388©W.ZHANG ET AL 3.00 IgM-D5 20 .Dashed ine which as determined able 2.M (AS)and blood.Data were shown as PCR Ct values. or ntially shed this th Date 0-05 Date 0-AS Dat 303 we successfully applied serology test a large population and showed which could greatly improved detectior 327 ent 300 31A y that the c rent strategy for the detection atient 20 of viral RNA in oral swabs used for 2019-nCoV diagno 3.6 sis is not perfect.The virus may be present in anal 238 negative. MERS 2 of infection -9 However.patients infected with 338 269 2019-nCoV may harbour the virus in the intestine at the early or late stage of disease.It is also worth to note no of the patients with nia blood had pos For molecular detection.we found 8 oral swab positive (50%)and 4 anal swabs (25%)in these 16 people on day 0.On day 5,we were only able to find thus posea threat toother people In. 4 oral s vabs positive (25%).In contras we ound e viral antibodies in near all patients,indicating serology hal swa 57. ing all swa should be considered for 2019-nCoV epidemiology.A me fron swab (/10.80%)on day shi from oral this trend appears to change on day 5.We found d Thi lied that more (6/8, 75%)anal swab positive than oral swab charge a patient purely based on oral swabs negative positive(4/8,50%).Another e reoccur who may still shed the virus by oral-fecal route ofthese 6 ere d cted r Above all,we anal swabs (Table 2).These data sug ested a shift from al tes more oral positive during early period(as indicated by In summary,we provide a cautionary warning that antibody titres)to more anal positive during later 2019-nCov may be transmitted through multipl period might happen routes Both molecular and serological tests are needed to definitively confrm a virus carrier Discussion Acknowledgements tion tools.Thisis the first molecular and serological We thank Dr.Danielle Anderson of Duke-NUS Medica study on this virus after the initial identification of 2019-NCov from 7 patients diagnosed with Virology. For molecular detection, we found 8 oral swabs positive (50%) and 4 anal swabs (25%) in these 16 people on day 0. On day 5, we were only able to find 4 oral swabs positive (25%). In contrast, we found 6 anal swabs positive (37.5%). When counting all swab positives together, we found most of the positives came from oral swab (8/10, 80%) on day 0. However, this trend appears to change on day 5. We found more (6/8, 75%) anal swab positive than oral swab positive (4/8, 50%). Another observation is the reoccur￾rence of virus in 6 patients who were detected negative on day 0. Of note, 4 of these 6 viral positives were from anal swabs (Table 2). These data suggested a shift from more oral positive during early period (as indicated by antibody titres) to more anal positive during later period might happen. Discussion Within 1 month of the 2019-nCoV disease outbreak, we rapidly developed molecular and serological detec￾tion tools. This is the first molecular and serological study on this virus after the initial identification of 2019-NCoV from 7 patients diagnosed with unidentified viral pneumonia [5]. We detected the virus in oral swabs, anal swabs and blood, thus infected patients can potentially shed this pathogen through respiratory, fecal–oral or body fluid routes. In addition, we successfully applied serology test a large population and showed which could greatly improved detection positive rate. We show that the current strategy for the detection of viral RNA in oral swabs used for 2019-nCoV diagno￾sis is not perfect. The virus may be present in anal swabs or blood of patients when oral swabs detection negative. In SARS-CoV and MERS-CoV infected patients, intestinal infection was observed at later stages of infection [7–9]. However, patients infected with 2019-nCoV may harbour the virus in the intestine at the early or late stage of disease. It is also worth to note none of the patients with viremia blood had posi￾tive swabs. These patients would likely be considered as 2019-nCoV negative through routine surveillance, and thus pose a threat to other people. In contrast, we found viral antibodies in near all patients, indicating serology should be considered for 2019-nCoV epidemiology. A possible shift from oral positive during early infection to anal swab positive during late infection can be observed. This observation implied that we cannot dis￾charge a patient purely based on oral swabs negative, who may still shed the virus by oral–fecal route. Above all, we strongly suggest using viral IgM and IgG serological test to confirm an infection, considering the unreliable results from oral swabs detection. In summary, we provide a cautionary warning that 2019-nCoV may be transmitted through multiple routes. Both molecular and serological tests are needed to definitively confirm a virus carrier. Acknowledgements We thank Dr. Danielle Anderson of Duke-NUS Medical School for critical review of this report. We thank National Virus Resource Center (NCRC) in Wuhan Institute of Virology. Figure 1. Serological detection of 2019-nCoV. Dashed line indicates cutoff, which was determined based on data from healthy controls. Table 2. Molecular detection of 2019-nCoV in swabs from two investigations. Samples were from oral swabs (OS), anal swabs (AS) and blood. Data were shown as qPCR Ct values. Date 0-OS Date 0-AS Date 5-OS Date 5-AS Patient 1 23.2 Patient 2 30.3 Patient 3 19.5 Patient 4 32.7 30.2 Patient 5 33.1 Patient 6 31.1 30.0 31.4 Patient 7 27.3 Patient 8 27.0 Patient 9 32.9 33.6 Patient 10 23.8 Patient 11 31.9 Patient 12 32.3 Patient 13 17.8 Patient 14 25.5 Patient 15 30.0 Patient 16 33.8 26.9 27.5 388 W. ZHANG ET AL