Background COVID-19 pandemic continues to be a priority in public health worldwide, and factors inherent to SARS-CoV-2 pathogenesis and genomic characteristics are under study. Investigations that evaluate possible risk factors for infection, clinical manifestations, and viral shedding in different specimens also need to clarify possible associations with COVID-19 prognosis and disease outcomes. Study design In this study, we evaluated SARS-CoV-2 positivity and estimated viral loads by real-time RT-PCR in stool, sera, and urine samples from 35 patients, with a positive SARS-CoV-2 RNA molecular test in respiratory sample, attended at a University COVID-19 referral hospital in Goiania, Goias, Brazil. Whole-genome sequencing was also performed in samples with higher viral load. Results The positivity index was 51.43%, 14.28%, and 5.71% in stool, sera, and urine specimens, respectively. The median viral load was 8.01 × 10 6 GC/g, 2.03 × 10 6 GC/mL, and 1.36 × 10 5 GC/mL in stool, sera, and urine, respectivelly. Of all patients, 88.57% had previous comorbidities, and 48.39% of them had detectable SARS-CoV-2 RNA in at least one type of clinical specimen evaluated by this study (stool, sera or urine). A higher viral load was observed in patients with more than two previous comorbidities and that were classified as severe or critical conditions. Samples with the highest viral loads were sequenced and characterized as B.1.1.33 variant. Conclusion We conclude that SARS-CoV-2 RNA is present in more than one type of clinical specimen during the infection, and that the most critical patients had detectable viral RNA in more than one clinical specimen at the same time point.

COVID-19 is characterized by diverse clinical outcomes that include asymptomatic to mild manifestations or severe disease and death. Infection by SARS-CoV-2 activates inflammatory and metabolic responses that drive protection or pathology.

The current SARS-CoV-2 pandemic has imposed new challenges and demands for health systems, especially in the development of new vaccine strategies. Vaccines for many pathogens were developed based on the display of foreign epitopes in the variable regions of the human adenovirus (HAdV) major capsid proteins (hexon, penton and fiber). The humoral immune response against the HAdV major capsid proteins was demonstrated to play a role in the development of an immune response against the epitopes in display. Through the immunoinformatic profiling of the major capsid proteins of HAdVs from different species, we developed a modular concept that can be used in the development of vaccines based on HAdV vectors. Our data suggests that different immunomodulatory potentials can be observed in the conserved regions, present in the hexon and penton proteins, from different species. Using this modular approach, we developed a HAdV-5 based vaccine strategy for SARS-CoV-2, constructed through the display of SARS-CoV-2 epitopes indicated by our prediction analysis as immunologically relevant. The sequences of the HAdV vector major capsid proteins were also edited to enhance the IFN-gamma induction and antigen presenting cells activation. This is the first study proposing a modular HAdV platform developed to aid the design of new vaccines by inducing an immune response more suited for the epitopes in display.