Up Close and Personal: The Challenges of Precision Medicine in Melanoma

The review by Griewank and colleagues (1) in this issue of the Journal gives a timely overview of the recent developments in genomics that have led to major advances in targeted therapy for melanoma. The authors describe how genomic analysis, including deep sequencing and comparative genomic hybridization, has led to the identification of many of the genetic alterations that drive melanoma initiation and progression. Although many genomic platforms have contributed to our understanding of melanoma biology, some have proven more useful than others. Techniques such as comparative genomic hybridi-zation and fluorescence in situ hybridization have proven utility in differentiating suspicious skin lesions such as blue nevi or spitzoid lesions from invasive melanomas (2) but little impact in the diagnosis of advanced melanoma. In their central argument, the authors surmise that most of the known and important genetic variants in melanoma will have been discovered within the next few years. Although this is likely to be true, a number of major obstacles remain in the path to improved targeted therapy for melanoma patients. Here we describe what we perceive to be the major impediments to improved targeted therapies for melanoma and propose strategies to overcome them. Massively parallel sequencing has given us the tools to genetically characterize hundreds if not thousands of melanomas rapidly and cheaply, yet important information on kinetic changes as tumors progress over time from in situ melanoma to primary to nodal metas-tasis to metastasis to subsequent metastasis are lacking (3,4). These sorts of studies would allow a filtering of the data on the hundreds of single nucleotide variants found in melanoma, many of which are ultraviolet damage–related C-T transitions and are not drivers of malignancy (3,4). A more pressing issue is determining how the many thousands of genetic changes found in a typical cutaneous melanoma interact to drive melanoma development. This will be especially important in melanomas that are BRAF/NRAS wild-type because sequencing studies to date have failed to identify any major oncogenic drivers in this subgroup (3,4). It seems likely that many of the genetic changes identified so far occur in common, overlapping pathways—perhaps defining a core set of processes that drive the oncogenic transformation of melanocytes. How these mutations interact and cooperate to drive the key signaling pathways required for melanomagenesis remains a major gap in our knowledge. In particular , it seems likely that the high mutational rate seen in a typical ultraviolet-induced melanoma …