Feline chronic gingivostomatitis (FCGS) is a pathology with a complicated therapeutic approach and with a prevalence between 0.7 and 12%. Although the etiology of the disease is diverse, feline calicivirus infection is known to be a predisposing factor. To date, the available treatment helps in controlling the disease, but cannot always provide a cure, which leads to a high percentage of refractory animals. Mesenchymal stem cells (MSCs) play a pivotal role in the homeostasis and reparation of different tissues and have the ability to modulate the immune system responses. This ability is, in part, due to the capacity of exosomes to play a part in intercellular cell communication. However, the precise role of MSC-derived exosomes and their alterations in immunocompromised pathologies remains unknown, especially in veterinary patients. The goal of this work was to analyze the proteomic profile of feline adipose tissue-derived MSCs (fAd-MSCs) from calicivirus-positive FCGS patients, and to detect possible modifications of the exosomal cargo, to gain better knowledge of the disease's etiopathogenesis. Using high-resolution mass spectrometry and functional enrichment analysis with Gene Ontology, exosomes isolated from the fAd-MSCs of five healthy cats and five calicivirus-positive FCGS patients, were pooled and compared. The results showed that the fAd-MSCs from cats suffering from FCGS not only had a higher exosome production, but also their exosomes showed significant alterations in their proteomic profile. Eight proteins were exclusively found in the exosomes from the FCGS group, and five proteins could only be found in the exosomes from the healthy cats. When comparing the exosomal cargo between the two groups, significant upregulation of 17 and downregulation of 13 proteins were detected in the FCGS group compared to the control group. These findings shed light on new perspectives on the roles of MSCs and their relation to this disease, which may help in identifying new therapeutic targets and selecting specific biomarkers.