key: cord-0893841-hkjl2pr4 authors: Schoen, M.; Lindenau, C.; Boeckers, A.; Altrock, C.-M.; Krys, L.; Nosanova, A.; Lang, N.; Reny, A.; Kroschel, J.; Pensel, E.; Grab, C.; Mayer, B.; Messerer, D. A. C.; Fassnacht, U.; Delling, J. P.; Horneffer, A.; Zernickel, M.; Debatin, K.-M.; Muench, J.; Kirchhoff, F.; Wirth, T.; Boeckers, T. M. title: Longitudinal SARS-CoV-2 infection study at Ulm University date: 2021-05-06 journal: nan DOI: 10.1101/2021.05.04.21256382 sha: 19bd5d489ea19b65edced057cf1cb5c40c98883d doc_id: 893841 cord_uid: hkjl2pr4 The practical course in macroscopic (gross) anatomy is an essential component of medical studies. The dynamic situation with high SARS-CoV-2 infection rates prior to the winter semester (02.11.2020 until 01.03.2021) confronted university institutions with the difficult question of how or whether essential practical courses in medical schools can be conducted in presence. The gross anatomy course at Ulm University took place with a comprehensive hygiene concept and accompanied by a longitudinal study. This included in particular SARS-CoV-2 pathogen detection (swab with RT-PCR) at neuralgic time points, as well as antibody testing at the beginning and at the end of the semester for both students and teaching staff. The first SARS-CoV-2 RT-PCR test prior to the gross anatomy course revealed two asymptomatic SARS-CoV-2 positive individuals of 327 students. All institute and student staff of this course tested negative at semester start (n=75). Antibodies to SARS-CoV-2 were detected in 6.4% of the anatomy course students (22 out of 345). The second SARS-CoV-2 RT-PCR test after the Christmas break was negative in all participants, including teaching staff (n=429). At the end of the course in mid-February 2021, seroconversion after infection was detected in only two students of the anatomy course who participated in both tests (0.6%, n=325). Also other semester cohorts of the first three years of study in human medicine and dentistry were invited. No further active SARS-CoV-2 infections at the start of the semester and seven seroconversions after infection (n=335) were detected after the semester in these cohorts. The data illustrate the likely preventive effect from the interaction of hygiene concepts, regular information on the pandemic and testing. Thus, this study demonstrates ways in which face-to-face teaching can be implemented for selected courses at universities, even with high national incidence rates. second SARS-CoV-2 RT-PCR test after the Christmas break was negative in all participants, including teaching staff (n=429). At the end of the course in mid-February 2021, seroconversion after infection was detected in only two students of the anatomy course who participated in both tests (0.6 %, n=325). Also other semester cohorts of the first three years of study in human medicine and dentistry were invited. No further active SARS-CoV-2 infections at the start of the semester and seven seroconversions after infection (n=335) were detected after the semester in these cohorts. The data illustrate the likely preventive effect from the interaction of hygiene concepts, regular information on the pandemic and testing. Thus, this study demonstrates ways in which face-to-face teaching can be implemented for selected courses at universities, even with high national incidence rates. Infection with SARS-CoV-2 is associated with mild to severe acute respiratory symptoms. In rare cases, the infection is asymptomatic. Particularly in severe disease courses, multiorgan disease may occur (Mokhtari et al., 2020 , Wu & McGoonan, 2020 . Young, healthy individuals show usually milder symptom than elderly individuals (Davies et al., 2020 , Wu & McGoonan, 2020 . Susceptibility to infection, however, increases sharply in the age range between 18 and 25 years (Davies et al., 2020) . The outbreak of the pandemic in spring 2020 led to discontinued or reduced classroom teaching at medical schools worldwide as well as in Germany (Richter-Kuhlmann, 2020) . For the past winter semester, extensive hygiene concepts were developed at Ulm University, as well as at many other universities worldwide, in order to be able to hold essential practical courses as conventional classroom teaching. In the early fall, mainly younger people were infected (Robert Koch Institute, daily situation report, 20 Oct 2020, in German: https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Situationsberichte/Okt_2020 /2020-10-20-de.pdf?__blob=publicationFile). In view of the sharp rise in infection numbers in fall, which indicated the onset of a severe second infection wave, we developed a longitudinal testing strategy with direct pathogen detection in addition to a hygiene concept already developed in spring and summer for the gross anatomy course. In order to gain more information on the incidence of infection in students, this study was supplemented with antibody detections and questionnaires at the beginning and end of the semester. A similar approach has already been proposed by American colleagues (Fung et al., 2020) . In a modeling of infection events, one to two tests with direct pathogen detection in students before and during a semester were also recommended (Rennert et al., 2020) . This study also included further semester cohorts from the first three years of medical and dental school. The voluntary study was reviewed and approved by the ethics committee of Ulm University (application number 405/20). All students of the first three years of medical school (semesters 1, 3, 5, and 6), as well as dental students of the first and second semester were invited to participate after informed written consent. In Germany, medical school lasts 6 years and is made up of a preclinical (first 2 years), clinical-theoretical (third year), and clinical phase. The core cohort of the study was the gross anatomy course with students and institute staff. Cohort sizes, participation rates with corresponding average length of practical courses with face-toface teaching can be seen in Table 1 and in Figure 1 in the appendix. Testing consisted of direct pathogen detection (pharyngeal swab with reverse transcription polymerase chain reaction (RT-PCR) analysis, rapid antigen test as self-testing) and serum antibody tests accompanied by questionnaires. The basic study design is outlined in Figure 2 . The swab tests with RT-PCR analysis were offered at neuralgic time points: before the start of the gross anatomy course (early November) and after the Christmas break (early January). At both time points, many participants spent the time before in their home environment. To protect the relatives and contacts of the study participants at home, we also offered rapid testing before Christmas break and after the end of the semester. To assess infection prevalence before and during the semester, serum antibodies to SARS-CoV-2 at the beginning of the semester (mid-November to early December) and at the end of the semester (mid-February to early March; approximately three months after the first testing time point) were determined. A paper-based questionnaire accompanied these two antibody testings. Students in the first-year cohorts (HM1, D1, and D2; see Table 1 ) received all antibody tests, but only the offer of a swab RT-PCR test at the beginning of the semester (late November) and no rapid tests. These cohorts mostly had few courses with conventional classroom teaching at the university. The same applies to fifth-and sixth-semester students (HM5 and HM6; see Table 1 ), who also did not receive swab RT-PCR testing. Most of these third-year cohorts had already been tested for SARS-CoV-2 infection with a swab outside of this study due to clinical courses. The cohort of the gross anatomy course was offered direct pathogen detection more frequently (second swab with RT-PCR, two antigen rapid tests) because of the long duration of the face-to-face classes and special conditions. The dissection of a body donor, even with a reduced number of students, requires an undercutting of the recommended spatial distance (see hygiene concept of the course of macroscopic anatomy in the appendix, table 2). The data from the tests were then calculated as percentages and corrected according to sensitivity and specificity. This testing offer was made to all anatomy course students, student tutors, and staff. The other semester cohorts were invited to the antibody determinations and questionnaires at the beginning (mid-November to early December) and end of the semester (mid-February to early March). In some cases a swab test was offered at the beginning of the semester (HM1, D1, and D2). In the case of underage persons (n=3), a swab test was offered and only executed by a study physician, but the result was not used in the data analysis. Study physicians were present at each round of testing. Hygiene concept in the gross anatomy course The hygiene concept for the gross anatomy course is summarized as an example in the appendix ( Table 2) . Central elements of the hygiene concept were contact tracing, reduction of group size per body donor (six students, one tutor; note: before pandemic ten to eleven students), prohibition of contact between groups, spatial distance (> 1.5 m where possible), mandatory mask (surgical or FFP), high air exchange rate, disinfection of teaching materials, and regular informational offerings during course time and via e-mails or as online teaching events. There was a maximum of 116 persons in total with teaching staff in dissection rooms covering 450 m 2 . Within the table groups, it was attempted to maintain as much distance as possible, but it was usually less than 1.5 m. Pharyngeal swab with SARS-CoV-2 RT-PCR analysis A naso-and oropharyngeal swab (same swab) was performed. At the request of the participants, in some cases only an oropharyngeal swab was performed. Exclusion criteria were symptoms suspicious for COVID-19, as these prohibited entry into the university (see first chapter in the results section, second paragraph). Medical students in the clinical phase were appropriately trained in the performance of this testing and instructed in essential hygiene measures. A dry swab from the company nerbe plus (Winsen, Germany) was used. The analysis was performed using the Cobas 6800 test system from Roche Diagnostics (Mannheim, Germany). According to the manufacturer, sensitivity and specificity are both at 100 % (correspondence with Roche Diagnostics, April 2021). Ten swab sticks were pooled for each test according to the multiple swab method (Schmidt, 2020) . The swabs were retained until a negative pool result was obtained; if the result was positive, the pool was dissolved. Roche Cobas gives positive or negative results for two different target regions: ORF1 a/b (nonstructural region unique to SARS-CoV-2) and the E gene (envelope structural protein used to detect pan-sarbecoviruses, but also used to detect SARS-CoV-2 virus). 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 May 6, 2021. ; https://doi.org/10. 1101 Antigen Point-of-care (POC) test The SARS-CoV-2 Rapid Ag from Roche Diagnostics (Mannheim, Germany) test was used. The test systems were separated and packed in sterile bags and offered for takeaway. Sensitivity and specificity is 99,03 and 98,65 %, respectively (correspondence with Roche Diagnostics, April 2021). For name-related data protection and hygiene reasons, this test was offered as a self-test. On a defined day, the participants were asked to perform the test on themselves. A self-made, detailed demonstration video and written instructions were provided. If the test result was unclear or positive, students could contact a study physician. This physician performed another antigen test at their home on the same day. A second swab with RT-PCR analysis was also performed. Blood was collected using a butterfly collection system and a serum monovette. Medical students (clinical phase) were instructed in the technique and hygiene measures. Elecsys Anti-SARS-CoV-2 S from Roche Diagnostics (Mannheim, Germany) was used as the test system. Serum was collected by centrifugation and immunological quantitative in vitro detection of antibodies (according to manufacturer's information on request mainly IgG, but also IgM and IgA) against the spike (S) protein receptor binding domain (RBD) of SARS-CoV-2 was performed. Further specifications of the manufacturer: quantitative test with linear measurement range of 0.4 -250 U/ml, traceable to Roche internal quantitative standard. The test has been validated on a large collective with representative patient groups and, according to the manufacturer, achieves a sensitivity of 98,8 % and specificity of 99.98% (correspondence with Roche Diagnostics, April 2021). An online calculation tool was used to calculate the corrected prevalence, taking into account sensitivity, specificity, and number of individuals analyzed: https://epitools.ausvet.com.au/trueprevalence (accessed date: 20.04.2021). The confidence interval (Blaker) was set at 0.95 for the corrected prevalence. Questionnaire A largely identical paper-based questionnaire (format from Evasys v8.1, evasys GmbH, Lüneburg, Germany) was offered at the beginning and end of each semester at the same times as the tests took place. Among other items, gender and vaccination status against SARS-CoV-2 was asked. In designing the questionnaire, we followed study protocols and questionnaires from the Robert Koch Institute seroepidemiological studies (Santos-Hövener et al., 2020) and the study protocol of a locally conducted parent-child infection study (Tönshoff et al., 2021) . Quantitative analysis followed using nominal and ordinal scales. Data analysis is currently ongoing. To comply with data protection regulation, all data were pseudonymized by an external institute (Center for Clinical Studies, Ulm University Hospital) and were only accessible to the study team in this form. Individual results were communicated via this office. Information was provided to raise students' awareness of the pandemic and infection prevention and control measures in the context of their medical courses. At the beginning of each day in the gross anatomy course, information on hygiene measures was communicated 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint according to the current situation. The participants were regularly informed about the interim results of the study, scientific-medical background of the pandemic and the tests performed. For the gross anatomy course students and tutors, for example, the anatomical conditions of the pharyngeal swab examination with swab sticks were demonstrated on a midsagittal section of a body donor's head at the tables and could be tried out by the students themselves. The different detection methods of SARS-CoV-2 antibodies, RT-PCR analysis, and the antigen test were shown on graphs in addition to a self-produced video on how to perform the rapid test. Due to the situation at Christmas and to protect the home environment of the students and staff, three face-to-face course days were canceled and held as an online event. That coincided with a federal lockdown. The hygiene concept was continuously adapted to the current requirements of state and university regulations and communicated to the students. All participants were offered a compensation option if they missed a course as a result of isolation due to a positive pathogen detection. We also provided support to participants if they tested positive outside of the study. For example, we swabbed after the end of the official isolation period (authors' note: e.g., ten days in case of an asymptomatic course), as a negative test result after given time intervals was a prerequisite by university for resuming courses with face-to-face instruction. At any point, participants could join or withdraw from the study. In particular, swab and rapid antigen tests were also offered for non-study participants after informed written consent for the test. The results of all participants are reported in Table 3 and Figure 3 (swab RT-PCR), and Table 4 and Figure 4 in the appendix (serum antibodies). No participant had a medical condition that would have precluded participation in the study. No complications due to testing occurred. A total of 853 students of all invited semester cohorts and 14 staff members of the anatomical institute participated in the serum antibody detections of the study. Participation rates of antibody testing are shown in Table 1 and Figure 1 in the appendix. They represented nearly 80% of the total number invited participants to the study (n=1103; see Table 1 ). The majority (63.8%) of the participants were female (those invited to the study 61.7%). The mean age was 21.5 years (3.4 years standard deviation) excluding personnel from the anatomical institute. The mean age of all invited students was 21.8 years, standard deviation 3.6 years. The 14 participating institute staff members (n=16) were predominantly university lecturers in the gross anatomy course. The largest invited cohort was the gross anatomy course, which included students (course participants, n=385), student tutors (n=66), and staff (n=16). In this cohort, most participated in both tests (gray areas of these three sub-cohorts in Figure 1 in appendix, see also Table 1 ). Only 6% of the anatomy course students did not participate in any antibody testing. Cohorts in other semesters were smaller and participation rates were lower. The amount of conventional classroom teaching of practical courses can be seen in Table 1 . Of note, the cohort of the gross anatomy course (students, tutors, staff) had to fall below 1.5 m distance within table groups due to the nature of this course. Therefore, this cohort was offered more tests with direct pathogen detection. The fifth semester study cohort 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint provided most of the student teaching staff (52 tutors in total) in the gross anatomy course. These are part of the anatomy course cohort for the purposes of this study. Three tutors from semester 6 were included in the anatomy course cohort. The management of the gross anatomy course was notified of 17 reports of COVID-19 symptoms in course students and tutors during the semester. 15 were quarantined. Nine SARS-CoV-2 infections were documented (three before the semester start), but none could be associated with university teaching events. No contact situation in which a person had to be quarantined as a result of attending the course was reported. There was one SARS-CoV-2 infection in a student tutor in the training week for tutors before the course began. All other tutors as well as the lecturers with contact were classified as category II contacts as defined by the Robert Koch Institute (Robert Koch Institute, in German: https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Kontaktperson/Management .html). An additionally offered swab with RT-PCR was generally negative. Prior to the swab RT-PCR testing for the anatomy course cohort, six individuals at the start of the semester and five individuals after the Christmas break were unable to participate because they either had a SARS-CoV-2 infection (n=3), symptoms suspicious for COVID-19, or were quarantined. In the swab test with RT-PCR for the gross anatomy course, 84.9% (n=327) of the course students, 90.9% (n=60) of the tutors, and 93.8% (n=15) of the staff participated. Two swabs of anatomy course students (0.6%) were positive in both target regions, none of the institute and tutor staff. Both positively tested individuals contacted the study team voluntarily and had an asymptomatic infection according to their own statements. We refrain from giving further details on age and sex for reasons of data protection. After ten days, they were offered a swab test with RT-PCR, since a negative test result at this point allowed the resumption of practical courses by the university administration. One individual was negative. The other individual tested positive in only one of the two target regions (E gene) with a now very high Ct (cycle threshold) value (> 34) and was therefore released from quarantine based on the publication by Wölfel et al. (Wölfel et al., 2020) . Among first-year human and dental students, 164 swabs were negative. All results are provided in Figure 3 and Table 3 . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We recorded high participation rates for the first antibody test of the gross anatomy course (see Table 1 and Figure 1 in the appendix). Twenty-two of the 345 anatomy course students were positive (6.4%), one person among the tutors (1.6%), and none among institute staff. Participation rates among the other cohorts were lower. Among the 155 tests of human medicine students of the first semester, nine tests were positive (5.8%). Among fifth semester students, ten of 111 tests (9%) were positive. The other cohorts were small and participation rates were not as high as in the larger cohorts. The results are summarized in Table 4 . total 46 out of 720 6.4 (6.5) 119 out of 818 (53) 8.1 (8.2) 9 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint After the Christmas break, a second swab test was offered before resuming the gross anatomy course. With a participation rate of over 90% among students, tutors, and staff, all 429 samples were negative (see Figure 3 and Table 3) . All results are shown in Table 4 . In the gross anatomy course, 35 of the 342 students were detected with antibodies, of which ten vaccinated cases had positive antibody status. 21 of the 22 cases positive at the semester start examination participated again and still had a positive status. Two positive cases with no indication of vaccination were not present at the semester start examination. Thus, only two individuals changed from a negative to a positive finding without having been vaccinated against SARS-CoV-2. Among tutors, five of 66 had a positive status. Of these, four were vaccinated. Because the person who tested positive on the first test participated again and was still positive, we did not see any seroconversions with 100% participation. Among the university lecturers, all 14 were seronegative. Results among the other cohorts were as follows: Of 212 first-semester human medicine participants, 38 were positive, including 20 after vaccination. Eight of the previously nine positive individuals participated again and were positive again, one of whom was vaccinated. Five individuals had seroconversion without vaccination, and five seropositive individuals participated only in the second round of testing. Among fifth-semester students, 35 of 132 were positive. Ten of these were so at the semester start test, with one person vaccinated. Thirteen vaccinated individuals were negative at the first test, five had not participated in that first test, and all 18 were positive at the second. Only one person was found to be seroconverted presumably after infection. Six other participants had participated only at the second time point and were positive at that time without vaccination. In the smaller cohorts (dentistry semesters 1 and 2, human medicine semester 6), there were no detected seroconversions despite increased participation rates. The sex ratio and age distribution of all SARS-CoV-2 antibody-positive individuals (n=66) at semester start or end without vaccination was similar to the overall student study cohort. 42 of the 66 subjects were female, and the mean age was 21.2 years (standard deviation 2.5 years). All results are shown in Table 4 and a summary over all cohorts in Figure 4 is provided in the appendix. It should be noted that some were only present at the start (T1) or end (T2) of the semester. Especially at the antibody test at semester end, significantly more people participated. Therefore, the percentages after T2, even after removal of those with vaccination, are only comparable to a limited extent with the percentages at the start of the semester (see Table 4 ). For a longitudinal comparison, only the individuals who participated at semester start and end (T1+T2) are considered in the next paragraph. Antibody status of the student cohorts between the start and end of the semester Overall, we detected only nine seroconversions (+1.36%) among the students who participated in both antibody tests (n=660) that could not be explained by vaccination. Figure 5 (left pie chart) shows the antibody results of the student participants in the study who appeared at both semester start and end test rounds. 6.7% were already positive at baseline; none had a loss of positive status at semester end. Another nearly 6% became positive after 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 May 6, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 vaccination against SARS-CoV-2. Five vaccinated individuals had a negative antibody result, but of these, either a complete indication of the vaccination date was not available or the first dose was vaccinated only a few days before the test day. As far as indicated in the questionnaire, most of those vaccinated were immunized with the Comirnaty vaccine from Biontech/Pfizer. It is interesting to compare the two larger cohorts of first semester medicine students and the anatomy course students (Figure 5 ; middle and right pie chart, respectively). Note that for the anatomy course nearly 85% of invitees participated in both testing time points, compared to only 40% for semester 1. Although semester 1 had fewer practical courses at university (see Table 1 ), and the gross anatomy course is accompanied by a closer contact between students (within a table group < 1.5 m, see also Table 2 in the appendix), more seroconversions occurred in semester 1. The results refer to those who participated in the tests at the beginning and end of the semester. Institute staff members are not included here. Student tutors are included in the diagram on the left. HM = human medicine. A total of four students reported abnormal test results after both rounds of testing (before Christmas break, end of semester; in total approx. 900 tests). One person had indicated that the swab was very bloody, two had a questionable positive result, and one had a positive result. For the last three individuals mentioned, a study physician performed a second rapid antigen test on the same day. The questionably positive tests were now negative, and the swabs with RT-PCR were also negative in each case. The person with a positive antigen test was also positive in the rapid test performed again. Pharyngeal swabs with RT-PCR from the same day and two days later were both negative. The study physicians considered this case as false positive. The study presented here started at the beginning of November and ended beginning of March. Over 60% of students in the anatomy course cohort reported that they felt significantly or very significantly more protected from infection with SARS-CoV-2 as a result of being offered the tests according to the questionnaire at semester end. Due to the consistently high participation rate of the anatomy course cohort (including teaching staff) with more than 400 participants in each test part (mostly more than 90%), we assume a representative sample of this sub-cohort. A self-selection bias cannot be excluded if, for example, a person deliberately did not participate after previous infection. However, the infection dynamics of the other semesters between the beginning and end of the semester did not differ fundamentally from that of the anatomy course, although fewer students participated in these groups, and their presence teaching times were also mostly lower. Students in the gross anatomy course actually had fewer seroconversions as result of an infection compared to participants of the other semester cohorts. It was particularly pleasing to note that there was no infection detected by SARS-CoV-2 antibody seroconversion during the semester in the tutors of the gross anatomy course. Most of them belonged to semester 5. In addition to their tutoring activities, they attended the face-to-face practical courses according to their semester affiliation. Among the larger cohorts, only in semester 5 seroprevalence was strikingly higher at the start of the semester (November). We were informed at the beginning of the semester that a mandatory RT-PCR swab test was performed by the university hospital prior to an examination course. Approximately ten out of 100 swabs were positive for SARS-CoV-2. This examination course was then cancelled. The high rate of previously detected SARS-CoV-2 infections most likely explains the higher seroprevalence detected later. This semester was tested for antibodies in early December as part of this study, so this testing most likely covered the post-infectious seroconversion. Many students with little face-to-face teaching at university reported that they would not be living in Ulm, which could explain the lower participation rates at the semester start examinations among these cohorts. At the end of the semester, we were able to pair the timing of blood collection with exams in presence. The steady increase in study participants was encouraging, with a maximum in the last round of testing for antibody determination, demonstrating the high positive acceptance among students. The study team was repeatedly told that the study was very positively received. The students showed an extremely high level of interest in protecting themselves and their fellow students, taking active steps against the pandemic such as following infection control measures and participating in the study, and thus ultimately helping to ensure that practical courses can be completed with the necessary presence. There was always the possibility to get in contact with the study management to discuss the individual or overall results. This offer was and still is frequently taken. A limitation of the study is the lack of comparison with a cohort of students who had only online classes during the winter semester. Few studies have investigated the incidence of SARS-CoV-2 infections in students. In most cases, only seroprevalence at a specific time point was determined (Tilley et al., 2020 , Tuells et al., 2021 . Another study examined antibodies to SARS-CoV-2 in a cohort of students at the beginning of the fall term in the United States. The students had high infection status, with over 30% seropositive findings. The study does not report the antibody status of students at the end of the semester (Arnold et al., 2021) . In the USA, there was a large increase in local 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint COVID-19 incidence in the vicinity of universities that may be associated with the onset of fall term (Leidner et al., 2020) . To our knowledge, a longitudinal study with direct SARS-CoV-2 pathogen and antibody detection during the second wave of infection of the pandemic is unique. The winter term and study largely coincided with the second severe wave of the pandemic in Germany. At the start of the semester (testing times from mid-November to early December), 6.4% (46 of 720) of all cohorts were seropositive. This value is difficult to rank in comparison to cohorts of the same age or in an overall population comparison. A survey of students in Los Angeles in May 2020 came up with a value of 4% (Tilley et al., 2020) , and at a Spanish university in July 2020, seroprevalence was less than 3% (Tuells et al., 2021) . However, incidences are not and have not been comparable to other countries. Even within Germany and in each population group, there are likely to be substantial differences. An interim report of the SERODUS I study found a seroprevalence of 3.1% among 18-30 year-old residents of Düsseldorf, Germany. The time of measurement was largely identical to the semester start surveys of this study (https://www.uniklinik-duesseldorf.de/fileadmin/Fuer-Patienten-und-Besucher/Kliniken-Zentren-Institute/Institute/Institut_fuer_Medizinische_Soziologie/Forschung/SeroDus/Feld-_und_Ergebnisbericht_SERODUS-I_SERODUS-II_03-02-2021_v01.pdf, in German). An at least regionally interesting comparison is a random sample survey of the seroprevalence of the Munich population by the Tropical Institute of the LMU University Hospital Munich in the KoCo19 study. Before Christmas, 3% were seropositive (http://www.klinikum.unimuenchen.de/Abteilung-fuer-Infektions-und-Tropenmedizin/de/COVID-19/KoCo19/index.html, in German). A meta-analysis estimated that between 0.79-3.67% (95% confidence interval) of the population in Germany had SARS-CoV-2 antibodies by August (Rostami et al., 2021) . The SeroTracker program, available online, estimates seroprevalence in nations and regions through a systematic review process. For the testing period (winter term), local prevalence of serum antibodies is estimated to be between 1.4% and 4.4% (https://serotracker.com/en/Explore). In the synopsis of these data, it can be assumed that the values we determined at the start of the semester were above average compared to the German population. This is presumably due to the more frequent social contacts, the housing situation (often shared flats), and the often mild expression of COVID-19 symptomsor even asymptomatic infectionsin this age group (Tilley et al., 2020) . Furthermore, not all individuals develop antibodies after SARS-CoV-2 infection and antibody levels correlate inversely with symptom expression (Weis et al., 2021) . In addition, antibody levels may decrease over time (Seow et al., 2020) . However, we detected a quantitative decrease in only a few participants within the 3-month measurement period (data not shown, analysis ongoing). Across all cohorts including Institute staff who participated in both antibody tests (n=671), approximately three months after the start of the semester, only nine participants changed from seronegative status to positive at the end of the semester, which cannot be explained by vaccination against SARS-CoV-2. In the gross anatomy course, only two such seroconversions were added to the 21 SARS-CoV-2 seropositive individuals among course students at the start of the semester (an increase of 0.6% in total, and 9,5% when compared to the 6,5% positive individuals at the semester start of those students who participated in both antibody tests at semester start and end). Thus, we had a strong opposite trend to the infection incidence in Germany. In the same period, the number of all infected persons detected in Germany increased by a approx. 200% (mid-November to end of February, data from the Robert Koch Institute, Germany, and John Hopkins University, USA). Of note, these are not seropositive individuals but persons with a direct detection of the pathogen. However, we assume a similar dynamic between acute infection and the and the subsequent formation of antibodies. What this opposing trend (9,5% among students of the 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint anatomy course to 200% in the German population; note: 20,5 % in all students in this study) is due to can only be assumed, especially since clusters of infections among students at the beginning of a semester have been described (Wilson et al., 2020) . It is possible that behavior during the semester break differed from that during the semester in terms of disciplined adherence to infection prevention and control measures. The gross anatomy course in particular is considered to be very learning intensive. Our impression was that the students felt that it depended on their behavior, individually and on the entire cohort, whether the semester could be carried out as planned. Moreover, we suspect a preventive effect of the tests and the hygiene concepts. The announcement of the tests and the comprehensive information provided about the study and the pandemic likely resulted in a change in behavior towards the pandemic situation. Considerations for a testing concept at universities during COVID-19 pandemic Since costs for testing are likely to play an important role for all university institutions, we recommend selective testing with a swab followed by RT-PCR analysis, especially at the beginning of the semester. The high sensitivity led to the detection of two infected individuals in our study population. It can be derived from our study data that students are more likely to be virus carriers prior to the start of the semester due to returning to the study site from their home environment and lower awareness of the issue prior to the start of the semester. Resource-sparing multiple swab method was sufficient to detect asymptomatically infected individuals in this study. Even then results were available no later than one day after swabbing. Rapid antigen tests were used as self-tests. They are less expensive than RT-PCR testing (factor of 4.5 in this study). The rapid antigen test used shows high quality in terms of sensitivity and specificity compared to other lateral flow tests (Kohmer et al., 2021) . According to the study participants, the handling in combination with written and video instructions was sufficient. However, control of use, timing of performance, and feedback of positive results can only be ensured by reliable information from study participants. It may also be that correct performance of the test is more common with knowledge of medical topics. Antibody tests provide a relatively cheap insight into the longer-term infection status of populations. However, we recommend their use primarily for study purposes, as the data are very valuable from a seroepidemiological perspective. It is also possible that feedback on infection incidence by antibody status to students will also generate increased awareness of the pandemic. Further infection epidemiologic studies at other study sites with other collectives would be very important, especially in light of the upcoming more infectious virus variants. Providing regular information on infection prevention and control, scientific and medical background of the pandemic, and testing might also decrease infection due to increased awareness of hygiene recommendations. We offer to all colleagues in the university environment to support them in the preparation of a study protocol and hygiene concepts, information offers (e.g., video instructions for swab tests, information material on tests), and in the performance of tests on the university campus in compliance with infection prevention and control measures. Likewise, it was very gratifying and moving for us as a study team to see how many students in the clinical phase were willing to help with the testing rounds. A frequent feedback on their motivation was that they wanted to contribute in dealing with the pandemic and that they could support their preclinical fellow students. Overall, we see in this study a very encouraging indication that even large face-to-face events with more than 100 people and practical courses that require distances between students less than 1.5 m are possible under certain conditions without an increased infection rate. The basic prerequisite is a good hygiene concept. Regular provision of information and selective testing 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint presumably lead to preventively effective changes in behavior and might make a decisive contribution to controlling the infection situation. In view of the severity of the pandemic, it cannot be concluded on the basis of the presented data that universities can be opened without restriction. It is necessary to determine which teaching programs are essential to be performed as face-to-face classes. These could then even have an infection-preventing effect with a combination of hygiene concepts, selective testing, and information offers for students. Furthermore, it should be considered whether these measures are sufficient with more infectious SARS-CoV-2 variants, which were not yet strongly prevalent in Germany during the period of the study. Percentages refer to all those invited for testing (see Table 1 ). AC = anatomy course, D = dentistry, HM = human medicine, T1 = antibody test semester start, T2 = antibody test semester end. 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint Overall, eight of the seronegative individuals at the end of semester reported prior vaccination. However, either a complete report of vaccination was not available (dose, timing) or the initial dose was vaccinated a few days prior to blood collection. T1 = antibody test semester start, T2 = antibody test semester end, Ø no participation. 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 May 6, 2021. ; https://doi.org/10.1101/2021.05.04.21256382 doi: medRxiv preprint SARS-CoV-2 Seroprevalence in a University Community: A Longitudinal Study of the Impact of Student Return to Campus on Infection Risk Among Community Members Age-dependent effects in the transmission and control of COVID-19 epidemics SARS-CoV-2 Viral and Serological Testing When College Campuses Reopen: Some Practical Considerations The Comparative Clinical Performance of Four SARS-CoV-2 Rapid Antigen Tests and Their Correlation to Infectivity In Vitro Opening of Large Institutions of Higher Education and County-Level COVID-19 Incidence -United States COVID-19 and multiorgan failure: A narrative review on potential mechanisms Modelling the impact of presemester testing on COVID-19 outbreaks in university campuses Digitales Wintersemester 2020/21: Studieren im Homeoffice (German) SARS-CoV-2 seroprevalence worldwide: a systematic review and metaanalysis Seroepidemiologische Studie zur Verbreitung von SARS-CoV-2 in der Bevölkerung an besonders betroffenen Orten in Deutschland -Studienprotokoll von CORONA-MONITORING lokal (German) Novel multiple swab method enables high efficiency in SARS-CoV -2 screenings without loss of sensitivity for screening of a complete population Longitudinal observation and decline of neutralizing antibody responses in the three months following SARS-CoV-2 infection in humans Multi-Organ Involvement in COVID-19: Beyond Pulmonary Manifestations A Cross-Sectional Study Examining the Seroprevalence of Severe Acute Respiratory Syndrome Coronavirus 2 Antibodies in a University Student Population Prevalence of SARS-CoV-2 Infection in Children and Their Parents in Southwest Germany Seroprevalence Study and Cross-Sectional Survey on COVID-19 for a Plan to Reopen the University of Alicante (Spain) Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Antibody response using six different serological assays in a completely PCR-tested community after a coronavirus disease 2019 outbreak-the CoNAN study Multiple COVID-19 Clusters on a University Campus -North Carolina Virological assessment of hospitalized patients with COVID-2019 The outbreak of COVID-19: An overview Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China A Novel Coronavirus from Patients with Pneumonia in China We thank the Robert Koch Institute for providing study protocols and questionnaire documents The Institute for Anatomy and Cell Biology would like to sincerely thank the following medical students for their voluntary and extraordinary dedication and commitment to sample collection in this study: Josef Anetzberger, Helene Banßhaf, Jonathan Behr, Jacqueline Berg, Amelie Bogenschütz, Carla Brendler, Noemi Damwerth, Osman Demir, Maryse De Molière, Simon Ehricke, Lotta Elonen, Moritz Embacher, Ali Fattom, Oliver Fetzer, Julia Haug, Marlene Heimbeck, Sophie Jauch, Dominik Karl, Bianka Kernl, Tibor Kelety, Tassja Kleiter, Carla MS and TMB designed and led the study. CL and MS organized and conducted the testings. AB, CMA, LK, AN, ME, BM, and MS analyzed the data. NL and AR helped to plan the study and performed name-related care for the data and communicated the individual results to study participants. JK directed the laboratory analyses of the antibody tests, EP directed the RT-PCR analysis of the pharyngeal swabs. CG, DM, ME, and AH assisted in organizing the study. UF and AB prepared the hygiene concept for the gross anatomy course. JPD assisted with sample collection. MZ and KMD assisted in the preparation of the study protocol. JM and FK advised on virological issues. TW and TMB facilitated funding for the study. The study was funded by the Faculty of Medicine of Ulm University. MS, FK, JM, and TMB received funding on COVID-19 (Sonderfördermaßnahme COVID-19) of the state of Baden-Württemberg, Germany. The authors have declared no competing interest.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. recurring information on hygiene rules, COVID-19 symptoms, behavior in case of infection, contact with infected persons, for risk groups, interim results of the study and on the pandemic 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.