October 2015, Vol. 4, No. 5
Genetic Testing for Melanoma: What’s State of the Art?
Genetic testing in metastatic melanoma should include testing for all actionable mutations in exon 15, said Grant McArthur, MB, BS, PhD, at the 2015 World Cutaneous Malignancies Congress. Testing at the time of disease progression is not yet ready for routine clinical use.
Test the Metastases
In metastatic melanoma, tumor heterogeneity in BRAF mutations requires testing the metastases whenever possible for optimal selection of targeted therapy, advised McArthur, Head, Cancer Therapeutics Program, Peter MacCallum Cancer Centre, Melbourne, Australia. He pointed to a case in which a patient with a primary tumor of the scalp that was wild-type BRAF V600E on Sanger sequencing had a parotid lymph node metastasis that was chimeric wild-type BRAF and BRAF V600E mutation, and a distant metastasis that was pure for BRAF V600E mutation. The patient had a sustained response (>2 years) to vemurafenib.
The Complex Genome of Melanoma
The complexity of melanoma tumors was demonstrated in a series of 34 primary melanomas (Clin Cancer Res. 2013;19:4589-4598). Exome sequencing revealed an average of >1000 single nucleotide variants in coding regions. Thirty-five percent of these primary tumors had a canonical BRAF V600E or V600K mutation, and 35% were NRAS mutant in exon 3. BRAF/NRAS wild-type tumors were more complex than BRAF-mutant tumors, with a higher mutation load.
“Under those common genes we all know about, there are a lot of other sequence variants in so-called melanoma genes,” said McArthur. “How much should we pay notice to that in our clinical practice?”
A landmark analysis from The Cancer Genome Atlas Network shows the emergence of NF1 mutations and a triple wild-type melanoma.
Mutation detection in a complex genome is important for choosing therapy and may also guide intensity of follow-up. Gene knockdown studies show that melanoma tumors are addicted to driver oncogenes such as BRAF and NRAS. “That’s a very useful paradigm when one thinks about why to test…find oncogenic events that the tumor’s addicted to that we can turn off with drugs,” he said. BRAF mutation status is profoundly predictive of response to BRAF inhibitors.
Overall, >50% of tumors have mutations in either BRAF or NRAS. Other newly identified oncogenes are NF1, RAC1, and KIT (rare in cutaneous melanoma at about a 2% prevalence). “Then there are some other recurring mutations like in ERBB4, which is probably not oncogene addiction like you get in BRAF or NRAS, but more research is needed in that space,” he said.
“What’s emerging in clinical trials is the potential for targeting NRAS initially,” said McArthur. Clinical trials in progress are using MEK inhibitors such as binimetinib, which has demonstrated clinical activity in phase 2, to target NRAS mutations.
“My personal guess is that we’re not going to get profound clinical benefit from the phase 3 trial of single-agent binimetinib therapy,” he said. “I think we’re going to need to use MEK inhibitors in combination in NRAS-mutant disease.”
In KIT-mutant tumors, a number of therapies approved for other indications (eg, imatinib) appear to improve overall survival in melanoma.
The response rate to KIT inhibitors in KIT-mutant melanoma is <30%, and the responses that do occur tend not to be durable. “However, if patients have hot spot recurrent mutations in KIT that occur in a number of defined areas in the cytoplasmic kinase domain of KIT, overall survival is better, or if they mutated to wild-type allele ratio, so the copy number of the genes was greater than 1, there was also better overall survival,” said McArthur. “So there may be some patients who do well with KIT inhibitors, but clinically, at the moment, we reserve this for cases that have failed immunotherapies.”
So far, there is little clinical utility to molecular testing to determine prognosis in melanoma, with the exception of monosomy chromosome 3 in uveal melanoma, which is a significant predictor of relapse-free and overall survival.
Molecular chimerism of melanoma tumors has been proposed, but no cases of loss of BRAF mutation in patients whose disease progresses on BRAF inhibitors has been described.
The preferred method for BRAF testing is next-generation sequencing, which also detects NRAS and KIT, in addition to NF1, the targeting of which is being explored in early-phase clinical trials.
Are Less Common Mutations Actionable?
The most common mutation in BRAF by far is the substitution of valine 600 by glutamic acid (V600E), which accounts for >85% of the BRAF mutations in melanoma.
A mutation in V600K is more common in patients who live in regions with high chronic ultraviolet exposure, with mutations in other codons in exon 15, particularly L597 and position 601E. Uncommon mutations are predicted preclinically to respond to MEK inhibitors. “These mutations that activate the [MEK] pathway do respond to the agents that we’re using for the common BRAF mutations,” he said. Not all of the less common mutations respond to BRAF inhibition, such as the I598_599ins mutation.
Testing on Progression?
The Achilles’ heel of targeted therapies is their poor durability of action. Therefore, next-generation sequencing, once the patient’s disease has progressed, and switching to a different therapy based on the result appear attractive. With multiple mechanisms of resistance, however, there are no therapeutic implications at this time to testing at progression.
“It seems to be better to give your best treatments up front and then combine immune and targeted agents either sequentially or simultaneously and not deal with that complex genome as it emerges at resistance,” McArthur said.
Identifying Targets in Anal Squamous Cell Carcinoma:
An Interview with Patrick Boland, MD, of Roswell Park Cancer Institute
At the recent ASCO Annual Meeting, Patrick Boland, MD, of Roswell Park Cancer Institute, submitted an abstract entitled Comprehensive Multiplatform Biomarker Analysis of 212 Anal Squamous Cell Carcinomas highlighting the need to identify novel potential targets and therapeutic options for less common malignancies. In his study, investigators analyzed 212 anal [ Read More ]
Comprehensive genomic profiling of tumors can identify more targeted treatment options than “hots pot” tests and enable innovative trial designs, said Roman Yelensky, PhD, at PMO Live 2015. Unvalidated diagnostic tests with poor accuracy represent a challenge to informed targeted treatment decisions for patients with cancer. Therapies targeting the genomic [ Read More ]