Diagnostic Lymph Node Pathology

Margaret Ashton-Key, Penny Wright, Dennis Wright

• Publisher: CRC Press
• Year: 2018
• ISBN:
  9781315352879
• Edition: 3
• Topic: Medicine
• Subtopic: Medical Theory, Practice & Reference

Diagnostic Lymph Node Pathology presents a logical and systematic approach to lymph node biopsies and guides general pathologists and haematopathologists alike through the maze of differential diagnoses, enabling them to reach an accurate diagnosis.

The book is presented in an easy-to-read format and contains boxes for quick reference detailing the clinical, morphological, immunohistochemical, and genetic features of each entity. Thoroughly revised, the third edition also includes more than 300 high-quality histological colour photographs along with updated references.

This edition covers both classical and less well-known features of individual disease processes. It is a valuable guide for both professionals and trainees.

1

Handling of lymph node biopsies, diagnostic procedures and recognition of lymph node patterns

Taking and handling of lymph node biopsies

Processing, sectioning and staining

Immunohistochemistry

Molecular techniques

Diagnosing the ‘undiagnosable’ biopsy

Where to begin? Recognizing lymph node patterns

Sinus architecture

Capsule

Reactive follicles

Overall growth pattern

Paracortex

Marginal zone

Necrosis

Apoptosis

Granulomas

TAKING AND HANDLING OF LYMPH NODE BIOPSIES

Suboptimal techniques in the taking and handling of lymph node biopsies are probably the biggest obstacle to achieving a correct diagnosis. All concerned with this process should bear in mind that the objective of the biopsy is to achieve a timely and accurate diagnosis on which the subsequent management of the patient can be based. Feedback information at multidisciplinary team meetings is a valuable means of achieving and maintaining a high diagnostic standard of lymph node biopsies. In the absence of such meetings, personal contact is needed to ensure that any shortcomings in the biopsy technique and handling of the specimen are rectified.

Lymph nodes should be selected for biopsy on the likelihood that they contain the pathological process. They should be dissected out whole, if possible, and with the capsule intact. Fragmented nodes may be more difficult to diagnose than intact nodes, depending on the pathological process involved. Traction artefacts are usually most severe when the biopsy tissue is very fibrotic or has to be taken from a confined space, such as the anterior mediastinum.

Needle biopsies are now more frequently used for the diagnosis of lymph node pathology. When possible, open lymph node biopsies should be used for superficial, accessible lymph nodes; however, needle biopsies have a lower morbidity than open biopsies and are of particular value in sampling mediastinal, abdominal and retroperitoneal lymph nodes, avoiding the need for more invasive procedures, such as laparotomy. These biopsies are usually taken by radiologists using ultrasound or computed tomography (CT) guidance. If fixed quickly, needle biopsies give good morphological preservation, which, together with immunohistochemistry, allows the precise identification of most common lymphomas. The technique may be less successful in the identification of non-neoplastic proliferations. The handling and preparation of needle biopsies is discussed in more detail in Chapter 11.

Fine needle aspiration (FNA) biopsies have their greatest value in the separation of carcinoma from lymphoma and for the identification of recurrences or for staging. The role of this technique for the primary diagnosis of lymphoma is limited and presents many pitfalls unless in the hands of an expert cytopathologist. Alternatively, the lymph node aspirates obtained via ultrasound-guided endoscopic (endoscopic ultrasound [EUS]) or endobronchial (endobronchial ultrasound [EBUS]) sampling may be processed entirely as clot preparations, without making aspirate smears, thereby maximising the material available for immunohistochemistry. Additional information may be obtained if a separate aspirate sample is sent for flow cytometry. Ideally, the clot preparations obtained will contain lymph node microbiopsies, and serial sections may be stained with haematoxylin and eosin (H&E) whilst the intervening spare sections may be used for immunostaining or other special stains (Figure 1.1).

Figure 1.1

Series of sections from a fine needle aspiration cell block preparation demonstrating the use of ‘spare sections’ between haematoxylin and eosin (H&E) levels for immunohistochemistry.

Logistics dictate that many laboratories receive their lymph node biopsies in fixative. In such cases, the volume of the fixative should be at least ten times that of the specimen. Whole lymph nodes should be sliced as soon as possible to allow rapid penetration of the fixative.

Ideally, lymph node biopsies should be received fresh in the laboratory immediately after excision. This requires good communication between the pathologist and surgeon or radiologist, to ensure that there is minimum delay in the specimen reaching the laboratory. A slice taken from one end of the node can be gently touched onto a clean glass slide, and air-dried and stained by one of the rapid Romanowsky techniques to provide a rapid cytological assessment. Pathologists experienced with this technique may be able to give a provisional cytological diagnosis from this when appropriate. The technique is most useful, however, in determining the subsequent handling of the specimen in the laboratory. The slice of lymph node used to make the imprint preparation may be frozen for subsequent molecular investigation. It should not be used for histology, if this can be avoided, since the process of making imprint preparations often causes traction artefacts in the tissue (Box 1.1).

BOX 1.1: Fresh whole lymph node biopsies

Slice using clean sharp blade; use slices as follows:

• Imprint cytology (tissue used to make imprints should not be used for histology) may be used for:
• Rapid evaluation
• Air-dried imprint slides stored for fluorescence in-situ hybridization (FISH) if required
• Histology and immunohistochemistry. Place slices in fixative; if using formalin fix for 12–24 hours
• Fresh tissue slices may be used for:
• Cytogenetics
• Molecular analysis
• Cell culture
• Microbiology

Fresh tissue may be sent for cytogenetic analysis and/or flow cytometry in appropriate cases. Frozen sections may be cut for morphology and immunohistochemistry when indicated. One or more slices of the node should be placed in fixative overnight or longer for histology and immunohistochemistry. Needle biopsies require a similar period of fixation. Fortunately, for diagnostic purposes a wide range of procedures, including immunohistochemistry, polymerase chain reaction (PCR) and fluorescence in-situ hybridization (FISH), can be performed on fixed tissue (Box 1.2).

BOX 1.2: Fixed whole lymph node biopsies

• Cut into 5-mm slices with a sharp scalpel as soon as possible after biopsy
• Place in fixative, at least ten times the volume of the specimen
• Leave in fixative for 12–24 hours if formalin is the fixative
• Tissue for long-term storage should be blocked in paraffin after fixation, not left in fixative

PROCESSING, SECTIONING AND STAINING

Laboratories should maintain quality control of their reagents and equipment to ensure adequate processing, cutting, staining and immunohistochemistry. Cell morphology is important in haematopathology and can easily be obscured or distorted by poor fixation, processing and sectioning. Section thickness has a marked influence on cytological and histological appearances. The optimum thickness is 3–5 µm.

H&E is the stain most widely used in histopathology and is often the only one used in lymph node diagnosis. The Giemsa stain can add another dimension to haematopathology and is the stain of choice for this subspeciality in much of mainland Europe. The Giemsa stain highlights basophilia and eosinophilia, and this aids the identification of blast cells, plasma cells, eosinophils and mast cells. When using this technique, care must be taken with the quality of the Giemsa stain used and the pH of the reagents; otherwise a section stained uniformly pale blue is obtained, which is of little diagnostic value.

The periodic acid–Schiff (PAS) stain is sometimes of value in haematopathology. It highlights intranuclear immunoglobulin M (IgM) inclusions (Dutcher bodies), basement membrane and ground substance, such as that seen around the blood vessels in angioimmunoblastic T-cell lymphoma. The reticulin stain can be of value in determining the overall structure of the lymph node, highlighting follicularity, sinus structure and blood vessels.

In some laboratories, all of the aforementioned stains are used as a ‘lymph node set’. However, there is now a tendency to move directly from the H&E section to immunohistochemistry, when the additional use of one or more of these stains could be of greater diagnostic value. If only a small amount of biopsy material is available, as with most core biopsies, avoid cutting ‘levels’ routinely but keep spare unstained sections for immunohistochemistry: re-cutting the block inevitably wastes valuable tissue.

IMMUNOHISTOCHEMISTRY

With experience and good histological preparations it is possible to diagnose many of the common lymphomas on morphology alone. In most practices, however, a substantial number of cases cannot be categorized precisely without the aid of immunohistochemistry. Even in biopsies that are diagnosable with reasonable certainty on morphology alone, such as diffuse large B-cell lymphoma (DLBCL), immunohistochemistry will often provide additional prognostic information. It has therefore become standard practice in many laboratories to perform confirmatory immunohistochemistry on all lymphoma biopsies. The cost of this is trivial when set against the need to obtain an accurate diagnosis and the expense of treatment. Unfortunately, in the developing world, where obtaining and maintaining antibodies is often difficult and no costs are trivial, diagnostic immunohistochemistry is not widely practised.

In our experience, it is not uncommon to receive biopsies from pathologists who have made a reasonable morphological diagnosis but have been confused by the subsequent immunohistochemistry. To avoid this pitfall, pathologists should be aware of the staining characteristics of antibodies and the specificity of their reactivity. The laboratory should be subject to ongoing quality control. For most antibodies used in haematopathology, there will be internal controls within the tissues being investigated (e.g., reactive B- and T-cells, histiocytes). For antigens not commonly expressed within normal and reactive tissues, such as anaplastic lymphoma kinase 1 (ALK-1), external control tissues are necessary. Beware the section that is uniformly blue; it usually indicates technique failure.

Immunohistochemical techniques have improved considerably in the past decade, with the production of increased numbers of robust antibodies and the development of techniques for antigen retrieval.

Is it a lymphoma?

Large cell lymphomas may resemble other anaplastic neoplasms. The leukocyte common antigen (LCA, CD45) shows membrane expression in almost all lymphoid cells and has only rarely been reported in non-haematopoietic cells. However, there are pitfalls. LCA may be absent on precursor (lymphoblastic) B- or T-cell lymphomas. Anaplastic large cell lymphomas (ALCLs) are also often LCA negative, as are classic Hodgkin/Reed–Sternberg cells. ALCLs frequently express epithelial membrane antigen (EMA), which together with LCA negativity may suggest an epithelial neoplasm. EMA is expressed, however, on plasma cells and on the cells of a number of lymphomas. Most epithelial neoplasms (carcinomas) express low molecular weight cytokeratins. In the rare cases in which low molecular weight cytokeratins have been reported in lymphoma cells, this has usually been in the form of a paranuclear dot. The majority of malignant melanomas can be identified with antibodies to S100 protein and HMB45.

Is it a B-cell lymphoma?

IMMUNOGLOBULINS

B-lymphocytes are defined by their ability to synthesize immunoglobulins, which should therefore provide the most reliable means of identifying these cells. In practice, they are not used for this purpose in most laboratories. The main reason for this is that plasma immunoglobulins cause diffusion artefacts, particularly in poorly fixed specimens, that are often confusing and obscure specific staining. Cells that appear positive for immunoglobulins owing to passive uptake usually show smooth cytoplasmic staining that is most intense at the cell membrane. Within the node, these cells often occur in broad bands corresponding to the advancing front of the fixative as it diffuses into the tissue. Immunoglobulin in synthetic cells often appears granular, owing to its accumulation within the endoplasmic reticulum, or as larger inclusions. Synthesized immunoglobulin also frequently manifests as paranuclear (Golgi) staining and as strong staining around the nucleus corresponding to immunoglobulin within the perinuclear space. IgM is synthesized by a large proportion of DLBCLs and, because of its large molecular size and relatively low concentration in the plasma, shows less diffusion artefact than other immunoglobulins. Surface IgD can be recognized in well-fixed paraffin-embedded tissues. Reactive mantle cells are positive, as are most cases of B-cell chronic lymphocytic leukaemia/small lymphocytic lymphoma (B-CLL/SLL) and mantle cell lymphoma.

In addition to their use as B-cell lineage markers, immunoglobulins can be used to imply clonality. Clonal (neoplastic) populations are monotypic (i.e., they express only one light chain). It should be noted that not all monotypic populations show clonal immunoglobulin rearrangements (see Castleman disease and paediatric nodal marginal zone lymphoma). Monotypia may be obscured by a background population of reactive cells, which usually express kappa and lambda light chains in the ratio of two to one. In general, reactive cells show more intense staining for immunoglobulins than neoplastic cells.

Although immunohistochemistry can be used to identify light chain restriction in cells expressing cytoplasmic immunoglobulin, it is difficult to identify surface light chains except in the best fixed tissues. In-situ hybridization (ISH) for kappa and lambda mRNA offers an alternative and more reliable technique for demonstrating light chain synthesis, which avoids the problems of background staining.

CD20 (L26)

CD20 is a non-glycosylated phosphoprotein expressed on the membrane of B-cells. Although widely regarded as such, it is not a perfect B-cell marker because it is not expressed in the earliest stages of B-cell differentiation and is lost as the B-cell undergoes plasma cell change. It is absent therefore on many B-lymphoblastic lymphomas and plasmacytic tumours. Staining for CD20 should be on the cell membrane; cytoplasmic, nuclear and nucleolar staining are non-specific. Because it is a surface membrane antigen, CD20 is often most strongly stained in biopsies that show some degree of shrinkage artefact and may appear less strongly stained in well-fixed tissues, such as needle biopsies. CD20 expression is of clinical importance because of the use of rituximab, an anti-CD20 antibody, in the treatment of lymphomas. Rarely, expression of CD20 may be lost in relapsed lymphomas after rituximab therapy.

CD20 has rarely been reported on T-cell lymphomas. It is also expressed on the epithelial cells of some thymic carcinomas (beware when diagnosing mediastinal large B-cell lymphoma).

CD79A

CD79 is a heterodimeric glycoprotein signal transduction molecule that associates with membrane immunoglobulin. Antibodies to the α-chain of the molecule (CD79a) provide an almost perfect B-cell lineage marker because CD79a is expressed throughout B-cell differentiation. It should be noted, however, that 50 per cent of T-lymphoblastic lymphomas express CD79a. Staining for CD79a is cytoplasmic and is strong on plasma cells. It is more strongly expressed on mantle cells than on germinal centre cells.

PAX5

PAX5 is a transcription factor that encodes the B-cell lineage specific activator protein (BSAP) that is expressed at earlier stages of B-cell differentiation but is usually not detected in plasma cells. It is a reliable marker of B-cell origin and is useful in the diagnosis of classical Hodgkin lymphoma. Rarely, aberrant expression may be seen in T-cell lymphomas, and PAX5 expression is also seen in small cell carcinoma and Merkel cell carcinoma.

OTHER MARKERS

Details of other markers of value in the diagnosis of B-cell lymphomas are as follows:

• Annexin A1—Is a member of the annexin family of calciu...

Table of contents

• Half Title
• Title Page
• Copyright Page
• Contents
• Preface to the third edition
• Preface to the second edition
• Preface to the first edition
• Acknowledgements
• Authors
• Abbreviations
• 1 Handling of lymph node biopsies, diagnostic procedures and recognition of lymph node patterns
• 2 Normal/reactive lymph nodes: Structure and cells
• 3 Reactive and infective lymphadenopathy
• 4 Precursor B- and T-cell lymphomas
• 5 Mature B-cell lymphomas
• 6 T-cell lymphomas
• 7 Immunodeficiency-associated lymphoproliferative disorders
• 8 Hodgkin lymphoma
• 9 Histiocytic and dendritic cell neoplasms, mastocytosis and myeloid sarcoma
• 10 Non-haematolymphoid tumours of lymph nodes
• 11 Needle (trephine) core biopsies of lymph nodes