Distinct antigens have been evaluated with diagnostic purpose for canine and human visceral leishmaniasis (VL), and variable sensitivity and specificity values have been obtained in the assays. In the present study, a Leishmania infantum hypothetical protein called LiHyG, which was identified in an immunoproteomics study in Leishmania infantum amastigote extracts by antibodies in VL dogs sera; was cloned, expressed, purified and evaluated as a recombinant protein (rLiHyG) for the diagnosis of canine and human disease. The recombinant amastigote-specific A2 protein (rA2) and a soluble L. infantum protein extract (SLA) were used as controls. For canine VL, the sensitivity values were of 100%, 57.29% and 48.57%, when rLiHyG, rA2 and SLA were used, respectively, while the specificity values were of 100%, 81.43% and 88.57%, respectively. In addition, AUC values were of 1.00, 0.72 and 0.65, when rLiHyG, rA2 and SLA were used, respectively, while accuracy was of 100%, 72.38% and 75.24%, respectively. For human VL, the sensitivity values were of 100%, 84.00% and 88.00%, when rLiHyG, rA2 and SLA were used, respectively, while the specificity values were of 100%, 58.75% and 73.75%, respectively. In addition, AUC values were of 1.00, 0.76 and 0.83, when rLiHyG, rA2 and SLA were used, respectively, while accuracy was of 100%, 64.8% and 66.6%, respectively. The prognostic role of rLiHyG in the human VL was also evaluated, by means of post-therapeutic serological follow-up with sera samples collected before and six months after treatment. Results showed that treated patients presented significant reductions in the anti-rLiHyG IgG, IgG1, and IgG2 antibody levels, with results being similar to those found in healthy subjects. Testing the rA2 protein and SLA as antigens, lower IgG, IgG1, and IgG2 levels were also found, although they were higher after treatment than those obtained for rLiHyG. In conclusion, results suggested that rLiHyG could be considered for future studies as a diagnostic and/or prognostic marker for canine and human VL.