br The most important objective of the different pro tocols
The most important objective of the different pro-tocols has been to achieve maximum likelihood of detecting metastases. Extending the distance between
steps in the sectioning protocol was undertaken so that the previously neglected part of broad lymph nodes was then assessed. In a preliminary assessment, steps were increased up to 300 mm, but an increased detection rate was not seen. This supports Cochran’s original PD-98059 that metastases are concentrated in the cen-tral plane through the hilum and the longest dimension of the node . However, wider steps could still be appropriate when the dissection of the lymph node has resulted in an asymmetric bisection of the node. In deciding that the procedure should identify metastases down to 0.1 mm, we opted for a minimum thickness of step of 0.05 mm (50 mm).
There is insufficient evidence to suggest that the EORTC protocol for the pathological evaluation of SLNs for melanoma should be fundamentally changed at this stage. It was decided to focus on the clarification and simplification of the existing procedures with the objective to reduce the technical workload thereby making it more sustainable in a routine diagnostic setting. We also decided to optimise both conventional sectioning and staining procedures including immuno-histochemistry and provide guidance on the description of the spatial localisation of melanoma deposits in a SLN.
The following changes to the existing protocol are proposed:
2. Sectioning protocol
The SLN is bivalved, so revealing the largest surface area of the lymph node in two pieces for sectioning. The maximum dimension is taken from the cut surface. Both pieces are placed face down in one cassette, except when the lymph nodes are too large, in which case each piece is placed in a separate cassette.
If in error the bivalving of a more rounded node produces markedly unequal pieces, the largest of them can be further sliced or bivalved producing three pieces in total. Alternatively the sectioning steps can be extended up to 400 mm. The sectioning protocol is determined at the grossing stage by the pathologist. This in turn is based on an estimate of the length of the short axis of the node. The estimation is determined by palpation of the node between finger and thumb with
adjacent ruler. This enables an approximation of this dimension which can be made virtually instantaneously. The steps between sections is determined by the formula given in Fig. 1 and is written on the cassette to convey to the sectioning technician. This incremental increase in steps is intended to ensure that a larger proportion of the SLN is assessed microscopically.
(a) In the sectioning protocol, two sections are cut at each step, except for step 2 where three extra sections are taken and retained unstained.
(b) At each level, a section is stained for S100 protein.
(d) One spare section is taken at all levels, except for step 2 where three spare sections are taken.
Figs. 1 and 2 illustrate the sectioning and staining protocol as above described.
In comparison with the previous protocol , the net result of these changes is a reduction of the total number of sections, a decrease of staining by five H&Es and two
Fig. 1. Revised EORTC protocol depicting sectioning of sentinel lymph node in melanoma. EORTC, European Organisation for Research and Treatment of Cancer.
Fig. 2. Revised EORTC protocol for staining sentinel lymph node in melanoma. EORTC, European Organisation for Research and Treatment of Cancer.
immunostains. The thickness of each section should be 3 mm. The unstained sections are retained at each level so that they can be used for further H&E or immunos-tains if needed.
3. Immunohistochemical staining
We have confidence in S100 protein staining for efficient recognition of melanoma metastases. Diaminobenzidine is the preferred chromogen because the red chromogen aminoethylcarbazole is not always well-localised.
S100 protein has a high sensitivity for melanoma detection because it is expressed in almost all primary and metastatic melanomas [21,22] but with a relatively low melanoma specificity because S100 protein expres-sion can also be observed in nodal naevus cells, dendritic reticular cells, Schwann cells and adipocytes. This can possibly result in a confusing background staining. However, the morphological features of the other cells potentially staining with S100 in lymph nodes are so distinctive the problem is largely theoretical. Short
experience is sufficient to produce high confidence and accuracy.
HMB-45 is a relatively specific marker for melano-cytes but lacks sensitivity because approximately 70e76% of melanoma are HMB-45 positive [21,22]. This stain might result in missed metastases.
Although more sensitive than HMB-45, Melan A is considered not sufficiently specific  and not quite as sensitive as S100 protein because it is not positive in all melanomas. It can also stain macrophages/melanoph-ages, whereas S100 protein and HMB45 do not as a rule.