A series of laboratory analyses were then conducted (Appendix). The cat’s owner collected 26 swab samples according to instructions received by telephone; 16 samples contained varying amounts of the SARS-CoV-2 genome (Table). Overall, positive samples were detected March 11–24. The cat was examined by veterinarians at the time of blood sampling on day 22 after onset of first symptoms. Clinical examination of the cat was unremarkable at that time, and auscultation of the thorax revealed no abnormalities. Results of a complete blood count and a serum biochemistry panel were within reference ranges. Presence of serum IgG was first sought by Western blotting of mock-exposed and SARS-CoV-2–exposed Vero E6 cells lysates. In convalescent-phase serum, 5 protein bands that were simultaneously absent from mock-exposed Vero E6 cell lysates were identified (Appendix Figure). Furthermore, the convalescent-phase serum was positive by double-epitope sandwich ELISA and for 2 of the 3 antigens tested by double-epitope luciferase assay (Table; Appendix Table). Whereas serum samples from 30 control cats and 10 control humans were negative by virus neutralization assay, the convalescent-phase serum samples from the cat and her owner were positive; endpoints were 1:512 for the cat and 1:128 for the human.

The cat at first showed general signs, then gastrointestinal signs, and finally respiratory signs, similar to those observed in humans. Subsequently examined samples from the cat revealed viral RNA persisting for about 10 days. With the exception of a vomitus fluid sample collected on March 13, the amounts of viral RNA were relatively low. For this reason, and despite the simultaneous presence of a compatible clinical syndrome and a suggestive chronology of events, we cannot automatically rule out passive contamination of the cat’s samples by its owner.

To confirm the hypothesis of a productive infection of the cat, we conducted a series of serologic analyses by using 4 different testing approaches and targeting distinct viral protein targets. All procedures converged toward the same result: the convalescent-phase serum from the cat contained immunoglobulins against SARS-CoV-2, which were absent from the serum from control cats. These antibodies target several distinct viral proteins, and they caused a total neutralizing effect up to a much higher dilution than those from the owner’s serum. This household cat was therefore productively infected with the SARS-CoV-2 virus excreted by its owner, and the infection caused a nonfatal but nevertheless severe disease, mainly of the respiratory system (Videos 2–6).

Public health officials are still learning about SARS-CoV-2, but no current evidence indicates that pets play a role in spreading the virus. Therefore, taking measures against companion animals that may compromise their welfare is not justified.