NEW YORK – Numerous actionable genomic mutations detected in melanoma tumors by researchers offer potential personalized treatment opportunities for patients, particularly for those who lack a BRAF V600 mutation or have resistance to immune checkpoint inhibitors.
While BRAF and MEK inhibitors have been shown to significantly prolong progression-free survival and overall survival in late-stage melanoma patients whose tumors bear a BRAF V600 mutation, the remaining 50 percent to 60 percent of patients with BRAF wild-type tumors or another BRAF mutation do not benefit from these approaches.
Additionally, although PD-1- and CTLA-4-targeted immune checkpoint inhibitors have been shown to improve overall survival in both BRAF wild-type and BRAF V600-mutated melanomas, only a minority of patients have durable responses and most eventually develop resistance. Combining immune checkpoint inhibitors with molecularly targeted therapies such as tyrosine kinase inhibitors, RAS/RAF/MAP kinase inhibitors, and PI3K inhibitors could potentially avert resistance if tools were available to tailor these regimens to specific gene variants in patients' cancers.
In a study published in the October issue of the Journal of Molecular Diagnostics, Australian researchers described how they used circulating tumor DNA (ctDNA) to predict outcomes for patients with stage III or IV melanoma, identify targetable tumor-derived mutations that could be used to guide treatment for BRAF-negative melanoma, and track patients' responses to immune checkpoint inhibitors.
Lauren Aoude, a surgical oncologist at the University of Queensland and one of the lead investigators of the study, said the intention of the research was to develop a minimally invasive test to track melanoma patients' disease status over time and guide treatment. In the study, the researchers isolated ctDNA in plasma samples drawn from 106 melanoma patients treated with curative intent and used Roche Diagnostics' 77-gene Avenio expanded sequencing panel to detect gene variants. The patient cohort included 98 patients with BRAF wild-type disease and eight patients with BRAF V600E or BRAF V600K mutations.
They detected 587 ctDNA variants in 90 of the 106 patients, or 85 percent. The samples had between zero and 26 variants each, and there was no association between cancer stage or ctDNA concentration and the number of variants in a sample. "I was surprised at how many actionable mutations we were actually able to see, and that these might create new treatment options for patients in the future," Aoude said.
Four patients who at baseline had a BRAF V600 variant had no detectable variant in their blood sample, indicating that the patients had responded to combination therapy with a BRAF and MEK inhibitor. Three additional samples had BRAF V600-positive tumors that had been missed in routine testing. NRAS or KRAS variants showed up in 22 percent of the patients and were mutually exclusive with BRAF V600 variants, except for one patient who had both.
The study authors noted that patients who were wild type for both BRAF and NRAS were more likely to have low ctDNA variant counts than BRAF V600 mutation- and NRAS mutation-positive patients. Ninety patients had ctDNA gene variants in targetable pathways such as VEGF, EGFR, PI3kinase/AKT, Bcl2/mTOR, ALK/MET, and CDK4/6.
"By looking at the types of variants, we could see that there were mutations that might be targeted by a range of different drugs," Aoude said, adding that this can open up future treatment options for patients.
Sandra Brosda, a postdoctoral fellow at Queensland and another investigator on the study, noted that longitudinal sampling in a subset of 16 patients revealed a close correlation between the emergence of genetic variants and the size of the tumor. These patients had one blood sample drawn at baseline and two during treatment. Nine patients had a third sample taken during treatment. In 9 out of 11 patients who were alive with no residual disease at the time of analysis, there were no detectable ctDNA variants at six and 12 months from the start of therapy.
"You could trace some of the variants over time … [and] if the variant disappeared, the tumor disappeared, as well," Brosda said. "Whereas, if the variant spiked up after a certain time, you could also see it on the scan that the tumor came back."
In the study cohort, patients with a ctDNA concentration of less than 10 ng/mL had improved disease-free survival compared to those with higher ctDNA levels. That difference was particularly striking in patients with stage IV disease. In that group, patients who had ctDNA concentrations of more than 10 ng/mL had significantly worse disease-free survival and progression-free survival at 6.82 months and 4.98 months, respectively, compared to patients with ctDNA less than 10 ng/mL who had a median disease-free survival of 42.85 months and 20.2 months median progression-free survival.
For both stage III and stage IV patients, there was a trend toward improved disease-free survival when no variants were detected, and there was a similar trend for stage III patients with just one or two variants. Generally, lower overall ctDNA levels correlated to a decrease in size of lung and adrenal metastases and lymph nodes. For three patients, an increase in ctDNA levels correlated with the growth of metastatic lesions.
One of the limitations of the study, according to the authors, is that the Avenio panel does not include the NF1 gene, which is mutated in 45 percent to 93 percent of cutaneous melanoma, meaning that some of the 16 patients with no detectable variants in their plasma samples could have had an NF1 mutation.
Senior author Vanessa Bonazzi, a research fellow at Queensland, said in a statement that a blood test based on ctDNA would allow doctors to monitor patients' responses to treatments and track their disease progression more closely than standard PET/CT scans, which patients typically undergo every six months. "If, for example, the patient's ctDNA profile shows a PIK3CA mutation from baseline, you could follow it in different blood samples from this patient over time, and down the track, you could use a PIK3CA inhibitor and hopefully be able to prevent potential recurrence of disease," she noted.
This type of test would be especially valuable for monitoring disease progression in patients who travel long distances to receive cancer care, Aoude pointed out. "A blood test can often be performed at a health clinic" that's closer, she said, saving the patient a lengthier trek to a hospital for a PET scan.
However, translating this work to a test for patients will require larger studies. Aoude hopes to see this type of ctDNA analysis included in an upcoming clinical trial so that eventually, with more robust data, such testing can be integrated into patient care outside of research.