Histologic tumour margin assessment – general information, terminology, techniques, limitations and artefacts
For surgery of a neoplasm with curative intent, the goal is to achieve a macroscopic margin around the tumour at surgery and a microscopically tumour-free margin. The assumption is that with an ‘adequate’ margin, local recurrence can be prevented. In reality, the matter is more complex. Neoplasms with neoplastic tissue histologically extending to the tissue border (often referred to as ‘dirty’ margin) may not recur, while in other cases neoplasms with microscopic tumour-free margins (often referred to as ‘clean’ margins) may locally recur. Possible reasons for this are discussed in the paragraphs below.
“Dirty” margins: Why hasn’t the tumour recurred?
The reason for non-recurrence of neoplasms with ‘dirty’ histologic margins may be a false positive result due to a tissue artefact (more explanations on artefacts below).
It could also be that a tumour was ‘shelled-out’. Histologically, there may be no non-neoplastic tissue visible surrounding the neoplasm, however, this does not mean that tumour cells have remained in situ. Even if some neoplastic tissue remains at the surgical site, it may be eradicated by the host immune response (possibly triggered by the surgical procedure).
For benign, well-circumscribed neoplasms, ‘shelling-out’ is generally expected to be curative and literature indicates, that for some ‘low grade’ malignancies, such as ‘low grade’ soft tissue sarcoma, in many cases local recurrence does not occur after ‘shelling-out’.
For some low grade malignancies with incomplete local resection, local recurrence may be delayed and may still occur after a longer observation period.
“Clean” margins: Why has the tumour recurred?
Only a small subsection of the tumour and its marginal tissue are assessed histologically. Therefore, while the margins may be “clean” in the examined sections, there may be neoplastic tissue at the surgeon-cut edge in areas not histologically assessed. This is particularly problematic in highly infiltrative neoplasms.
Some neoplasm (e.g. high grade canine cutaneous mast cell tumours) tend to form satellite tumours (local metastasis) in the vicinity, which may not be resected with the primary tumour.
Theoretically, there may also be a residual pro-tumorigenic microenvironment at the excision site. This could predispose to novel malignant transformation at the site or induce “homing” signals that draw circulating neoplastic cells to the tumour environment.
Interpretation of the Margin Assessment
The histologic assessment of the margins must be interpreted in conjunction with clinical and surgical features, including location of the mass, the surgical margin and demarcation at surgery. Additionally, the specific tumour biology is important with some types of neoplasms having higher potential for recurrence regardless of the histologic tumour free distance. Therefore, correlation with tumour type and grade is critical. This information, along with other clinical factors, will need to be integrated to formulate the treatment plan, including the advisability of revision surgery, radiotherapy, and chemotherapy.
Outlined below is information on the terminology and techniques that we use in our lab for histologic margin assessment. Additionally, there are limitations of histologic margin assessment of which pathologists and clinicians should be aware.
Terminology
Histologic margin: The tissue that separates the neoplastic tissue from the tissue border. Often, the tissue border is referred to as the margin (as in the neoplasm extends to the margin); however, this is a wrong use of terminology. Please be aware that we only refer to the tissue border as the ‘surgeon-cut edge’ if the tissue was inked by the surgeon. This is because we cannot know if the tissue border we ink in the lab is the actual surgeon-cut edge (more explanations see below).
Histologically tumour-free distance: the width of the margin.
Margin dimensions: the deep margin is the tissue underlying the mass. We refer to the margins to the sides of the mass as horizontal. You may also see these margins referred to as lateral. We prefer to use the term “horizontal” because lateral is also a descriptor that may more precisely indicate location (as in lateral and medial). If the sample can be orientated and a more specific location can be indicated for the horizontal margins, we use the usual terminology (cranial/ rostral, caudal, ventral, dorsal, lateral, medial, etc.).
NB: Using sutures is a helpful method to provide orientation. In order to provide the most accurate assessment of orientation, please indicate the orientation in at least two different planes. For example, for a mass on the left lateral thorax, place 1 suture to indicate the dorsal horizontal border and 2 sutures to indicate the cranial tissue border. We will be able to provide you with a more precise assessment of the histologic tumour free distance at specific parts of the sample with this information.
Cruciate sections: A sample is sectioned cross-wise resulting in a transverse and longitudinal plane of section (see also below in techniques).
Shaved/ ‘en-face’ margins: The peripheral tissue of a sample is shaved off and embedded with the surface down in the block to be cut first (see also below in techniques).
Techniques for histologic margin assessment
Cruciate sections
This is the technique most commonly used, especially for ellipsoid excisions of skin tumours. A section is cut transversly through the narrowest area of the sample and then longitudinal sections are cut perpendicularly through the wider ends of the sample. This allows for examination of one plane of section through each of the 4 horizontal dimensions of the sample and examination of the deep margin.
Serial slicing and serial transverse section
For some samples, serial transverse slicing is the more appropriate technique to apply. Examples of samples that are generally serially sliced are small ellipsoid excisions, mammary strips or scar revisions. Depending on the size of the sample all (small ellipsoid excisions) tissue is subsequently processed or representative transverse sections are processed.
Shaved/ ‘en-face’ margins
As the cruciate and serial transverse sections allow for the examination of only a small proportion of the surface area of the tissue border, shaving of the deep and horizontal tissue border may be applied as an additional method and at an extra cost (HIMASK) charge code. This is particularly useful for neoplasms that appear poorly demarcated/ where infiltrative growth is suspected.
We do not shave neoplasms that have been narrowly excised (‘marginal excision’, ‘shelled-out’) because inevitably, neoplastic tissue will be observed in the shaved sections and no additional useful information is added. If shaving was requested, but is not thought to be useful or possible, we will inform you and remove the charge for shaving.
Techniques for complex samples
Complex sample types (such as whole extremities, maxillectomies, mandibulectomies, enterectomies, liver or lung lobes, digits, etc.) mostly require more bespoke methods to allow for histologic margin assessment. This is often a combination of representative sections through the margin into the mass and shaving of tissue borders. This is included in our charge for complex samples (HBEET).
Inking of samples
Inking of tissue borders of a sample can be very helpful for tissue orientation, as well as ensuring that the tissue border on a histologic slide represents the surgeon-cut edge created at surgery.
Inking is best performed by the surgeon. This will help the laboratory technicians and pathologists orientate in particular complex samples. Importantly, any artefact that may be introduced to the margin tissue between surgery and macroscopic examination in the lab can be spotted much easier. Lastly, ink sticks better to fresh tissue than formalin fixed tissue and does not ‘bleed’ into the tissue as much.
Any samples where margin assessment is required that do not come to the laboratory pre-inked are inked in the lab; however, the caveats outlined above need to be kept in mind.
Limitations
We can only report the absence of neoplastic cells/ tissue at the tissue border in the planes of tissue that we examine. The assumption is that for most benign tumours and well-demarcated malignant tumours, the absence of neoplastic cells/ tissue at the tissue border correlates well with the completeness of local excision.
For infiltrative neoplasms (e.g. feline injection site sarcoma) and neoplasms that tend to form satellite tumours (e.g. high grade canine cutaneous mast cell tumours), the relationship between the absence of tumour tissue at the tissue border and completeness of local excision is unknown.
We cannot know if neoplastic cells have remained in situ, even if neoplastic cells are observed at the tissue border histologically. For neoplasms that have been excised with no surrounding non-neoplastic margin (‘shelled out’, ‘marginal excision’)) it is impossible to say if neoplastic cells have been left behind or not. We assume that there is a higher likelihood that tumour cells remain in situ if tumour tissue is observed at the tissue border in tumours that are less demarcated and/or infiltrative.
Artefacts
As for all diagnostic tests, we must be aware of artefacts that may occur during processing, which may cause false interpretation of data. Knowing that these artefacts can occur allows us to take pre-emptive measures to reduce their clinical impact.
Tissue shrinkage: This is an expected artefact that changes the extent of the surgical margin compared to the histologically tumour free margin. Most of the shrinkage occurs immediately after surgery due to contraction of the tissue. Some tissue shrinkage is caused by formalin-fixation. The extent of the tissue shrinkage is unpredictable. Tissues may shrink between 10-50% and tissue components may be differently affected (e.g. skeletal muscle and fibrous connective tissue would be expected to shrink more than adipose tissue). This tissue shrinkage may also cause distortion of the sample, which can change the original extent of the margins. This can somewhat (but not completely) be prevented by suturing together the skin and the deep fascial plane (if one could be taken).
Distortion of samples can also be caused by using unsuitable containers. This is most apparent if samples get squeezed into containers that are too small. In addition to the distortion of the tissue borders, placing samples in small containers affects the speed of fixation. Poor fixation can cause other tissue artefacts.
Transport induced: Adipose tissue is very prone to laceration, which can happen during transport of the sample. Unless the surgeon-cut edge has been immediately inked after surgery, it is nearly impossible to differentiate the ‘true’ from a ‘false’ surgeon-cut edge in the lab.
All of these factors may lead to reporting of unexpectedly narrower margins or in the worst case false positive reporting of neoplastic tissue at the tissue border.
If you have any further questions or would like us to arrange a CPD on histologic margin assessment for your team, please complete this short form or contact your nearest laboratory.
References
Bray et al.: Evaluating the relevance of surgical margins. Part one: The problems with current methodology. Vet Comp Oncol 21 (2023).
Bray et al.: Evaluating the relevance of surgical margins. Part two: Strategies to improve prediction of recurrence risk. Vet Comp Oncol 21 (2023)
Milovancev and Russel: Surgical margins in the veterinary cancer patient. Vet Comp Onc 15 (2017).
Kamstock et al: Recommended Guidelines for Submission, Trimming, Margin Evaluation, and Reporting of Tumor Biopsy Specimens in Veterinary Surgical Pathology. Vet Path 48 (2011).