key: cord-1041623-rd0nt1z9 authors: Zurita Cruz, N. D.; Martin Ramirez, A.; Rodriguez Serrano, D. A.; Gonzalez Alvaro, I.; Roy Vallejo, E.; De la Camar, R.; Fontan Garcia-Rodrigo, L.; Cardenoso Domingo, L. title: Usefulness of real-time RT-PCR to understand the kinetics of SARS-CoV-2 in blood: a prospective study. date: 2022-03-08 journal: nan DOI: 10.1101/2022.03.07.22271764 sha: a8e29a5f640bfca0217781e8739ebdd20f90ab79 doc_id: 1041623 cord_uid: rd0nt1z9 Background SARS-CoV-2 viral load and kinetics assessed in serial blood samples from hospitalised COVID-19 patients by RT-PCR are poorly understood. Methods We conducted an observational, prospective case series study in hospitalised COVID-19 patients. Clinical outcome data (Intensive Care Unit admission and mortality) were collected from all patients until discharge. Viremia was determined longitudinally during hospitalisation, in plasma and serum samples using two commercial and standardised RT-PCR techniques approved for use in diagnosis of SARS-CoV-2. Viral load (copies/mL and log10) was determined with quantitative TaqPath TM COVID-19 test. Results SARS-CoV-2 viremia was studied in 57 hospitalised COVID-19 patients. Persistent viremia (PV) was defined as two or more quantifiable viral loads detected in blood samples (plasma/serum) during hospitalisation. PV was detected in 16 (28%) patients. All of them, except for one who rapidly progressed to death, cleared viremia during hospitalisation. Poor clinical outcome occurred in 62.5% of patients with PV, while none of the negative patients or those with sporadic viremia presented this outcome (p<0.0001). Viral load was significantly higher in patients with PV than in those with Sporadic Viremia (p<0.05). Patients presented PV for a short period of time: median time from admission was 5 days (Range=2-12) and 4.5 days (Range=2-8) for plasma and serum samples, respectively. Similar results were obtained with all RT-PCR assays for both types of samples. Conclusions Detection of persistent SARS-CoV-2 viremia, by real time RT-PCR, expressed as viral load over time, could allow identifying hospitalised COVID-19 patients at risk of poor clinical outcome. Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) was first described in 63 December 2019 in Wuhan, China (1). As of 19th of January 2022, 5,542,359 deaths were reported 64 to WHO(2) . 65 While most COVID-19 patients present mild disease, others develop a severe disease (3). 66 Identifying patients at risk of developing severe COVID-19 is an unmet need to improve the 67 management of these patients and prevent morbidity and mortality. Some risk factors of poor 68 prognosis, such as elevated concentrations of interleukin (IL)-6, contribute to risk stratification 69 of . 70 Detection of SARS-CoV-2 RNA by real-time reverse transcription polymerase chain reaction 71 (RT-PCR) in nasopharyngeal swabs is the gold standard for COVID-19 diagnosis (7). However, 72 no consistent association has been reported between the viral load in these samples and disease 73 severity (8). 74 Recently, detection of SARS-CoV-2 RNA in blood samples (viremia) has been considered as a 75 potential predictor of poor prognosis, due to its association with rapid deterioration and death (9-76 13). Nevertheless, there is a lack of reports analysing viral load quantification over time in 77 sequential samples, which could allow understanding viremia kinetics and its relationship with 78 the patient clinical outcome. 79 The objectives of this prospective study were to analyse the kinetics of SARS-CoV-2 viremia, by 80 qualitative and quantitative RT-PCR methods using standardised commercial reagents in 81 sequential samples collected longitudinally from COVID-19 patients during hospital admission, 82 and to determine its relationship with disease clinical course. 83 84 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 This is a prospective longitudinal case series study, conducted at Hospital Universitario de La 87 Princesa, Madrid (Spain), from November 2020 to January 2021. Fifty-seven consecutively 88 admitted patients with COVID-19 diagnosed by RT-PCR from nasopharyngeal swabs were 89 included. All patients gave their informed consent, and the research protocol was approved by the 90 Institutional Ethics Review Board (register number 4267). We followed guidance from The 91 Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) standards for 92 observational research, and those of the Updated List of Essential Items for Reporting Diagnostic 93 Accuracy Studies (STARD) (14) . 94 In addition to sociodemographic variables, clinical characteristics and outcome variables 96 (Intensive Care Unit [ICU] admission and mortality during hospitalisation) were collected. "Poor 97 Outcome" was considered when the patient had at least one of the two clinical variables described 98 above. Other relevant variables collected were symptom onset, hospital admission, hospital 99 discharge, death and sample collection dates. All data collected was entered in an anonymized 100 database with access only to the research team. 101 First, a blood sample (serum and plasma) was collected within the first 24-36 hours upon 103 admission; subsequently, additional blood samples were collected every 48-72 hours during the 104 first week. Later, samples were collected twice a week until discharge. Samples were frozen at -105 80ºC until RT-PCR performance. 106 All blood samples were assessed by two qualitative RT-PCR methods: Cobas®SARS-CoV-2 108 Test, Roche Diagnostics (cobas®-test), an automated method for SARS-CoV-2 detection; and 109 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Thermocycler manufacturer specifications (15). Detailed description of standard curve 120 determination is shown in supplementary material. Samples were considered quantifiable when 121 mean Ct in the duplicate test for each gene was ≤37 and standard deviation (SD) was <0.5. All 122 results not fulfilling these criteria and/or those with detection in only one duplicate, were 123 considered positive, but not quantifiable. To estimate a value for quantification limit, median of 124 all non-quantifiable values was calculated, obtaining 17.75 copies/mL for serum samples and 125 14.89 copies/mL for plasma samples. For the genes that met quantification criteria, mean quantity 126 was calculated from the duplicates, expressed as copies/mL and logarithm with base 10(log10). 127 Two positive controls (corresponding to 20,000 and 200 copies) and two negative controls were 128 added in each run in duplicate. 129 Moreover, we studied the number of genes (N, S or ORF) that were amplified in each plasma 130 sample tested by q-RT-PCR, as an indication of the association of viremia with the presence of 131 The presence of SARS-CoV-2 RNA was detected by qualitative RT-PCR methods and viral load 134 was determined by qTaqPath-test in plasma and serum of each patient collected throughout the 135 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Fifty-seven patients consecutively admitted to our hospital were included in this study ( Figure 1 A total of 598 samples (298 serum and 300 plasma samples) were assessed ( Figure 1 ). 150 151 152 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. respectively) and those obtained by cobas®-test (35.34±2.12 and 35.16±2.33, respectively) did 162 not show significant differences (p>0.5). Pearson correlation analysis of mean Ct showed 163 correlation coefficients (r) of 0.86 and 0.76 for plasma and serum samples, respectively ( Figure 164 2). Results of the correlation of Ct obtained by both techniques in serum samples are consistent 165 with findings described in our previous series(16). 166 167 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Viremia (PV), viremia was detected in two or more consecutive determinations, n=16 (28%); ii) 199 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. viremias assessed by the 2 qualitative techniques (expressed a Ct) and viral load assessed by the 218 q-TaqPath test (expressed as copies/mL) were determined in plasma and serum samples. Kinetics 219 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 curves represent the change of these parameters in individual PV patients during hospitalisation 220 for the qualitative ( Figure 6A ) and quantitative ( Figure 6B ) determinations. Curves obtained for 221 each individual patient showed similar kinetic behaviour regardless of the technique or the type 222 of sample used for the analysis. These results suggest that PV could be monitored using plasma 223 or serum samples and both, qualitative and quantitative RT-PCR methods. 224 225 226 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. both groups was analysed with the χ2 test. 246 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 No difference was detected in age (p=0.2) or sex (p=0.4) between patients with different viremia 248 The viral load had a normal distribution when it was expressed as log10. Regarding viral load in (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 The hypothesis of this study was that detectable SARS-CoV-2 viremia in successive samples over 265 time (persistent viremia) could identify hospitalised COVID-19 patients with high risk of poor 266 clinical outcome. To verify this, we studied methods for qualitative detection of SARS-CoV-2 267 RNA and for quantification of viral load in plasma and serum, using commercially available RT-268 PCR kits, marked with CE and FDA authorisation. In addition, the association of viremia 269 evolution with patient outcome was also evaluated. reproducible. However, some discrepancy values were detected in isolated cases where the Ct 281 value or viral load was lower than those in the paired sample from the same blood extraction 282 (plasma vs. serum). We hypothesised that those discrepancies could be due to a loss of sample 283 integrity.(24) However, the fact that different techniques can be used to analyse the kinetics of 284 the SARS-CoV-2 viremia increases the versatility of this determination when it comes to its 285 implementation in clinical microbiology laboratories. 286 Our previous results suggested that the Ct value is a good approximation for the stratification of 287 severely ill patients(13). However, the variation due to the sample type and handling, the 288 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 technique used, and the intrinsic variability of RT-PCR makes it necessary to develop a 289 standardised method to obtain more reproducible results(25). Even in this series, both assays 290 detected a higher number of positives in plasma than in serum. In this sense, it is well known that 291 the quantification of viral load in plasma has been standardised in other pathologies associated to 292 viral infections such as HIV, HCV, HBV, CMV(26-29), and provides a more precise and reliable 293 monitoring of viremia than Ct values. For this reason, we propose the use of SARS-CoV-2 viral 294 load from plasma samples for the analysis of viremia kinetics. 295 Different authors have quantified viremia using new technologies such as Droplet Digital PCR 296 (18, 31) using internally developed methods with primer design (17), or ultrasensitive quantitative 297 RT-PCR(32). In this context, the objective of this work was to assess the usefulness of a 298 quantification method using RT-PCR to monitor SARS-CoV-2 viral load in hospitalised COVID-299 19 patients. We used commercial and standardised reagents with CE and FDA marking for 300 authorisation, making a standard curve and taking advantage of the options offered by the 301 QuantStudio 5 thermal cycler interpretation software. Taken all together, these characteristics 302 provide a robust and reproducible quantification method that allows monitoring viremia in 303 hospitalised COVID-19 patients. Also, we agree with other authors on the need to develop 304 standardised techniques for quantifying viral load in blood, with FDA or CE approval, that could 305 contribute to improve the management of patients with 21) . 306 The detection of SARS-CoV-2 RNA in blood has been called RNAemia by some authors (24, 30, 307 31). However, we refer to this determination as viremia since, in almost 80% of the samples 308 studied in this study, 2 or 3 SARS-CoV-2 genes were detected; the presence of these genes could 309 be correlated with the presence of viral particles (32, 33). In addition, detection of RNA in 310 successive blood samples would indicate the presence of the virus. 311 The present study has certain limitations. First, the number of patients included in the study was 312 moderate (57 patients). Second, although two RT-PCR techniques were performed, only the 313 TapPath™ test allowed us to quantify the viral load (in copies/ml and log10 viral load). 314 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 Nucleic Acid and Organ Damage in Coronavirus 2019 Patients: A Cohort Study. Clinical 410 Infectious Diseases Detection of SARS-CoV-2 RNA in serum is associated with increased 418 mortality risk in hospitalized COVID-19 patients STARD 2015: An Updated List of Essential 421 Items for Reporting Diagnostic Accuracy Studies. British medical journal 351 Pub. no. MAN0010408) Revision B.0 Evaluation of two RT-PCR techniques for SARS-CoV-2 RNA detection in serum for 427 microbiological diagnosis SARS-CoV-2 RNA Load Predicts Outcome • OFID • 1 SARS-CoV-2 Blood RNA Load Predicts 431 Outcome in Critically Ill COVID-19 Patients No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted Dynamics of Blood Viral Load Is 434 Strongly Associated with Clinical Outcomes in Coronavirus Disease Viral load dynamics and disease 439 severity in patients infected with SARS-CoV-2 in Zhejiang province, China Clinical Outcome in COVID-19 Patients SARS-CoV-2 RNA in plasma 447 samples of COVID-19 affected individuals: a cross-sectional proof-of-concept study 448 Clinical Infectious Diseases High Frequency 451 of SARS-CoV-2 RNAemia and Association with Severe Disease 452 SARS-CoV-2 viral load 457 (SARS-CoV-2) RNA in Blood of Patients with Coronavirus Disease Severe Acute Respiratory Syndrome Coronavirus 2 Normalized Viral Loads and 465 Subgenomic RNA Detection as Tools for Improving Clinical Decision Making and Work 466 Quantification of Viral Load: Clinical 468 Relevance for Human Immunodeficiency Virus, Hepatitis B Virus and Hepatitis C Virus 469 Infection Standardization of Hepatitis C Virus RNA Quantification Molecular-Based Methods for Quantifying HIV Viral Load Evaluation of CMV viral load using TaqMan CMV quantitative PCR and comparison with CMV antigenemia in heart and lung 476 transplant recipients Relationship Between Serum Severe Acute Respiratory Syndrome Coronavirus Nucleic Acid and Organ Damage in Coronavirus 2019 Patients: A Cohort Study In summary, we conclude that persistent SARS-CoV-2 viremia in blood samples may be 315 This research did not receive any specific grant from funding agencies in the public, commercial, 334 or not-for-profit sectors. 335The work of ER-V has been funded by a Rio-Hortega grant CM19/00149 from the Ministerio de 336Economía y Competitividad (Instituto de Salud Carlos III) and co-funded by The European 337Regional Development Fund (ERDF) "A way to make Europe" 338 339 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 24 is associated with increased disease severity and mortality. Nature communications 458 11:5493-. 459 Jacobs JL, Mellors JW. Detection of Severe Acute Respiratory Syndrome Coronavirus 2 460 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 8, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022