Trephine interpretation

As with aspirate interpretation where you have a system of assessing cellularity, then looking at megakaryocytes, then erythroid and myeloid/lymphoid lines - so it's the same with trephine interpretation. Because the examiners are aware that samples are normally reported by histopathologists, the pathology will be very distinctive to the eye and there should be clues in the wording of the question to point you towards a particular diagnosis.

This would be a suggested method:

1. Assess the architecture of the trephine (something not available to you on the aspirate). This should be done at low power - - ideally start with the x4 lens, but if that is not available a x10 objective will do. Oil lenses should not be used to examine trephines - the area of interest should be obvious at x4/x10 and then at x40 the precise abnormality should be apparent.

2. Examine:

the bone (is there evidence of increased or decrease in bony trabeculae?)
cellularity (changes in age will mean reduction in cellularity)
the stroma (supporting architecture of the marrow) - looking at blood vessels (for example for amyloid endovascularly)
- the stroma consist of the fibroblasts, fat cells, osteoblasts and osteoclasts in addition to the vasculature

3. Only when you've done this should you examine the haematopoietic cells.

Immature myeloid cells are found paratracebularly (e.g. promyelocytes and myelocytes); and they mature as they are further from the bony trabeculae. More mature cells (neutrophils, eosinophils) will be found near the sinusoids. Blasts are hard to see on a trephine sample and require staining with immunohistochemistry to conclusively identify them.

Erythroid islands are often clustered around a macrophage and are more mature the further away from the centre of the island. These are non-paratrabecular and non-sinusoidal.

Megakaryocytes are non-paratrabecular and peri-sinusoidal - and should not be found in clusters (clustering is pathological).

Plasma cells are frequently found near capillaries. Reactive lymphoid aggregates can occur, and frequently increase with age.

Lymphoproliferative disorders in bone marrow

Burkitt's lymphoma and T cell lymphomas may be more easily diagnosed on aspirates. Otherwise, the distribution of the abnormal infiltrate gives a key to what type of lymphoproliferative disorder it will be as follows:

Paratrabecular

follicular lymphoma (uncommon to form follicles in marrow)
mantle cell lymphoma
lymphoplasmacytic lymphoma
marginal zone lymphoma
NOT chronic lymphocytic leukaemia

Nodular (may abut the trabecula but not spread along it)

nodular sclerosing Hodgkin's disease
CLL or other small B cell lymphoma
reactive lymphoid aggregates

Interstitial

T cell lymphomas (commonly can be interstitial and nodular)
hairy cell leukaemia
normal T cells (will be scattered)
normal plasma cells (with some also perivascularly)

Diffuse

chronic lymphocytic leukaemia
hairy cell leukaemia (often packed pattern)
diffuse large cell lymphomas (commonly found throughout the marrow if present)

Random

high grade NHL
non-nodular sclerosing Hodgkin's disease (typically classical - which is associated with a stromal response - reactive or fibrotic. Look for the stromal response before examining for Reed-Sternberg cells)

Any combination

mantle cell lymphoma
lymphoplasmacytic lymphoma

Lymphoplasmacytic lymphoma in comparison with other lymphomas will have the presence of increased numbers of mast cells.

Guidance on lymphoproliferative disease interpretation

Interstitial lymphocytosis being reactive versus malignant is hard to tell without further clinical information or confirmation of a clonal population by immunohistochemistry or immunophenotyping
Perisinusoidal or intrasinusoidal are suspect
Paratrabecular infiltrates are neoplastic unless proven otherwise
Random irregular patches are likely neoplastic
Plasma cells should not aggregate apart from around perivascular spaces (normal plasma cells do NOT express CD20, 56 or light chain restriction). They WILL express CD138, 79a. They may lose CD79a in myeloma.
- Always look for Amyloid with Congo red staining (apple-green birefringence)
- MUM-1 can be useful for monitoring of low level myeloma
Infiltrates can appear paratrabecular if there is crush artefact, so be wary of this

Differential between follicular lymphoma and other paratrabecular lymphoma: follicular lymphoma is usually a solid lymphoma; others have blood borne component usually and should be reflected in the FBC.

Amyloidosis: look for a number of thickened blood vessels as well as plasma cells

Lymphoplasmacytic lymphoma: will attract mast cells, versus myeloma which does NOT attract mast cells.

Hairy cell leukaemia: looks like a pale infiltrate because of abundant cytoplasm... Some of the cells can look a little spindly. TRAP or BRAF can be useful; also have significantly increased reticulin

Hodgkin's: CD30, MUM-1, EBER if initial node positive, Reed-Sternberg cells are often CD45 negative

For previously treated patients with rituximab: you can test for CD79a or PAX-5

Small cell lymphoma: investigate CD5, CD23, cyclinD1, bcl6 (for FL, rather than bcl-2 in the marrow, as there are no germinal centres in the marrow), consider Ki-67 to look at proliferative index. Bcl-2 is positive in T and B cells in marrow; plus in reactive or neoplastic cells.

Other pathological features in trephine examination

Clustering of megakaryocytes may be seen in the following conditions:

myeloproliferative disorders
regenerating blood counts post chemotherapy
HIV

Fibrosis is a very eye catching feature on a trephine very visible at x40 and often at x10. Ensure you have reviewed a few slides of fibrotic change within a marrow either on a course or with your local histopathologist. Common features which you might find is so called 'streaming' and dilated sinusoids.

Haematological causes include:

Myelofibrosis
Hairy cell leukaemia (hence the experience of a 'dry tap' when taking an aspirate)
Hodgkin's disease
Systemic mastocytosis (the cells compared with hairy cell leukaemia will have granules - plus the exam question will likely lead you in the right direction to distinguish between the two).
- mastocytosis can present in the paratrabecular or perivascular areas with some spindly looking cells. Consider this if there is new trabecular formation and fibrosis of the marrow.

Non-haematological causes include:

bony disease
SLE
HIV myelopathy
granulomatous disease
other non-haematological neoplasms (may be associated with bone marrow necrosis)
- common findings with non-haematological neoplasms are a focal stromal reaction, clumping of cancer cells (check to make sure that these done appear to be erythroid islands - review another part of the biopsy to confirm)
- consider possibilities of breast; prostate; GI malignancy; lung and renal origins
- suspect a neuroectodermal tumour in children with any metastasis
- Useful stains to learn might be:
  - Carcinoma: cytokeratin stain
  - Prostate: check PSA
  - Melanoma: S100
  - (Haematology: CD45)

With the non-haematological causes there may simply be an increase in reticulin staining rather than an outright fibrotic picture.

Necrosis may also be a feature of cancer - on an H&E stain there will be large empty acellular pink spaces in between the trabeculae and fat cells, and loss of the osteocytes within the bony trabeculae (empty spaces of white where the cells should sit).

Google Sites

Report abuse