Intrinsic Limitations
a) Pathologist handling
After surgical removal, the specimen is fixed to preserve cell morphology and tissue architecture before being visually examined during the grossing step. The method of sectioning is debated in HN cancers, as there are no standardized procedures unlike other sites such as the colon or breast. Generally, specimens can be cut using the “radial or perpendicular” (right angle) or the “parallel or en-face” (shave margin) method. Only the first method allows for measuring the microscopic distance in millimetres between the inked margin and the tumour [
9]. Despite the apparent linearity and simplicity of the grossing step procedures, it is important to highlight the potential for mistakes during the initial stages of sampling and the subjectivity that can exist among pathologists in the subsequent histologic interpretation [
10].
Regardless of the method of grossing tumour used, each of the subsequent histological processing steps (dehydration, clearing and infiltration, microtomy cutting, haematoxylin and eosin staining, and placement on the microscope) theoretically has the potential to cause tissue alterations that can impact microscopic analysis [
6].
Therefore, there is an urgent need for more concrete guidelines to better define the criteria for correct and standardized pathological margin assessment and reporting. Additionally, there is a general consensus on the specific need to establish a group of dedicated HN pathologists collaborating with surgeons from the time of surgical tumour removal through the initial stages of tissue processing. This collaboration could facilitate the development of specialized expertise and promote a closer working relationship between surgeons and pathologists to address the mutual challenges in interpreting these highly complex specimens.
b) Tissue shrinkage
The necessity to resect approximately 1–1.5 cm of normal tissue together with the tumour [
7] is primarily due to the well-known phenomenon of shrinkage, which occurs immediately after resection and to a lesser extent after formalin fixation. It also appears to vary within different regions of a single specimen [
6], less intra-tumoral and more pronounced in the marginal region: consequently, surgical margins tend to move closer to the tumour mass [
11].
This raises the question: Is specimen shrinkage predictable? And how can we utilize this information in margin assessment? An interesting recent study by Burns et al. [
12] found an average shrinkage of 26% (post-resection and post-processing) and suggested adjusting margin measurements accordingly. This proposal could potentially eliminate the need for larger resections in situations where they are unachievable due to tumours located in functionally or cosmetically important areas. Furthermore, it could help avoid the necessity for re-resection or adjuvant treatment in cases with close or positive margins. However, it should be noted that this remains a proposal at present.
c) Minimal residual disease and precancerous field
Even when surgical margins are histologically tumour-free, local recurrence may still occur in 10–30% of cases. These unexpected local relapses can be attributed to two different mechanisms: minimal residual disease (MRD) and the persistence of a precancerous field [
13].
MRD refers to the persistence of small tumour cell clusters in the margins after surgery. These clusters may be too small to be identified through routine histopathology, making their detection in a single or a few pathological sections within a large tissue volume extremely challenging or even impossible. The presence of these unrecognized clusters along the cutting line can eventually lead to recurrence at the surgical site, even if the margin was initially classified as clear [
13].
Immunohistochemistry (IHC) can aid in detecting residual isolated cells or small clusters, as used in sentinel node biopsies analysis [
14].
Precancerous field, involves the persistence of genetically mutated cells that are phenotypically indistinguishable from normal cells. As a result, they are not visible to the naked eye during resection or histopathological margin analysis. Various techniques, including immunohistochemistry and the detection of copy number changes, have limited clinical use currently [
13].
Extrinsic Limitations
a) Surgical tools and thermal damage of margins
New surgical instruments using radiofrequency or ultrasound for cutting and haemostasis have emerged in the last decades, replacing cold blades, ligatures, or clips. However, all these instruments can potentially cause damage to nearby tissues due to the lateral spread of thermal energy. The extent of lateral thermal spread varies depending on the type of instrument, power setting, and application time [
15].
Electrosurgical devices, such as the monopolar scalpel and bipolar forceps, are the most commonly used instruments in HN surgery. The monopolar scalpel remains the gold standard and is widely used worldwide, despite being shown to generate the highest temperatures and cause the greatest tissue damage, up to 1.5 cm [
15]. On the other hand, newer generation instruments, such as the Electrothermal Bipolar Vessel Sealing System, Harmonic Scalpel, and CO2 Laser, produce minimal thermal damage (in the order of micrometres) [
15,
16]. According to Mannelli et al. [
17], thermal damage to specimens can lead to histopathological mistakes with potential therapeutic and prognostic implications. The possible consequences of thermal damage include: (i) false positives with indications for re-excision or adjuvant radiotherapy due to the loss of healthy tissue caused by tissue damage, resulting in a reduced readable distance between the margin and the tumour, or due to thermal changes that can mimic tumour characteristics, making it difficult to distinguish from cancer cells; (ii) false negatives with the absence of indications for re-excision or adjuvant radiotherapy, as thermal injury can partially destroy small cancerous or pre-cancerous cell clusters along the margin, falsely indicating complete resection. Thus, the extent of thermal injury caused by the instruments used for tumour resection is of utmost importance, as the pathologist cannot take into account the areas of epithelium affected by thermal injury [
15‐
17].
It should be noted that new generation instruments were introduced in the early 2000s, while the trials that definitively established the need for adjuvant therapy in case of adverse features were based on patients surgically treated between 1994 and 2000 [
2,
3]. It can be speculated that during those years, the most commonly used instrument (although not specifically mentioned in the methods section) was the monopolar scalpel, with the potential consequences mentioned above. One might wonder if the use of newer generation tools causing less thermal damage would have yielded the same results in those studies….
b) Different kind of surgical resection and frozen section sampling
Recent decades have seen a shift from the traditional “en-bloc” resection, which involves removing the tumour along with a margin of 1–2 cm of normal appearing tissue, to the so-called “piecemeal” resection [
6,
18]. In classical en-bloc resection, margins are primarily assessed during the final pathological examination, and the pathologist may include the peripheral part of the tumour in the section to determine the microscopic distance in millimetres between the inked margin and the tumour front [
9,
19]. On the other hand, in piecemeal resection, radicality is achieved step by step through the removal of additional tissue strips from the tumour bed until a tumour-free margin is obtained during frozen section (FS) analysis [
6,
19]. In this type of surgery, it can be challenging for the pathologist to accurately assess the true distance between the tumour front and the inked margins, as the tumour is not always present in every tissue strip. Consequently, calculating the total length of the margins by adding the widths of multiple tissue strips seems to be impractical [
5,
9].
Given these considerations, the strict metric assessment proposed by current guidelines appears difficult to apply to piecemeal resection, even though it is a well-established surgical technique.
FS offers real-time information on the completeness of tumour resection. Despite the widespread use of FS [
20] some open problems still exist. First, point sample technique results in FS that may not be representative of the entire margins. On the contrary, if FS are collected as tissue strips for superficial margins and as a bowl for the deep ones, surgical margins can be examined in their entirety [
21]. Moreover, it is not clearly defined which site is more correct to collect FS from, whether from the tumour bed or the surgical specimen. Some studies suggest that the specimen driven approach is more predictive for local control [
22]; however, a recent survey by Bulbul et al. demonstrated an almost similar distribution between the two methods among the interviewed surgeons [
20]. A survey by the American Head and Neck Society [
23], although a bit dated, showed that 76% of the respondents collected FS from the surgical bed. The risk of this approach is the difficulty in relocating the site to enlarge after a positive FS. However, a precise surgical specimen orientation and continuous communication between the surgeon and pathologist could overcome this issue [
21]. Finally, the prognostic impact of FS remains debated, with some studies showing worse outcomes for initially positive margins, regardless of whether re-resection guided by FS was performed [
24], while others did not find such an association [
25].