Clinical Implications of Cell-Free DNA in Managing BRAF V600E Mutation-Positive Colorectal Cancer

Abstract

Background/Objectives: BRAF^V600E-mutant colorectal cancer (CRC) is associated with poor prognosis, and despite the introduction of BEACON therapy, significant treatment challenges remain. This study investigates the clinical utility of BRAF^V600E in cell-free DNA (cfDNA BRAF^V600E) as a biomarker for real-time treatment monitoring in metastatic cases and for evaluating minimal residual disease (MRD) after curative resection. Methods: This single-center, prospective observational study included 37 patients with BRAF^V600E-mutant CRC treated at Nippon Medical School Hospital between April 2017 and June 2024. Patients were divided into two cohorts: Cohort 1 (Stage IV cases): Evaluated cfDNA BRAF^V600E for treatment monitoring. Cohort 2 (Stage I–III curatively resected cases): Assessed cfDNA BRAF^V600E for recurrence risk prediction. Blood samples were collected before and during treatment and analyzed using droplet digital PCR (ddPCR) to measure cfDNA BRAF^V600E levels. Results: Cohort 1 (Stage IV, n = 14): Pre-treatment cfDNA BRAF^V600E was detected in 93% of cases. Patients with a decrease in cfDNA BRAF^V600E variant allele frequency (VAF) after chemotherapy had significantly longer overall survival (511 vs. 189 days, p = 0.03) than those without a decrease. Cohort 2 (curatively resected, n = 23): cfDNA BRAF^V600E was detected in 4/23 patients (17.4%) at 1 month post-surgery. cfDNA BRAF^V600E showed better recurrence prediction compared to CEA (100% vs. 18.8%, p = 0.004). Among the seven patients who experienced recurrence, those with postoperative cfDNA BRAF^V600E positivity had significantly shorter disease-free survival compared to cfDNA BRAF^V600E-negative patients (179 vs. 840 days, p = 0.04). Conclusions: These findings support cfDNA BRAF^V600E as a promising biomarker for monitoring treatment response and MRD detection in BRAF^V600E-mutant CRC, reinforcing its role in guiding personalized treatment strategies and postoperative surveillance.

1. Introduction

Advances in drug therapy and surgical techniques have extended the median overall survival (OS) of patients with advanced and recurrent colorectal cancer beyond 30 months. However, colorectal cancer with the BRAF^V600E mutant accounts for 5–10% of all colorectal cancers and has shown poor response to conventional chemotherapy. The median OS of patients with BRAF^V600E-mutant CRC is reported to be approximately 10 months, compared to about 30 months in BRAF^V600E wild-type cases [1,2], indicating a significantly worse prognosis. In particular, some cases exhibit rapid disease progression, and the transition to second-line treatment is often difficult. The combination therapy of a BRAF inhibitor and an anti-EGFR antibody (encorafenib + binimetinib + cetuximab or encorafenib + cetuximab, hereafter referred to as the BEACON regimen) has demonstrated efficacy, establishing a new therapeutic option for previously treated patients with unresectable advanced and recurrent BRAF^V600E-mutant CRC [3,4]. However, despite the introduction of the BEACON regimen, the poor prognosis of BRAF^V600E-mutant CRC remains unchanged, necessitating the development of more effective multidisciplinary treatments and therapeutic strategies.

Two major clinical challenges remain in the treatment of BRAF^V600E-mutant CRC.

1.1. Real-Time Evaluation of Treatment Response

The monitoring of treatment efficacy in BRAF^V600E-mutant CRC has traditionally relied on tumor markers (CEA, CA19-9) and imaging assessments. However, these indicators have limitations in sensitivity and specificity, making real-time disease monitoring difficult. In particular, evaluating resistance to the BEACON regimen and determining the eligibility for conversion surgery (curative resection after chemotherapy) require a more sensitive and objective biomarker that can accurately and promptly assess treatment efficacy and guide therapeutic decision-making.

1.2. Minimal Residual Disease (MRD) Evaluation After Curative Resection

Several studies have reported that the prognosis of BRAF^V600E-mutant CRC remains worse than that of BRAF^V600E wild-type CRC even after curative resection [5,6]. Furthermore, subgroup analyses of the MOSAIC trial have suggested that the additional benefit of oxaliplatin-based adjuvant therapy is limited in patients with BRAF^V600E mutations. Pathological evaluation alone has limitations in detecting MRD (minimal residual disease). Therefore, high-sensitivity molecular tools are required to complement pathological assessments. Given the high recurrence risk in BRAF^V600E-mutant CRC, optimizing adjuvant therapy strategies and follow-up protocols based on MRD evaluation is crucial.

Liquid biopsy is a non-invasive technique enabling repeated sampling and analysis of tumor-derived molecular information from body fluids such as blood [7,8,9,10,11]. We have previously reported the utility of cell-free DNA (cfDNA) in monitoring the treatment response of colorectal cancer [12]. cfDNA has a shorter half-life (several hours) than tumor markers (CEA, CA19-9), allowing for more real-time assessment of tumor dynamics [13]. Therefore, cfDNA BRAF^V600E is expected to serve as a promising biomarker for treatment monitoring and MRD assessment in BRAF^V600E-mutant CRC.

This study aims to evaluate the clinical utility of cfDNA BRAF^V600E for treatment monitoring in Stage IV patients and MRD assessment in patients after curative resection of BRAF^V600E-mutant CRC.

2. Materials and Methods

2.1. Study Population and Design

This study was a single-center prospective observational study of patients with BRAF^V600E-mutant colorectal cancer. Patients who underwent surgery or chemotherapy at Nippon Medical School Hospital between April 2017 and June 2024 were included. Patients were enrolled if BRAF^V600E mutation was detected in surgical specimens or endoscopic biopsy samples. Patients with high-frequency microsatellite instability (MSI-H) were excluded. The study population was classified into two cohorts according to the study objectives:

• Cohort 1 (Stage IV cases): Prediction of treatment response using cfDNA;
• Cohort 2 (Stage I–III curatively resected cases): Minimal residual disease (MRD) assessment using cfDNA

Patients with synchronous malignancies in other organs were excluded to ensure cfDNA analysis reflected the primary tumor and maintained data consistency. The primary tumor location was classified as right-sided (cecum to transverse colon) or left-sided (splenic flexure to the rectum).

2.2. Blood Sample Collection and cfDNA Extraction

For Stage IV patients, blood samples were collected simultaneously with routine hematological and biochemical tests for treatment evaluation and subsequently at approximately 4-week intervals, in line with tumor marker (CEA, CA19-9) assessments during follow-up. For curatively resected cases, blood was collected preoperatively and at one month postoperatively. Blood was collected using BD Vacutainer EDTA tubes (Becton Dickinson and Company, Franklin Lakes, NJ, USA) and processed within 3 h by centrifugation to obtain plasma. Plasma samples were stored at −80 °C until cfDNA extraction.

cfDNA was extracted from 1 mL of plasma using Maxwell^® RSC cfDNA Plasma Kits (Promega, Madison, WI, USA). The cfDNA concentration was measured using the Qubit quantification assay (Thermo Fisher Scientific, Waltham, MA, USA).

2.3. Droplet Digital PCR (ddPCR)

cfDNA samples were adjusted to 1000 ng/mL and analyzed using QX200 Droplet Digital PCR (Bio-Rad Laboratories, Hercules, CA, USA). The presence of the BRAF^V600E mutation was determined using variant allele frequency (VAF) with a detection threshold of ≥0.1%. In this study, cfDNA BRAF^V600E refers to the VAF of the BRAF^V600E mutation detected in plasma cfDNA, calculated as the proportion of mutant alleles relative to total BRAF^V600E alleles.

2.4. Statistical Analysis

Comparisons between groups were performed using Fisher’s exact test or the Mann–Whitney U test. Survival analysis was conducted using the Kaplan–Meier method, and group comparisons were performed using the log-rank test. A p-value < 0.05 was considered statistically significant. Statistical analyses were conducted using SPSS software version 27.0 (IBM Japan Ltd., Tokyo, Japan).

2.5. Ethical Considerations

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Nippon Medical School Hospital (Approval No. 28-01-694). Written informed consent was obtained from all patients.

3. Results

Among the 60 patients with BRAF^V600E-mutant CRC identified during the study period, 23 MSI-H cases were excluded, leaving 37 eligible patients. Of these, 14 were Stage IV (Cohort 1), and 23 were Stage I–III curatively resected cases (Cohort 2). The baseline characteristics of each cohort are shown in

Table 1. Patient characteristics of Cohort 1.

Cohort 1: Stage IV Colorectal Cancer with BRAFV600E Mutation
Age (y/o)	69	(59.0–71.5)
Sex [M/F]	10:4
Primary site [right/left]	9:5
Tumor size (mm) 1	47.5	(40.3–55)
Histology type [tub/por/muc] 2	5:5:4
Distal metastasis [liver/lung/LN/dissemination/bone] 3	6:2:10:1
Primary tumor resection [yes/no]	10:4
Surgical procedure (ileocecal resection 3:right hemicolectomy 4:partial resection 3)
First-line treatment regimen [mFOLFOX6 + Bv/FOLFOXIRI + Bv/mFOLFOX6]	5:6:3
Pre-treatment CEA (ng/mL)	6:2	(3.5–18.5)
Pre-treatment CA19-9 (U/mL)	87	(2.1–2249.4)
Pre-treatment cfDNA volume (ng/mL)	123	(101.5–145)
	median	(Interquartile range)

¹ Tumor size in resected cases only. ² Main histological component. ³ Multiple sites possible.

and

Table 2. Patient characteristics of Cohort 2.

Cohort 2: BRAF V600E Mutation-Positive Colorectal Cancer with Curative Resection
Age (y/o)	72	(64.5–82.5)
Sex [M/F]	11:12
Primary site [right/left]	18:5
Tumor size (mm)	39	(30–57)
pT [1b/2/3/4a]	2:2:14:5
pN [0/1a/1b/2a/2b/3]	11:4:3:3:1:1
Histology type [tub/por/muc] 1	12:5:6
Surgical procedure [ICR/RHC/partial resection/Sigmoidectomy/LAR]	10:8:4:1
Laparoscopic/open	18:5
Adjuvant chemotherapy [Yes/No]	9:14
regimen [CAPOX 5: FOLFOX 1: UFT/UZEL 3]
Pre-treatment CEA (ng/mL)	4.7	(2.1–8.5)
Pre-treatment CA19-9 (U/mL)	12.7	(7.8–27.7)
Pre-treatment cfDNA volume (ng/mL)	138	(82–151)
	median	(Interquartile range)

¹ Main histological component.

.

3.1. Cohort 1: Stage IV Cases (n = 14)

3.1.1. First-Line Treatment Response and cfDNA

Among the 14 Stage IV patients, 13 received chemotherapy, while 1 patient was not eligible for aggressive treatment due to poor performance status. The first-line chemotherapy regimens were as follows: FOLFOXIRI + Bevacizumab (n = 6), mFOLFOX6 + Bevacizumab (n = 5), and FOLFOX6 (n = 3).

Pre-treatment cfDNA BRAF^V600E was detected in 13/14 (93%) of Stage IV cases.

3.1.2. Overall Survival (OS) Based on cfDNA BRAF^V600E Status

After chemotherapy initiation, patients were divided into two groups based on cfDNA BRAF^V600E VAF reduction: VAF reduction group (n = 5) and non-reduction group (n = 9).

The median OS was significantly longer in the VAF reduction group compared to the non-reduction group (median OS: 511 days vs. 189 days, log-rank test, p = 0.03) (Figure 1).

3.1.3. Comparison of cfDNA and Tumor Markers in Clinical Practice

Case 1: Conversion Surgery After Initial Treatment

A patient with lower colon cancer and multiple peritoneal metastases showed a partial response (PR) to first-line chemotherapy, and underwent conversion surgery. At the time of partial response (PR) on imaging, the CEA level showed an unexpected rise, whereas cfDNA BRAF^V600E remained negative. The tumor was surgically removed, and the patient has remained disease-free for more than 10 months after surgery (Figure 2A).

Case 2: Timing of Regimen Change

A patient with multiple peritoneal dissemination showed little response to treatment initially and the disease progressed rapidly. While the CA19-9 level decreased over time with primary chemotherapy, there was almost no change in cfDNA BRAF^V600E.

On imaging evaluation, peritoneal dissemination progression and the appearance of lung metastasis were observed, and the patient was transferred to the BEACON regimen, but died of disease about 40 days later (Figure 2B).

3.2. Cohort 2: Stage II–III Curatively Resected Cases (n = 23)

Among the 23 patients who underwent curative resection, 9 received adjuvant chemotherapy (regimens: CAPOX:5, FOLFOX:1, UFT/UZEL:3). A total of seven patients (30.4%) developed recurrence, including one with Stage IIa, five with Stage IIIb, and one with Stage IIIc disease. The most common recurrence sites were peritoneal dissemination (n = 4), liver (n = 2), lung (n = 2), and lymph nodes (n = 2), with some cases showing multiple recurrence sites.

3.2.1. Postoperative cfDNA BRAF^V600E and Recurrence-Free Survival (RFS)

Preoperatively, cfDNA BRAF^V600E was positive in 1 of 23 cases (4.3%). At 1 month postoperatively, cfDNA BRAF^V600E was positive in four cases (17.4%). Among the seven recurrence cases, four had postoperative cfDNA BRAF^V600E positivity.

3.2.2. Comparison with Conventional Tumor Markers

Table 3. Prediction of recurrence using blood samples taken one month after curative resection.

	Relapse	No Relapse		Relapse	No Relapse
BRAFV600E positive in cfDNA	4		CEA ≥5.0 ng/mL	2	3
BRAFV600E negative in cfDNA	3	16	CEA <5.0 ng/mL	5	13
p = 0.004			p = 0.58

summarizes the recurrence prediction performance of cfDNA BRAF^V600E compared to CEA. cfDNA BRAF^V600E positivity at 1 month postoperatively was associated with a significantly higher recurrence rate (100% vs. 18.8%, p = 0.004). The sensitivity and specificity for recurrence prediction were 57.1% and 100% for cfDNA BRAF^V600E, compared to 28.6% and 81.3% for CEA, respectively. The positive predictive value (PPV) was 100% for cfDNA BRAF^V600E, whereas it was 40.0% for CEA. The negative predictive value (NPV) was 84.2% for cfDNA BRAF^V600E and 72.2% for CEA.

3.2.3. Postoperative cfDNA BRAF^V600E Status and DFS in Recurrent Seven Cases

Among the seven patients who developed recurrence, postoperative cfDNA BRAF^V600E status was evaluated for its association with DFS (Figure 3). The median DFS was significantly shorter in the cfDNA BRAF^V600E-positive group (n = 4; 179 days) compared to the cfDNA BRAF^V600E-negative group (n = 3; 840 days) (p = 0.04, log-rank test). All four cases with cfDNA BRAF^V600E positivity experienced recurrence within 6 months after surgery, whereas recurrence occurred beyond 12 months in the cfDNA BRAF^V600E-negative group.

4. Discussion

This study demonstrated that cfDNA BRAF^V600E may serve as a useful biomarker for treatment monitoring and MRD assessment in BRAF^V600E-mutant CRC. In unresectable cases, cfDNA BRAF^V600E was effective in the real-time evaluation of treatment response, while in curatively resected cases, it was useful for assessing recurrence risk. Previous studies have investigated cfDNA BRAF^V600E as a biomarker in colorectal cancer [14] and its role in predicting resistance to anti-EGFR therapy [15]. However, this study further builds on these findings by demonstrating the potential utility of cfDNA BRAF^V600E in guiding treatment decisions and postoperative surveillance, particularly in relation to MRD assessment and conversion surgery selection.

For unresectable BRAF^V600E-mutant CRC, the current standard treatment involves Triplet + Bv or Doublet + Bv as first-line therapy, followed by the BEACON regimen in the second-line setting. The ANCHOR CRC trial evaluated the BEACON regimen as a first-line therapy, reporting an objective response rate (ORR) of 47.8% (95% CI: 37.3–58.5%), thereby meeting its predefined efficacy criteria [16]. Building on this, the BREAKWATER trial, a phase III study comparing encorafenib + cetuximab ± chemotherapy versus chemotherapy alone in the first-line setting, demonstrated a higher ORR in the experimental arm (61%) compared to the control group (40%), further supporting the efficacy of targeted combination therapy [17]. Additionally, for resectable BRAF^V600E-mutant metastatic colorectal cancer, the NEXUS trial, a phase II multicenter study, is currently evaluating the efficacy and safety of perioperative encorafenib + binimetinib + cetuximab. The findings from this trial are expected to provide valuable insights into the role of targeted therapy in the perioperative setting, potentially refining treatment strategies for this aggressive disease subtype [18]. As the treatment landscape for BRAF^V600E-mutant CRC continues to evolve, integrating molecularly targeted approaches with personalized strategies will be crucial in improving patient outcomes and redefining the standard of care.

BRAF^V600E-mutant CRC is aggressive, and conventional tumor markers (CEA, CA19-9) or imaging often fail to accurately assess treatment efficacy [19]. Therefore, a more sensitive and reproducible biomarker is needed. cfDNA BRAF^V600E can be obtained noninvasively and provides real-time tumor dynamics, making it well suited for treatment response assessment and resistance monitoring. In particular, preoperative evaluation of resectability is crucial in cases where conversion surgery is considered after a favorable response to chemotherapy or in locally advanced disease. A decrease in the variant allele frequency (VAF) of cfDNA BRAF^V600E mutations has been associated with a better prognosis after resection, suggesting that cfDNA could be an objective indicator for conversion surgery. To further validate the clinical utility of liquid biopsy, the BEETS trial (JACRRO CC-18) has completed data collection, investigating the role of liquid biopsy in metastatic BRAF^V600E-mutant CRC treated with the BEACON regimen [20]. Moving forward, the findings from this study are expected to establish liquid biopsy as a valuable tool for treatment monitoring and personalized therapy in clinical practice.

With the recent inclusion of RAS and BRAF^V600E mutation testing under health insurance coverage, an increasing number of cases are being diagnosed as BRAF^V600E mutation-positive based on the analysis of resected specimens. In this study, 7 out of 23 patients who underwent curative resection experienced recurrence, and among those who were cfDNA BRAF^V600E-positive postoperatively, disease progression after recurrence was notably rapid. This suggests that postoperative minimal residual disease (MRD) assessment is particularly critical in BRAF^V600E-mutant CRC compared to BRAF^V600E wild-type cases. Moreover, there remains a need to reconsider follow-up strategies, adjuvant chemotherapy criteria, and administration regimens, as it is unclear whether the current approach for BRAF^V600E wild-type colorectal cancer is equally applicable to BRAF^V600E-mutant cases. At the same time, advances in genomic analysis have revealed that BRAF^V600E-mutant CRC is not a homogeneous entity. Notably, the BM1 subtype frequently aligns with the CMS4 classification, which is associated with poor prognosis [21,22]. Moreover, recent studies suggest that classifying patients into CD (controlled disease) and UD (uncontrolled disease) groups based on initial treatment response may provide additional prognostic value [23,24]. Moving forward, further exploration of the relationship between disease control patterns (CD vs. UD) and genetic subtypes could lead to refined treatment strategies and more precise prognosis predictions.

Limitation

This study was conducted as a single-institution observational study with a small sample size, and many cases were treated before BEACON therapy was approved. Although Figure 1 and Figure 2 show statistically significant survival differences, the small sample size warrants cautious interpretation of the log-rank test results, as statistical significance may be overestimated in limited cohort comparisons. To generalize these findings, verification in a larger, multi-institutional setting is necessary.

However, the clear visual differences in survival curves based on cfDNA BRAF^V600E status suggest that it has significant potential as a clinically relevant biomarker. Given the real-time advantages of cfDNA BRAF^V600E, its integration into clinical practice for treatment monitoring and MRD assessment warrants further investigation.

This study builds on previous research demonstrating the potential of cfDNA BRAF^V600E as a biomarker in colorectal cancer. By further validating its role in treatment monitoring, postoperative MRD assessment, and conversion surgery selection, these findings contribute valuable insights into the clinical aspects of BRAF^V600E-mutant CRC, which requires a multidisciplinary treatment approach. Future large-scale studies will be crucial to fully establish cfDNA BRAF^V600E as a standard component of precision medicine in colorectal cancer care.

5. Conclusions

This study suggests the potential utility of cfDNA in the treatment of BRAF^V600E-mutant CRC, particularly for treatment monitoring and MRD assessment. As precision medicine in colorectal cancer continues to advance, further developments in diagnostic testing and targeted therapies for BRAF^V600E-mutant CRC are anticipated. With accumulating evidence, improvements in treatment outcomes for BRAF^V600E-mutant CRC are expected, ultimately enhancing patient prognosis and therapeutic strategies.