Impact of the COVID-19 pandemic on the prevalence of influenza A and respiratory syncytial viruses elucidated by wastewater-based epidemiology

Since the COVID-19 pandemic, a decrease in the prevalence of Influenza A virus (IAV) and respiratory syncytial virus (RSV) has been suggested by clinical surveillance. However, there may be potential biases in obtaining an accurate overview of infectious diseases in a community. To elucidate the impact of the COVID-19 on the prevalence of IAV and RSV, we quantified IAV and RSV RNA in wastewater collected from three wastewater treatment plants (WWTPs) in Sapporo, Japan, between October 2018 and January 2023, using highly sensitive EPISENS™ method. From October 2018 to April 2020, the IAV M gene concentrations were positively correlated with the confirmed cases in the corresponding area (Spearman's r = 0.61). Subtype-specific HA genes of IAV were also detected, and their concentrations showed trends that were consistent with clinically reported cases. RSV A and B serotypes were also detected in wastewater, and their concentrations were positively correlated with the confirmed clinical cases (Spearman's r = 0.36–0.52). The detection ratios of IAV and RSV in wastewater decreased from 66.7 % (22/33) and 42.4 % (14/33) to 4.56 % (12/263) and 32.7 % (86/263), respectively in the city after the COVID-19 prevalence. The present study demonstrates the potential usefulness of wastewater-based epidemiology combined with the preservation of wastewater (wastewater banking) as a tool for better management of respiratory viral diseases.


H I G H L I G H T S G R A P H I C A L A B S T R A C T
• IAV and RSV were detected from wastewater with highly sensitive EPISENS™ methods. • Wastewater IAV and RSV concentrations positively correlated with reported cases. • Typing of IAV HA gene in wastewater enables identification of prevalent subtypes. • Wastewater IAV and RSV concentrations decreased after the beginning of pandemic. • Wastewater banking allows for elucidating impact of the pandemic on viral epidemics.

A B S T R A C T A R T I C L E I N F O Editor: Damia Barcelo
Keywords: Wastewater-based epidemiology COVID-19 Influenza A virus RSV Typing Wastewater banking Since the COVID-19 pandemic, a decrease in the prevalence of Influenza A virus (IAV) and respiratory syncytial virus (RSV) has been suggested by clinical surveillance. However, there may be potential biases in obtaining an accurate overview of infectious diseases in a community. To elucidate the impact of the COVID-19 on the prevalence of IAV and RSV, we quantified IAV and RSV RNA in wastewater collected from three wastewater treatment plants (WWTPs) in Sapporo, Japan, between October 2018 and January 2023, using highly sensitive EPISENS™ method. From October 2018 to April 2020, the IAV M gene concentrations were positively correlated with the confirmed cases in the corresponding area (Spearman's r = 0.61). Subtype-specific HA genes of IAV were also detected, and their concentrations showed trends that were consistent with clinically reported cases. RSV A and B serotypes were also detected in wastewater, and their concentrations were positively correlated with the confirmed clinical cases (Spearman's r = 0.36-0.52). The detection ratios of IAV and RSV in wastewater decreased from 66.7 % (22/33) and 42.4 % (14/33) to 4.56 % (12/263) and 32.7 % (86/263), respectively in the city after the COVID-19 prevalence. The present study demonstrates the potential usefulness of wastewater-based epidemiology combined with the preservation of wastewater (wastewater banking) as a tool for better management of respiratory viral diseases.

Introduction
Influenza A virus (IAV) and respiratory syncytial virus (RSV) are major pathogenic respiratory viruses that are routinely monitored through clinical surveillance due to their significant impacts on public health (Centers for Disease Control and Prevention (CDC), 2023; Griffiths et al., 2017;World Health Organization, 2023a, 2023b. IAV is classified into many subtypes and several of them (e.g., H1N1, H1N1pdm, H3N2) infect humans of all ages, causing an estimated 290-650 thousand deaths per year worldwide (World Health Organization, 2023c). RSV is divided into two major antigenic subtypes (A and B types), causing significant respiratory symptoms, especially for infants and elderly people (Griffiths et al., 2017). Although RSV results in more than three million hospitalizations annually worldwide (World Health Organization, 2019), no effective vaccinations or therapeutics have been developed for RSV. The incidence of these two respiratory viruses typically increases in winter months in temperate regions (Li et al., 2019;Obando-Pacheco et al., 2018), but the COVID-19 pandemic has disrupted the typical seasonality of these viruses in both the northern Feng et al., 2021;van Summeren et al., 2021;Weinberger Opek et al., 2021) and southern hemisphere (Eden et al., 2022;Huang et al., 2021;Tempia et al., 2021). In Japan, the clinically confirmed cases with influenza and RSV between 2020 and 2021 remarkably decreased to <0.01 % (World Health Organization, 2023a) and 10 % (National Institute of Infectious Disease, 2023) of reported cases between 2019 and 2020, respectively. This phenomenon, which was observed globally, is possibly associated with non-pharmaceutical interventions (NPIs) against COVID-19 such as lockdown, interprovincial travel ban, and masking (Eden et al., 2022;Emborg et al., 2022;Tempia et al., 2021). In addition, the COVID-19 pandemic may have caused medical facilities to become overwhelmed, leading to potential underestimation of reported cases of the two respiratory viruses. This underestimation is difficult to prove through ordinary passive clinical surveillance methods.
Wastewater-based epidemiology (WBE) is a promising tool for monitoring infectious disease prevalence at the population level, overcoming drawbacks of clinical surveillance such as dependency on test capacity. To date, numerous studies have demonstrated the applicability of WBE for tracking the prevalence of SARS-CoV-2 in a community (Amman et al., 2022;Duvallet et al., 2021;Karthikeyan et al., 2022;Medema et al., 2020). Moreover, in our previous study, we demonstrated the ability to estimateCOVID-19 followinf five-days through WBE. We used a mathematical model and the highly sensitive EPISENS-M method to detect SARS-CoV-2 RNA from wastewater. This method was effective when clinically reported cases exceeded 0.69 per 100,000 in the corresponding area (Ando et al., 2023). Recently, in preparation for subsequent infectious respiratory disease pandemics in the post-COVID-19 era, researchers have explored the applicability of WBE to IAV Mercier et al., 2022;Wolfe et al., 2022) and RSV . Several studies have shown a positive correlation between the concentrations of IAV and RSV in wastewater and clinically confirmed cases in a community Mercier et al., 2022;Wolfe et al., 2022). However, a limited number of studies have traced the circulations of subtypes of IAV (Mercier et al., 2022) and RSV in a community via WBE, although understanding their epidemic subtypes is essential for global public authorities to distribute effective vaccinations for respiratory viruses. Furthermore, to the best of our knowledge, no studies have reported the longitudinal patterns of influenza virus and RSV concentrations in wastewater, including pre-COVID-19 pandemic periods.
The objective of the present study is to investigate the influence of the COVID-19 pandemic on the prevalence of IAV and RSV, including their subtypes through WBE. To achieve this goal, we measured the concentrations of IAV and RSV RNA in influent wastewater collected at three wastewater treatment plants (WWTPs) located in Sapporo City, Japan, between October 2018 and January 2023. We observed a remarkable decline in the viral RNA concentrations follwing the introduction of SARS-CoV-2 to the city.

Wastewater sampling
Between October 26, 2018, and January 12, 2023, a total of 296 grab influent wastewater samples (1 L each) were collected at three WWTPs located in Sapporo City (population: approximately 1.96 million), Japan. The three WWTPs are connected to combined sewer systems. The catchment areas of the WWTPs are adjacent to each other (Fig. 1). From 2018 to April 2020, 33 wastewater samples were collected monthly from WWTP A, which serves approximately 10 % of the city's population. From May 2020 to January 2023, 263 wastewater samples were collected weekly at two WWTPs (WWTP B; n = 133, WWTP C; n = 130), serving approximately 20 % of the population in the city. The wastewater samples were collected in sterile plastic bottles and immediately transported to the laboratory on ice. The samples were processed on the day of collection.

Validation tests of the two variations of the EPISENS™ methods
Two variations of the EPISENS™ methods (i.e., EPISENS-M and EPISENS-M_K) were used in this study. To compare their sensitivity and viral recovery yields, 1 L of influent wastewater collected on November 6, 2022 from WWTP A was seeded with influenza H1N1pdm (A/California/ 07/2009; ATCC VR-1894) and RSV A2 (ATCC VR-1540) to obtain the final concentrations of 6.07 × 10 5 /L and 1.02 × 10 6 /L, respectively. The seeded wastewater samples were processed with the two detection methods described in Section 2.3.

The EPISENS-M method
The wastewater samples collected between May 2020 and January 2023 were processed using the Efficient and Practical virus Identification System with Enhanced Sensitivity for Membrane (EPISENS-M) method (Ando et al., 2023). Briefly, 300 mL of each sample supplemented with 25 mM MgCl 2 were filtered through electronegative membranes (90-mm diameter 0.90-μm pore size; Merck Millipore, Billerica, USA; Catalog no. AAWP-09000). A quarter of the membrane was subjected to RNA extraction using a PowerWater bead beating tube (RNeasy PowerWater Kit (Qiagen, Hilden, Germany), QIAcube Connect platform and RNeasy PowerMicrobiome Kit (Qiagen)). The extracted RNA volume was adjusted to 50 μL.
The extracted RNA sample was subjected to one-step reverse transcription, followed by preamplification (RT-Preamp) using the iScript™ Explore One-Step RT PreAmp Kit (Bio-Rad Laboratories, Hercules, CA, USA). The components of the mixture and primers are summarized in the supplementary materials (Tables S1, S2, S3). The thermal cycling conditions of the RT-Preamp were 25°C for 5 min, 45°C for 60 min, and 95°C for 3 min followed by 10 cycles of 95°C for 15 s and 58°C for 4 min. A tenfold serial dilution of plasmid DNA containing the amplification region sequence of the qPCR assay (US CDC N1, Integrated DNA Technologies) was included to generate a standard curve at concentrations ranging from 10 5 to 10 0 gene copies (GC) per RT-Preamp tube. Nuclease-free water was used as a negative control in the RT-Preamp reaction.
We used four IAV (i.e., IAV M, IAV H1, IAV H1pdm, IAV H3) (Nakauchi et al., 2011), three RSV (i.e., RSV A, RSV B, RSV AB) (Hu et al., 2003;Hughes et al., 2022), and PMMoV (Haramoto et al., 2013;Zhang et al., 2006) assays to measure the concentration of viral RNA in wastewater with qPCR. PMMoV was used as a normalizer of virus RNA concentrations in this study Wolfe et al., 2022). Each qPCR assaywas performed in a total reaction volume of 25 μL along with primers, probes, and nucleic acid samples (Table S3). PCR amplification was performed with an ABI PRISM 7500 or 7500 Fast Sequence Detection System (Thermo Fisher Scientific). The thermal cycling conditions for PMMoV were as follows: 50°C for 2 min and initial denaturation at 95°C for 10 min to activate the DNA polymerase, followed by 45 cycles of denaturation at 95°C for 3 s and annealing and extension at 55°C for 30 s (Ando et al., 2022). The thermal conditions for IAV assays are as follows: 50°C for 2 min and initial denaturation at 95°C for 15 min followed by 45 cycles of denaturation at 95°C for 15 s and annealing and extension at 56°C for 75 s (Nakauchi et al., 2011). The thermal conditions for RSV assays are as follows: 50°C for 2 min and initial denaturation at 95°C for 10 min followed by 45 cycles of denaturation at 95°C for 15 s and annealing and extension at 60°C for 60 s (Hu et al., 2003). Tenfold serial dilutions of gBlocks for PMMoV, IAV, and RSV (10 5 to 10 1 GC per reaction) were used for the quantification of viral GC numbers in the PCR tubes. The slope, Y-inter, and R 2 of each qPCR assay are summarized in Table S4 and Fig. S1. Nuclease-free water was included in every reaction, and none was positive for all target genes. Amplification data were collected and analyzed using Sequence Detector software version 2.0 (Applied Biosystems 7500).

The EPISENS-M_K method for archived electronegative membranes
The wastewater samples collected between October 2018 and April 2020 were originally intended for the detection of enteric viruses and thus concentrated with an electronegative membrane method with acid rinse followed by alkaline elution according to Katayama et al. (2002). In our previous work during that period, 100 mL of each sample supplemented with 25 mM MgCl 2 was filtered through electronegative membranes (90-mm diameter 0.45-μm pore size; Merck Millipore, Billerica, USA; Catalog no. HAWP-09000), followed by filtration of 200 mL of 0.5 mM H 2 SO 4 and 10 mL of 1 mM NaOH (pH 10 to 11). The remaining membranes that had been stored in −80°C were analyzed as archived samples in the present study along with the EPISENS™ protocol, named as the Efficient and Practical virus Identification System with Enhanced Sensitivity for Membrane coupled with EPISENS-M_K. Specifically, a quarter of the membrane was directly subjected to RNA extraction, followed by RT-Preamp and qPCR as described in Section 2.3.1 (i.e., direct RNA extraction from the membrane and the latter processes were identical to the EPISENS-M method).

Clinical data and statistical analyses
Clinically confirmed cases with influenza and RSV in Sapporo City are reported based on sentinel surveillance. Clinical data on the confirmed numbers of infected with influenza and RSV were downloaded from the city website (City of Sapporo, 2023a, 2023b). Unfortunately, the data on confirmed cases with IAV and each serotype of RSV have not been published in Sapporo City. The data on clinically confirmed cases with SARS-CoV-2 in Sapporo City are shown in the Supplementary Fig. S2. All statistical analyses were performed using R Statistical Computing Software version 4.1.2 (R).

Validation of the detection methods
In this study, concentrations of viral RNA in wastewater collected between October 2018 and January 2023 were determined with two variations of the EPISENS™ method, namely EPISENS-M_K and EPISENS-M. To validate these two methods for the quantification of IAV and RSV RNA in wastewater, a seeding experiment was conducted. Indigenous PMMoV RNA was detected from all virus-seeded samples at the concentrations of 1.07 Â 10 8 (range: 9.77 Â 10 7 to 1.18 Â 10 8 GC/L, n = 3) and 5.32 Â 10 7 (range: 4.44 Â 10 7 to 6.27 Â 10 7 GC/L, n = 3) GC/L with the EPISENS-M_K and EPISENS-M methods, respectively. Both IAV and RSV were detected at similar Ct values and corresponding viral RNA concentrations between the EPISENS-M_K and EPISENS-M methods. IAV was de- We also confirmed that PMMoV RNA was detected in all influent samples collected during the study period (n = 296) with relatively stable concentrations regardless of the detection methods used, WWTP, and seasons (Table 1, Fig. S3). The measured concentrations were 8.00 Â 10 7 (range: 1.24 Â 10 7 to 2.38 Â 10 8 ; n = 33) at WWTP A from 2018 to April 2020, and 4.20 Â 10 7 GC/L (range: 1.01 Â 10 7 to 7.94 Â 10 8 ; n = 133) at WWTP B and 3.89 Â 10 7 GC/L (range: 3.84 Â 10 6 to 1.40 Â 10 8 ; n = 130) at WWTP C from May 2020 to January 2023.

Detection of IAV
To track the circulation of IAV since October 2018, a total of 296 influent wastewater samples were examined with the two methods using four qPCR assays targeting the M gene and HA gene (H1, H1pdm, and H3 types) of IAV. The M gene, which is conserved among IAV strains, was detected in 66.7 % (22/33) of the samples between October 2018 and April 2020 (Table 1, Fig. S4A), but only from 0.79 % (2/254) of the samples between May 2020 and November 2022. In contrast, the M gene was detected from all samples (n = 10) since 13 December 2022 (Table 1, Fig. S4A). The concentrations of the M gene in wastewater increased in the winter season (from October to April). The concentrations normalized with PMMoV fluctuated at the range of <4.86 Â 10 −6 to 2.10 Â 10 −4 from October 2018 to May 2019, <3.17 Â 10 −6 to 1.70 Â 10 −4 from October 2019 to April 2020, and <1.83 Â 10 −6 to 3.62 Â 10 −5 from December 2022 to January 2023, corresponding to the weekly clinically confirmed influenza cases in Sapporo City ranging from 1 to 2264, from 0 to 1779, from 15 to 452, respectively ( Fig. 2A). The normalized concentrations correlated well with the clinically confirmed cases of influenza (Spearman's r = 0.61, p = 0.0003) (Fig. S5A).
To estimate the prevalence of subtypes in the IAV-positive samples, the wastewater collected from October 2018 to April 2020 (n = 33) and from December 2022 to January 2023 (n = 10) were subjected to type-specific detection of HA genes (i.e., H1, H1pdm, and H3 types). The change in the epidemic subtype between 2018 and April 2020 was observed by WBE, as well as clinical surveillance (Table 1, Figs. 2B, S4B). In 2018/2019 season, the normalized concentrations of the H3 type (range: <6.28 Â 10 −5 to 1.69 Â 10 −4 , positive ratio: 50.0 % (7/14)) was higher than those of the H1pdm type (range: <1.10 Â 10 −5 to 8.16 Â 10 −5 , positive ratio: 42.9 % (6/14)), whereas the H1pdm type were dominantly detected at a concentration of <2.90 Â 10 −6 to 1.61 Â 10 −5 without the detection of H3 type in 2019/2020 season. The clinically confirmed cases of H1N1pdm and H3N2 were up to 5 and 12 per week, respectively in 2018/ 2019 season, and 9 and 1 per week, respectively in 2019/2020 season. In 2022/2023 season when clinically confirmed H3N2 cases were up to four per week, H3 type was dominantly detected at the concentration of <7.41 Â 10 −7 to 1.09 Â 10 −4 without the detection of H1pdm type. H1 type was detected in none of the samples collected in the period when no confirmed H1N1 cases were reported.

Detection of RSV
To determine the prevalence of RSV since 2018, influent wastewater samples collected in Sapporo City were examined using three assays for RSV (i.e., the RSV-A, the RSV-B, and the RSV-AB) ( Table 1, Figs. 3, S6). In Sapporo City, confirmed cases with each serotype of RSV were not reported. Both RSV A and B were detected in 42.4 % (14/33) of the identical samples between October 2018 and April 2020 and 3.04 % (8/263) of the samples between May 2020 and January 2023. From 2018 to April 2020, the positive ratio of the RSV A (72.7 %) was almost comparable to that of RSV B (54.5 %) (Fisher's exact test P = 0.20), whereas the relationship changed after the COVID-19 epidemic began in Sapporo City. From May 2020 to January 2023, the positive ratio of RSV A (39.2 %: 103/263) was significantly higher than that of RSV B (6.84 %: 18/263) (Fisher's exact test P < 2.3 Â 1.0 −16 ). In addition, PMMoV-normalized concentrations of the N gene of RSV A were greater (range: <1.06 Â 10 −6 -5.91 Â 10 −4 ) than that of RSV B (range: <1.01 Â 10 −6 -3.07 Â 10 −5 ) (Fig. 3A).

Discussion
Global clinical surveillance has reported a decrease in the prevalence of influenza and respiratory syncytial virus (RSV) since the beginning of the COVID-19 pandemic Feng et al., 2021;Huang et al., 2021;National Institute of Infectious Disease, 2023;Tempia et al., 2021;van Summeren et al., 2021;Weinberger Opek et al., 2021;World Health Organization, 2023a). However, this phenomenon has not been examined with WBE, which reflects the prevalence of infectious diseases in sewer catchments, regardless of clinical testing capacity or any other potential biases. To date, only a few studies have demonstrated the applicability of WBE for tracking the prevalence of influenza virus and RSV Mercier et al., 2022;Wolfe et al., 2022). Here, we highlight the potential application of WBE to understand the circulation of the viruses, including their subtypes, and eventually elucidated the prevalence of the viruses before and during the COVID-19 pandemic in Sapporo City owing to wastewater banking and the two variations of the EPISENS™ method.
The EPISENS-M_K method showed comparable detection sensitivity and quantifiability of IAV and RSV RNA in wastewater to the EPISENS-M method. This is probably because enveloped viruses tend to remain on the membrane even after exposure to H 2 SO 4 and NaOH, as suggested previously by Haramoto et al., 2009 using cyprinid herpesvirus, an envelope virus. This previous study reported an adsorption efficiency of over 87 % for the virus, but <1.5 % recovery in the elution process (Haramoto et al., 2009). Since IAV and RSV are also enveloped viruses, it is reasonable to assume that the seeded IAV and RSV remain on the membrane after the chemical exposure and were recovered directly from the membranes with the EPISENS-M_K method at comparable sensitivity and recovery yields to the EPISENS-M method. The results suggest a minimum difference between the two methods in quantifying IAV and RSV RNA in wastewater.
Through these two methods, the prevalence of IAV, including its HA type, was successfully traced with WBE. Between October 2018 and April 2020, the concentrations of the IAV M gene peaked in the winter ( Fig. 2A), which is consistent with the typical seasonality of clinically confirmed cases in the corresponding area and Japan (Bloom-Feshbach et al., 2013). In terms of subtypes, the epidemic subtype determined by clinical surveillance in each season was detected at higher concentrations than the other subtypes (Fig. 2B). Monitoring an epidemic subtype of IAV in a season is pivotal for making an effective vaccination on IAV in response to seasonal variations in epidemic subtypes. Since an epidemic subtype in a region is likely to spread worldwide, accurate understanding the epidemic subtype in a region with WBE coupled with clinical surveillance contributes to preventing public health crises in other regions by facilitating effective vaccination preparation.
Wastewater surveillance confirmed the co-circulation of RSV A and RSV B before the COVID-19 pandemic and the dominance of RSV A during the COVID-19 epidemic in Sapporo City (Fig. 3A), but the concentrations of RSV RNA (A and B types) did not show similar trends to weekly clinically confirmed cases (Figs. 3B, S5B). This latter observation is presumably attributed to two possible scenarios. The first possible scenario The theoretical limit of detection (LOD) is 49.3 GC/L for the EPISENS-M method and 148 GC/L for the EPISENS-M_K method under the assumption that the recovery efficiency of viral RNA in the whole process is 100 %. ND: not determined. NA: not analyzed. is the discrepancy of corresponding regions monitored by WBE and clinical surveillance. We investigated the concentrations of viral RNA in wastewater at WWTPs whose service coverage is approximately 10-20 % of the city's population (as detailed in Section 2.1), which might not represent the actual prevalence and trend in the entire city. The second scenario is the different groups of the infected monitored types in wastewater, respectively. Red and yellow bars denote the confirmed reported number of the infected with H1N1pdm and H3N2 subtypes, respectively. In this study period, no influenza patient infected with H1N1 subtype was reported in Sapporo City.
by the two surveillances at the city level. Clinical passive surveillance tests symptomatic RSV patients, the majority of whom are considered infants (aged under 3 years) or elderly people (Staadegaard et al., 2021). However, WBE could fail to capture the major group of symptomatic people whom clinical surveillance monitors because they wear diapers and rarely excrete the viruses in sewage. Although our results demonstrated the potential of WBE to mirror the prevalence of RSV, including its subtypes, further research is needed, especially concerning the relationship between RSV concentration and RSV incidences. Since there are no effective vaccinations and therapeutics for RSV, understanding the actual prevalence is pivotal for public authorities to prevent RSV epidemics. Wastewater surveillance for respiratory viral diseases confirmed the disappearance of IAV and RSV since COVID-19 prevalence in Sapporo City (Figs. 2, 3, S2). These findings are consistent with clinical surveillance in the corresponding area and clinical observations around the globe (Eden et al., 2022;Feng et al., 2021;World Health Organization, 2023a). Our findings support the notion that IAV and RSV have actually been absent in the community since the COVID-19 pandemic. The decreased prevalence of these viruses worldwide is probably caused by NPIs mitigating the spreading of SARS-CoV-2. Indeed, the spreading of RSV and influenza in a community was observed after relaxing moving restrictions and reopening schools worldwide (Eden et al., 2022;Emborg et al., 2022;Fourgeaud et al., 2021;Tempia et al., 2021;World Health Organization, 2023a), as COVID-19, influenza, and RSV are respiratory infectious diseases and share common infection routes or are airborne. Interestingly, RSV has become an epidemic in Sapporo since 2021, unlike IAV (Figs. 2,  3). In addition, the co-circulation of RSV A and RSV B was disturbed after the COVID-19 pandemic (Fig. 3A), which is also clinically observed in Australia (Eden et al., 2022). These phenomena remain poorly understood, but might be due to viral interference.
Importantly, our findings about the circulation change of IAV and RSV were accomplished owing to the preservation of wastewater samples (hereafter called "wastewater banking") as archival community-level public health information. Wastewater banking provides opportunities to analyze samples retrospectively to obtain important public health insights at the population level; for example, when a pathogenic virus invades or emerges in a community, how effective public health interventions work for mitigating the prevalence of infectious disease, or whether the characteristics of infectious disease (e.g., epidemic level, genetic distribution, seasonality) change compared to the previous seasons. This concept can be applied to other targets in wastewater, such as chemical substances and antimicrobial resistant bacteria. In this study, we succeefullydetectedPMMoV, IAV and RSV from membranes frozen for at least a few years. These results demonstrated the utility of wastewater banking to obtain novel insights related to improving our health, although the stability of viral nucleic acid under storage, including frozen membranes, should be investigated as suggested by a previous study (Cutrupi et al., 2021).
The present study has several limitations in interpreting data and proposes further studies to understand the actual prevalence of respiratory infectious diseases with WBE. First, we cannot directly discuss the change in the concentrations of IAV RNA and RSV RNA during the study period because sampling locations have changed since May 2020. Although the catchment areas of the three WWTPs are adjacent to each other (Fig. 1), we cannot exclude the possibility of localization of those infected with IAV or RSV in the catchment. Second, we could not identify the potential Fig. 3. The dynamics of the concentration of respiratory syncytial virus (RSV) RNA in influent wastewater and weekly clinical confirmed RSV cases in Sapporo City. The plots denote the wastewater concentrations of viral RNA normalized by PMMoV. The circle plots denote the wastewater concentrations at WWTP A as determined by the EPISENS-M_K method between October 2018 and April 2020, while square and triangle plots denote the wastewater concentrations at WWTP B and WWTP C, respectively, as determined by the EPISENS-M method between May 2020 and January 2023. ND refers to the non-detection of viral RNA from wastewater. The first infected person with SARS-CoV-2 was clinically confirmed on February 14, 2020 in Sapporo City. (A) The blue and yellow (with varied darkness) plots denote the RNV N gene concentrations of the RSV A type and RSV B type, respectively. In Sapporo City, confirmed cases with each serotype of RSV were not reported. (B) The red plots denote the RSV RNA concentrations with an assay detecting both RSV A and RSV B serotypes, and the grey bars denote the confirmed reported number of the infected with RSV in Sapporo City.
of WBE as a leading indicator to clinical surveillance results for respiratory viruses by using correlation analysis considering lag time (Wu et al., 2022) or mathematical models (Ando et al., 2023) that were applied to COVID-19. This is because clinically confirmed cases of influenza and RSV are reported weekly, unlike COVID-19, which has been reported daily. Further research is needed to examine the relationship between the concentrations of the virus RNA in sewage and clinically confirmed cases.

Conclusions
• In this study, we successfully traced concentrations of IAV and RSV, including their subtypes in wastewater, with two variations of the EPISENS™ methods (EPISENS-M_K and EPISENS-M) that demonstrate the applicability of WBE to respiratory viruses. This finding underscores the usefulness of WBE for combating respiratory viral diseases, especially during the COVID-19 pandemic and in subsequent endemic phases. • We observed the dramatic change in IAV and RSV activity since the beginning of COVID-19 and elucidated the actual impact of COVID-19 on the circulations of the respiratory infectious disease viruses. This finding is of scientific and societal importance, and further research is necessary to examine this phenomenon and move toward a better understanding of environmental viral ecology and formulating effective strategies against the prevalence of respiratory viral diseases in the anticipated COVID-19 pandemic era. • The change of the IAV and RSV before and during COVID-19 activity was successfully achieved owing to wastewater banking. Wastewater banking allows for retrospective analysis to discover novel insights related to the improvement of public health and thus, we strongly recommend preserving wastewater samples.

Data availability
The data that has been used is confidential.

Declaration of competing interest
Ryo Iwamoto and Yoshinori Ando are employees of SHIONOGI & Co., Ltd. Masaaki Kitajima who received research funding from SHIONOGI & Co., Ltd. and AdvanSentinel, Inc. and patent royalties from SHIONOGI Co., Ltd. Satoshi Okabe received research funding from SHIONOGI Co., Ltd. and AdvanSentinel, Inc. Hiroki Ando and Warish Ahmed have no competing interests to declare.