Diagnostic Uncertainty and the Epidemiology of Feline Foamy Virus in Pumas (Puma concolor)

Abstract

Feline foamy virus (FFV) is a contact-dependent retrovirus forming chronic, largely apathogenic, infections in domestic and wild felid populations worldwide. Given there is no current ‘gold standard’ diagnostic test for FFV, efforts to elucidate the ecology and epidemiology of the virus may be complicated by unknown sensitivity and specificity of diagnostic tests. Using Bayesian Latent Class Analysis, we estimated the sensitivity and specificity of the only two FFV diagnostic tests available—ELISA and qPCR—as well as the prevalence of FFV in a large cohort of pumas from Colorado. We evaluated the diagnostic agreement of ELISA and qPCR, and whether differences in their diagnostic accuracy impacted risk factor analyses for FFV infection. Our results suggest ELISA and qPCR did not have strong diagnostic agreement, despite FFV causing a persistent infection. While both tests had similar sensitivity, ELISA had higher specificity. ELISA, but not qPCR, identified age to be a significant risk factor, whereas neither qPCR nor ELISA identified sex to be a risk factor. This suggests FFV transmission in pumas may primarily be via non-antagonistic, social interactions between adult conspecifics. Our study highlights that combined use of qPCR and ELISA for FFV may enhance estimates of the true prevalence of FFV and epidemiological inferences.