Merkel cell carcinoma: if no breslow, then what?

Merkel cell carcinoma (MCC) is a skin tumor of neuroendocrine origin with a mortality of 33%, the highest of any cutaneous malignancy. Within the United States the estimated annual incidence is increasing and is estimated at 0.44 per 100,000 persons, resulting in roughly 1,000 new cases per year [1]. Several factors may be contributing to this trend, including changes in the major risk factors for this disease: the increasing age of the general population, higher rates of immunosuppression, and increasing sun exposure. Additional likely contributors to the trend include improved detection and reporting. While reported rates have tripled between 1986 and 2001, MCC remains approximately 50-fold less common than melanoma [1]. This rarity has hampered our ability to study this disease systematically. To date there are no controlled trials to assess treatment and outcome. Further obscuring matters is the lack of uniform agreement on staging and treatment. While its high mortality and increasing incidence have brought greater attention to MCC recently, our ability to accurately predict prognosis based on histology remains limited. Evidence to date supports sentinel lymph node biopsy (SLNB) as the best prognostic indicator. Allen et al. [2] found a significant 5-year disease-specific survival difference between patients clinically staged as node-negative (75% survival) and those pathologically node-negative (97%) by SLNB (P1⁄4 0.009). Further supporting the role of SLNB, Gupta et al. [3] found the 3-year recurrence rate three times higher in patients with a positive SLNB (60%) as compared to those with a negative biopsy (20%) (P1⁄4 0.03). While it is thus clear that SLNB provides important prognostic information, it is certainly not an ideal indicator. In the Allen series 3% of those with a negative biopsy went on to die of MCC and in the Gupta series 20% had recurrence within 3 years. Additionally, SLNB is not performed on all patients. Indeed, technical difficulties may arise that can thwart a successful node biopsy. This is particularly a problem in the head and neck region (the most common site for MCC) where the facial nerve and parotid anatomy often complicate the procedure. Furthermore, for patients in whom a wide surgical excision has been performed, the ability to identify the true sentinel lymph node is often compromised by disruption of the relevant draining lymphatics. Given the above, it would be very helpful to have independent pathologic predictors. MCC has been likened to melanoma on several fronts: both are cutaneous neoplasms with aggressive biologic behavior and are associated with sun exposure and advancing patient age. The role of Breslow depth as a prognostic indicator for melanoma is well established. However, for MCC a reproducible histologic prognostic indicator has not emerged in the literature. In this issue of the Journal of Surgical Oncology, Goldberg et al. evaluated the role of histologic depth as a potential prognostic indicator of disease-free and overall survival for patients with MCC. Paraffin-embedded sections were evaluated for tumor thickness and the charts retrospectively reviewed for pertinent clinical data on 60 MCC patients. No correlation was found between tumor thickness and disease-free or overall survival in patients lacking nodal involvement, arguing that ‘‘Breslow depth’’ will not become a useful independent predictor for MCC as it has for melanoma. Numerous parameters have thus far been evaluated for MCC in the search for a histologic prognostic indicator. Overall tumor size (clinical diameter) was reported by several authors as having significant prognostic capabilities; however, larger studies have failed to find this association [2,3]. Growth patterns, the associated lymphocytic inflammation, cell size, mitotic activity, necrosis, perineural invasion, and a wide array of