Only in the last 2 decades or so have sufficient data become available to statistically associate high mitotic rate with survival in melanomas.
In 1953, Allen and Spitz published data showing the negative impact of high mitotic rate on survival in patients with primary melanoma.1,2Only in the last 2 decades or so have sufficient data become available to statistically associate high mitotic rate with survival in melanomas.
In 2011, Thompson and colleagues published data from the American Joint Committee on Cancer melanoma staging database to show the independent prognostic relevance of primary tumor mitotic rate compared with other features of stages 1 and 2 melanoma. Data from more than 13,000 patients were analyzed, and they found mitotic rate to be second only to tumor thickness as an independent predictor of survival.2
Research into the molecular characteristics of melanoma has been ongoing. While time to diagnosis may be a reason for poor outcomes with melanoma, evidence is emerging that there is an aggressive subtype.3
Jönsson et al in 2010 analyzed global gene expression in stage 4 melanoma from 57 patients. They identified 4 subtypes on the basis of characteristic gene signatures, namely, high-immune response, proliferative, pigmentation, and normal-like. The proliferative subtype had a low expression of immune response genes but an elevated expression of genes associated with the cell cycle, for example, E2F1, BUB1, and CCNA2.BRAF/NRASmutations were significantly more common in the proliferative subtype tumors (P<.01) where no double wild-type were found but with a high frequency of CDKAZA homozygous deletions. This subtype was associated with significantly poorer survival versus the other subtypes (P = .003).4Harbst et al later confirmed that these 4 subtypes could also be identified in primary melanoma. In their analysis they demonstrated high-grade and low-grade forms of primary melanoma. The high-grade melanomas were thicker, had mitotic rates ≥6 /mm2, and were more often nodular in form.3
Given the advances in microscopic and molecular techniques, the obvious problem confronting clinicians is the decision on whether to biopsy. After all, mitotic rates and gene signatures can only be investigated following a biopsy.5More information is needed about the appearance of high-grade primary melanomas, where they are most commonly found, and in whom.
An analysis recently published in the Journal of the American Medical Association, by Shen and colleagues, offers some answers to these questions. They performed a cross-sectional analysis of 1441 patients with 1500 primary invasive melanomas at a single hospital in Australia.6The aim was to assess the clinicopathologic associations of high mitotic rate to enable clinicians to earlier identify aggressive primary melanomas.
Patients were grouped according to a range of mitotic rates (mitoses per mm2), and researchers noted corresponding demographics, phenotypic markers, tumor presentation, and histopathologic characteristics.
Univariate analysis detected a strong association between older age, male sex, and increasing mitotic rate. In the highest mitotic rate group (≥10/mm2), 45% of cases were ≥70 years of age. Furthermore, univariate analysis found that high mitotic rate melanomas were significantly more likely to be located on the head and neck (P= .005), amelanotic (P<.001), and have an increasing growth rate (P<.001). Amelanosis was found in 50% of melanomas with a mitotic rate ≥5/mm2. The authors concluded that the association with growth rate possibly explained why high mitotic rate melanomas were significantly (P<.001) more likely to be detected by the patient than by a physician or others. Indeed, that particular association was also found in multivariate analysis (P= .007).
Amelanosis and growth rate continued to be significantly associated with high mitotic rate in the multivariate analysis (P<.001). High mitotic rate was associated with thicker melanomas, nodular melanomas, and more than half of melanomas with mitotic rates ≥10/mm2were ulcerated. Most (73%) of low mitotic rate melanomas were of the superficial spreading type.
The overall multivariate analysis showed statistically significant positive associations of nodular tumor type, ulceration, Clark level 5 invasion, and thickness >4 mm (allP<.001) with high mitotic rate. In summary, Shen concluded that high mitotic rate primary cutaneous melanoma is more likely to be encountered on the head or neck of elderly males with a history of excessive sun exposure and have an aggressive histologic appearance.
Writing in an accompanying editorial, Samuel J Balin, MD, at the David Geffen School of Medicine at the University of California, Los Angeles, and Raymond L. Barnhill, MD, at the division of dermatopathology, department of pathology and laboratory medicine, University of California Los Angeles Medical Center, Los Angeles, stated that the study provides another important source of data for clinicians to use when deciding whether or not to take a biopsy. They emphasized that an important discovery emerging from the analysis was that a history of blistering sunburns, a family history of melanoma, and existence of dysplastic nevi in the melanoma were associated with low mitotic rates.
While one interpretation may be that melanomas developing from preexisting nevi are not as aggressive as those of de novo origin, another explanation may underline the importance and consequences of regular screening. Patients with those characteristics may have more frequent checkups, making it more likely that melanomas will be found and removed before more serious disease develops.5
Research into the molecular architecture of melanomas is contributing greatly to an understanding of their biology, identifying aggressive and less aggressive forms, and leading to the development of new treatment strategies. The approach of Shen and colleagues will contribute to the body of knowledge essential to promote earlier diagnosis of high mitotic rate melanomas with poor prognosis.