Fine Needle Aspiration of Lymph Nodes

Fine needle aspiration (FNA) of superficial and deep-seated lymph nodes is a well-established and safe method for assessing lymphadenopathy in adult and pediatric patients. Patients with primary or recurrent lymphoma frequently undergo FNA, and this diagnostic modality is even more common in suspected reactive or metastatic lymphadenopathy. Using FNA as a first-line procedure has obvious benefits, such as rapid turnaround time, low cost, and low morbidity. The cytomorphologic diagnosis is relatively straightforward in the majority of cases of reactive lymphadenopathy and non-hematopoietic metastatic disease. The diagnostic accuracy of FNA in lymphoma cases is variable and dependent on lymphoma type and concurrent use of ancillary studies. The latter significantly enhance diagnostic sensitivity and specificity beyond that obtained with cytomorphologic evaluation alone. Nevertheless, the inability to assess architectural features and limited immunohistochemistry can be challenging. Therefore the application of FNA to establish a primary diagnosis of lymphoma is controversial. Currently, with few exceptions, FNA has been used predominantly as a screening tool with a final diagnosis and lymphoma classification often requiring lymph node excision or biopsy. Only in cases in which excision or biopsy is medically contraindicated, diagnostic decisions must be based on the FNA specimen alone. The FNA without a follow-up biopsy is more commonly used for diagnosis of progression, transformation, or recurrent disease in patients with a previously documented history of lymphoma, and to procure fresh material for specialized studies such as genetic testing for targeted therapy.

The effectiveness of FNA as a diagnostic procedure in hematopoietic neoplasms is dependent on a dedicated multidisciplinary team of highly specialized experts in cytopathology, hematopathology, and experienced aspirators. This integrated approach will ensure the procurement of an adequate FNA specimen and analysis by appropriate cytomorphologic, immunophenotypic, and molecular techniques. A detailed clinical history should be reviewed at the time of FNA. An on-site evaluation provides information regarding the cellularity of a sample, guides selection of additional testing, and, in cases with scant cellularity, allows one to prioritize ancillary studies. If an adequate sample is available and there is suspicion of a lymphoproliferative process, material should be reserved for flow cytometry and cell block for immunohistochemistry or molecular studies. When a definitive diagnostic immunophenotype is not provided by flow cytometry or other ancillary studies, such as immunocytochemistry/immunohistochemistry, a lymph node excision is recommended.

This chapter focuses on the cytomorphologic diagnoses of the most common reactive and neoplastic lymphoid proliferations, and provides a guide for the optimal processing and evaluation of cytologic samples obtained by fine needle aspiration of lymph nodes ( Table 2-1 ).

Table 2-1

Approach to Diagnosis of Lymphoma on FNA: Recommendations and Limitations

Process	Recommendations/Limitations
Sample collection	On-site evaluation of specimen adequacy by a pathologist or cytotechnologist is highly recommended. If on-site evaluation is not available, perform additional passes until solution is cloudy. Perform multiple passes (>3) from different parts of lymph node. Prepare cell block for immunocytochemistry (critical in cases with rare neoplastic cells that may be difficult to immunophenotype by FC, such as cHL).
Case sign-out	Correlate cytomorphology with ancillary studies. Consider a possibility of false-positive and false-negative results of ancillary studies. In challenging cases or in cases with discrepant results, discuss cytomorphology with flow cytometrist, which may prompt additional gating. If cytologic findings do not correlate with clinical presentation, recommend lymph node excision or biopsy.
Evaluation of high-grade B-cell lymphoma and high-grade transformation	Large lymphocyte count >20% is highly predictive of large cell lymphoma (in practice, >25%). If large cell count is >25% but <50%, correlate cytomorphology with clinical and immunophenotypic findings. Biopsy/excision may be required in discrepant cases.

cHL, classical Hodgkin's lymphoma; FC, flow cytometry.

Specimen Collection and Processing

The proper handling and processing of a lymph node aspirate is imperative for an accurate diagnosis. In general, at least three separate passes should be executed. Non-aspiration technique (to minimize bleeding and sample mixing with peripheral blood) can be applied for cytologic evaluation of lymph nodes. On-site Coulter counters can be used to ensure collection of a minimum of 10 million cells for adequate flow cytometry. An on-site assessment for specimen adequacy should be performed by a pathologist. This is most easily accomplished with a Wright-Giemsa–type stain (usually a Diff-Quik [DQ]) carried out on an air-dried smear, which allows a detailed visual evaluation of cytoplasm and nuclei of the lymphoid cells, which is imperative for classification. Although DQ stain, comparable to the Romanowsky and Giemsa stains used in clinical hematology laboratories, is generally preferable for cytologic evaluation, some authors think that alcohol-fixed slides stained with Papanicolaou (Pap) stain should also be prepared to provide the enhanced nuclear detail. Techniques that use alcohol fixation with Pap staining, though, including monolayer technologies, are insufficient for demonstration of cytoplasmic features and should not be used as the only stain when hematopoietic processes are evaluated. If desired, these approaches can be used in addition to air-dried Giemsa-stained material. Air-dried Giemsa-stained cytospins may be particularly helpful because the cell morphology on the cytospin may be superior to that on the smear owing to the flattening and enlarging effect of cytocentrifugation ( Fig. 2-1 ). Sample collection for monolayer preparations is easy and may be a viable alternative, especially when on-site assessment is not available. However, such preparations are known to increase the risk of false-negative diagnoses and should not be used without accompanying FNA smears.

Figure 2-1, A, Smear of chronic lymphocytic leukemia/small lymphocytic lymphoma showing mostly small, atypical lymphoid cells with scant cytoplasm and round, slightly irregular nuclei with occasional nuclear clefts. B, Cytospin preparation of the same case showing the flattening and enlarging effect of atypical lymphoid cells, accentuating the nuclear irregularity and clefts (arrows) (Diff-Quik).

Once a differential diagnosis is formulated based on cytomorphology and clinical history, a portion of a sample should be placed in cell culture media such as RPMI. From this aliquot, cells can be submitted directly for flow cytometry and molecular diagnostics. A cell block or cytospin can also be prepared for immunocytochemistry/immunohistochemistry, fluorescence in situ hybridization (FISH), or in situ hybridization for Epstein-Barr virus (EBV) with the EBV-encoded small RNA (EBER) probe.

Ancillary Studies

Immunocytochemistry

Immunocytochemistry (ICC) can be performed on air-dried cytospins or smears on charged slides that have been stored desiccated and refrigerated, and are post-fixed in acetone before staining. The staining protocols used for air-dried cytospins are similar to those used for frozen section material (see Chapter 4 ). Cell block sections can also be used for immunohistochemical studies with a staining protocol similar to that used for tissue sections. If cellular material is limited, it may be preferable to prepare cytospins rather than attempting a cell block with potentially insufficient material. Of note, alcohol fixation may preclude the performance of some lymphoid markers.

ICC on cytospins may be as effective as flow cytometry for the immunophenotyping of cytologic specimens and may be particularly suited for samples with an insufficient number of cells for flow cytometric analysis. One distinct benefit of ICC on cytospins is the detailed visualization of cell size in conjunction with immunophenotypic staining patterns, particularly with mixed populations of cells.

Flow Cytometry

Flow cytometry (FC) is an indispensable ancillary technique that significantly enhances the sensitivity and specificity of lymphoma diagnosis in cytologic material. The combination of FC and cytomorphology can lead to a specific lymphoma classification with a rapid turnaround time of less than 24 hours. With the use of commonly available multiparameter FC instruments, numerous markers can be analyzed simultaneously, allowing for a comprehensive immunophenotyping of even paucicellular samples. This obviates the need to prioritize markers on the basis of a cytomorphologic triage. Optimally, several million cells are needed to analyze an expression of 10 to 20 antigens. In scant samples, immunophenotyping can be performed in as few as 50,000 cells provided that the cell viability is not compromised. In these cases and in those with a suspicion of rare types of lymphoma, close communication between the cytologist and the hematopathologist performing the FC may help to design a more specific panel of antibodies.

Sample type and processing can significantly influence cellular yield and viability. In our experience, FNA yields a superior material for FC in comparison with core needle biopsies (CNBs), which are often procured at the same time. The FC sample should be placed in RPMI with 10% fetal bovine serum immediately after procurement and stored at 4° C. Samples of select, highly proliferative lymphomas will deteriorate rapidly even if stored in a protective media; thus, whenever possible, a specimen should be stained promptly to avoid cell loss. If rapid analysis is not feasible, a stained sample can be fixed and analyzed the next day.

Results of FC have to be interpreted in a context of cytomorphology and clinical information because of a possibility of false-negative or false-positive results. A false-negative immunophenotype occurs most commonly in limited paucicellular samples because of necrosis and fibrosis, or cell loss during processing. A malignant clone may also be challenging to identify in cases in which neoplastic cells are accompanied by a rich reactive background or aggregating with reactive lymphocytes, as seen in classical Hodgkin's lymphoma. False-positive results may be encountered in reactive lymphoid processes with a skewed kappa-lambda light chain ratio, which is primarily seen in follicle center lymphocytes. Minor immunophenotypic abnormalities should not be interpreted as evidence of a clonal process unless supported by cytomorphologic findings and clinical presentation. Additional studies may be warranted when a false-negative or false-positive result is suspected.

When a definitive diagnostic immunophenotype is not provided by FC and other ancillary studies (such as immunocytochemistry or immunohistochemistry) are not conclusive, a lymph node excision is recommended. The latter may also be required in cases in which there is a discrepancy between clinical presentation, cytomorphology, and FC immunophenotype.

Molecular Studies

Molecular studies are most commonly performed in FNA cases in which a conclusive diagnosis cannot be reached with a combination of cytomorphology and immunophenotyping. T-cell and B-cell receptor gene rearrangements, including IGH and kappa light chain, as well as PCR-based assays for Epstein-Barr virus and human herpesvirus 8, can be performed on fresh or archival FNA samples (via slide scrape lysates). Similar to FC, the results of the molecular studies need to be interpreted in the context of morphologic findings and clinical data because of potential false-positive and false-negative assays.

Interphase FISH has added tremendously to the diagnostic specificity of FNA. Numerous commercially available probes allow for interrogation of lymphoma type–specific rearrangements. Cytospins or smears prepared from FNA material are ideally suited for FISH because scoring of fluorescent signals is easier in a cell population dispersed as a monolayer. Both fresh, ethanol-fixed, and archival (Pap and Diff-Quik stained) slides can be used. Adequacy of hybridization varies in published reports, with the highest success rates at approximately 95%. FISH with lymphoma type–specific probes is predominantly used for the classification of lymphoid neoplasms. A demonstration of genetic abnormalities by FISH is also an indication of clonality and can support a diagnosis of lymphoma in cases with equivocal cytomorphology and/or immunophenotyping. Of note, only select translocations, such as IGH/CCND1, are specific for a particular entity and allow for a definitive classification in a context of appropriate cytomorphology and immunophenotype. Many other genetic alterations can occur in several histologically and immunophenotypically defined entities, and thus are not diagnostic of a specific lymphoma type.

The importance of saving additional unstained smears for molecular assays has been emphasized by some cytopathologists, and the triage of FNA material based on on-site evaluation has been proposed. It remains to be seen whether FNA with stepwise application of ancillary techniques including molecular studies is the most accurate and cost-effective approach in patients with newly diagnosed lymphoma, particularly involving peripheral lymph nodes that are amenable to excision.

Non-Neoplastic Aspirates

Lymph node enlargement can be secondary to lymphadenitis, an inflammatory or infectious process or reactive lymphoid hyperplasia secondary to a variety of immune stimuli. Lymphadenitis is broadly divided into acute and chronic (granulomatous and non-granulomatous). The inflammatory or infectious processes can be readily identified in aspirated samples based on the composition of the cellular population. The presence of atypical lymphoid cells in an otherwise inflammatory background, however, raises the possibility of lymphoma.

Aspirates of reactive hyperplasia are diverse and diagnostically challenging ( Fig. 2-2 ). The pattern and distribution of the lymphoid population vary according to the stage of the reactive process and the primary lymph node compartment affected by it—lymphoid follicles or paracortex. Paracortical hyperplasia is characterized by a polymorphous population of lymphoid cells, ranging from small lymphocytes to immunoblasts, and other inflammatory cells including plasma cells, histiocytes, and eosinophils. Follicular center cells associated with tingible body macrophages and follicular dendritic cells (FDCs) predominate in follicular hyperplasia. The lymphoid cells are frequently seen in aggregates, enmeshed in a network formed by the FDCs and their processes. Some lymphomas are associated with a polymorphous background and may mimic an inflammatory process. A high proliferative activity favors lymphoma but can also be seen in some reactive conditions, such as infectious mononucleosis.

Figure 2-2, Reactive lymphoid hyperplasia.

Aspirates of Lymphoid Neoplasms

The WHO (World Health Organization) classification includes a wide variety of B-cell, T-cell, and histiocytic-dendritic cell neoplasms. Reviewing the cytologic features of all tumor types is beyond the scope of this chapter. Our goal is to focus on the cytomorphologic features of the most common types of B- and T-cell lymphomas encountered in clinical practice.

Mature B-Cell Neoplasms

The most common cell types seen in B-cell lymphomas are centrocytes, centroblasts, and immunoblasts ( Fig. 2-3 ). The range of cytologic appearances is exceedingly broad and reflects the spectrum of B-cell differentiation. Immunophenotypic and molecular features useful in the differential diagnosis are listed in the chapters reviewing individual disease entities, and this information will not be duplicated in the following paragraphs, unless unique to cytologic preparations.

Figure 2-3, Common cellular components of B-cell lymphomas.

Diffuse Large B-Cell Lymphoma, Not Otherwise Specified

Cytomorphology

Diffuse large B-cell lymphoma (DLBCL) is characterized by the presence of a significant number of discohesive large lymphoid cells ( Figs. 2-4 and 2-5 ). Smear preparations showing cohesive clusters of large lymphoid cells mimicking carcinoma cells may also be seen. Cytoplasmic fragments, so-called lymphoglandular bodies, are usually abundant. The majority of cells on FNA smear preparations are centroblasts. These cells have a vesicular chromatin pattern, distinct nuclear membranes, prominent nucleoli, and basophilic cytoplasm. The immunoblastic variant of diffuse large B-cell lymphoma shows a predominance of lymphoid cells (immunoblasts) with large round nuclei, single prominent nucleoli, and abundant plasmacytoid or clear to pale cytoplasm. Atypical large cells may display pleomorphic multilobated nuclei, similar to anaplastic large cell lymphoma. In cell block preparations, the presence of “sheets” of large lymphoid cells may be an indication of a primary diagnosis of large cell lymphoma or a transformation of a small cell lymphoma.

Figure 2-4, Diffuse large B-cell lymphoma, not otherwise specified.

Figure 2-5, Diffuse large B-cell lymphoma, not otherwise specified.

Differential Diagnosis

The differential diagnosis includes Hodgkin's lymphoma, Burkitt's lymphoma, histiocytic sarcoma, myeloid sarcoma, malignant melanoma, seminoma, and metastatic carcinoma. Distinctive cytologic features of non-lymphoid malignancies include the following:

Metastatic carcinoma—presence of atypical cells in clusters and absence (usually) of lymphoglandular bodies in the background
Metastatic melanoma—presence of pigment as well as intranuclear cytoplasmic inclusions
Seminoma—presence of a “tigroid” background on DQ with scattered small, mature lymphocytes; there may be multinucleated giant cells
Myeloid sarcoma—lack of lymphoglandular bodies; cytoplasmic granules may be present, including Auer rods (rarely); nuclear chromatin is finely distributed, with prominent and usually central nucleoli in blasts; myeloid maturation may be present.

Follicular Lymphoma

Cytomorphology

Follicular lymphoma aspirates comprise a mixed population of centrocytes and centroblasts in varying proportions ( Fig. 2-6 ). It is important not to confuse centroblasts with FDCs, a normal occupant of lymphoid follicles. FDCs have oval to coffee bean–shaped nuclei, with smooth nuclear membranes and indistinct cytoplasm ( Fig. 2-7 ). The atypical lymphoid cells may be seen in tight clusters, fragments of follicles, or adherent to the FDCs. Follicular structures may also be seen in reactive hyperplasia. Tingible-body macrophages may be seen on occasion but are less frequent than in reactive lymph nodes.

Grading

Although architectural assessment cannot be achieved on aspirates, grading of FNA samples using the counting method of Mann and Berard on entire smears or solely on follicular structures has been the subject of investigation using Pap-stained or Pap- and DQ-stained cytologic material. The investigators agree that the discrimination of large centrocytes from centroblasts is facilitated by use of the Pap stain. Although Sun and coworkers were able to discriminate intact follicular structures in smears and use them for a centroblast count, Young and colleagues were unable to make this discrimination reliably on any material other than cell blocks and used the entire smear for centroblast counting.

The WHO classification does not require the distinction of grades 1 and 2, which historically was challenging in both tissue sections and cytologic preparations. In the 2004 study by Sun and coworkers, a minimum of 200 cells was counted in 6 to 10 intact lymphoid follicular structures at 40× magnification. The number of large cells or centroblasts was expressed as a percentage of the total number of cells counted within the follicles and graded accordingly. In grade 3, they identified 48.4 ± 7.5% centroblasts, which is readily distinguished from significantly fewer centroblasts in grades 1 and 2 (9.7 ± 2.9% and 24.7 ± 5.6%, respectively). Recently, Brandao and coworkers were able to grade follicular lymphoma on Pap-stained monolayer preparations by counting the number of centroblasts in 300 lymphoid cells or 10 high-power fields.

Ancillary Studies

Both FC and interphase FISH for IGH/BCL2 rearrangement can be used to support a diagnosis of follicular lymphoma in cytologic preparations. The sensitivity and specificity of FC in detecting a neoplastic population varies with the assay and can approach 94% to 100%. Similarly, a recent study showed 81% sensitivity and 100% specificity for detection of the IGH/BCL2 rearrangement on archival Pap-stained cytologic smears. Of note, t(14:18)(q32;q21) can also be detected on Giemsa-stained smears by PCR-based techniques; however, it is generally less sensitive than the FISH-based approach.

Differential Diagnosis

Included in the differential diagnosis are reactive hyperplasia, mantle cell lymphoma, marginal zone lymphoma, small lymphocytic lymphoma, and diffuse large B-cell lymphoma, not otherwise specified.

Mantle Cell Lymphoma

Cytomorphology

Mantle cell lymphoma aspirates often show a monotonous population of small to intermediate-sized lymphoid cells with delicate nuclear clefts, dispersed or finely stippled chromatin, inconspicuous nucleoli, and distinct pale or basophilic cytoplasm ( Fig. 2-8 ). Two variants of mantle cell lymphoma—blastoid and pleomorphic—have potential clinical significance. The blastoid variant exhibits intermediate-sized to large lymphoid cells with enlarged, slightly irregular nuclei, evenly distributed chromatin, and small nucleoli ( Fig. 2-9 ). The cytoplasm on DQ-stained material is scant and pale blue. Apoptotic and lymphoglandular bodies may be present in the background. In the pleomorphic or anaplastic variant, the atypical lymphoid cells are larger, with more nuclear irregularity and hyperchromasia.

Figure 2-9, Blastoid variant of mantle cell lymphoma.

Ancillary Studies

The t(11;14)(q13;q32) is present in the vast majority of cases and can be detected by FISH on cytospins of FNA material. Immunohistochemistry for SOX11 may be informative in cases negative for IGH/CCND1 rearrangement.

Differential Diagnosis

The differential diagnosis includes reactive hyperplasia, follicular lymphoma, marginal zone lymphoma, small lymphocytic lymphoma, and lymphoblastic lymphoma.

Marginal Zone Lymphoma

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here