Telemedically Assisted Sampling of COVID-19 Patients - Is the Sampling Quality Sufficient

Overview

The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) pandemic has resulted in more than 3.8 billion registered tests, 275 million positive cases, and 5 million deaths worldwide. Early and regular testing has been an important pillar of secondary prevention since the beginning. However, this pandemic has also fostered solutions in the form of e telemedicine with enormously increased applicability. The question of whether telemedically supervised testing with SARS-CoV-2 Rapid Antigen Tests is non-inferior to the same tests being carried out by trained personnel in test centers is still unanswered. With this study, the investigators aim to compare and evaluate the reliability and sampling quality of telemedically guided self-performed rapid tests for professional use compared to professional sampling by healthcare personnel. Our hypothesis is that, applying a strict standard operating procedure (SOP, attached), guided oropharyngeal + nasal (OP+N) self-sampling (GSS) is non-inferior to nasopharyngeal (NP) or OP+N sampling performed by health care professionals (HCP), and that guided OP+N sampling is superior to unsupervised OP+N self-sampling (USS).

BACKGROUND AND PROJECT RATIONALE The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) pandemic has resulted in more than 3.8 billion registered tests, 275 million positive cases, and 5 million deaths worldwide. Early and regular testing has been an important pillar of secondary prevention since the beginning. However, this pandemic has also fostered solutions in the form of e telemedicine with enormously increased applicability. Several emerging issues, such as the use of technology by older adults, the availability of high-capacity Internet access, and support for telemedicine, can be overcome in the near future to increase the benefits of telemedicine. The question of whether telemedically supervised testing with SARS-CoV-2 Rapid Antigen Tests is non-inferior to the same tests being carried out by trained personnel in test centers is still unanswered. With this study, the investigators aim to compare and evaluate the reliability and sampling quality of telemedically guided self-performed rapid tests for professional use compared to professional sampling by healthcare personnel. Our hypothesis is that, applying a strict standard operating procedure (SOP, attached), guided oropharyngeal + nasal (OP+N) self-sampling (GSS) is non-inferior to nasopharyngeal (NP) or OP+N sampling performed by health care professionals (HCP), and that guided OP+N sampling is superior to unsupervised OP+N self-sampling (USS). PROJECT OBJECTIVES AND DESIGN 2.1 Hypothesis and primary objective Our hypothesis is that, applying a strict standard operating procedure (SOP, attached), guided oropharyngeal + nasal (OP+N) self-sampling (GSS) is non-inferior to nasopharyngeal (NP) or OP+N sampling performed by health care professionals (HCP), and that guided OP+N sampling is superior to unsupervised OP+N self-sampling (USS). OP+N GSS is non-inferior to nasopharyngeal (NP) sampling performed by HCP OP+N GSS is non-inferior to OP+N performed by HCP OP+N GSS is superior to unsupervised OP+N USS The investigators are also interested to know if the difference between the sampling techniques is the same depending on the SARS-CoV-2 viral load (low, medium, high). Another secondary endpoint will be in how many people a correctly executed nasopharyngeal sampling is possible. Due to lack of comfort, deviation of septum, as well as nasal polyps in a small percent of the population, the investigators hypothesize that in a majority of people it is not possible to correctly execute a nasopharyngeal sampling. 2.2 Primary and secondary endpoints The Ct-value is defined as the number of cycles required for a signal to cross the detection threshold and can be used as a surrogate for sampling quality, as it allows to quantify the amount of material that is being extracted. Primary endpoints: Concordance of positive/negative samples in antigen tests with GSS, USS, and HCP-based samples. Secondary endpoints: Housekeeping gene Ct comparison based on sampling method between samples based on GSS, USS, or HCP ORF1ab/RdRp-, E-, N-, and S genes Ct value comparison based on sampling method between samples based on GSS, USS, or HCP Qualitative analysis of perceived sampling accuracy established with a short form Detected variant of concern Case report form (CRF) assessing basic demographics and variables of interests as indicated separately in the CRF form Prevalence of variants of concern Immunization status Travel destinations in the passed three months n of people in which a nasopharyngeal sampling is possible correctly (ratio) 2.3 Project design This is a single-center, randomized, controlled study at a Swiss test station. 2.3.1 Piloting In order to assess the sampling quality by HCP or telemedically guided self-sampling, Ct values obtained in RT-PCR analysis of a COVID-19 assay, a housekeeping gene can be used for reference. Hence, for the first part of the study, the investigators will recruit 10 people in which the investigators will perform an OP+N sampling at two-time points, 15 minutes apart, and compare the Ct values of these two values to calculate the standard deviation and see if the produced SD\*1.96/√(Sample size) equals the margin of error. Our 95% confidence interval (CI) will be standard deviation (SD)+/- the margin of error. Alternatively, the investigators will pool confirmed RT-PCRs of two/three swab samples into one and spike them with SARS-CoV-2 standard. Next, the investigators do as many repeats of the same sample on RT-PCR comparing housekeeper and SARS-CoV-2 genes. The advantage here is that the investigators will have controlled starting biological material and the resultant SD / 95% CI will concern only the analytical and laboratory process variability. With the first approach, the investigators will test SD of the swab procurer too, along with the interference coming from the analytical process (lab, instruments, sample storage). 2.3.2. Randomized telemedical and HCP sampling After giving informed consent by signing the consent form, patients will be randomized to either OP+N GSS - or to OP+N USS. After performing either procedure, they will continue to be sampled by HCP OP+N and HCP NP. The investigators will start the assessment with the self-sampling methods in order to reduce bias that could occur if they experience the HCP's sampling. Written instructions on how to perform a self-test will be provided, representing the instructions that come along with a commercially available rapid test set. An instructional online video tutorial or similar would also account as the latest state of the art.30 However, this has not yet been implemented comprehensively with all available test sets that are currently being distributed across Switzerland. Furthermore, to avoid bias in the sampling procedure, exclusively one HCP at the testing station will be designated as the study's sampling expert. The HCP will perform the NP and OP+N sampling sequentially in a previously randomized order. The telemedical guidance will be exclusively done by the same HCP in order to minimize interoperability differences. The standardized telemedical sampling will be conducted at the same test station, in a separate empty room in which a computer will be placed with instructions on how to connect for the telemedical consultation. During each sampling procedure, the patients will either undergo OP+N GSS or OP+N USS for PCR analysis, as well as for antigenic testing. This means each participant will be sampled a total of six times within 15 minutes from different locations in the upper respiratory tract. Concordance of results between sampling by GSS vs. USS vs. HCPs will be assessed. As the difference in time between the samplings can influence the Ct value, the sampling process will only be started when the HCP and the telemedical supervisor are ready. The three different samples will be stored in the test station refrigerator between 2°C-8°C before transport. A grouping of samples will ensure that temperature changes are consistent between samples from the same participant. PROJECT POPULATION AND STUDY PROCEDURES 3.1 Project population, inclusion and exclusion criteria All patients, regardless of symptoms, who present themselves at the Test-station SARS-CoV-2 testing will be offered to participate in the study. Study inclusion criteria; Written informed consent Patients visiting the test-station for a SARS-CoV-2 test Consent to comply with the study protocol Age of 18 or older List the study exclusion criteria: Refusal to comply with the study protocol Digital illiteracy (i.e. patients who would never use telemedicine as they do not feel competent enough to use the necessary tools) Not German or English speaking People trained in the use of rapid diagnostic tests (HCP, biologists) will be excluded to reduce bias30 3.2 Recruitment, screening and informed consent procedure All costs for tests performed within the framework of the study will be financed by the study. A payable travel certificate will be issued for unchanged costs. Participation in the study and spending time on it is voluntary, free of charge and at no cost to the study participants. However, each study participant will receive a voucher for free PCR testing at a later date. Participants will not receive financial compensation for the participation in this study. However, each study participant will receive a voucher for free PCR testing at a later date. The investigators shall end recruiting either when reaching the necessary sample size or no later than October 30th, 2022. STATISTICS AND METHODOLOGY 4.1. Statistical analysis plan 4.1.1 Randomization To ensure allocation concealment, randomization will be performed with a block of 170 in a 1:1 allocation ratio by a web-based independent randomization software (Random Allocation version 1.0). Patients will be randomly assigned to either OP+N GSS or to OP+N USS. Instruction on the assigned procedure will be kept in sealed opaque envelopes labeled with sequential study numbers and opened right before swab-testing. 4.1.2 Sample size The SD of the mean of Ct values for RT-PCR analysis of multiple samples is not clearly defined in the literature and varies between studies.34,35 If there is truly no difference between the standard and experimental treatment, then 85 patients per group are required to be 90% sure that the lower limit of a one-sided 97.5% confidence interval (or equivalently a 95% two-sided confidence interval) will be above the non-inferiority limit of -3. We, therefore, estimated a standard error of six that accounts for this variance. In addition, the limit of non-inferiority was set at three according to clinical significance. Results will be stratified by symptom status, age, and sex. 4.1.3 Hypothesis Primary outcome NP HCP vs OP+N HCP H0: μe - μs ≤ -d -\> μ(NP HCP ct) - μ(OP+N HCP ct) ≤ -3 H1: μe - μs \> -d -\> μ(NP HCP ct) - μ(OP+N HCP ct) \> -3 OP+N HCP vs OP+N GSS H0: μe - μs ≤ -d -\> μ(OP+N HCP ct) - μ(OP+N GSS ct) ≤ -3 H1: μe - μs \> -d -\> μ(OP+N HCP ct) - μ(OP+N GSS ct) \> -3 OP+N HCP vs OP+N USS H0: μe - μs ≤ -d -\> μ(OP+N HCP ct) - μ(OP+N USS ct) ≤ -3 H1: μe - μs \> -d -\> μ(OP+N HCP ct) - μ(OP+N USS ct) \> -3 Secondary outcome: We will stratify the primary outcome by viral load (low, medium, high). 4.1.4. Statistical test A variant of a paired student t test or Wilcoxon test (if non-normal distribution) with exchange of null and alternative (relaxed) hypotheses will be used as a non-inferiority test to assess differences in cycle threshold (Ct, considered as a log2-transformed measure of the amount of viral RNA load) between the NP and OP/N samples (HCP and GSS). Stratification for possible confounders will be done using logistic regression. In order to show agreement between the approaches and estimate the limits of agreement and those are not too large, different graphical methods, such as Bland Altman diagrams will be used to show agreement between the approaches. 4.2. Handling of missing data Analysis is based on intention-to-treat assumption. Collected data until drop-out will be included for further analysis.