Multicenter Clinical Validation of the Molecular BD Max Enteric Viral Panel for Detection of Enteric Pathogens

The conventional methodology for gastrointestinal pathogen detection remains time-consuming, expensive, and of limited sensitivity. The objective of this study was to evaluate the performance of the BD Max enteric viral panel (Max EVP) assay for identification of viral pathogens in stool specimens from individuals with symptoms of acute gastroenteritis, enteritis, or colitis.


Appendix A: Further details on the reference method
Reference method testing consisted of: • Nucleic acid extraction of pre-screened, and quantified, clinical positive stool specimens, virus, viral culture fluid, or dilutions of target DNA • Two (2) alternate rtPCR assays • Sanger Sequencing.
Quantified samples were extracted on the Roche MagNA Pure LC 2.0 using the Total Nucleic Acid Isolation Kit -High Performance according to the instrument user manual and kit protocol.
Following extraction or gBlock dilution, the purified viral nucleic acid, or synthetic DNA, was tested with different Master Mixes representing two (2) categories of assays, the non-sequencing assay group, and the sequencing assay group.
Detection for these assays is achieved through the use of fluorescently labeled probes. The sequencing assays group consisted of ten (10) melt-based real-time rtPCR assays. rtPCR detection was achieved through the use of Melt Temperature Analysis with an intercalating dye.
rtPCR for both sequencing and non-sequencing assays was performed on the Bio-Rad CFX96 or Bio-Rad CFX96 Touch™ Real-Time PCR Detection Systems according to the instrument Instruction Manual. For the Sequencing assay group, potentially positive samples were identified and then further processed for Sanger Sequencing. Sanger Sequencing was performed on the Advanced Biosystems® GA3500xL Genetic Analyzer according to the User Guide.

Reference Method Validation
The reference method consists of twenty-one (21) assays that test for multiple genotypes, each, of the five (5) viral targets of interest (Norovirus, Rotavirus, Adenovirus, Sapovirus, and Astrovirus). Ten (10) assays, for which, viral targets/genotypes had a previously characterized limit of detection (LoD), were utilized for this study. A representative strain of each organism target was used in this study (Table A1). Each strain was tested and analyzed individually. The remaining viral genotypes and assays were used to evaluate inclusivity.
LoD titration and determination was tested using a limiting dilution approach and followed by confirmation. LoD evaluation studies were performed using MagNA Pure LC 2.0 DNA extraction from a target-negative, unpreserved, stool matrix using the parameters described below. Unpreserved stool matrix represents the "worst-case" sample matrix as the enteric viral targets under investigation are more susceptible to degradation-as compared to Cary-Blair preserved samples. Therefore, both unpreserved and Cary-Blair preserved specimen types were considered validated through this study.
Amplification and detection were performed using a total of four (4) CFX96™ platforms. Ct Score, EP (fluorescence height-end of the reaction curve) and baseline values were generated using the Bio-Rad CFX Manager (version 2 or later) software.

Acceptance Criteria
Twenty (20) replicates were tested to confirm the putative LoD for each strain. LoD confirmation was achieved when the proportion positive was ≥ 95%. No further testing was performed if LoD confirmation resulted in 95% (19/20) proportion positive during initial confirmation testing. If the proportion positive was below 95%, additional testing at a higher level was required.
External positive and negative controls were included on each extraction run, subsequent PCR amplification and sequencing (when applicable), and were required to meet expected results for a positive sample to be considered valid. Successful sample extraction was required on the MagNA Pure LC 2.0.
For a potentially positive sequenced sample to be considered positive for its intended assay target, it was required to achieve the following criteria: • Sample score QV20 ≥ 90% for both the forward and reverse sequence reads; and • BLAST E-value of the consensus read ≤ 10-30 with at least 90% query coverage and 95% identify for the intended assay target.

Conclusion
Limit of detection (LoD) analysis was completed for 21 master mixes used in the sequencing and non-sequencing reference method assays (inclusive for all five viral targets). Following extraction or gBlock dilution, the purified viral nucleic acid, or synthetic DNA, was tested with different Master Mixes representing two (2) categories of assays, the non-sequencing assay group, and the sequencing assay group.
The non-sequencing assays group consisted of eleven (11) TaqMan®-based real-time rtPCR assays. Detection for these assays is achieved through the use of fluorescently labeled probes.
The sequencing assays group consisted of ten (10)  A total of 70 specimens comprised of viral and gBlock targets were evaluated in the study. Each strain was initially tested in triplicate. If the required proportion positive percentage (100% for three replicates) failed to be met, the target(s) of interest were titrated higher until the proportion positive was ≥95%.
Alternative PCR Inclusivity non-sequencing Sixty-one (61) of the 70 samples were available for testing on non-sequencing assays, of which there were 41 samples of that confirmed at ≥95% at 3x LoD. The remaining samples were titrated upwards. Seven (7) more were confirmed at 30x LoD, five (5) more were confirmed at 300x LoD, and four (4) more required titration to 3000x LoD. The remaining four (4) samples were not detected. A Summary of the non-sequencing results can be found in Table A3.
Alternative PCR Inclusivity sequencing Sixty (60) of the 70 samples were available for testing on sequencing assays, of which there were 39 samples that confirmed at 3x LoD. The remaining samples were then titrated up to 30x LoD where seven (7) more samples were confirmed. Nine (9) samples confirmed at 300x LoD, and finally two (2) were confirmed at 3000x LoD. The remaining three (3) samples were not detected. A Summary of the non-sequencing results can be found in Table A4.
The EVP Alternate PCR Reference Method assays were shown to be inclusive of all tested genotypes.   One Astrovirus Type 7 viral stock was not detected at 528x LOD. c One Astrovirus MLB1 gBlock was detected at 30x LOD. d One Astrovirus MLB2 gBlock was detected at 30x LOD. e One Astrovirus MLB3 gBlock was detected at 30x LOD. f One Norovirus GI.3 gBlock was detected at 300x LOD and another GI.3 gBlock was detected at 3000x LOD. g One Norovirus GI.7 clinical stool was detected at 30x LOD. h One Norovirus GII.1 clinical stool was detected at 30x LOD. i One Norovirus GII.3 clinical stool was detected at 300x LOD. J One Norovirus GII.12 clinical stool was detected at 30x LOD. k One Norovirus GII.17 clinical stool was detected at 30x LOD. L One Norovirus GII.6 clinical stool was detected at 0.9x LOD. m One Rotavirus (unknown) clinical stool was detected at 300x LOD. n One Rotavirus A viral stock was detected at 54x LOD. o One Sapovirus GI clinical stool was detected at a concentration of 2.5x LOD; one Sapovirus GI clinical stool was detected at a concentration of 57x LOD; one Sapovirus GI clinical stool was not detected at a concentration of 2529x LOD. p One Sapovirus GII clinical stool was not detected at 302x LOD. q One Sapovirus GIV clinical stool was detected at 300x LOD.