The article from Shain et al, published in the New England Journal of Medicine in November 2015, presents new data about the genetic evolution of melanoma from precursor lesions, showing different evolutionary trajectories for different melanoma subtypes.
It is known that there are distinct pre-malignant lesions that evolve to melanoma. The exact mechanism and sequence in which this occurs is not well known, and the genetic characterisation would be an important asset with clinical and probably therapeutic implications.
In this paper, the authors try to uncover and understand the nature of genetic evolution of melanoma, going from the precursor to the invasive lesions.
With this in mind, authors defined an intermediate category of melanocytic neoplasia. These intermediate lesions are characterised by the presence of more than one pathogenic genetic alteration. The authors also tried to clarify the “dysplastic nevus” concept, which still remains controversial as an autonomous entity.
In this study, a total of 37 formalin-fixed, paraffin-embedded melanocytic neoplasms, enriched for melanomas with histologically distinct precursors, were retrieved from three different centres. A total of 150 spatially distinct areas of the samples were subjected to next generation sequencing and targeted sequencing with a panel of 293 cancer-related genes.
These samples were analysed by eight dermatopathologists that included the lesion in one of the following four categories: benign, intermediate but probably benign, intermediate but probably malignant and melanoma.
According to the authors, a high rate of inter-observer agreement was shown with a good definition of malignant and benign lesions. However, in a group of the analysed lesions, even when the morphologic criteria overlapped, there was no agreement between the pathologists. This is one of the limitations of histopathology analyses that could be overcome by genetic profile.
Interestingly, melanomas with BRAF V600E mutations were shown to arise from benign nevi and the melanomas with NRAS, BRAF V600K or K601E mutations were, unexpectedly, associated with intermediate lesions or melanoma in situ. These findings show that lesions classified as intermediate by histopathological characteristics, also had genetic features between benign and malignant lesions, which could justify the difficulty of its morphologic characterisation.
Melanomas with BRAF V600E mutations were considered to be common in the sun-damaged skin of young patients, whereas melanomas with NRAS, BRAF V600K or K601E mutations were observed in chronically sun-damaged skin of older patients. The biologic differences between benign and intermediate precursors, most likely reflect the different pathway between lesions that arise from skin chronically sun damaged and skin with no sun-damage.
Authors’ findings suggest that all precursor lesions were initiated by mutations of genes that are known to activate the MAPK pathway, followed by activation of telomerase and disruption of the G1-S checkpoint.
In every analysed case, a mutation known to activate the MAPK pathway was found to be the initiated oncogene. The descendent neoplasms had other mutations that were not present in the precursor lesion and that probably occurred later in the evolution, contributing to progression.
Somatic mutations in melanoma oncogenes (BRAF, NRAS, GNAQ and GNA11) were already present in benign nevi, suggesting that they arise early during progression. Later-stage melanoma showed other mutations such as TERT, CDKN2A, TP53, genes encoding SWI/SNF subunits and PTEN among others.
Finally, the authors proposed a progression model of melanomas in chronically sun-damaged skin.
In this model, point mutations and copy number alterations that increase in number and time function are related to UV radiation in the entire spectrum from benign lesions to invasive melanoma. Driver genes -BRAF, NRAS, TERT - are present in benign and intermediate lesions, as well as in melanoma in situ. Other mutations/alterations in driver genes such us CDKN2A and genes encoding for SWI/SNF subunits and TP53 and PTEN only appear in invasive melanomas. The exact sequence, in which these mutations appear, could allow us to better classify melanoma lesions and gives us more accurate information on the natural history of the disease. These genetic alterations are also important for defining targetable alterations/”drugable” mutations.
Although this study shows that UV radiation is a major mutagen factor, implied in both initiation and progression of melanoma, this is still not universally accepted. However, the presence of UV induced mutational signature of several driver mutations supports this theory. Therefore, sun protection should reduce melanoma risk, especially among people with a high nevus count.
Melanocytic lesions transitioned from linear to branched evolution at later stages of progression, showing tumour heterogeneity. Polyclonal lesions are expected to be more resilient to the immune system and therapeutically approaches, which may explain the therapeutically success of surgical removal.
- Is it possible to define a genetic signature for each melanoma stage?
- Could these genetic alterations be “drugable” in the future?
- Since it seems that polyclonal lesions can more effectively evade immune system, should all melanocytic lesions be surgically removed earlier or tighter surveillance should be recommend? What should be the criteria for early surgical removal?
- Is UV radiation really a modifiable risk factor in the development of cutaneous melanoma?
- Does the use of sunscreen confer the same of protection to all (?) or, since people with a high nevus count, which have higher risk, would also beneficiate the most?
Teresa Amaral and Ioanna Tampouri declare no conflict of interests.
The content of this article reflects the personal opinions of the authors and is not necessarily the official position of the European Society for Medical Oncology.