PT - JOURNAL ARTICLE AU - Claudio M. Verdun AU - Tim Fuchs AU - Pavol Harar AU - Dennis Elbrächter AU - David S. Fischer AU - Julius Berner AU - Philipp Grohs AU - Fabian J. Theis AU - Felix Krahmer TI - Group testing for SARS-CoV-2 allows for up to 10-fold efficiency increase across realistic scenarios and testing strategies AID - 10.1101/2020.04.30.20085290 DP - 2020 Jan 01 TA - medRxiv PG - 2020.04.30.20085290 4099 - http://medrxiv.org/content/early/2020/05/13/2020.04.30.20085290.short 4100 - http://medrxiv.org/content/early/2020/05/13/2020.04.30.20085290.full AB - Background Due to the ongoing COVID-19 pandemic, demand for diagnostic testing has increased drastically, resulting in shortages of necessary materials to conduct the tests and overwhelming the capacity of testing laboratories. The supply scarcity and capacity limits affect test administration: priority must be given to hospitalized patients and symptomatic individuals, which can prevent the identification of asymptomatic and presymptomatic individuals and hence effective tracking and tracing policies. We describe optimized group testing strategies applicable to SARS-CoV-2 tests in scenarios tailored to the current COVID-19 pandemic and assess significant gains compared to individual testing.Methods We account for biochemically realistic scenarios in the context of dilution effects on SARS-CoV-2 samples and consider evidence on specificity and sensitivity of PCR-based tests for the novel coronavirus. Because of the current uncertainty and the temporal and spatial changes in the prevalence regime, we provide analysis for a number of realistic scenarios and propose fast and reliable strategies for massive testing procedures.Findings We find significant efficiency gaps between different group testing strategies in realistic scenarios for SARS-CoV-2 testing, highlighting the need for an informed decision of the pooling protocol depending on estimated prevalence, target specificity, and high-vs. low-risk population. For example, using one of the presented methods, all 1·47 million inhabitants of Munich, Germany, could be tested using only around 141 thousand tests if the infection rate is below 0·4% is assumed. Using 1 million tests, the 6·69 million inhabitants from the city of Rio de Janeiro, Brazil, could be tested as long as the infection rate does not exceed 1%.Interpretation Altogether this work may help provide a basis for efficient upscaling of current testing procedures, taking the population heterogeneity into account and fine grained towards the desired study populations, e.g. cross-sectional versus health-care workers and adapted mixtures thereof.Funding German Science Foundation (DFG), German Federal Ministry of Education and Research (BMBF), Chan Zuckerberg Initiative DAF and Austrian Science Fund (FWF).Evidence before this study The concept of group testing goes back to mathematical ideas developed in the 1940’s and has already been successfully implemented for various infectious diseases but also in non-medical settings such as testing for failures of electronic components. The issue of group testing for SARS-CoV-2 has been addressed in a number of very recent papers and preprints including feasibility studies of different laboratories and a few methodological overview papers. Nevertheless, to the best of our knowledge, no study provided a comprehensive comparison contrasting hierarchical testing, array testing, and informative testing strategies based on combined groups for stratified populations and relying on up-to-date data about the accuracy of PCR-based test – all of them feasible to be implemented in the current pandemic. Moreover, a discussion of massive informative testing strategies for pandemic scenarios, employing combined pools consisting of high-risk and low-risk individuals, was still missing in the public health literature.Added value of this study We analyse sensitivity, specificity, and throughput of group testing schemes in a series of scenarios tailored to realistic prevalence rates for SARS-CoV-2 in stratified populations and to the characteristics of the qRT-PCR tests used to diagnose COVID-19. Our analysis yields a comprehensive characterisation of a wide range of pooling schemes, broken down by population characteristics, that can serve as a guideline to be queried by testing entities to meet their settings. In particular, our findings demonstrate that a promising strategy to test asymptomatic or presymptomatic individuals in conjunction with high-risk individuals such as healthcare workers is to amalgamate them together in a suitable way, and we provide adequate pool configurations. Such strategies had not been explored for SARS-CoV-2 testing as of yet. We also provide insights on group testing under constraints, i.e. when the number of stages, maximum group size and false negative rate of the whole method are restricted to a certain range of realistic values. We introduce efficient paralleliz-able non-adaptive test procedures for simplified and fast large-scale test design in case of severe shortage of test components. We develop an intuitive web application that can be used by any researcher working on the front line of testing procedures to visualize all the different strategies and to design pooling schemes in an flexible manner according to their specific prevalence scenario and test configuration.Implications of all the available evidence Testing for SARS-CoV-2 presents new challenges to authorities such as rapidly changing prevalence estimates and bottlenecks in testing capacity. In this context, applying an appropriately chosen group testing procedure can allow for up to 10-fold increase of the feasible throughput, such that the existing testing capacities can be used to test a significantly increased number of individuals. As a consequence, adequate tracing and quarantine strategies to reduce community transmission can be established and valuable epidemiological studies relying on accurate prevalence rates can be performed for the ongoing pandemic situation.Competing Interest StatementThe authors have declared no competing interest.Funding StatementCMV gratefully acknowledge support by German Science Foundation (DFG) within the Gottfried Wilhelm Leibniz Prize under Grant BO 1734/20-1, under contract number PO-1347/3-2 and within Germany's Excellence Strategy EXC-2111 390814868. CMV and FK gratefully acknowledge support by German Science Foundation in the context of the Emmy Noether junior research group KR 4512/1-1. TF and FK gratefully support funding by German Science Foundation (project KR 4512/2-2). FJT gratefully acknowledges support by the BMBF (grant# 01IS18036A and grant# 01IS18053A) and by the Chan Zuckerberg Initiative DAF (advised fund of Silicon Valley Community Foundation, 182835). JB and DE gratefully acknowledge support by Austrian Science Fund (FWF) under grants I3403-N32 and P 30148. PG and PH declare that no external funding was received and DSF acknowledges support from a German Research Foundation (DFG) fellowship through the Graduate School of Quantitative Biosciences Munich (QBM) [GSC 1006 to D.S.F.] and by the Joachim Herz Stiftung. Author DeclarationsAll relevant ethical guidelines have been followed; any necessary IRB and/or ethics committee approvals have been obtained and details of the IRB/oversight body are included in the manuscript.YesAll necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).YesI have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.YesFor comparative visualization and querying of the precomputed results we provide an interactive web application. The source code for computation is open and freely available. https://gitlab.com/hararticles/group-testing-simulations http://www.group-testing.com/