BRAFV600-mutant ctDNA as a Prognostic Biomarker in Resected Stage III Melanoma

1. ctDNA detection was found to have a relapse HR of 2.91 in the placebo and HR 2.98 in the treatment arm.

2. At 3 months, ctDNA detection had a sensitivity of 59%, a specificity of 100%, a positive predictive value of 100%, and a negative predictive value of 71%.

Evidence Rating Level: 1 (Excellent)

Study Rundown: There is a lack of reliable biomarkers to differentiate high-risk from low-risk patients with melanoma after surgical resection of stage II–IV disease, which creates uncertainty in treatment decisions and leads to both overtreatment and undertreatment. The detection of circulating tumour DNA (ctDNA) offers a promising method to monitor minimal residual disease (MRD) and personalize therapy. This study further investigates this in a randomized phase III trial with post-surgical treatment with dabrafenib plus trametinib vs placebo. The primary endpoint was recurrence-free survival (RFS), and secondary endpoints included overall survival (OS), as well as biomarker analysis (ctDNA, interferon gamma [IFNG] gene expression and tumour mutational burden) at baseline and recurrence. It was found that ctDNA detection in the placebo group correlated with RFS at 3.71 months vs 24.41 months with HR 2.91 (significant). In the treatment arm, this was seen at 16.59 months vs 68.11 months with HR 2.98 (significant). ctDNA detection was also associated with OS (placebo group HR 3.35 [significant] and treatment group HR 4.27 [significant]. For predictive ability, in placebo, baseline ctDNA detection was found to have an RFS HR 3.55 (significant) compared to IFNG (less than the median) with HR 1.49 (significant), however both had similar HR in the treatment group with ctDNA HR 2.47 (significant) vs IFNG HR 2.04 (significant). When comparing the predictive ability of baseline ctDNA vs tumour mutational burden, in placebo, baseline ctDNA had an HR 2.63 (significant) vs HR 1.67 (significant) with tumour mutational burden, and in treatment, this was found to be HR 2.50 (significant) vs 1.34 (non-significant), respectively. Yielding a positive ctDNA assay at recurrence had no association with having more metastatic sites (non-significant). It was found that the sensitivity of ctDNA to detect recurrence at 3 months was 59%, the specificity was 100%, the positive predictive value was 100%, and the negative predictive value was 71%. Patients with molecular relapse or persistently positive ctDNA have a worse RFS (8.31 months and 5.32 months, respectively) compared with patients who had undetectable ctDNA after a positive baseline result or a durable undetectable result (19.25 months and NA months, significant). The strength of this study included its methodology, and the limitations included the sample size for biomarker analysis. Overall, this study found some potential value of measuring ctDNA in adjuvant therapy in patients with BRAF V600-positive stage III melanoma.

Click to read the study in Lancet Oncology

Click to read an accompanying editorial in Lancet Oncology

Relevant Reading: Prediction and monitoring of relapse in stage III melanoma using circulating tumor DNA

In-Depth [randomized controlled trial]: This international, double-blinded, phase III trial enrolled adults who were post-resection of BRAFV600-mutant stage III melanoma and randomized them (1:1) to dabrafenib plus trametinib (n=438) or placebo (n=432). ctDNA was obtained at baseline, then at 3-month intervals until 1 year, and again at recurrence. The median follow-up time for the biomarker analysis was 60 months for the therapy group and 58 months for the placebo group. The baseline biomarker dataset comprised 69% of the total population. At baseline analysis, 92% of patients had BRAFV600E mutations and 8% had BRAFV600K mutations, and 13% of pre-treatment samples had BRAF V600-mutant ctDNA. It was found that ctDNA detection in the placebo group correlated with RFS at 3.71 months (95% CI, 2.39-6.89) vs 24.41 months (17.28-43.13) with HR 2.91 (95% CI, 1.99-4.25, p<0.0001). In the treatment arm, this was seen at 16.59 months (95% CI, 12.02-26.80) vs 68.11 months (50.36-NR) with HR 2.98 (1.95-4.54, p<0.0001). ctDNA detection was also associated with OS (placebo group HR 3.35 [2.01-5.55, p<0.0001] and treatment group HR 4.27 [2.50-7.27, p<0.0001]. For predictive ability, in placebo, baseline ctDNA detection was found to have an RFS HR 3.55 (95%CI, 2.24-5.63) compared to IFNG (less than the median) with HR 1.49 (95%CI, 1.01-2.17), however both had similar HR in the treatment group with ctDNA HR 2.47 (95%CI, 1.41-4.33) vs IFNG HR 2.04 (95%CI, 1.33-3.22). When comparing the predictive ability of baseline ctDNA vs tumor mutational burden, in placebo baseline ctDNA had an HR 2.63 (95%CI, 1.53-4.55) vs HR 1.67 (95%CI, 1.05-2.63) with tumor mutational burden, and in treatment this was found to be HR 2.50 (1.43-4.34) vs 1.34 (0.86-2.08), respectively. Yielding a positive ctDNA assay at recurrence had no association with having more metastatic sites (p=0.40). It was found that the sensitivity of ctDNA to detect recurrence at 3 months was 59%, the specificity was 100%, the positive predictive value was 100%, and the negative predictive value was 71%. Patients with molecular relapse or persistently positive ctDNA have a worse RFS (8.31 months and 5.32 months, respectively) compared with patients who had undetectable ctDNA after a positive baseline result or a durable undetectable result (19.25 months and NA months, p<0.0001). Overall, this study found some potential value of measuring ctDNA in adjuvant therapy in patients with BRAF V600-positive stage III melanoma.

Image: PD

©2025 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.