Implementing the Lolli-Method and pooled RT-qPCR testing for SARS-CoV-2 surveillance in schools: a pilot project

The SARS-CoV-2 pandemic implies major challenges for our society and has a strong impact on various aspects of social life [1, 2]. Children and adolescents have been markedly affected by school closures, ranging from impacts on education to mental health [3, 4]. Closing educational institutions is likely to exacerbate social disparities, as children from less advantaged backgrounds tend to be most affected [5]. Therefore, keeping schools open is of utmost concern [6].

Children often show mild courses of SARS-CoV-2 infection or may even present without symptoms [7]. However, they still may be infectious with high viral loads [8]. Furthermore, recent observations indicate a shift of infection pattern toward the younger unvaccinated population, as SARS-CoV-2 vaccines have only recently been licensed for children between 5 and 11 years of age [9]. Thus, systematic testing of the general, asymptomatic population was proposed early in the pandemic for limiting transmissions [10] and needs to be established in schools for surveillance and infection control. However, conducting regular testing poses a substantial challenge to laboratory capacities [10].

From November to December 2020, the multicentre intervention study “Bundesweites Forschungsnetz Angewandte Surveillance und Testung” (B-FAST) was initiated with the intention of developing comprehensive and scalable surveillance strategies. Within this controlled randomized study with 3,970 participants [11], we implemented the Lolli-Method, a non-invasive self-sampling method for saliva samples, combined with RT-qPCR pooled testing as a highly sensitive and broadly accepted method for SARS-CoV-2 screening at educational institutions. The diagnostic sensitivity of the Lolli-Method was shown to be 93.9% when the viral load of corresponding Np-/Op-swabs was > 10³ copies/ml¹². Sample collection consists of sucking on a swab (like a lollipop). The swabs of the entire class are jointly being analyzed for SARS-CoV-2 by pooled RT-qPCR in the laboratory. However, the presence of medical staff for supervision of the sample collection was mandatory during the study period for regulatory reasons. Thus, it could not be determined whether the method also proves to be applicable in everyday school life without the additional support of medical staff. This pilot project aimed at determining the feasibility of the Lolli-Method, implementable in all types of schools.

With this pilot project, we present a feasible and accepted testing strategy, suitable for everyday use in educational institutions without requiring additional on-site medical staff. The Lolli-Method is a child-friendly and safe method of sampling. Students, also those of young age, are able to perform sampling independently without assistance. Surveillance strategies using pooled samples have been previously described in study settings leading to a scalable and resource-efficient screening tool [13‐16]. It allows the simultaneous sample collection of up to 30 individual swabs (e.g., an entire school class), providing a time and cost-effective testing method for SARS-CoV-2 surveillance [11] in educational institutions. During the project period, we detected seven positive pooled tests (0.9%), identifying seven SARS-CoV-2-positive students from elementary schools. As our project was performed in March 2021 where incidence rates of SARS-CoV-2 were decreasing in Germany, the low rate of case detection is not surprising [17, 18] (Table 1). In most schools, the testing was carried out only once a week per student. To minimize the number of missed infections, it is conceivable that testing should be performed more often [12].

The aim of the present project was to assess the feasibility of the surveillance program based on the Lolli-Method and RT-qPCR pool analysis in terms of acceptance and logistics. Even though the initial set-up appears to be relatively complex, since test days have to be determined for each school, routes for the laboratory specimen transport have to be planned and test material has to be delivered, we hereby prove feasibility once the structures are established. Results of our online survey indicate, that the Lolli-Method can easily be integrated in everyday school life without disrupting the teaching schedule (Fig. 3). Sweeney-Reed et al. assessed the acceptance of gargle samples taken at home combined with pooled tests [19]. When asked about the preferred location for sample collection, most participants preferred sample collection at school. One reason could be, that the presence of teachers ensures the test is carried out correctly.

Within this pilot project, an individual lolli swab of each participating student was collected, but only analyzed if the pooled test turned out to be positive for SARS-CoV-2. This approach not only has the disadvantage that many swabs had to be discarded, but also that it may not be feasible in case of supply shortages, which have been common during the pandemic. An alternative, and thus more resource-saving option, would be to perform a single test on the following day in case of a positive pooled test to identify the SARS-CoV-2 positive individual. The result of a positive pooled test, as well as the result of the subsequent analyzed individual sample, was available on average 6–8 h after the samples had reached the laboratory. Thus, respective schools were notified on the same day to prevent students from a SARS-CoV-2-positive pool coming to school the following day.

An alternative, widely used method for regular screening is based on rapid antigen detection tests (RADTs) [20, 21]. In contrast to the Lolli-Method, RADTs provide immediate results. Arguably, the delay of provided results in the RT-qRCR-pooled tests may be seen as a limitation of this screening method as potentially infectious students remain at school. On the other hand, RT-qPCR-pool testing reduces stigmatization as a student’s positive tests results are not immediately disclosed to their classmates leading to an increased acceptance of this surveillance method. In addition, we have recently compared the sensitivity of RADTs with that of the Lolli-Method obtained by RT-qPCR pool tests [12, 22, 23]. The Lolli-Method has proven to be more sensitive and therefore has the potential to detect SARS-CoV-2-positive individuals before they become highly infectious [12, 22]. Furthermore, handling of RADTs appears to be more difficult than the Lolli-Method, especially for young children that may need help from school staff.

Some errors occurred during sample collection that partly hindered correct processing of the samples. This may be attributable to lack of experience of teaching staff in handling of medical equipment or limited understanding of the procedure. However, those difficulties were fully eliminated after the first week of testing, demonstrating that the sample collection using the Lolli-Method can be conducted by non-medical staff with some practice. In addition, uncertainties in labeling samples may be avoided using pre-labeled testing material. Providing sufficient information and timely test results is key to build trust in the project among teachers as well as students and their families. Ultimately, a medical procedure is transferred to a school setting.

We were able to demonstrate that the Lolli-Method as a screening strategy for SARS-CoV-2 in the daily school routine is feasible and can be applicable on a large scale. Based on the experience gained from the pilot project, the test concept was implemented as a SARS-CoV-2 screening program at elementary schools and special needs schools in North-Rhine Westphalia [12]