FLOW CYTOMETRY AND SUPPORTIVE TECHNIQUES FOR THE DIAGNOSIS AND CHARACTERIZATION OF FELINE LYMPHOPROLIFERATIVE DISEASES

Flow cytometry (FC) is an immunophenotyping technique routinely applied in human medicine to assess the origin of hematopoietic malignancies, even if this is not its only use in human species. In veterinary medicine, its use in clinical practice mainly concerns with the evaluation of hematopoietic malignancies. However, this diagnostic tool is more widely and commonly applied in canine oncology, whereas in feline species there are many limitations for an extensive and routine use of this technique (Wilkerson et al., 2012; Guzera et al., 2016). One of the most relevant limitations for feline species is the poor availability of feline species-specific and cross-reactive antibodies to the leukocyte clusters of differentiation (CDs). This doesn’t allow a sufficiently detailed investigation of neoplastic cells. Another important limitation for the species concerns with the localization of lymphomas: great part of lymphomas in cat are of intestinal origin hence intra-abdominal (Lowerens et al., 2005). This makes lesions harder to be sampled, especially when it is a homogeneous infiltrations of the bowel or without a significant involvement of regional lymph nodes. The sampling is even not encouraged by the frequent need of general anaesthesia because of the temperament of feline patients. The aim of this PhD project is to consider different aspects of diagnosis of lymphoma and leukaemia in feline species, using flow cytometry as main tool, supported by more widely used and strong techniques. Due to the paucity of data in literature on the topic we planned to start to evaluate analytic feseability and the evaluation of normality data then going on with the evaluation of some specific lymphoma subtypes and the potential clinic-pathological meaning of some candidate markers that could be inserted in FC panels for cats. To this aim, four studies will be illustrated. The first one is a retrospective study about pre-analytical factors possibly affecting the quality of samples submitted for flow cytometry and thus their likelihood of being processed. Between 2009 and 2016, samples of suspected lymphoma with primary lesions/lymph nodes/effusions available for flow cytometry were selected from FC service’s database of the University of Milan. Pre-analytical variables that were considered were related to the patient, the sampling procedure, the lesion and the clinician who performed the sampling. At their arrival in the lab, gross aspect and cellularity of the samples were assessed. Total nucleated cell count (TNCC) came up to be the variable that mostly affected the likelihood of a sample to be processed for FC and TNCC itself was influenced by caliber of the needles used for sampling procedure (21 G being the most performing). 22% of samples analysed were not conclusive; differences in cellularity between thoracic/abdominal and peripheral samples were not identified. No side effects following sampling were reported by vets, except in one case. The second study is a prospective collection of feline non-neoplastic lymph nodes, analysed by FC, cytology and histopathology. The aim of the study was to describe the lymphocyte subsets in feline non-neoplastic lymph nodes to create the basis for comparison in neoplastic samples. Sixteen lymph nodes from 11 patients were collected, cellular suspensions were obtained and cytological smears were done. A half of each lymph node was preserved in order to perform histology. The results observed for the FC and cytological analysis were very similar to those described in the dog, with a higher proportion of medium size lymphocyte. Histological examination revealed the hyperplastic nature of 5 samples which considered separately from the others, showed a significantly higher expression of CD8 (p=0.008). Further studies are needed to understand the potential differences between non-neoplastic and neoplastic lymph nodes in feline species. The third study is a retrospective collection of feline mediastinal masses, analysed by FC and cytology. The aim of the study was to compare the immunophenotype of lymphoma and non-lymphomatous lesions in order to assess if FC could reliably support the distinction between these two entities. 19 cases were finally collected: 13 lymphomas and 6 non-lymphomatous lesions. Among lymphomas, the most common immunophenotype detected was CD4+CD8+ double positive T-cell, while in non-lymphomatous lesions, lymphocyte population was composed by heterogenous T-lymphocyte subsets, except in one case in which CD4+CD8+ subset was dominant, reaching 78.8%. According to our results, FC is not enough to discriminate these two entities, since in cats, CD4+CD8+ double positive lymphomas are likely very common, thus the cut-off proposed for canine species (which was highly specific for thymomas) was not applicable to cats. Anyway, a larger caseload would be warranted, and histopathology should be available for every sample in order to reach a final diagnosis. The fourth study is a retrospective analysis of pan-leukocyte markers on feline WBC populations in healthy, reactive and neoplastic samples. The aim of the study was to depict the pattern of expression of CD18 and CD44 on WBC subclasses on peripheral blood (PB) of healthy cats, and to provide preliminary data on possible variations with different functional states (resting, reactive and neoplastic) similarly to what already done in the dog. Samples from 16 healthy cats and 21 cats with different pathological conditions were tested by FC for CD18 and CD44 expression. In healthy cats, both molecules were expressed at higher level on monocytes, medium level on PMNs and lower levels on lymphocytes. CD18-Median Fluorescence Index (MFI) discriminated well the three population, whereas CD44-MFI mostly overlapped between monocytes and PMNs. Reactive lymphoid cells had higher CD18 expression compared to resting lymphocytes, whereas no difference was detected in CD44-MFI between the two groups. Both molecules were variably expressed on the neoplastic cells from different individuals, but CD44-MFI tended to be higher than in resting and reactive lymphoid cells. Overall, the results of this PhD project lead to account FC as an “immature” technique for the study of hematopoietic malignancies in feline species, that would need further investigations, mainly concerning the development and application of new markers, as well as a greater cooperation from clinicians in order to set better clinicopathological and clinical assessment on a greater number of patients, being lymphoma a quite common neoplastic disease in feline species.