis; sinusitis, Aspergillus sinusitis.
DNA was extracted from 0.5 ml of serum using the Qiagen EZ1 advance XL DSP virus kit, following the manufacturer's instructions, and was eluted in 60 μl. Positive (serum sample containing 10 ge of A. fumigatus DNA) and negative (serum sample only) extraction controls were included in each run. At the time of extraction, a well-validated in-house and an internal control PCR were performed (11). A galactomannan enzyme immunoassay (GM-EIA) (Bio-Rad, United Kingdom) was performed, following the manufacturer's instruction and using a positivity index of 0.5.
PathoNostics AsperGenius PCR amplification.
For both the analytical and clinical studies, the AsperGenius species and resistance PCRs were performed on the Qiagen RotorGene Q high-resolution melt instrument, following the manufacturer's instructions and using 5 and 10 μl of DNA template in a final reaction volume of 25 μl for the species and resistance assays, respectively.
Statistical evaluation.
Analytical analysis of the AsperGenius species PCR for testing of serum samples involved determining the LOD (defined as the lowest concentration with a 100% reproducibility of detection), the range of DNA concentrations over which amplification was linear, and the efficiency of PCR amplification over this range, calculated from the gradient of the slope using the following equation: 10−1/slope − 1. For the AsperGenius resistance PCR, only the LOD was determined. Further analysis was performed to correlate AsperGenius species and resistance performance so that the quantification cycle (Cq) value generated by the A. fumigatus assay could be used as a guide to the likelihood of success when the resistance assay is performed. All of the calculated analytical parameters are influenced by the extraction process required to retrieve DNA from the serum sample and may vary from parameters calculated on the basis of dilutions of Aspergillus genomic DNA.
To determine the clinical accuracy of the AsperGenius species PCR results, the positivity rates in samples originating from cases were compared to the false-positivity rates in control samples. To determine the clinical performance (sensitivity, specificity, positive and negative likelihood ratios, and diagnostic odds ratio) of the AsperGenius species assay, 2 × 2 tables were constructed, using both proven/probable invasive fungal disease (IFD) and proven/probable/possible IFD as true cases and patients with no evidence of fungal disease (NEF) as the control population. For all patients, multiple (≥2) samples were tested. To provide performance data relevant for both screening strategies to exclude disease and diagnostic approaches to confirm disease, a patient was considered to be PCR positive when either a single PCR-positive result or multiple (≥2) PCR-positive results were found. Given the case-control study design and the artificially high prevalence of proven/probable IA (28.6%), predictive values were not used. For each proportionate value, 95% confidence intervals (CIs) and, when required, P values (Fisher's exact test; P = 0.05) were generated to determine the significance of the difference between rates. Receiver operating characteristic (ROC) curve analysis was performed for positive clinical samples to identify an optimal crossing point threshold (Cq) when PCR positivity was determined.
RESULTS
Analytical performance of the AsperGenius species assay.
When A. fumigatus genomic DNA extracted from serum samples was detected using the Qiagen EZ1 DSP virus kit, the LOD for both the A. fumigatus and Aspergillus species assays was 25 genomes/sample. At 10 and 5 genomes/sample, the reproducibility was 87.5% (95% CI, 52.9 to 97.8), and at 1 genome/sample, the reproducibility was 37.5% (95% CI, 13.7 to 69.4) (Table 1). Use of the bioMérieux easyMAG protocol for DNA extraction lowered the LOD to 10 genomes/sample, although reproducibility at 5 genomes/sample and 1 genome/sample was 33.3% (1/3; 95% CI, 6.2 to 79.2), and no false positivity was noted. The reproducibility of detection of A. terreus genomic DNA for both the A. terreus and Aspergillus species assays was 100% (3/3; 95% CI, 42.9 to 100) for all concentrations (50, 10, and 5 genomes/sample) extracted from serum samples using the Qiagen EZ1 kit (data not shown).
For the A. fumigatus assay, amplification was linear across the entire range (1 to 10,000 genomes/sample) of testing of DNA extracted by both the easyMAG and EZ1 assays (Fig. 1A). The PCR efficiency using DNA extracted from serum samples using the EZ1 kit was 72.6% compared to 97.1% for DNA extracted by the easyMAG assay, and mean Cq values were obtained significantly more quickly using easyMAG extracts (P = 0.0018).
FIG 1.
Standard curves for the PathoNostics AsperGenius assay. A. fumigatus (A) and Aspergillus species (B) assay testing A. fumigatus genomic DNA extracted from serum samples by the Qiagen EZ1 and bioMérieux easyMAG automated extractors.
For the Aspergillus species assay, when DNA extracted from samples loaded with A. fumigatus genomic DNA was tested by both the easyMAG and EZ1 assays, the amplification was linear from 5 to 10,000 genomes/sample (Fig. 1B). The PCR efficiency using DNA extracted from serum samples by the EZ1 kit was 106% compared to 124% for DNA extracted by the easyMAG assay, and mean Cq values were obtained significantly more quickly using easyMAG extracts (P = 0.0013).
Analytical performance of the AsperGenius resistance assay.
The 100% LOD for resistance markers were 75, 100, 500, and 75 genomes/sample for L98H, T289A, TR34, and Y121F, respectively (Table 3). However, nonreproducible detection was achievable for all markers at 10 genomes/sample, and for Y121F and T289A, detection to 5 genomes/sample was possible. For TR34 and Y121F, successful amplification occurred on one occasion when DNA extracts from a sample containing 1 genome were tested, although the subsequent melt curve did not provide a definitive melt temperature, suggesting that nonspecific amplification had occurred. Typical melt temperature curves are shown in Fig. 2. Predictably, with greater fungal burdens, the quality of the melting curves improved, although at lower fungal burdens when detectable amplification was apparent, the melting curves were easily discernible for all markers, apart from Y121F (Fig. 2c).
TABLE 3.
Performance of the Pathonostics AsperGenius resistance assay in testing of serum samples (0.5 ml) spiked with various quantities of A. fumigatus genomic DNA and extraction using the Qiagen EZ1 DSP virus kit
| Fungal load (gea/serum) | L98H |
T289A |
TR34 |
Y121F |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Reproducibility | Mean Cq | Mean Tm | Reproducibility | Mean Cq | Mean Tm | Reproducibility | Mean Cq | Mean Tm | Reproducibility | Mean Cq | Mean Tm | |
| 10,000 | 3/3 | 29.7 | 61.4 | 3/3 | 28.2 | 63.0 | 3/3 | 29.8 | 63.9 | 3/3 | 25.9 | 62.9 |
| 1,000 | 3/3 | 32.6 | 61.3 | 3/3 | 31.6 | 62.9 | 3/3 | 32.8 | 64.2 | 3/3 | 28.7 | 62.8 |
| 500 | 3/3 | 33.4 | 61.6 | 3/3 | 32.1 | 63.1 | 3/3 | 34.6 | 64.5 | 3/3 | 29.5 | 63.1 |
| 100 | 5/5 | 37.7 | 61.0 | 5/5 | 36.2 | 62.7 | 4/5 | 37.1 | 64.4 | 5/5 | 34.9 | 63.6 |
| 75 | 5/5 | 38.7 | 61.2 | 3/5 | 36.2 | 62.9 | 4/5 | 37.5 | 64.4 | 5/5 | 35.5 | 63.6 |
| 50 | 3/5 | 38.9 | 60.9 | 5/5 | 36.4 | 62.7 | 1/5 | 38.0 | 64.0 | 4/5 | 34.8 | 63.5 |
| 25 | 1/5 | 38.4 | 61.0 | 0/5 | 1/5 | 39.6 | 64.5 | 1/5 | 36.1 | 62.7 | ||
| 10 | 1/8 | 37.1 | 62.0 | 1/8 | 36.5 | 63.0 | 1/8 | 36.7 | 65.0 | 1/8 | 37.8 | 63.0 |
| 5 | 0/8 | 3/8 | 36.2 | 63.1 | 0/8 | 3/8 | 36.7 | 63.3 | ||||
| 1 | 0/8 | 0/8 | 1/8 | 45.0 | 1/8 | 38.5 | ||||||
ge, genome equivalents.
FIG 2.
Representative melt temperature (Tm) peaks for the AsperGenius assay. Shown are the L98H mutation (a), T289A mutation (b), Y121F mutation (c), and TR34 mutation (d) from simulated serum samples with burdens ranging from 10,000 to 1 Aspergillus fumigatus genome using the Qiagen Rotor-Gene Q high-resolution melt instrument. For each curve, Bin A represents the Tm for the specific mutation, while Bin B represents the Tm for DNA extracted from simulated serum samples containing A. fumigatus genomic DNA with the wild-type sequence. The reproducibility of detection at various concentrations of DNA is shown in Table 3. For each target, the lowest genomic burden generating a distinguishable Tm peak is indicated by an arrow and associated burden.
This analytical performance was used to determine a minimum fungal burden in a serum sample that might permit successful amplification of the regions containing the potential resistance markers. For reproducible detection of these markers, a burden of >75 genomes/sample was required, corresponding to a Cq value of ≤33 cycles when detected by the A. fumigatus assay. Nonreproducible detection of resistance markers could be expected when burdens between 10 and <75 genomes/sample, corresponding with Cq values between >33 and <38 cycles, were tested with the A. fumigatus assay.
Clinical performance.
The true sample positivity rates, representing the number of AsperGenius positive results associated with extracts originating from proven/probable cases, were 46.3% (25/54; 95% CI, 33.7 to 59.4) and 48.1% (26/54; 95% CI, 35.4 to 61.2) for the A. fumigatus and Aspergillus species targets, respectively. Twenty-four of the 25 A. fumigatus results and 24/26 of the Aspergillus species assay results were concomitantly positive. For samples taken within ±2 weeks of radiological evidence, the positivity rates were 56.4% (22/39; 95% CI, 41.0 to 70.7) and 59.0% (23/39; 95% CI, 43.7 to 72.9) for the A. fumigatus and Aspergillus species targets, respectively. Eighty-eight percent of both the positive A. fumigatus and Aspergillus species results were within 2 weeks of radiological confirmation, with 60% and 61.5% positive within 1 week. The false positivity rates, representing the number of positive results associated with extracts originating from controls, were 1.4% (1/70; 95% CI, 0.25 to 7.7) and 4.3% (3/70; 5% CI, 1.5 to 11.9) for the A. fumigatus and Aspergillus species targets, respectively. One A. fumigatus false positive was also positive by the Aspergillus species assay. No samples (n = 18) from possible patients (n = 2) were positive by either target.
For both A. fumigatus and Aspergillus species assays, the true positivity for proven/probable IA cases even when possible cases are included, was significantly greater than the false positivity associated with the control population (P = < 0.0001 for all instances).
There was one case of non-A. fumigatus disease detected by the Aspergillus species PCR, but no positive results were generated by the A. terreus assay.
The performance of the AsperGenius species assay is shown in Table 4. The characteristics described represent a combination of the results for the A. fumigatus- and the broad-range Aspergillus species assays; as in a clinical scenario, a positive result in either assay carries significance. Currently, there is no specified threshold for the AsperGenius assays when serum samples are tested, and the data in Table 4 are calculated without a positivity threshold.
TABLE 4.
Clinical performance of the AsperGenius species assay in testing of serum samples from hematology patients with proven/probable IA (n = 14), possible IA (n = 2), and no evidence of fungal disease (n = 33)
| Parametera | Results for population: |
|||
|---|---|---|---|---|
| Proven/probable vs NEF |
Proven/probable/possible vs NEF |
|||
| Single positive threshold | Multiple (≥2) positive thresholds | Single positive threshold | Multiple (≥2) positive thresholds | |
| Sensitivity (n/N, % [95% CI]) | 11/14, 78.6 (52.4–92.4) | 9/14, 64.3 (38.8–83.7) | 11/16, 68.8 (44.0–85.8) | 9/16, 56.3 (33.2–76.9) |
| Specificity (n/N, % [95% CI]) | 30/33, 90.9 (76.4–96.9) | 33/33 100, (89.6–100) | 30/33, 90.9 (76.4–96.9) | 33/33 100, (89.6–100) |
| LR +tive | 8.64 | >643b | 7.56 | >563b |
| LR -tive | 0.23 | 0.36 | 0.34 | 0.44 |
| DOR | 37.6 | >1,786b | 22.2 | >1,279b |
LR +tive, positive likelihood ratio; LR -tive, negative likelihood ratio; DOR, diagnostic odds ratio.
To overcome infinity, the parameter was determined using a specificity value of 99.9%.
The mean Cq values for true-positive samples were 38.1 (SD, ±3.2) and 34.4 cycles (SD, ±1.5) for the A. fumigatus and species assays, respectively. The mean Cq values for false positives were 41.6 (n = 1) and 40.8 (SD, ±2.3), for the respective assays and were later than Cq values for true positives, although the numbers were limited. The ROC curve analysis showed a threshold of 39 cycles to be optimal, and using it retained the overall sensitivity of 78.6%, but increased the specificity to 100%.
The internal control target was successfully amplified for all samples (mean Cq, 31.6; SD, ±1.7), although for 15 samples, the Cq value was ≥2 cycles later than the mean value. Five of these samples were AsperGenius PCR positive, and the delay in the internal control Cq was likely a result of a competitive PCR. The other 10 samples were all AsperGenius PCR negative, and the delay might represent a degree of PCR inhibition, although further analysis is required to confirm the expected range of Cq values when serum samples are tested.
Determining resistance in clinical serum samples.
With use of the Cq guidelines as determined in the analytical study of the AsperGenius resistance assay, 12 samples (7 proven/probable IA cases) had positive A. fumigatus PCRs with Cq values indicating sufficient DNA for a resistance PCR to be performed. Seven of the 12 samples showed successful amplification for at least one of the four loci targeted, and 19 regions were successfully amplified (6/12 L96H, 5/12 T289A, 2/12 TR34, and 6/12 Y121F). All 7 patients had at least 1 region associated with resistance amplified, 4 patients had 3/4 regions amplified, 2 patients had 2 targets amplified, and a single patient had only 1 target successfully amplified. Of the patients with proven/probable IA in this study, 50% had at least 1 region associated with azole resistance successfully amplified directly from serum, although no mutations associated with azole resistance were detected.
DISCUSSION
Treatment options for IA are limited, with triazoles (voriconazole) considered the first-line therapy. The development of azole resistance is a major concern, potentially causing treatment failure or a delay in effective therapy that can result in increased mortality rates (12). Difficulties in the diagnosis of IA have resulted in the use of empirical or prophylactic strategies, resulting in overuse of antifungal agents and promotion of resistance.
To prevent overuse of antifungals, diagnostic performance needs to be improved. Several recent studies have shown the benefits of nonculture tests (GM-EIA and Aspergillus PCR) for enhanced diagnosis of IA, with high sensitivity reducing antifungal expenditure and combined positivity providing a degree of etiological certainty (1–3). While these strategies overcome the performance limitations of the classical diagnosis, they fail to deliver information concerning antifungal susceptibility, previously only available on the limited occasions when Aspergillus isolates were cultured. The development of a commercially manufactured assay that not only detects and differentiates potentially resistant Aspergillus species but also allows the user to infer some information with respect to azole susceptibility in A. fumigatus may be of major clinical benefit.
The first article describing the performance of the PathoNostics AsperGenius species and resistance assays showed excellent performance for the testing of BAL fluid specimens (8). To date, direct-from-specimen amplification of CYP51 regions potentially associated with azole resistance has been limited to BAL fluid, cerebrospinal fluid (CSF), or tissue specimens (8, 13–15). If the use of less invasive samples (e.g., blood) could provide similar performance, then the utility of the assay might be enhanced. In the study of Spiess and colleagues (13), no amplification was possible for DNA extracted from 25 blood samples that were PCR positive by the generic Aspergillus assay.
The current article describes the first successful attempt to determine azole resistance in testing of serum samples and utilizes a commercially available real-time PCR, obviating the need to perform postamplification processes, such as DNA sequencing.
Assay positivity was associated with cases, and the false-positive rates were low (4.3%). With a positivity threshold of 39 cycles, sensitivity and specificity were 78.6% (95% CI, 52.4 to 92.4) and 100% (95% CI, 89.6 to 100), respectively, providing a confident means of diagnosing IA. While the sensitivity was comparable to that for the previous meta-analyses of Aspergillus PCR for testing of blood samples, the specificity was superior, possibly because the meta-analyses included both whole blood and serum/plasma PCR data, whereas this study only utilized serum, previously shown to provide better specificity (16–19). The sensitivity of the assay is likely to improve through prospective testing, as has been shown for other commercial Aspergillus PCR assays (5, 20). No cases of A. terreus disease were available for testing, and further evaluation is required for this and other non-A. fumigatus species. For the clinical evaluation, all DNA extractions were performed using the Qiagen EZ1 assay, as is routine practice, resulting from previous contamination associated with the bioMérieux easyMAG platform (21). In the current analytical study, the easyMAG platform was evaluated to increase methodological diversity when the AsperGenius assay was used, and no contamination was evident, although numbers were limited. The DNA extracted using the easyMAG platform improved PCR efficiency and generated Cq values significantly more quickly, as noted previously (21). It is possible that if the previous easyMAG assay contamination was temporal or batch related, then the clinical performance of the AsperGenius assay will improve if DNA was extracted using this platform.
Successful amplification of regions associated with azole resistance directly from serum samples was achieved in 50% of proven/probable cases. The likelihood of success can be guided by the Cq value of the A. fumigatus assay, and this prevents the wastage of reagents, DNA extracts, and time. Of the four resistance regions targeted by the assay, four patients had at least three successfully amplified. In addition, resistant types can be identified with only one amplified mutation per resistance type. Although not 100% reproducible, the detection of azole resistance markers direct from serum samples is possible, and reproducibility could be improved by testing multiple replicates of each extract. Detection could be further enhanced by extracting DNA from larger initial sample volumes (≥1 ml) by using greater DNA template volumes (>10 μl) and through prospective evaluation.
In summary, the real-time AsperGenius assay provides good clinical performance for the detection of IA with the added ability to detect azole resistance in A. fumigatus directly from noninvasive samples. While the limited burden available in circulatory samples will prevent 100% successful amplification of potential resistance markers, early resistance screening may benefit some patients and directly targeted BAL fluid testing may overcome the limitations. By combining a commercial real-time PCR assay with internationally developed recommendations for DNA extraction, a standardized approach can be achieved across centers to provide essential large-scale prospective validation and potentially represents a significant advancement in the diagnosis and management of IA.
ACKNOWLEDGMENTS
P.L.W. is a founding member of the EAPCRI, received project funding from Myconostica, Luminex, and Renishaw Diagnostics Limited, was sponsored by Myconostica, MSD, and Gilead Sciences to attend international meetings, and provided consultancy for Renishaw Diagnostics Limited. R.A.B. is a founding member of the EAPCRI, received an educational grant and scientific fellowship award from Gilead Sciences and Pfizer, is a member of the advisory board and speaker bureau for Gilead Sciences, MSD, Astellas, and Pfizer, and was sponsored by Gilead Sciences and Pfizer to attend international meetings. R.B.P. declares no conflicts of interest.
REFERENCES
- 1.Morrissey CO, Chen SC, Sorrell TC, Milliken S, Bardy PG, Bradstock KF, Szer J, Halliday CL, Gilroy NM, Moore J, Schwarer AP, Guy S, Bajel A, Tramontana AR, Spelman T, Slavin MA. 2013. Galactomannan and PCR versus culture and histology for directing use of antifungal treatment for invasive aspergillosis in high-risk haematology patients: a randomised controlled trial. Lancet Infect Dis 13:519–528. doi: 10.1016/S1473-3099(13)70076-8. [DOI] [PubMed] [Google Scholar]
- 2.Aguado JM, Vázquez L, Fernández-Ruiz M, Villaescusa T, Ruiz-Camps I, Barba P, Silva JT, Batlle M, Solano C, Gallardo D, Heras I, Polo M, Varela R, Vallejo C, Olave T, López-Jiménez J, Rovira M, Parody R, Cuenca-Estrella M. 2015. Serum galactomannan vs a combination of galactomannan and PCR-based Aspergillus DNA detection for early therapy of invasive aspergillosis in high-risk hematological patients: a randomized controlled trial. Clin Infect Dis 60:405–414. doi: 10.1093/cid/ciu833. [DOI] [PubMed] [Google Scholar]
- 3.Barnes RA, Stocking K, Bowden S, Poynton MH, White PL. 2013. Prevention and diagnosis of invasive fungal disease in high-risk patients within an integrative care pathway. J Infect 67:206–214. doi: 10.1016/j.jinf.2013.04.020. [DOI] [PubMed] [Google Scholar]
- 4.De Pauw B, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE, Calandra T, Pappas PG, Maertens J, Lortholary O, Kauffman CA, Denning DW, Patterson TF, Maschmeyer G, Bille J, Dismukes WE, Herbrecht R, Hope WW, Kibbler CC, Kullberg BJ, Marr KA, Muñoz P, Odds FC, Perfect JR, Restrepo A, Ruhnke M, Segal BH, Sobel JD, Sorrell TC, Viscoli C, Wingard JR, Zaoutis T, Bennett JE, European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group, National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. 2008. Revised definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis 46:1813–1821. doi: 10.1086/588660. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.White PL, Perry MD, Moody A, Follett SA, Morgan G, Barnes RA. 2011. Evaluation of analytical and preliminary clinical performance of Myconostica MycAssay Aspergillus when testing serum specimens for diagnosis of invasive aspergillosis. J Clin Microbiol 49:2169–2174. doi: 10.1128/JCM.00101-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.White PL, Hibbitts SJ, Perry MD, Green J, Stirling E, Woodford L, McNay G, Stevenson R, Barnes RA. 2014. Evaluation of a commercially developed semi-automated PCR-SERS assay for the diagnosis of invasive fungal disease. J Clin Microbiol 52:3536–3543. doi: 10.1128/JCM.01135-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Torelli R, Sanguinetti M, Moody A, Pagano L, Caira M, De Carolis E, Fuso L, De Pascale G, Bello G, Antonelli M, Fadda G, Posteraro B. 2011. Diagnosis of invasive aspergillosis by a commercial real-time PCR assay for Aspergillus DNA in bronchoalveolar lavage fluid samples from high-risk patients compared to a galactomannan enzyme immunoassay. J Clin Microbiol 49:4273–4278. doi: 10.1128/JCM.05026-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chong GL, van de Sande WW, Dingemans GJ, Gaajetaan GR, Vonk AG, Hayette MP, van Tegelen DW, Simons GF, Rijnders BJ. 2015. Validation of a new Aspergillus real-time PCR assay for direct detection of Aspergillus and azole resistance of Aspergillus fumigatus on bronchoalveolar lavage fluid. J Clin Microbiol 53:868–874. doi: 10.1128/JCM.03216-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Huttner A, Emonet S, Harbarth S, Renzi G, Kaiser L, Schrenze J. 2014. Polymerase-chain reaction/electrospray ionization-mass spectrometry for the detection of bacteria and fungi in bronchoalveolar lavage fluids: a prospective observational study. Clin Microbiol Infect 20:O1059−O1066. doi: 10.1111/1469-0691.12749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.White PL, Mengoli C, Bretagne S, Cuenca-Estrella M, Finnstrom N, Klingspor L, Melchers WJ, McCulloch E, Barnes RA, Donnelly JP, Loeffler J, European Aspergillus PCR Initiative (EAPCRI). 2011. Evaluation of Aspergillus PCR protocols for testing serum specimens. J Clin Microbiol 49:3842–3848. doi: 10.1128/JCM.05316-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.White PL, Linton CJ, Perry MD, Johnson EM, Barnes RA. 2006. The evolution and evaluation of a whole blood polymerase chain reaction assay for the detection of invasive aspergillosis in hematology patients in a routine clinical setting. Clin Infect Dis 42:479–486. doi: 10.1086/499949. [DOI] [PubMed] [Google Scholar]
- 12.van der Linden JWM, Snelders E, Kampinga GA, Rijnders BJA, Mattsson E, Debets-Ossenkopp YJ, Kuijper EJ, Van Tiel FH, Melchers WJG, Verweij PE. 2011. Clinical implications of azole-resistance in Aspergillus fumigatus, the Netherlands, 2007-2009. Emerg Infect Dis 17:1846–1854. doi: 10.3201/eid1710.110226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Spiess B, Postina P., Reinwald M, Cornely OA, Hamprecht A, Hoenigl M, Lass-Florl C, Rath P-M, Steinmann J, Miethke T, Lauten M, Will S, Merker N, Hofmann W-K, Buchheidt B. 2014. Incidence of Cyp51 A key mutations in Aspergillus fumigatus—a study on primary clinical samples of immuno-compromised patients in the period of 1995-2013. PLoS One 9:e103113. doi: 10.1371/journal.pone.0103113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhao Y, Stensvold CR, Perlin DS, Arendrup MC. 2013. Azole resistance in Aspergillus fumigatus from bronchoalveolar lavage fluid samples of patients with chronic diseases. J Antimicrob Chemother 68:1497–1504. doi: 10.1093/jac/dkt071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Denning DW, Park S, Lass-Florl C, Fraczek MG, Kirwan M, Gore R, Smith J, Bueid A, Moore CB, Bowyer P, Perlin DS. 2011. High-frequency triazole resistance found in nonculturable Aspergillus fumigatus from lungs of patients with chronic fungal diseases. Clin Infect Dis 52:1123–1129. doi: 10.1093/cid/cir179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Arvanitis M, Ziakas PD, Zacharioudakis IM, Zervou FN, Caliendo AM, Mylonakis E. 2014. PCR in diagnosis of invasive aspergillosis: a meta-analysis of diagnostic performance. J Clin Microbiol 52:3731–3742. doi: 10.1128/JCM.01365-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mengoli C, Cruciani M, Barnes RA, Loeffler J, Donnelly JP. 2009. Use of PCR for diagnosis of invasive aspergillosis: systematic review of and meta-analysis. Lancet Infect Dis 9:89–96. doi: 10.1016/S1473-3099(09)70019-2. [DOI] [PubMed] [Google Scholar]
- 18.Springer J, Morton CO, Perry M, Heinz WJ, Paholcsek M, Alzheimer M, Rogers TR, Barnes RA, Einsele H, Loeffler J, White PL. 2013. Multicenter comparison of serum and whole-blood specimens for detection of Aspergillus DNA in high-risk hematological patients. J Clin Microbiol 51:1445–1450. doi: 10.1128/JCM.03322-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.White PL, Barnes RA. 2009. Aspergillus PCR, p 373–388. In Latge JP, Steinbach WJ (ed), Aspergillus fumigatus and aspergillosis, 1st ed ASM Press, Washington, DC. [Google Scholar]
- 20.Torelli R, Sanguinetti M, Brunella E, de Carolis E, de Pascale G, Caira M, Bello G, Antonelli M, Pagano L, Fadda G. 2011. Detection of Aspergillus DNA in bronchoalveolar lavage fluid samples from high risk patients using a commercial real-time PCR assay, abstr P071. Abstr Trends in Med Mycol 5; Valencia, Spain, 2−5 October 2011. [Google Scholar]
- 21.Perry MD, White PL, Barnes RA. 2014. Comparison of four automated nucleic acid extraction platforms for the recovery of DNA from Aspergillus fumigatus. J Med Microbiol 63:1160–1166. doi: 10.1099/jmm.0.076315-0. [DOI] [PubMed] [Google Scholar]