Canine lymphosarcoma: Overcoming diagnostic obstacles and introduction to the latest diagnostic techniques

Lymphosarcoma, or malignant lymphoma, is a neoplasm that is commonly diagnosed in dogs, accounting for up to 24% of all canine neoplasms (1). In general, initial diagnosis of lymphosarcoma is made through cytologic evaluation of a simple fine needle aspirate of an affected organ (most notably, lymph nodes), or by histologic evaluation of paraffin-embedded sections of a formalin-fixed biopsy. However, diagnostic challenges, some of which may be routine or complicated, are encountered on occasion. This article briefly highlights some of these challenges and provides information on how they may be overcome. Some of the latest diagnostic methods are introduced; these are used to help diagnose lymphosarcoma when cytologic or histopathologic findings are inconclusive.

Cytologic evaluation of a fine needle aspirate is a commonly used and valuable method for diagnosing canine lymphosarcoma. In the hands of a skilled pathologist, diagnostic accuracy of this method is high, provided the smear preparations are of good quality (2). A smear preparation of good quality will contain a high cellular yield of intact cells with an even spread of a monolayer of cells. This will allow individual cell morphology to be seen clearly. Recognition of lymphosarcoma in a cytologic smear requires identification of lymphoid cell populations with monotonous, uniform morphology (Figure 1A,B). The process begins in the hands of the clinician when the specimen is being collected.

Figure 1A.

Lymph node aspirate. Reactive lymphoid hyperplasia — a mixed population of reactive lymphoid cells, with predominance of small lymphocytes (arrowheads) and smaller proportions of large lymphoid cells (arrows). Bar = 25 μm.

Figure 1B.

Lymph node aspirate. Lymphosarcoma — a monotonous population of large neoplastic lymphoid cells that approximate the diameter of a mature neutrophil (arrowhead). Bar = 25 μm.

A frequent problem that may be encountered with fine needle aspiration of lymph nodes, and may preclude definitive diagnosis of lymphosarcoma, is failure to retrieve sufficient numbers of intact lymphoid cells. This may result from missing the intended target tissue upon aspiration, for example, aspiration of salivary glands that are mistaken for mandibular lymph nodes, or it may simply be due to inadequate yield of cells upon aspiration of the node. Even when adequate numbers of cells are obtained, poor preservation of lymphoid cells is frequently encountered. This may result from excess negative pressure applied to the syringe upon aspiration, or excess pressure on the slides during smear preparation; lymphoid cells, in particular, neoplastic lymphoid cells, are fragile and may rupture easily. To optimize the proportion of intact cells, many clinicians have adopted the “syringe-free” method of lymph node aspiration; the needle, without a syringe, is introduced into the node with several redirections of the needle before it is pulled out. This method generally yields abundant tissue, and as a syringe is only used to force air through the needle to discharge the contents onto slides for smear preparation, minimal trauma occurs to the lymphoid cells. When preparing pull smears, only the weight of the top slide is required to spread the cells across the bottom slide — any additional pressure often results in crushing of cells. In order to quickly determine that the target tissue was truly aspirated and to ensure that the specimen yield is adequate, many clinicians opt to stain 1 of the prepared air-dried slides with Diff-Quik (VWR Scientific of Canada, Mississauga, Ontario) for immediate and brief microscopic screening before sending all slides to a pathologist for interpretation. A final obstacle that is infrequently encountered with cytologic smears is poor staining due to prior exposure of the air-dried smears to formalin vapors from a nearby open formalin jar, even if the exposure is only a few seconds. Care must be taken to avoid this situation.

Additional diagnostic techniques are warranted in cases where good quality specimens were obtained, but cytologic results are inconclusive. In such cases the logical next step is to biopsy the lesion for histopathologic evaluation, which has the advantage of providing architectural features of the affected tissue. Depending on the tissue targeted, different types of biopsy collection may be indicated. With lymph nodes, it is best to surgically extirpate and submit the entire node, such that the entire architecture of the node can be assessed for neoplastic effacement. For internal organs, surgical biopsies are an excellent choice if the option is feasible for the patient. For example, a full thickness surgical wedge biopsy is optimal for gastrointestinal lesions. For the liver or kidney, surgical biopsy has the advantage of direct visualization of a lesion and allows a larger sample to be obtained. If surgery is not an option; however, ultrasonographic guidance is very helpful in targeting suspicious lesions for biopsy. With the spleen, obtaining representative biopsy samples can be challenging and often a splenectomy and submission of the entire spleen for histologic evaluation increases the chance of identifying a representative lesion. Bone marrow biopsy is typically performed during collection of a fine-needle aspirate of the marrow.

These traditional methods are often the initial approach taken in diagnosing lymphosarcoma because of the benefits of high diagnostic accuracy, fast turnaround time, and low expense. Inconclusive results, however, may be encountered from time to time using these methods, despite excellent specimen quality. This has prompted the development of more sophisticated diagnostic methods to help differentiate between a population of reactive and neoplastic lymphoid cells. Lymphocyte morphology can be quite ambiguous in many types of non-neoplastic lymphocyte proliferations, which can be ignited by antigenic stimuli from infections or autoimmune disorders, allowing overlap with morphology typically encountered with neoplastic lymphoid cells. Likewise, the sensitivity of detecting an early or insidious neoplastic population amongst a population of non-neoplastic cells is limited with cytologic and histologic methods. A particularly frustrating situation is how to categorize monotonous populations of lymphoid cells in splenic aspirates, as reactive splenic nodules can have a clear predominance of non-neoplastic lymphoblasts (2). The following tests have been developed in recent years to address these problems, and to expedite the appropriate diagnosis where time is of the essence.

Immunophenotyping of lymphoid cells is used to identify combinations of the many cell surface markers such as CD 79a (B-cell lymphocyte marker) and CD3 (T-cell lymphocyte marker), by using a wide array of labeled antibodies that detect a specific surface marker. A polyclonal, or reactive population of lymphoid cells would be expected to have many different combinations of surface markers, highlighting different subsets of B cells and T cells within a proliferating population. Conversely, a monoclonal, or neoplastic population of lymphoid cells would be comprised of cells with a homogeneous phenotype where the markers are identical or near identical for all cells (3). Furthermore, identification of a uniform lymphocyte population in an inappropriate location (predominance of B cells in thoracic effusion, where T lymphocytes should normally predominate if the fluid is simple chyle) and identification of aberrant surface markers (loss of surface markers that are expressed in normal lymphocytes) can lead to a diagnosis of lymphosarcoma (3). As no test is 100% sensitive and specific, interpretation by a pathologist requires skill in recognizing bi- or oligo-clonal neoplastic populations, recognizing the potential for rare, chronic infections (such as Ehrlichia canis) that can lead to monoclonal expansions, and recognizing methodology problems that may spuriously mask detection of a normal surface antigen (3). This technology has been employed using unstained air-dried smears (immunocytochemistry) and unstained paraffin embedded sections (immunohistochemistry) to differentiate a B-cell from a T-cell lymphosarcoma in a previously confirmed lymphosarcoma case. Flow cytometry is a valuable method that uses immunophenotyping to identify surface marker combinations in scores of thousands of cells within a fluid medium such as blood (Figure 2), cerebrospinal fluid, or third-space effusions in order to obtain information about the cell populations that, when combined with other diagnostic findings, can help differentiate a reactive from a neoplastic process (3).

Figure 2.

Blood smear. Two large lymphoid cells with visible nucleoli — the morphology is suspicious for circulating neoplastic lymphoid cells. Bar = 25 μm.

Another method gaining in popularity is the determination of a clonal, and therefore neoplastic, population of lymphoid cells by polymerase chain reaction (PCR) for antigen receptor rearrangement (PARR). This technique requires harvesting deoxyribonucleic acid (DNA) from air-dried, unstained or previously stained slides, or fluid media from the lesion, and applying primers to detect and amplify specific genes that encode immunoglobulin antigen receptors in B cells and T-cell receptors in T cells (3,4). Lymphocyte differentiation normally involves rearrangement of these genes in response to the multitudes of self and foreign (infectious, environmental) antigens that the immune system encounters during development. The result is an unfathomably large and variable population of lymphoid cells that can appropriately effect a response to the showering of uncountable foreign antigens encountered throughout a lifetime, while allowing an anergic response to self antigens. Genetically, these lymphoid cells are each represented by a specific complementary-determining region that has its own length and sequence. The goal of PARR is to amplify this region from many cells and electrophoretically separate the DNA bands on polyacrylamide gel to highlight a variety of PCR products (“ladder appearance” on the gel) which suggests a polyclonal lymphoid population, or 1 to 2 uniform band(s) of PCR product, suggesting a monoclonal or biclonal neoplastic population (3). As primers are available for B- and T-cell genes, the method can not only distinguish a neoplastic population amongst a heterogeneous population of reactive cells (Figure 3), but in some cases differentiate between B- and T-cell origin (4). This test has high sensitivity and specificity; one study showed that PARR was 2.5 times more likely to detect neoplastic lymphoid cells in peripheral blood than light microscopic evaluation alone (3). This method is not 100% sensitive or specific, and false negatives can occur due to lack of primers that hybridize to all possible sequences that neoplastic cells may contain; however, there has been significant advancement in this field (4). False negatives can also result from the presence of tumor cells that do not have antigen receptor rearrangement, such as “natural killer” cells, or insufficient yield of DNA from a sample (4). False positive results infrequently occur in chronic rickettsial infections where a minimally diverse antigenic stimulus can lead to oligo- or monoclonal, yet reactive lymphocyte proliferation (4). Likewise, a monoclonal band can arise if DNA from only a few, normal lymphoid cells are sampled or the DNA quality is poor (4).

Figure 3.

Splenic aspirate. Clusters of uniform lymphoid blast cells (arrowheads) are a minority amongst a heterogeneous population of hematopoietic cells and are not thought to represent proliferating rubriblasts or myeloblasts. PCR for antigen receptor rearrangement (PARR) would be helpful in highlighting clonality of these cells, leading to a diagnosis of lymphosarcoma. Bar = 25 μm.

Less widely available, yet emerging tests that will likely play a larger role in detecting all tumor types in the future include detection of chromosomal abnormalities by karyotyping and genome hybridization, and gene expression profiling (4).

To conclude, newer diagnostic techniques have advantages of increased sensitivity and specificity; however, they are not meant to replace more traditional, less expensive methods. These techniques are best used after traditional methods and clinical findings have allowed a significant degree of suspicion of a lymphoid neoplasm, yet have not resulted in a satisfactory diagnosis. Clinicians are encouraged to discuss the necessity and regional availability of these techniques with a clinical pathologist on a case-by-case basis.