Pooling for SARS-CoV-2 Control in Care Institutions

Background Workers and residents in Care Homes are considered at special risk for the acquisition of SARS-CoV-2 infection, due to the infectivity and high mortality rate in the case of residents, compared to other containment areas. The role of presymptomatic people in transmission has been shown to be important and the early detection these people is critical for the control of new outbreaks. Pooling strategies have proven to preserve SARS-CoV-2 testing resources. The aims of the present study, based in our local experience, were (a) to describe SARS-CoV-2 prevalence in institutionalized people in Galicia (Spain) during the Coronavirus pandemic and (b) to evaluate the expected performance of a pooling strategy using RT-PCR for the next rounds of screening of institutionalized people. Pooling have to resources and time with an increase in testing of the 69% for an incidence rate of SARS-CoV-2 infection of 10% or less [4,5,6,7], but it could be associated with a decrease in detection [8,9]. Main limitations could be the preanalytical step, the sample viral load or the increase of the limit of detection of the individual sample [10]. The rationale in this study is to develop a new strategy based on initial individual identication of positive coronavirus cases in order to organize low prevalence clusters, followed by a serial pooling strategy testing of these clusters, in order to control areas free of virus circulation, allowing them to be fully operative.

Results Distribution of SARS-CoV-2 infection at Care Houses was uneven. As the virus circulation global rate was low in our area, the number of people at risk of acquiring the infection continues to be very high.
In this work, we have successfully demonstrated that pooling of different groups of samples at low prevalence clusters, can be done with a small average delay on quanti cation cycle (Cq) values.
Conclusions A new surveillance system with guaranteed protection is required for small clusters, previously covered with individual testing. Our proposal for Care Houses, once prevalence zero is achieved, would include successive rounds of testing using a pooling solution for transmission control preserving testing resources. Scale-up of this method may be of utility to confront larger clusters to avoid the viral circulation and keeping them operative.
Surveillance of Care Homes has been critical to limit the mortality rate in Galicia (Spain). Direct viral detection by real time RT-PCR was useful to identify people with potential SARS-CoV-2 transmission risk. Limited stocks and restrictions in test capacity prevented a higher number of RT-PCR tests per day.
Pooling strategies have proven to preserve SARS-CoV-2 testing resources and time with an increase in testing capability of the 69% for an incidence rate of SARS-CoV-2 infection of 10% or less [4,5,6,7], but it could be associated with a decrease in detection [8,9].
Main limitations could be the preanalytical step, the sample viral load or the increase of the limit of detection of the individual sample [10].
The rationale in this study is to develop a new strategy based on initial individual identi cation of positive coronavirus cases in order to organize low prevalence clusters, followed by a serial pooling strategy testing of these clusters, in order to control areas free of virus circulation, allowing them to be fully operative.  Pool positivity assessment According to other authors [11], for prevalence between 1 and 2%, sensitivity 95% and speci city 100%, the optimal pool size would be between 25 and 16 samples and the optimal sub pool size would be between 4-5 samples. In order to minimize the false negative factor for pooled testing recently de ned [12]and to standardize the pooling method, pools of twenty samples (P20) and sub pools of ve samples (SP5) were selected.
Test performance of twenty-six P20 and fourteen SP5 was studied. Each pool included one positive Mean delay in the Cq values (Cq pool-Cq positive sample) was 5.02 cycles for the P20 and 2.85 cycles for the SP5 ( Table 1). The N gene was not detected by Allplex™2019-nCoV assay in one speci c sample independently of pooling or individual testing.

Proof of concept
Samples from Care Homes selected by prevalence were retrospectively tested in pools using the following algorithm: P20, SP5 when positive, individual analysis when positive. A rst simulation was performed with 100 samples from 2% (2/100) prevalence Care Homes. Five P20 were tested. As 2 positive pools were obtained, 8 SP5 were processed. Two SP5 were positive, so 10 samples were tested individually.
Two samples were positive. Number of tests was reduced 77% (0.23 tests per individual).
A second simulation included 60 samples from 1.7% (1/60) prevalence institutions. Three P20, 4 SP5 and 5 individual samples were tested. One sample was positive. Number of tests was reduced by 80% (0.20 tests per individual).

Discussion
A global SARS-CoV-2 seroprevalence of 5% in Spain [13] and a global viral prevalence around 3% at Care Houses reported in the present study, suggest that the number of people at risk of acquiring the infection continue to be very high. The role of presymptomatic individuals in transmission has been shown to be important [14, 15,16] and their early detection seems critical to prevent further outbreaks. To control the spread of the virus, it is essential to detect as many infected individuals as possible, as quickly as possible to trace down and test possible contacts [17].
We performed the screening of 306 Care Homes (25,386 determinations) in workers and residents using individual testing by RT-PCR. With a prevalence < 2% for more than 85% people in Care Houses, pooling could achieve maximum usefulness. After reviewing the literature, and due to the absence of accumulated experience with this type of strategy for SARS-CoV-2, a more conservative sizepools of 20 and sub pools of 5 samples were chosen [4] [5] [18].
Two tests authorized by the Food and Drug Administration were available at our laboratory. Both have shown suitable speci city and sensitivity for clinical diagnosis, but speci c studies will be required for assessing their performance in pooling conditions. The choice of the Allplex™2019-nCoV Assay was due to the exibility and adaptability in the automation process useful for future interventions. Additionally, although it has been established a moderate mutation rate of SARS-CoV-2 [19,20,21], the possibility of detecting three targets could increase the possibilities of detection [20,22].
As previous studies [8,9,18], our results using pools showed an increase of 3-5 cycles in the Cq value between pooled tests and individual positive samples. It has not carried out loss of sensitivity in pools for samples with Cq value within the rst three quartiles observed in our population. A 100% sensitivity was also achieved when testing Care Homes with prevalence around 2%, reducing until 80% the number of tests.

Proposed methodology
Here there is our proposal for introducing the pooling strategy in Care Institutions: When an institution with prevalence zero is characterized, successive rounds of pooling testing would be the option for transmission control. The maximum interval between rounds would be adjusted to avoid the loss of detection of infected people who could be in a phase of low viral load. The incubation period has been reported to be highly variable with an estimated average of 5-6 days [23,24,25]. Viral load could be detectable 2-3 days before onset for a median of 20 days after symptoms onset, although levels drop signi cantly after day 10 [15,26]. In this study we have focused on demonstrating that any pool containing individual samples from highly infectious people would be detected. Samples with Cq values associated with low viral load were excluded (fourth quartile) as this would not be the expected situation. In fact, the expected arrival of new cases would be of presymptomatic or initial symptomatic contagious individuals with viral loads reaching their peak, as multiple rounds of screening are being considered and positive and symptomatic people are excluded from pooling testing. In this situation, considering a doubling time higher than 14 days, test rounds of 14 days will allow us to detect any new highly infectious person. This would allow us to introduce highly effective contact tracing and case isolation.
This strategy would be enough to control new outbreaks of COVID−19. The frequency of rounds should also be adjusted by local doubling time and serial interval in the case that a positive appears in the cluster. It seems reasonable to adapt time for new testing according to virus circulation, local rate of infection or risk severity and population tolerance of the sampling method.
Limitations of this study were the limited number of samples included. Testing more negative samples would allow us to assess speci city and the risk of contamination along the processing.
This work has shown the prevalence of SARS-CoV-2 in Spanish Care Homes during the Coronavirus pandemic. Prevalence differences shown between Institutions should address the interventions for viral transmission control. Few studies have assessed the performance of pooling for SARS-CoV-2 detection by rRT-PCR in real conditions, especially when aiming to keep areas free of virus circulation to be operative and functional.

Conclusions
Sample pooling could be a new testing strategy relevant for maintaining low level or no transmission among institutionalized people. Our proposal for Care Houses, once prevalence zero is achieved, would include successive rounds of testing using a pooling solution for transmission control preserving testing resources. Scale-up of this method may be of utility to confront larger clusters to avoid the viral circulation and keeping them operative. Further studies with self-sampling methods, modular systems and more speci c pooling strategies will be necessary for the process improvement. Availability of data and materials

Abbreviations
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. All data and materials were obtained working under Servizo Galego de Saúde, Consellería de Sanidade of Xunta de Galicia (institutions belonging to our National Health Public System) and are under their regulations.