Lymphoproliferative disease in cats and dogs: a comparison of immunophenotyping by flow cytometry and clonality testing.

This thesis is a retrospective study on canine and feline lymphoproliferative diseases advanced diagnostic techniques. In particular, we focus our attention on immunophenotyping by Flow Cytometry (FC) and clonality assays, comparing the diagnostic sensitivity and specificity of the latter in the course of canine leukaemias and feline lymphoproliferative diseases. In Chapter 1, an overview of lymphoproliferative diseases was given: the importance of immunophenotyping in the classification of such diseases, and its prognostic value have been well established in the last decade of research. However, immunophenotype is still unable to distinguish in some cases between reactive and neoplastic disorders, making clonality assays a useful, complementary tool in the diagnostic process. PCR for antigen receptor rearrangement (PARR) is a molecular biology technique aimed to amplify receptor genes which are clonally rearranged in the course of a clonal neoplastic expansion. Due to the potential application of PARR to many kinds of source material and due to the significant advantages offered by FC as a secure, cost and time effective technique, we believe that FC and PARR could be a valid alternative to the more expensive and complicated biopsy with histology and immunohistochemistry, which to date is considered the gold standard in lymphoproliferative diseases diagnostics. In Chapter 2, we optimised the protocols for extraction of sufficient DNA from our samples: due to the retrospective nature of our study, we believed that having optimised protocols using the minimum volumes of source material possible with the best performances, could be useful for our purposes. We optimised protocols for whole EDTA blood (used in Chapter 4 for the canine clonality assays) and cytological slides (used in Chapter 5 for the feline clonality assay). The first experiments on blood were performed using two kits (GenElute Mammalian Genomic DNA Miniprep Kit by Sigma Aldrich and the QIAMP Mini and Blood kit from Qiagen): we selected the Sigma kit protocol using as source material sample volumes containing a white blood cell count (WBC) of 9x106/mL and optimising the elution step. Extraction from cytological slides was optimised combining three protocols and creating a new original one, with optimised cell lysis and elution steps. The protocols developed were used for further experiments in the succeeding chapters. In Chapter 3 we established feline positive controls as internal controls for the following experiments run in Chapter 5. We selected four feline patients, diagnosed with lymphoma by histology and immunophenotyped by IHC. After extraction of DNA from FFPE tissue samples, we performed PARR using the primer set intended to be used in the following experiments in order to select the samples which were positive to the single primers. Due to the immunophenotype and the positive clonal amplification of the TCRG gene, one sample was selected as a positive control for T cell clonality. One other patient was selected, according to the same criterion, as positive control for B cell clonality testing. However, DNA obtained from two feline lymphoma cell lines (MS-4 and FT-1, for B and T cell clonality respectively) were used in the present study along with our established positive controls. In Chapter 4 the value of PARR in comparison with FC in the course of canine leukaemias was investigated. Twenty-nine cases were selected: of those, nineteen were diagnosed with leukaemia, and ten showed a raised WBC attributable to infection or immune-stimulation. The first group comprised: eight dogs diagnosed with T-ALL (of which, one was showing evidence of LGL leukaemia), four AUL, three B-ALL, two biphenotypic leukaemias, one T-CLL and one AML. PARR was performed using two primers directed to the TCRG locus, and two directed to the IGH gene. Separation and visualisation of PCR products were performed by Agarose gel electrophoresis. Overall sensitivity and specificity of clonality testing using FC as gold standard were 73% and 100% respectively. Neoplasia was detected in approximately 63% of the neoplastic cases diagnosed by FC; all the non-neoplastic cases were confirmed as reactive by PARR. PARR confirmed phenotype in 50% of the cases of B and T cell neoplasia; moreover, two AUL was diagnosed as T cell neoplasia, and the double phenotypic leukaemias clonally rearranged just the TCRG locus. However neoplasia was not detected in high rates, and cell lineage did not match between the two techniques in many cases: for this reason, PARR should not be considered as a diagnostic tool by itself but has to be always integrated into a more comprehensive diagnostic process. In Chapter 5, clonality testing was performed on feline lymphoma/leukaemia and reactive lymphoproliferative disorders samples. Of the thirty-seven cases retrieved from the Dick White Referrals (DWR) laboratory archive, only twenty-one cases provided sufficient amounts of good quality DNA from the cytological slides available. Thirteen cases were diagnosed as having neoplasia (seven T cell neoplasia and six B cell neoplasia), in the peripheral blood (four cases), affecting the gastrointestinal tract (four cases), or peripheral lymphnodes (three cases) or other locations (two cases). Five cases were diagnosed as having a reactive process. In three cases a clear-cut distinction between neoplasia and reactive hyperplasia was not possible by the FC analysis only. Clonality testing was performed using a primer set comprising one primer directed to the TCRG locus and five primers directed to the IGH gene. Agarose, PAGE and capillary electrophoresis (CE) were performed and compared. If DNA was available for the reactive cases, detection of IGH gene rearrangement was performed using additional primers. Clonality testing on our feline case series showed an overall sensitivity and specificity of 63.6% and 100% respectively, and neoplasia was detected in 42,5% of the cases of neoplasia diagnosed by FC. PAGE and CE showed a good concordance, but CE has to be preferred especially in ambiguous cases where a clonal population is present within a polyclonal background. Our results showed the potential of PARR in refining the diagnosis of lymphoma achieved by FC, but highlights as well the weakness of the technique in terms of low sensitivity, possibly due to incomplete gene coverage of the primers used. Moreover, it confirmed the importance of using high resolution and highly sensitive techniques for visualisation of PCR products, such as automated capillary electrophoresis.