Surgical Margins for Lentigo Maligna, 2004

Bub and her colleagues¹ make an important contribution to our fund of knowledge with this report on surgical excision of lentigo maligna and lentigo maligna melanoma using staged excision with rapid permanent sections. The value of this study is that it provides scientific evidence to help clinicians address the current clinical problems of surgical margins for head and neck melanoma.

There are many problems confronting the surgeon planning treatment of lentigo maligna and lentigo maligna melanoma of the head and neck. First, it is more difficult for the surgeon to establish clinical margins in these areas compared with melanoma of the trunk or proximal extremities. Age- and sun-related skin changes such as solar lentigines, actinic keratoses, and seborrheic keratoses camouflage tumor-specific skin changes and make it more difficult for the clinician to outline the tumor margin.

Second, it is difficult to excise 1- or 2-cm margins for invasive melanoma without compromising cosmetic or functional results. Yet these are the current recommendations in the literature for invasive melanoma and are deemed to be the standard of care by some. In truth, the recommended margin of 1 to 2 cm is based, for the most part, on data obtained from excision of melanoma on the trunk and proximal extremities. These reports have always concluded that the smallest of the studied surgical margins was as effective as the widest margin and does not influence overall survival.^{2 - 3} As a result, most surgeons recommend a surgical margin on the head and neck "as wide as possible" up to the recommended margin. The issue of the proper surgical margin is still vague.

Third, we know that very narrow surgical margins on the head and neck (as well as the hands, feet, and genitalia) result in high rates of inadequate excision and true local recurrence. Published results of treatment of head and neck melanoma including melanoma in situ show recurrence rates up to and exceeding 20%. This is significantly greater than the low rates of less than 3% on the trunk and proximal extremities where wide margins can be taken. This higher rate of failure in the head and neck is due to the problem of poorly defined margins as well as narrow surgical margins and inadequate excision.

What margin is necessary? While we know that the width of the surgical margin does not influence survival, excisions with wide margins effectively remove the entire tumor, while narrow margins may result in inadequate excision of the melanoma and local recurrence. More than 10 years ago the National Institutes of Health (NIH) Consensus Panel recommended a 5-mm margin for melanoma in situ or lentigo maligna and 1-cm margins for melanomas less than 1 mm in thickness.⁴ However, the melanoma in situ recommendation was written as a "consensus only" based on history and practice patterns of the Panel members. More recently, the American Academy of Dermatology Guidelines of Care stated that the goal of melanoma surgery was "to assure complete excision confirmed by histologically negative margins."⁵ No definite measured margin was given for melanoma in situ other than the goal of clear histologic margins.

Bub et al¹ have provided us with data and guidelines for excision of melanoma and melanoma in situ of the head and neck. Their data confirm the majority of other previously published reports as well. In this report, they excised melanomas (92% on the head and neck) and then studied the surgical margins by sampling approximately each millimeter of the tissue margin with rapid (24-hour) paraffin-embedded permanent sections. A pathologist interpreted the slides and a dermatologic surgeon performed the surgery needed to obtain negative histologic margins.

The results of this study tell a story. First, 5-mm margins are not adequate for melanoma in situ or invasive melanoma despite the outdated NIH Consensus Panel report of 1992. A 6-mm margin would have resulted in inadequate excision in 15% of their patients. This can likely be interpreted to mean that a 6-mm surgical margin might lead to a 15% recurrence rate. This conclusion is reasonable because 5 or 6 mm is a common margin on the head and neck to minimize cosmetic and functional deformity; and it is reasonable because the recurrence rate with common margins on the head and neck approximate 15%. Certainly a 15% recurrence rate is less than acceptable and we should strive for better.

What should our goal be? The real goal is zero recurrence from inadequate surgery, but realistically that goal is not easily achieved. With surgeons doing their best to visually establish tumor margins and then excise as much as possible without doing too much harm, the historical result is a 15% to 20% recurrence rate. But when surgeons work with pathologists to more carefully examine the margin, then recurrence rates go down. It is clear that the more closely we examine the surgical margin, the less likely we will leave melanoma behind in the surgical wound. Visual examination of the margin with margins as wide as convenient, along with standard histologic sampling (breadloafing) examining less than 0.1% of the margin, is the historical standard that leads to a 15% to 20% recurrence rate. Many other methods for more complete margin examination have been described. Examining a greater percentage of the margin will lead to fewer false-negative examination reports and fewer recurrences. Other methods to examine surgical margins include "enface," permanent sections using rush permanents, the square technique using strips of skin surrounding the melanoma and examining the margins with rapid permanent sections, enface frozen sections, or even traditional Mohs surgery. These techniques examine closer to 100% of the surgical margin and offer very low recurrence rates of 0.5% to 5%.

This study used radial rather than enface sectioning and permanent sections rather than frozen sections, and examined or sampled each millimeter of the margin (still, <1% of the true margin). The findings confirm most other reported data when surgical margins have been examined more closely, and this study showed that margins of almost 1 cm (6-9 mm) are necessary to completely excise melanomas that are less than 2 cm in diameter (3 cm²). For melanomas greater than 2 cm in diameter (3 cm²), a margin of 8 to 12 mm was necessary to excise melanomas completely to achieve at least a 95% success rate. To translate these data to the width of a surgical margin, the surgeon must take at least 8 to 12 mm of normal-looking skin if the method for examining the margin is traditional histologic examination with permanent sections. If one of the more complete techniques for examining the margin is used, then the surgeon can begin with more narrow margins to excise tumor-positive skin where necessary.

Finally, this study touches on the last but not least important current clinical issue in the treatment of melanoma. That issue is the definition of a positive surgical margin. There is no general agreement among dermatopathologists about how much atypia or how much hyperplasia constitutes a positive margin. Some pathologists would suggest that single-cell atypia alone is enough to warrant further excision. Others might require confluent hyperplasia or pagetoid spread yet others may require nesting. These authors used a definition that required solitary units or nests of atypical melanocytes with extension of atypical melanocytes along adnexal structures. This definition may be more stringent than other definitions. How do we decide what definition is correct? How do we know what type or extent of melanocytic atypia or proliferation is biologically important? The ultimate answer to the question of what needs to be excised—the gold standard—is determined by the recurrence rate after surgery using each definition of a positive margin. Bub et al report a recurrence rate of 5% with their definition, which suggests that their definition may be reasonable. Other definitions such as confluent hyperplasia and nesting, visible by high-quality frozen section with Mohs surgery, has been reported to result in a 0.5% recurrence rate at 5 years and also suggests that excising nonconfluent hyperplasia or single isolated atypical melanocytes may be unnecessary.⁶ An analogy in the treatment of squamous cell carcinoma is that isolated atypical keratinocytes or minor actinic keratoses at the margin need not be excised to achieve a high cure rate for squamous cell carcinoma.

Altogether, this report should help clinicians by:

Bub and colleagues are to be congratulated for providing additional scientific evidence to help us fill the voids in the current clinical problems of managing melanomas of the head and neck.