The diarrheal virus qPCR panel consists of two assays simultaneously detecting Norovirus GI and GII, Rotavirus (A) and Adenovirus (assay 1: NRA_UCT), and Sapovirus, Astrovirus and Enterovirus A-D (assay 2: SAE_UCT). In the following text, we refer to the diarrheal virus qPCR assay as NRA_SAE_UCT assay. To compile the assay, previously published primer/probe sets detecting Norovirus GI [17], Norovirus GII [17, 18], Rotavirus [19], Adenovirus [20, 21], Sapovirus ( [22] and own design), Astrovirus [23] and Enterovirus [24] were selected and modified (table S1). The assays were selected based on compatibility with the UCT Master Mix for cobas5800/6800/8800 systems (Roche, Mannheim, Germany) and a universal PCR run profile, as well as a good clinical performance as described in previous studies [17, 19,20,21,22,23,24,25,26]. Alignments of relevant sequences were collected from Genbank and compared to published primer/probe sequences using Geneious 9.0 (Dotmatics, Boston, USA, data of the alignment 01/2023). Primers and probes were adjusted to increase inclusivity according to recent data bank queries. To prevent formation of primer dimers formations in the multiplexed reaction, primers were modified with 2’O-methyl RNA bases at the 3’-region as indicated in table S1. Probes are either double-quenched, internally quenched, with a C3-spacer at the 3’-end and/or contain locked nuclear acids (LNA; +). For Sapovirus we implemented two different probes to cover Sapovirus genogroup I, II and V. To reduce enhanced background levels and to improve fluorescence levels, these two probes were designed using different fluorescent dyes (Atto 425 and Atto 550, table S1) and were detected on two different channels (1 and 3, table S2). Primers and probes were custom-made by Ella Biotech (Fuerstenfeldbruck, Germany), Biomers (Ulm, Germany) and Integrated DNA Technologies (IDT) (Coralville IA, USA). All oligos used in this study are listed in table S1.

Cobas omni utility channel reagent kits were prepared according to manufacturer’s instructions. Briefly, primer stock (500 µM concentration) and probe stock solutions (100 µM concentration) were added to 10 ml MMX-R2 master mix to obtain final concentrations in the PCR reaction as indicated in table S2. The PCR master mix was filled into utility channel cassettes which were then loaded onto the system to run tests. The cobas5800/6800/8800 system provides a spike-in RNA full-process control, which is added automatically during extraction and is detected in channel 5 (the respective sequence is not disclosed by the manufacturer; see table S2 for the full run protocol).

Technical performance evaluation for the NRA_SAE_UCT assay was performed according to the new European Union In Vitro Diagnostics Regulation (2017/746 EU IVDR). To evaluate the analytical sensitivity (lower limit of detection, LoD) and linearity of the NRA_SAE_UCT assay clinical samples which were positive for Norovirus GI and GII, Rotavirus, Adenovirus, Sapovirus, Astrovirus and Enterovirus, were used as standards. For Norovirus GI linearity a quantitative synthetic Norovirus GI RNA standard (no. VR-3234S) obtained from ATCC (Manassas, VA, USA) was used.

For quantification of standards from clinical samples, nucleic acids were extracted from selected stool samples on a MagNA-Pure 96 instrument (Roche Diagnostics) and standards were quantified by digital-PCR using the Qiacuity (Qiagen, Hilden, Germany) digital-PCR system according to manufacturer’s instructions. The unit of the standards is digital copies/ml (dcp/ml).

LoD was determined by serial two-fold dilution of standard suspensions using at least eight dilution steps and 21 repeats per dilution. Dilutions were prepared using a Hamilton IVD STARlet liquid handler (Hamilton, Bonaduz, Switzerland).

Linearity was assessed by ten-fold serial dilution according to CLSI guidelines (i.e. linear fit vs. higher-order polynomal fit) of each enteric virus standard (at least six dilution steps, n = 5 per dilution) and calculated using Validation Manager software (Finbiosoft, Espoo, Finland).

The within-run and between-day precision was determined by testing on three different days and by using one high positive, one low positive and one negative clinical sample for each enteric virus in triplicates. Within-lab precision was calculated as sum of squares of precision components. Precision was calculated as standard deviation (SD) with coefficient of variation (CV %) according to ANOVA statistics using Validation Manager (Finbiosoft).

For inclusivity 13 EQA samples (INSTAND, Dusseldorf, Germany) from the gastrointestinal virus (#430/ 2023 and 2024) or Norovirus panel (#381/ 2023 and 2024) were analysed using the new NRA_SAE_UCT assay. Moreover, a DNA ultramer (5’- GAT TTG GCC CTC GCC ACC TAC AAT GCC TGG TTC GTA GGT GGT ACA GCT CCA GAC CCA GAG CGC CCC ACT GAA GGT GCA CCC AAA TTA GTG TTT GAG ATG GAG GGC AAT − 3’) was created (Ultramer, manufactured by IDT) and used to verify inclusivity of Sapovirus variants (accession number LC504373 [27]) that are detected by the probe used in channel 1 (Atto 425) but not by the probe used in channel 3 (Atto 550) (see also table S1). For analysis, the ultramer was dissolved in 1 ml of QuantiTect Nucleic Acid Dilution Buffer (Qiagen) and diluted 1:1000 in Roche cobas PCR medium.

An exclusivity study was performed using 26 isolates of different common enteric bacteria (e.g. E. coli, E. faecium, E. gallinarum) and 12 clinical samples containing viruses (e.g. EBV, HSV-1/2, VZV).

For clinical validation, 243 clinical stool samples from routine testing were directly measured using NRA_SAE_UCT assay on cobas 5800/6800/8800 systems and compared to the Allplex GI-Virus Assay (Seegene, Duesseldorf, Germany), except for Entereovirus where the LightMix Kit (TIB molbiol, Berlin, Germany) was used. Both CE-IVD assays were used according to the manufacturer’s instructions. Discrepant results obtained with the Allplex GI-Virus Assay (Seegene) were resolved using LightMix Kits for Adenovirus, Norovirus GI and GII, Rotavirus, Sapovirus and Astrovirus (TIB molbiol).

Prior to analysis, a swab was directly dipped into the clinical stool sample, transferred to one tube of cobas PCR medium (4.2 ml) (Roche Diagnostics, Rotkreutz, Switzerland) and vortexed (direct swab stool sample preparation method [28]). Samples that returned invalid after analysis were diluted 1:10 with cobas PCR medium (Roche) and re-analysed.

The study was conducted according to the guidelines of the Declaration of Helsinki. This work was conducted in accordance with § 12 of the Hamburg hospital law (§ 12 HmbKHG). The use of anonymized remnant diagnostic samples from patients was approved and informed consent was waived by the ethics committee of the Hamburg Medical Association (PV5626).

Figures were prepared using Validation Manager (Finbiosoft) and GraphPad Prism 10.2.2. (Boston, USA).