Optimization of the STARlet workflow for semi-automatic SARS-CoV-2 screening of swabs and deep respiratory materials using the RealAccurate Quadruplex SARS-CoV-2 PCR kit and Allplex SARS-CoV-2 PCR kit

Abstract

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic triggered the implementation of large-scale screenings in the health care and in the general population. Consequently, medical laboratories have to apply lean laboratory management to design workflows that are able to process large batches within short turnaround times while maintaining flexibility to use different SARS-CoV-2 reverse transcription polymerase chain reactions (RT-PCRs) and to be able to process a variety of clinical samples. We validated two SARS-CoV-2 PCR assays on the STARlet workflow: Allplex SARS-CoV-2 PCR kit and RealAccurate Quadruplex SARS-CoV-2 PCR kit. Furthermore, we optimized and validated the STARlet workflow for semi-automatic screening for SARS-CoV-2 in upper respiratory swabs and deep respiratory materials (sputa, bronchoalveolar lavage, and aspirate). Strikingly, guanidine-containing lysis buffers allow for easy processing and can enhance sensitivity of SARS-COV-2 screening since sampling in these buffers may preserve viral transcripts as evident by the higher copy numbers of the SARS-CoV-2 N gene. Moreover, using the principles of lean laboratory management, several bottlenecks that are typical for medical laboratories were addressed. We show that lean laboratory management resulted in significant reduction of the turnaround times of the SARS-CoV-2 PCR in our laboratory. This report thus describes a useful framework for laboratories to implement similar semi-automated workflows.IMPORTANCEThe SARS-CoV-2 pandemic triggered the implementation of large-scale screenings in the health care and in the general population. Consequently, medical laboratories had to adapt and evolve workflows that are able to process large batches within short turnaround times while maintaining flexibility to use different assays and to be able to process a variety of clinical samples. We describe how the need for increased outputs and greater flexibility was addressed with respect to clinical samples and assays (Allplex SARS-CoV-2 PCR and RealAccurate Quadruplex SARS-CoV-2 PCR). Strikingly, we found that upper respiratory swabs collected in guanidine-containing lysis buffers both improved the ease of processing as well as enhanced the sensitivity of the SARS-CoV-2 screening. This report thus describes a useful framework for laboratories to implement and optimize similar semi-automated workflows.

Keywords: SARS-CoV-2; laboratorium automation; lean management; molecular diagnostics; turnaround time; validation.

Fig 1

(A–D) Comparison of the Seegene relative to the GeneXpert (Gx) SARS-CoV-2 assays [monoplex SARS-CoV-2 (A and B) and SARS/Flu/RSV (C and D) in original assay (A and C) and with additional RdRp target (B and D)]. The dotted line represents the isometric line where both test reports equal Ct values.

Fig 2

Plot of RAQ SARS-CoV-2 N gene Ct values against Seegene Allplex N gene Ct values. The correlation between the two Ct values is described as 0.9325x + 5.372.

Fig 3

Overview of the initial workflow (left) and the desired workflow (right). Each step was analyzed with respect to what value each step of the diagnostic workflow added for the patient. Steps that could be optimized are colored orange. Process steps in green are considered optimized or optimal.

Fig 4

(A) Comparison of SARS-CoV-2 E gene Ct values obtained after spiking PP-TL+ (red) or UTM (blue) with SARS-CoV-2 relative to the corresponding Ct value seen in eSwab medium. The dotted line represents ideal comparison. N = 5 apparatuses. (B) Effect of temperature on SARS-CoV-2 Ct values in PP-TL+ after storage for 7 days at −70°C and at room temperature. (C) Effect of time on SARS-CoV-2 Ct values in PP-TL+ if samples are stored at room temperature and measured on days 0, 3, 7. N = 2 independent dilution series represented by either black or transparent dots.

Fig 5

(A) Trendline and 95% CI of the N gene (blue) and E gene (red) concentration in the original material (ddPCR copies/mL); the equation for each is Log₁₀ E gene load = 1/–3.396*(Ct E gene-45.45) and Log₁₀ N gene = 1/–3.599*(Ct N gene-46,84). (B) Quantified numbers in clinical samples taken in eSwab (blue) or in PP-TL+ (orange). (C) Ratio of copy numbers for the E gene and N gene relative to the ORF1a.

Fig 6

(A) Descriptive figure of the four selected periods wherein the turnaround time of the SARS-CoV-2 PCR was measured and the process optimizations that were implemented between the selected periods. (B and C) Violin plots of turnaround time for (B) deep respiratory materials (clinic’s) and (C) respiratory swabs (first-line health care providers; i.e., general practitioners and care homes). The violin plot shows the distribution of the turnaround times (length) combined with the frequency density for each possibility (width).