Favoritism or impartiality? Do the four genomic subtypes of melanoma have a bias toward certain mutated genes and gene pathways, or do they welcome all mutations equally?
Answering that question has been especially difficult because of cutaneous melanoma’s high mutation rate — the profusion of misspelled, severed, out-of-place, missing-in-action, or overabundant genes found in melanoma skin tumors. This roar of genetic error makes it particularly difficult to discern the whisper of gene mutations that actually drive the disease. Detecting them would require canvasing all the molecular changes in the disease — an undertaking that would involve a deep analysis of tumor tissue samples from more than 1,000 patients. Until now, no analysis has involved more than a few hundred tissues.
For a new study in Nature Genetics, Dana-Farber researchers assembled the largest-to-date molecular dataset on melanoma and used it to uncover new details about the genomic subtypes of the disease. They found that each subtype has a “preference” for certain mutated genes and pathways, and that some of these alterations may make the tumors more susceptible to immunotherapy within the context of their respective subtype. The findings open the way toward even deeper dives into the molecular roots of melanoma and, potentially, to new therapies for the disease.
“Discoveries about the genomic landscape of melanoma have led to the development of effective targeted drugs for the disease and to strategies for determining which patients are most likely to benefit from some forms of immunotherapy,” says the study’s senior author, Eliezer Van Allen, MD, of Dana-Farber. “Still, only a subset of patients derive lasting benefit from these agents. To identify new targets for therapies, we need a more comprehensive picture of the genomic alterations that underlie the disease.”
A challenging prospect
The genomic disarray in melanoma reflects the cause of the disease: damage from ultraviolet light wreaks havoc on the genetic material in melanocytes, skin cells responsible involved in pigment production.
“Our initial goal in this study was to find all the genetic alterations driving melanoma,” says study lead author Jake Conway, a graduate student in Van Allen’s lab. “It’s challenging because of the high background mutation rate in melanoma and because existing statistical approaches to identifying driver mutations don’t work well in those conditions.”
For the new study, researchers analyzed the genomes of the largest number of melanomas — 1,048 — yet examined in a single study. To overcome the problem posed by melanoma’s massively mutated genome, they took a “consensus-based approach,” using multiple analytical techniques, such as focusing on “hotspot” regions of the genome where driver mutations tend to congregate, probing the functions of key genes, and developing a new statistical framework that can handle the spectrum of genomic alterations found in melanoma.
Previous molecular studies have enabled researchers to identify four genomic subtypes of cutaneous melanoma, named for the genes that are mutated, or not mutated, within them. Three — BRAF, (N)RAS, and NF1 — are named for mutated genes in the MAP kinase signaling pathway, which carries signals from outside the cell to DNA in the nucleus. The fourth, triple wild-type (or TWT), is named for the absence of those mutations. (On a genomic level, TWT is markedly different from the other three, with a pattern of mutations not associated with damage from ultraviolet light.) The subtypes have different clinical features and tend to respond better to some therapies than others.
“Our question was whether the subtypes have the same underlying driver mutations, or whether each subtype has a distinctive set of such mutations and, if so, what are the therapeutic implications of these differences?” Conway remarks.
Their analysis showed that the subtypes do favor certain gene alterations:
- The BRAF and (N)RAS subtypes often harbor sets of mutated genes that make the tumors especially vulnerable to immunotherapies known as checkpoint inhibitors, which unleash a potent immune system attack on cancer.
- In the BRAF subtype, the preferential alteration of TGF-β pathway, and of the genes MECOM and BMP5 in particular, increases tumors’ susceptibility to checkpoint inhibitors.
- Mutations in the genes for the SWI/SNF protein complex were a frequent feature of the (N)RAS subtype and are a sign that the tumor will respond well to immunotherapy.
- 15-20% of TWT tumors have a pattern of genetic alterations known as Mutational Signature 3, an idiosyncrasy found in only 1% of other melanomas. The signature indicates a reduced ability to repair certain kinds of DNA damage. Tumors with this kind of deficiency are often treated with platinum-based chemotherapy.
The findings are an example of the kind of insights that can be gleaned from a deep analysis of large numbers of tissue samples, the study leaders say. In addition to identifying potential new targets for therapy, such research can illuminate the complex biology at work within specific cancer types.