Introduction

Colorectal cancer (CRC) with liver metastases presents a significant clinical challenge as a leading cause of cancer-related mortality worldwide [1]. Liver metastases occur in approximately 25%–50% of CRC cases, and recurrence rates remain alarmingly high at 50%–70% despite curative-intent surgery [2]. Accurate prognosis and stratification of recurrence risk are critical for improving patient outcomes. However, existing prognostic models, which rely on clinicopathological factors, such as tumor size, number of metastases, and RAS mutation status, are insufficient to capture the complex biology underlying recurrence. The role of adjuvant chemotherapy (ACT) in this population remains controversial. Although ACT has demonstrated potential benefits in certain subsets of patients, its efficacy is inconsistent, and ACT administration is usually not guided by the use of robust predictive biomarkers [3–5]. This underscores the urgent need for innovative tools to refine postoperative risk stratification and optimize ACT strategies to improve survival outcomes.

Circulating tumor DNA (ctDNA) is a non-invasive biomarker for minimal residual disease and relapse detection after curative treatments of solid tumors. An earlier pioneering study by Tie et al. demonstrated the clinical utility of ctDNA by showing its role in detecting minimal residual disease and predicting recurrence in stage II colon cancer [6]. In addition, the GALAXY trial, part of the CIRCULATE-Japan platform, has been pivotal in establishing the clinical utility of ctDNA as a prognostic biomarker in CRC [7,8]. By employing tumor-informed ctDNA assays, GALAXY showed that postoperative ctDNA positivity is strongly associated with recurrence risk and guides ACT administration. This trial provided compelling evidence that ctDNA can stratify patients by recurrence risk more effectively than conventional clinicopathological factors. Using data from the GALAXY study, Kataoka et al. demonstrated that ctDNA positivity correlates with inferior disease-free survival (DFS) in patients with resected colorectal liver metastases, further emphasizing its role in guiding ACT [9]. These studies employed tumor-informed approaches, which, although highly effective, require tumor tissue for assay customization. This requirement can delay clinical decision-making and restrict their applicability in some clinical settings.

To address the limitations of tumor-informed methods, we have initiated the CASSIOPEIA study, a prospective observational trial that employs a plasma-only ctDNA assay. Unlike tumor-informed assays, this approach eliminates the need for tumor tissue, streamlining the diagnostic process, reducing turnaround time, and maintaining high sensitivity and specificity. Utilizing the advanced Plasma-Safe-SeqS next-generation sequencing method, this assay detects mutations at allele frequencies as low as 0.04% for 10,000 genome equivalents (GEs) of DNA input, making it particularly advantageous in clinical scenarios requiring rapid decision-making or where tumor tissue is unavailable [10]. This level of sensitivity is comparable to that of tumor-informed assays while offering a more practical and clinically implementable solution. To date, the Plasma-Safe-SeqS platform in a fully tumor-naïve design has not been prospectively validated in published studies of CRC. However, the platform has been incorporated into two prospective clinical trials. One is the ongoing ASCOLT trial (NCT00565708), a multicenter randomized study evaluating adjuvant aspirin in patients with resected high-risk CRC, where ctDNA analysis using Plasma-Safe-SeqS has been presented in a conference abstract [11]. The other is a multicenter prospective study in patients with Stage II CRC, presented at the 2022 ASCO Annual Meeting [12], which demonstrated the feasibility of Plasma-Safe-SeqS for minimal residual disease detection after curative resection. However, full peer-reviewed results from this study are not yet available. To the best of our knowledge, our study is the first to use a peer-reviewed protocol to evaluate this platform in patients with CRC liver metastases and aims to generate foundational feasibility data on its clinical utility in a tumor-naïve application.

The CASSIOPEIA study is based on the hypothesis that plasma-only ctDNA assays can reliably detect recurrence earlier than standard surveillance in patients with CRC with liver-only metastases following curative hepatectomy, offering critical insights into postoperative management. Specifically, we propose that this approach enables early identification of high-risk patients and facilitates tailored surveillance and treatment strategies, overcoming the logistical and operational challenges associated with tumor-informed methods. This prospective, exploratory proof-of-concept protocol is designed to generate the foundational data necessary to plan a subsequent large clinical study.

The CASSIOPEIA study aims to evaluate the efficacy of plasma-only ctDNA assays for detecting recurrence early in patients with CRC and liver-only metastases post-curative resection. By eliminating the need for tumor tissue, this approach simplifies diagnostic workflows and minimizes turnaround time, while maintaining high sensitivity. Focusing on a homogeneous cohort of liver-only metastases ensures robust evaluation of ctDNA prognostic value. This streamlined and scalable method is expected to bridge the gap between advanced molecular diagnostics and routine clinical practice, addressing key challenges identified in previous studies.

Materials and methods

Study design

The CASSIOPEIA study is a single-center, prospective observational trial designed to evaluate the clinical utility of plasma-only ctDNA assays for detecting early recurrence in patients with CRC and liver-only metastases undergoing curative hepatectomy. A SPIRIT schedule and overview of the study design can be found in Figs 1 and 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. SPIRIT schedule of enrolment, interventions, and assessments for the CASSIOPEIA study.

https://doi.org/10.1371/journal.pone.0335591.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Study design and follow-up protocol for the CASSIOPEIA study.

This figure outlines the integration of ctDNA testing into clinical follow-up to assess recurrence dynamics and survival outcomes. The study enrolls patients with CRC liver metastases undergoing curative hepatectomy. Plasma samples for circulating tumor DNA (ctDNA) analysis are collected preoperatively and at predefined intervals postoperatively (4, 12, 24, 36, and 48 weeks). Tumor-related gene mutations are detected using the Plasma-Safe-SeqS gene panel. Routine computed tomography (CT) imaging and tumor marker (TM) evaluations are conducted in parallel with ctDNA testing to monitor recurrence. Abbreviations: CRC, Colorectal cancer; CT, Computed tomography; ctDNA, Circulating tumor DNA; TM, Tumor marker.

https://doi.org/10.1371/journal.pone.0335591.g002

Recruitment and study timeline

Patient recruitment began on December 1, 2024, and is expected to continue until December 1, 2025, with a total of 10 patients planned for enrollment. The follow-up period extends for 5 years postoperatively, resulting in a final follow-up date of December 1, 2030.

1. • Recruitment period: December 1, 2024–December 1, 2025
2. • Follow up and data collection period: December 1, 2024–December 1, 2030
3. • Final analysis and results expected: By December 2030

As of the submission of this manuscript, patient recruitment and data collection are currently ongoing, and no final analyses have been conducted.

Plasma samples are collected preoperatively and at predefined postoperative intervals (4, 12, 24, 36, and 48 weeks). The study plans to enroll 10 patients. Given the small sample size, no formal power calculation or hypothesis testing is planned. This study is conducted based on a prospective exploratory, proof-of-concept protocol to explore ctDNA dynamics, survival outcomes, and assay implementation. The results generate sufficient pilot data and inform the design of future larger-scale, statistically powered multicenter studies. Eligibility criteria include patients with histopathological confirmed CRC and liver-only metastases planned for curative hepatectomy. To confirm the absence of extrahepatic disease, all patients undergo contrast-enhanced CT scans of the chest, abdomen, and pelvis, along with abdominal ultrasonography, prior to enrollment. Additional imaging, such as liver MRI or PET-CT, may be performed as clinically indicated. Participants must meet specific conditions, such as an Eastern Cooperative Oncology Group performance status of 0 or 1, and provide informed consent. Exclusion criteria include active double cancers, pregnancy, or conditions deemed unsuitable by the attending physician. The lower age limit is set at 20 years in accordance with the study protocol. Patients with CRC younger than 20 years are extremely rare and often present with hereditary or biologically distinct features. To maintain the clinical homogeneity of the study cohort and ensure reproducibility and reliability of ctDNA dynamics, such cases are excluded. Detailed eligibility and exclusion criteria are outlined in Table 1. A homogeneous cohort of patients with liver-only metastases ensures a robust evaluation of ctDNA dynamics and recurrence prediction.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Eligibility and exclusion criteria for the CASSIOPEIA study.

https://doi.org/10.1371/journal.pone.0335591.t001

Discussion

The CASSIOPEIA study represents a significant advancement in the clinical application of ctDNA as a non-invasive biomarker for early recurrence detection in patients with resected CRC liver metastases. By exclusively utilizing plasma-derived ctDNA without requiring tumor tissue, this study simplifies the diagnostic process, addressing critical challenges associated with current ctDNA methodologies, such as the need for high-quality tumor DNA, extended turnaround times, and logistical issues. This approach enhances the applicability of ctDNA analysis for integration into routine clinical practice and expands its scalability. Moreover, this streamlined methodology could facilitate earlier detection of recurrence, enabling timely therapeutic interventions that potentially improve survival outcomes for patients.

To our knowledge, this study is the first to employ a 14-gene panel using Plasma-Safe-SeqS technology for ctDNA analysis in patients with CRC liver metastases following curative resection. The Plasma-Safe-SeqS platform provides a highly sensitive method for detecting tumor-derived genetic mutations. The platform can detect as few as four mutant molecules, with sensitivity varying by input DNA amount. At 10,000 GE of DNA input, the detection limit is 0.04%, whereas at 4,100 GE, as used in this study, the sensitivity is 0.1% [10]. This high sensitivity ensures precise monitoring of recurrence risk and offers a streamlined workflow that eliminates the dependency on tumor tissue.

Currently, ctDNA methodologies predominantly rely on tumor-informed assays, which require prior knowledge of tumor-specific mutations derived from tumor tissue [6,7,9]. Although tumor-informed approaches are highly sensitive, they introduce practical limitations. These include the necessity for high-quality tumor DNA, which may not always be available, especially in cases where patients have received treatment at multiple institutions or where tumor samples degrade over time. Additionally, the requirement for tumor DNA sequencing and personalized assay design extends the turnaround time, delaying actionable clinical decisions. These challenges restrict the application of tumor-informed assays in broader clinical settings.

In contrast, tumor-naïve approaches, such as the plasma-only methodology employed in this study, overcome these limitations by directly analyzing ctDNA from plasma. This approach accelerates diagnostic workflows, captures broader tumor heterogeneity, and eliminates logistical dependencies on tumor tissue [13]. These advantages make plasma-only methodologies a practical and scalable option for routine clinical use. However, a consensus on whether tumor-informed or tumor-naïve assays are superior remains elusive to date. Each method has distinct advantages and limitations, requiring further studies to determine their optimal use and how they can complement each other in clinical practice. Furthermore, broader operational considerations must be addressed to facilitate the real-world implementation of plasma-only ctDNA assays. These include evaluating cost-effectiveness, particularly in resource-limited settings, and ensuring accessibility across diverse healthcare systems. Regulatory approvals are also a critical step to enable widespread adoption, alongside efforts to standardize protocols for ctDNA testing. Additionally, integration into routine clinical workflows requires collaboration between healthcare providers, laboratories, and policymakers to establish guidelines for when and how ctDNA testing should be employed. Addressing these challenges is essential for translating the findings of this study into real-world applications and ensuring the scalability and sustainability of this approach.

This study focuses on homogenous patients with CRC liver-only metastases, a population at high risk for recurrence after curative resection. Liver metastases are the most common site of distant spread in CRC and significantly contribute to recurrence and mortality. Despite advancements in surgical and systemic therapies, recurrence rates remain alarmingly high, underscoring the need for innovative surveillance strategies that extend beyond traditional imaging and tumor markers [2]. Traditional imaging methods, such as CT scans, often lack the sensitivity to detect micrometastatic disease, whereas tumor markers such as carcinoembryonic antigen are nonspecific and may lead to delayed detection of recurrence. These limitations highlight the potential of ctDNA-based monitoring to provide earlier and more precise detection of recurrence, addressing a critical gap in the management of high-risk populations.

In addition to recurrence detection, ctDNA offers significant potential to guide the use of adjuvant chemotherapy in the postoperative setting. Identifying ctDNA-positive patients postoperatively enables the targeted administration of adjuvant therapy, ensuring high-risk patients benefit from treatment while sparing ctDNA-negative patients from unnecessary toxicity [9]. Currently, clinical trials investigating adjuvant chemotherapy in patients with CRC with resected liver metastases have yielded inconclusive results, with limited benefit observed in certain studies [3,4]. Consequently, routine administration of adjuvant chemotherapy for all patients cannot be uniformly recommended. By integrating ctDNA status into decision-making, this study aims to refine postoperative treatment strategies and enable personalized care.

One of the primary strengths of this study is its focus on dynamic ctDNA monitoring at predefined intervals. This design provides a detailed understanding of ctDNA kinetics and their temporal patterns, which could help identify critical windows for intervention. The inclusion of pre- and postoperative plasma samples facilitates the evaluation of concordance in genetic mutations, offering insights into tumor evolution, mechanisms of resistance, and the potential impact of surgical intervention on ctDNA dynamics. Additionally, by assessing the DFS and OS rates in association with landmark ctDNA status, the study explores the broader implications of ctDNA status on long-term clinical outcomes, providing a foundation for future intervention strategies and treatment optimization.

Despite its strengths, this study has some limitations. As a single-center observational trial, the findings may not be immediately generalizable to broader or more diverse clinical settings. Multicenter trials with diverse patient populations could help validate the broader applicability of plasma-only ctDNA methodologies and account for variations in clinical practices across institutions. Additionally, the relatively small sample size restricts the ability to perform detailed subgroup analyses or achieve statistically significant conclusions. Nevertheless, the primary aim of this study is to provide essential proof-of-concept data, which serve as a foundation for designing larger multicenter trials to validate ctDNA-based surveillance and treatment strategies. The Plasma-Safe-SeqS platform offers high sensitivity, detecting mutations with a mutant allele frequency as low as 0.04% for 10,000 GE of DNA input (cutoff is 4MM, input quantity range is 1,000 to 20,000 GE). However, its sensitivity is somewhat lower than that of tumor-informed methods such as Signatera, which has a limit of detection of ~0.01% and sensitivity of >95% [7–9,14]. In addition, its performance has not been directly benchmarked against other ctDNA detection technologies, such as Guardant Reveal or tumor-informed approaches. Furthermore, the reliance of the platform on a standardized 14-gene panel may limit its capability to detect rare mutations in highly heterogeneous tumor profiles. Future comparative studies between plasma-only and tumor-informed ctDNA methodologies are crucial to establish their respective advantages and optimize their clinical applications. Additionally, the integration of advanced approaches such as whole-exome sequencing or whole-genome sequencing could provide a more comprehensive understanding of tumor heterogeneity and uncover novel biomarkers for recurrence and treatment response [15]. These techniques may play a pivotal role in future research, enabling the refinement of ctDNA analysis and expanding its clinical utility.

In conclusion, the CASSIOPEIA study represents a significant step forward in validating a plasma-only ctDNA approach for postoperative surveillance and therapeutic decision-making in patients with CRC liver metastases. By addressing the challenges associated with tumor-informed assays, this study demonstrates the clinical utility of integrating ctDNA analysis into routine workflows. Beyond recurrence detection, ctDNA analysis holds promise for transforming postoperative management by identifying patients who could benefit from adjuvant chemotherapy. These findings are expected to advance personalized medicine and improve outcomes for patients with CRC liver metastases. Moreover, this study establishes a foundation for incorporating plasma-only ctDNA analysis into clinical guidelines, paving the way for its standardized use in postoperative surveillance.

Supporting information

Acknowledgments

The authors express their gratitude to all patients and staff involved in the study for their valuable contributions. The authors also acknowledge Editage (www.editage.com) for providing professional English language editing services. This study is conducted in accordance with ethical guidelines.