Integration of SARS-CoV-2 testing and genomic sequencing into influenza sentinel surveillance in Uganda, January to December 2022

ABSTRACT The Uganda Virus Research Institute, National Influenza Center laboratory integrated SARS-CoV-2 polymerase chain reaction testing and genomic sequencing into the influenza surveillance program that was established in 2007. A total of 7,698 nasopharyngeal/oropharyngeal (NP/OP) swab samples were collected and analyzed from ILI/SARI sentinel sites across the country from January to December 2022. All samples were tested for influenza and SARS-CoV-2. Of these, 252 (3.3%), 162 (2.1%), and 589 (7.7%) were positive for influenza A, influenza B, and SARS-CoV-2, respectively. Out of 414 influenza-positive samples, 122 (29.5%) were AH1pdm09, 130 (31.4%) were AH3, and 162 (39.1%) were B-Victoria. All SARS-CoV-2 sequenced samples were of the Omicron variant, with subvariants of concern known to evade the immune system being detected, such as BQ.1 and XBB.2. Other SARS-CoV-2 positive samples collected from other health centers in the community outside the surveillance sites were included into SARS-CoV-2 genomic sequencing with similar patterns with respect to variants. In all, the ILI/SARI surveillance system has shown to be an efficient, cost-effective, and sustainable program, providing a ready platform to monitor the circulation of SARS-CoV-2 in communities at the national level while remaining vigilant for the persistent threat of influenza. The integration of SARS-CoV-2 detection and genomic surveillance into the influenza surveillance program will strengthen the laboratory response capacity, as well as facilitate the timely release of SARS-CoV-2 genomic information to be used to complement the multiple response strategies for COVID-19 pandemic mitigation. IMPORTANCE Respiratory pathogens cause high rates of morbidity and mortality globally and have high pandemic potential. During the SARS-CoV-2 pandemic, influenza surveillance was significantly interrupted because of resources being diverted to SARS-CoV-2 testing and sequencing. Based on recommendations from the World Health Organization, the Uganda Virus Research Institute, National Influenza Center laboratory integrated SARS-CoV-2 testing and genomic sequencing into the influenza surveillance program. We describe the results of influenza and SARS-CoV-2 testing of samples collected from 16 sentinel surveillance sites located throughout Uganda as well as SARS-CoV-2 testing and sequencing in other health centers. The surveillance system showed that both SARS-CoV-2 and influenza can be monitored in communities at the national level. The integration of SARS-CoV-2 detection and genomic surveillance into the influenza surveillance program will help facilitate the timely release of SARS-CoV-2 information for COVID-19 pandemic mitigation and provide important information regarding the persistent threat of influenza.

onset within the last 10 days.The case definition of SARI is an acute respiratory infection with a subjective or measured temperature of ≥38°C and a cough with onset in the last 10 days requiring hospitalization (11).
The influenza surveillance program comprises a network of 16 sentinel sites geographically spread across the country (Fig. 1).The selected sentinel sites include general hospitals and health centers that provide treatment to all age groups of patients and are in areas with high population density and high trade, where influenza transmis sion would likely be more rapid during a pandemic.The hospitals offer both outpatient and inpatient care, whereas the health centers offer only outpatient care.The surveil lance is conducted in 11 districts representing 4 geographically distinct regions across Uganda (Fig. 1).The UVRI-NIC laboratory therefore conducts and plays an important role in the COVID-19 pandemic response by providing a ready platform to monitor the circulation of SARS-CoV-2 in communities at the national level.The population of the districts in Uganda was reported in 2012 when the initial sentinel surveillance system was established (see Fig. 1 in reference 12).The sites in this system represent major population centers and areas of high trade (12).The ILI/SARI surveillance program is also an efficient, cost-effective, and sustainable platform in response to the COVID-19 pandemic while remaining vigilant for a persistent threat of influenza.
Our work describes the successful integration of SARS-CoV-2 detection and genomic sequencing into the ILI/SARI surveillance system of the WHO UVRI-NIC laboratory for the period of January to December 2022 (Epi week 50).Our main goals were to identify and track the number and proportion of influenza and SARS-CoV-2 infections at the surveillance sites and in other community health centers throughout Uganda, as well as determine SARS-CoV-2 variants through whole-genome sequencing by continuously sampling patients.This integration strengthens the laboratory response capacity, as well as facilitates the timely release of SARS-CoV-2 genomic information to be used to complement the multiple response strategies for COVID-19 pandemic mitigation.

Sample collection and processing
The study used nasopharyngeal (NP) and/or oropharyngeal (OP) swabs collected between January and December 2022 (Epi week 50) from patients visiting ILI/SARI sites and other community health centers outside the sentinel site surveillance program.The influenza surveillance program is a network of 16 sentinel sites located through out the country (Fig. 1).Samples from other health centers in Uganda represented community-level cases throughout the country.In total, 7,698 samples were collected from ILI/SARI sentinel surveillance sites and 23,167 samples were collected from other health centers throughout Uganda (community-level cases).All samples were tested for SARS-CoV-2, but only the 7,698 samples collected from surveillance sites were also tested for influenza.
RNA was extracted from 140 μL of NP/OP swab specimens in a viral transport medium (VTM) using the QIAamp Mini Viral RNA Extraction Kit from Qiagen (Halde man, Germany) according to the manufacturer's instructions.The RNA was tested on the ABI 7500 Real-Time PCR platform for influenza and SARS-CoV-2 using the CDC influenza SARS-CoV-2 (Flu SC2) multiplex Real-Time Reverse Transcription Polymerase Chain Reaction (rRT-PCR) (CDC Influenza Virus Real-Time RT-PCR Influenza A/B Typing Panel (VER 2) (RUO) (Catalog No. FluRUO-14) as per CDC protocol (13).All samples were tested for human RNAse P (RP) gene as a housekeeping gene.The RP gene serves as an internal control for the Flu SC2 Multiplex assay that is compared to other targets for the interpretation of an individual specimen.Briefly, RP should be positive for clinical specimens in the absence of a signal for one of the viral targets.A negative RP in the presence of a positive result for one of the viral targets, the viral target should be considered valid.However, if all viral targets and RP are negative, the test should be considered invalid (13).A second CDC multiplex rRT-PCR was performed for all influenza A positives, to subtype for influenza AH1pm09 and AH3 using the CDC Influenza Virus Real-Time RT-PCR Influenza A (H3/H1pdm09) Subtyping Panel (VER 3) (RUO) (Catalog No. FluRUO-15), whereas influenza B positives were genotyped for influenza B Victoria and Yamagata using the CDC Influenza B Lineage Genotyping Panel (RUO) (Catalog No. .

Whole-genome sequencing of SARS-CoV-2
SARS-CoV-2 PCR-positive samples from the surveillance sites with high viral load (CT < 30) were selected for whole-genome sequencing.Additional SARS-CoV-2-positive samples collected from other health centers in Uganda were also selected for wholegenome sequencing.Briefly, sequence libraries were generated from SARS-CoV-2-posi tive samples using the Swift Normalase Amplicon Panel (SNAP) SARS-CoV-2 kit (14) according to the manufacturer's protocol and sequenced on the Illumina MiSeq platform (Illumina, San Diego, CA), using the MiSeq Reagents Kit v3 (600 cycles).

Surveillance sites
Community health centers a They are split up by those obtained from surveillance sites and other community health centers throughout Uganda.
accepted to GISAID are referenced here for which we describe the variants.We report Pangolin lineages from GISAID as of 9 August 2023.
The monthly influenza and SARS-CoV-2-positive cases for 2022 as captured by the ILI/SARI surveillance program are shown in Fig. 2. The positivity of SARS-CoV-2 among ILI/SARI first peaked in January at 22% decreasing in February and March before peaking again in June at 20% and staying at less than 10% between August and December 2022.By contrast, influenza activity remained low, only peaking in March at 17% and decreasing to less than 10% from May to December 2022.Early in the year, influenza subtype AH3 was the predominate A subtype until May, at which point AH1pdm09 was predominate the rest of the year.AH3 did not appear again until October.Influenza B Victoria was found from February until October.
A total of 22 cases of co-infections from ILI/SARI surveillance sites were detected during the study period: 2 cases with SARS-CoV-2/AH3; 9 cases with SARS-CoV-2/ B-Victoria; and 11 cases with SARS-CoV-2/AH1pdm09.The outcome of the 22 SARS-CoV-2/influenza co-infected patients is not known, as the cases were not followed up.
We also tested a total of 23,167 samples collected from other community health centers outside the sentinel sites and found an overall positivity of 3.8% for SARS-CoV-2.These samples were only tested for SARS-CoV-2.A similar trend in positivity by month was observed when ILI/SARI sentinel site data were compared to SARS-CoV-2 testing in samples collected from health centers outside the sentinel site surveillance program at UVRI (Fig. 3).Both sets of samples show peaks in January and again in June 2022.
Samples were chosen for whole-genome sequencing based on the PCR CT val ues of samples collected during the reporting period.In total, 49 samples from sentinel surveillance sites and 97 samples from other health centers (community) were successfully assembled into SARS-CoV-2 consensus genomes and publicly shared through GISAID (https://www.gisaid.org),an initiative founded on sharing influenza virus sequencing data (Table 1).Samples failing consensus did so because of greater than 50% ambiguous bases.All 146 genomes were classified as Omicron (Fig. 4).The Omicron subvariant BA.2.31.1 ranked as the predominant SARS-CoV-2 strain during the reporting period for samples collected from other health centers (19/97) (Fig. 4).For surveillance sites, BA.1.1 (9/49) and BA.2.31.1 (9/49) were the predominate strains throughout the year.Generally, there were similar percentages between surveillance sites and other community health centers for each of the variants identified during our sampling period (Fig. 4).

DISCUSSION
Our main objective was implementing a comprehensive surveillance program for influenza and SARS-CoV-2 following the WHO recommendations to identify and track the proportion and subtypes of influenza and variants of SARS-CoV-2 infections in the community.During the study period, we identified more influenza A viruses compared to influenza B. Our surveillance also demonstrated a possible inverse relationship between influenza and SARS-CoV-2 activity (Fig. 2).However, this observation may be biased as we were not able to account for non-pharmaceutical interventions that were implemented during COVID-19 upsurge, leading to a reduction in influenza transmission.All influenza B viruses belonged to the Victoria lineage, similar to results in Uganda in 2009-2010 (19).
During its inception, our program detected more SARS-CoV-2 cases than influenza in January 2022.This could be attributed to the time of integration of SARS-CoV-2 testing and genomic sequencing into ILI/SARI surveillance in January 2022, during the peak of the third wave of COVID-19 in Uganda.SARS-CoV-2 cases peaked during June 2022 with a similar percent positivity in January.We observed an increase in influenza activity during the early months of 2022, which was different from the seasonality of influenza previously reported from Uganda (12).Previous data on seasonality showed influenza activity peaking from May to November with most cases appearing in October and November (12).However, in our study, influenza cases declined to less than 10% from May through December 2022 (Fig. 2).Rather, the highest number of influenza cases occurred in March and April 2022.Such variable and limited information regarding influenza in certain regions of the world emphasizes the need for a surveillance system to track influenza infections and their spread.As shown here, the timing of respiratory infections can change during a pandemic because of measures to reduce the spread of similarly transmitted pathogens (8).These patterns demonstrate the utility of detecting and continuously tracking respiratory pathogens throughout the country.
Another objective of our surveillance program was to understand SARS-CoV-2 variants in the community by continuously sampling ILI/SARI patients reported from a defined catchment area.To fulfill this objective, SARS-CoV-2-positive samples from ILI/SARI sites and other health centers outside the surveillance program, and with CT values < 30, were sequenced on the Illumina MiSeq platform to determine the circulat ing variants (Fig. 4).We successfully sequenced and got consensus genomes from 146 SARS-CoV-2 samples, which were deposited in GISAID (accession numbers are found in Table 1).All the variants detected from January to December 2022 at both surveillance sites and other community health centers were Omicron.Omicron was first identified in November 2021 and quickly spread, reaching the United States in December 2021 (20).The sentinel surveillance sites used here were able to pick up Omicron variants.There were 10 samples from surveillance sites and 12 samples from other community health centers that were collected from patients in January 2022.The vast majority of these were BA.1.1 (20/22).
Our study also reports co-infections in both out-patients and hospitalized patients with a predominance of out-patient samples.The sensitivity and robustness of our surveillance program demonstrate its ability to detect influenza cases when COVID-19 activity was at its peak in January and June 2022.The trends in the rise and fall of COVID-19 cases in Uganda, seen in our network, coincided with the national trends.Other studies have reported co-infections of SARS-CoV-2 and influenza globally (21)(22)(23)(24)(25). Co-infections of SARS-CoV-2 and influenza have also been highlighted as a cause of concern due to the demonstrated worsening of the clinical picture in such cases.A study in Saudi Arabia reported mortality and ICU admissions in influenza A/SARS-CoV-2 co-infections (24), and triple co-infections with influenza A and B, and SARS-CoV-2 have also been reported (26).
To the best of our knowledge, this is the first report from Uganda on SARS-CoV-2/ influenza co-infection cases identified through sentinel surveillance in a community setting.Our efforts of parallel testing for influenza and SARS-CoV-2 are in line with WHO recommendations for integrated testing for influenza and COVID-19.We report the circulation of influenza viruses in the community as well as hospital settings across the entire country, even during the COVID-19 pandemic, with AH3 being the dominant influenza virus A subtype that circulated in Uganda in 2022.Overall, this study reports a low prevalence of co-infections of SARS-CoV-2 and influenza.
There are advantages to using specified surveillance sites for monitoring infectious diseases throughout Uganda.When other health centers in the broader community that are not in the surveillance network send samples for processing, there are many instances of not receiving appropriate metadata that is important for decision-making, such as patient demographics or location-specific data.At surveillance sites, this can be better controlled, and the right metadata can be collected for public health officials to track disease spread and outbreak progression in the country.The lack of metadata also makes it hard to share with the scientific community, for example, uploading to GISAID.Another benefit of surveillance sites is sample integrity.We ensure the samples are kept frozen, and the cold chain is maintained until they get to the laboratory and stored at −80°C until processing.The surveillance sites have more control over the whole process from sample collection to sequencing.Other community samples were often delivered at room temperature in envelopes and therefore some samples failed to give good results.
This study had limitations.We present data from only 1 year up to Epi week 50 when SARS-CoV-2 genomic sequencing was integrated into ILI/SARI surveillance.We initiated testing only influenza viruses with SARS-CoV-2 and have not included other respiratory viruses, particularly RSV.However, our study reports SARS-CoV-2 genomic data from a geographically distributed network of 16 influenza sentinel sites throughout Uganda as well as other health centers outside the ILI/SARI sentinel surveillance network.In the future, additional respiratory viruses can be tested for at these already established surveillance sites.
As the world progresses toward COVID-19 endemicity, it will be important to strengthen surveillance capacities for other respiratory pathogens, which may lead to significant morbidity and mortality, and overburden healthcare systems in the future.Therefore, adopting and implementing cost-effective and reliable testing methods for the diagnosis of respiratory viruses with epidemic and pandemic potential will help in monitoring the trends and seasonality of those viruses for the prevention of epidemics and pandemics.Uganda will continue to benefit from the established sentinel surveil lance site network throughout the country, not only for influenza surveillance but also for other respiratory pathogens and future pandemic pathogens.

FIG 2
FIG 2 The number of positive cases of SARS-CoV-2 (orange) and influenza (broken up into AH1, AH3, and Influenza B) at the surveillance sites throughout Uganda from January 2022 to December 2022 (Epi week 50).The lines denote the percent positive cases for all influenza cases (solid line) and SARS-CoV-2 (dashed line).